Pages

April 18, 2012

ODAC Released

odac closeupCAREFREE ENJOYMENT: I’ve been listening almost exclusively to the ODAC for several weeks now and I’ve been smiling a lot—not so much at the ODAC but at the music. When you know your audio gear is genuinely transparent it opens a worry-free window into the music. And, while I suffer from expectation bias just like everyone else, I’ve run a second blind listening test and can report the O2+ODAC held its own against the $1600 Benchmark DAC1. They both are audibly transparent.

I’M NOT SELLING ANYTHING: Some have claimed because others may profit from the ODAC I’m profiting from it. But that’s not the case. My involvement was simply to help design and measure it--just like the DIY open source O2 amplifier. YoYoDyne and various other vendors are responsible manufacturing and selling the ODAC. They’re the ones taking the financial risks, so they’re also the only ones to profit. This is an entirely not-for-profit blog and will remain so. You can’t buy anything from me.

LIKE O2 LIKE ODAC: The O2 amplifier was created as a simple, low cost, minimalist design delivering 100% transparent performance into nearly any headphone. The ODAC takes the same minimalist approach to transparency. But it’s not quite as simple as it looks in the photo above. It has around 60 components and many of those were carefully chosen through a lot of detailed measurements and trial and error.

NO SNAKE OIL REQUIRED: Many audiophiles want to believe more elaborate or exotic DACs offer higher fidelity. The ODAC demonstrates you do NOT need any of these for 100% transparent performance:

  • Asynchronous USB
  • UAC2 (USB Audio Class 2) Support
  • Asynchronous Sample Rate Conversion (ASRC),
  • Minimum Phase Filtering (no pre-ringing)
  • Non-oversampling NOS DAC chips
  • Dual DAC chips
  • Balanced Outputs
  • Vacuum Tube Stages
  • Elaborate and/or High Current Power Supplies

THE PROOF: I can confidently say none of the above are required for DAC audio nirvana—i.e. having your DAC disappear from the signal chain. Some of the above, like NOS designs and tubes, degrade fidelity. I’m confident because detailed measurements and blind listening tests verify the ODAC’s real world performance. I’ve even tried several different PCs and the ODAC’s performance is relatively consistent between them. So while some of the above might improve a few measurements, if they don’t improve the sound, it’s like taking 4 pills to get rid of your headache when 2 do the job nicely. Once the headache is no longer perceptible, more pain reliever doesn’t help anything. I encourage others to compare the ODAC blind against other DACs, at any price, that measure reasonably well.

odac proto march 2012ASSEMBLED BOARD: As explained in the ODA/ODAC and ODAC Update articles the ODAC is not DIY friendly. This is true for most 24 bit USB DACs. Neither the USB interface nor the DAC chips are available in small quantities. Both require signed OEM agreements. minimum purchase quantities, and the USB chip also requires custom programming. The ODAC has many extremely tiny 0603 surface mount components and uses a fine pitch 48 pin IC making it very tedious to build by hand. Given all that, the ODAC requires commercial automated assembled in fairly large batches. DIYers can add the ODAC board to the O2, build it into a stand-alone DAC, or add it to the upcoming Objective Desktop Amp (ODA).

NO S/PDIF: To hopefully avoid most of the “how come you didn’t add S/PDIF” or “when will you add S/PDIF” questions, please see the previous ODA/ODAC and ODAC Update articles. Sorry, but it’s not possible.

PARTNERS REVEALED: Because of the above requirements, the ODAC development has been a joint effort with YoYoDyne Consulting. George has supplied other open source USB DAC boards at reasonable prices and seems to be well respected on diyAudio. He’s taking considerable financial risk having a large batch of ODAC boards assembled and hence is coordinating all the manufacturing and distribution. I conducted all the measurements, refined the design, and helped optimize the PCB layout. Just to be clear, I’m not getting any money from the ODAC. Here are the ODAC resources so far:

AVAILABILITY: YoYoDyne is estimating boards will be available before the end of May assuming nothing goes wrong. It might take a few days longer to get boards to the UK and Europe. For other information please see the links above.

SCHEMATIC: YoYoDyne has received approval from Tenor to release the schematic once the production boards are verified and released to the various vendors above. YoYoDyne is also interested in a possible future version that’s more DIY friendly.

GUILT-FREE VOLUME ADJUSTMENT: The majority of USB DACs only support 16 bits over USB. That means when you turn down the volume in software you’re getting less than 16 bits of resolution. At background music levels you might only be listening to 11 or 12 bit audio. But the ODAC has a 24 bit USB interface and enough dynamic range to allow guilt-free use of software volume controls.

OTHER DETAILS: There’s a lot more in the previous ODAC articles, but in summary, the ODAC is designed to fit inside the standard O2 enclosure in place of the batteries. A few internal wires need to be soldered to connect the ODAC’s output to the O2’s input jack. Other line level sources can still be used with the O2. It will also fit inside the upcoming ODA. And it can be used standalone with either the on board 3.5mm output jack or panel mounted RCA output jacks. The USB connector is a standard USB-Mini-B as used on the FiiO products, the Sansa Clip, cameras, etc.

ODAC rev 1 and rev 2FOUR REVISIONS: The ODAC has been through four lengthy revisions—two of the earlier boards are shown to the right. Despite the fact we started with essentially the reference design from the datasheets, the devil was in the details. The  first version played music and sounded OK. Many companies and DIYers that “design by ear” would have stopped there. But that first version didn’t come close to delivering what the DAC chip is capable of. Each revision cycle took at least several weeks, cost hundreds of dollars, and involved countless hours of work. But, in the end, it resulted in much better performance compared to where we started..

REAL WORLD PERFORMANCE: Most DACs and PC sound interfaces priced under $200 fail to come close to the published specs for the chips they use. The FiiO E10’s Wolfson DAC chip is rated at 117 dB of dynamic range but the E10 only delivers a modest 98 dB. The power supply, PCB layout, grounding scheme, I2S waveform fidelity, clock quality, and more, often degrade the performance to well below the manufacturer’s carefully optimized reference design used for the datasheet specs. This is especially true for USB sourced signals and DACs running from USB power. But the ODAC, despite being USB powered, managed to come very close to the “Holy Grail” datasheet specs. See the Tech Section for more.

THE REAL NUMBERS: Here are the real numbers (versus the April Fool’s Day numbers I published two weeks ago) and they all meet the requirements for audible transparency. The letter following many of the results is the same letter grade (A-F) I’ve used in previous reviews with A being the best, and F being a “Fail”:

Measurement ODAC FiiO E10 DAC1 Pre
Freq. Response 10 hz – 19 Khz 24/44 +/- 0.1 dB A +/- 0.1 dB A +/- 0.1 dB
THD+N 100 hz 0 dBFS 0.0029% A 0.005% A 0.0009%
THD+N 20 hz –1 dBFS 0.003% A 0.004% A 0.0009%
THD+N 10 Khz –1 dBFS 0.003% A 0.004% A 0.0007%
IMD CCIF 19/20 Khz –3 dBFS 0.0011% A 0.013% B 0.0005%
IMD SMPTE –1 dBFS 0.0004% A 0.004% A 0.0004%
Noise A-Weighted dBu 24/44 --102.8 dBu A -98.3 dBu C -105.4 dBu
Dynamic Range –60 dBFS A-Wtd --111.1 dBr A -97.6 dBr C -110.9 dBr
Linearity Error -90 dBFS 24/44 0.0 dB A 0.0 dB A 0.2 dB
Crosstalk 0 dBFS Line Out 100K --93.5 dB A N/A -106 dB
USB Jitter 11025 hz J-test 24/44 Excellent Very Good Excellent
Maximum Output Line Out 100K 2.0 Vrms 1.65 Vrms 2.5 Vrms (1)

NOTE 1: Maximum output of DAC1 is configured with internal jumpers

BOTTOM LINE: The ODAC has been released to production and will hopefully be available by the end of May. I’ll soon be publishing more detailed measurements, results of listening tests, etc. I’m confident the ODAC is audibly transparent. And, especially when it’s installed inside the O2 or future ODA, it offers a level of performance that’s difficult to find without spending substantially more. It also offers detailed measurements and blind listening tests to back up its performance which is something very few other DACs offer at any price. And, paraphrasing from credit card commercials, worry free enjoyment of music can be priceless.

 


TECH SECTION


odac drawing p6 coordinates

SPECIFICATIONS: Here are some ODAC basics:

  • Audio Formats: 16/44, 16/48, 16/96, 24/44, 24/48, 24/96
  • Interface: USB Audio Class 1
  • Native Driver OS Support: Windows XP & Later, OS X x86, Linux
  • Operating Systems Tested: XP, Vista, Win7 (32 & 64), OS X Snow Leopard, Ubuntu 9.1 32 bit
  • Line Output: Approximately 2 Vrms into 5K ohms or higher
  • Dynamic Range: > 110 dB A-Weighted
  • Distortion: < 0.005%
  • Dimensions: 49 x 58mm (see drawing to right)

BLIND EVIDENCE: So far I’ve run two relatively informal blind tests with the ODAC. The latest one used special software on the PC to play the same track on both my Benchmark DAC1 Pre and simultaneously on the ODAC plugged into the same PC (both connected via USB and running at 24/44). The ODAC was connected to an O2 headphone amp, and a switchbox allowed the headphones to be rapidly switched between the DAC1 and the O2+ODAC. The two sources were carefully level matched (using their respective volume controls) using a test signal and wideband DMM. I tried both my Sennheiser HD650 and Denon AH-D2000 headphones with a variety of well recorded favorite tracks. One other listener and I could not reliably tell which was playing.

FUTURE BLIND TESTS: I’ll hopefully be running a more comprehensive and rigorous blind test in the future. But, ultimately, it’s best to have listeners who expect to hear a difference, and someone else who understands the technical issues to supervise the test, perform level matching, sync the sources, etc. It would be ideal, for example, to have Mike and/or Lieven from Headfonia be a listener and George from YoYoDyne oversee the test. But sometimes geography and other factors restrict what’s realistic. If anyone is willing to help coordinate such a blind test, please contact me privately with the link in the right hand column.

USB INTERFACE: There are only a few USB interface chips capable of 24 bit operation. The best option we could find is the Tenor TE7022. It’s used in the Violectric USB 24/96 and several other commercial 24 bit USB DACs. Notably, it does not require proprietary drivers to work with any popular operating system including Windows because, like the Benchmark DAC1, it’s a USB Audio Class 1 interface. It also has respectably low jitter. The XMOS solution requires an expensive license and proprietary windows drivers and offers no audible benefit. The TAS1020B is being discontinued, requires extensive firmware, and also offers no audible benefit. A custom microcontroller creates even more hurdles with no audible benefit.

THE DAC CHIP: As I’ve explained elsewhere you can get transparent performance from most of the better DAC chips on the market from a half dozen chip vendors. What’s most important is choosing one that’s best suited to the particular application. In this case, that means running from a single 5 volt USB power supply, having a buffered voltage output (to avoid needing a single-supply op amp), and operating properly without using a microprocessor. Just those three requirements narrow the choices considerably. We chose the ESS Sabre ES9023 which is used in a lot of commercial designs (it’s an improved version of the popular ES9022). Unlike TI, Burr Brown, Analog Devices, etc, ESS specializes in audio chips and they did a nice job with the ES9023’s feature set and specs.

DIRTY LIES: Many popular “boutique” DACs (especially those being sold on eBay) mislead their fans by quoting only the chip specs for their entire DAC. Basically that amounts to cheating and lying. It further implies either the company is incapable of making the proper measurements or the real measurements were bad enough they didn’t want to share them. The implication is a DACs performance is solely determined by the chip used. But the opposite is usually more true. The implementation matters far more than the chip. the FiiO E10 is just one of many examples. The HA-Info I’ll soon be comparing to the ODAC is another.

CHIP ENVY: Any serious audio engineer designing DACs and making proper measurements knows the implementation matters more than the chip used. The first version of the ODAC, which closely followed the ESS datasheet, only managed about 98 dB of dynamic range. The distortion was also much higher than listed on the datasheet and the jitter was somewhat disappointing. Unless you exactly duplicate the chip manufacturer’s reference design, right down to the PC board layout (which I’ve yet to see any manufacturer do), you really don’t know what you’ll get. To measure all the important parameters you need more than just RMAA. You need a real audio analyzer with performance substantially better than the DAC being measured.

DESIGN BY EAR: Detailed and credible published measurements are missing with most “boutique” DACs being sold including those from Schiit Audio, Audio-GD, AMB, Twisted Pear, Burson, and NuForce. Where’s the credible evidence they got it right? A lot of these companies try to claim specs don’t matter, and they instead design by ear, but that method is seriously flawed (see: What We Hear). It would be like designing a car engine without a dynamometer and having no idea how much horsepower and torque it produced, how fuel efficient it was, etc. Given all the proven problems with sighted listening, and how our ears and brains work, those who claim to design by ear are very likely getting it wrong. Put another way, they’re often designing products with far lower fidelity than they’re otherwise capable of.

odac bench smt partsLESSONS LEARNED: If this project has taught me anything, it’s that getting much better than 16 bit (96 dB) performance can be challenging. The first version of the ODAC, despite following the reference design, only had about 98 dB DNR. That’s about the same as the FiiO E10. The photo to the right shows a few dozen assorted surface mount parts that were laboriously swapped out one at a time and measurements repeated dozens of times using the dScope. Some improvements were far from intuitive. Audiophile preferred polyphenylene capacitors performed worse than less expensive types. Additional filtering on the digital power supply dramatically increased jitter. Chasing down the last few dB of dynamic range the chip is capable of proved to be especially challenging. When it was said and done, the DNR went from 98 dB to over 111 dB. That’s a huge difference and something the design-by-ear crowd would have never achieved.

THE AUDIOPHILE WAY: All too many small or “boutique” audiophile manufactures and DIYers seem to just slap trendy chips on a board, listen to their creation expecting it to sound good (so that’s what they hear), and call it good. Many don’t even follow the reference design. Instead they include a bunch of “audiophile upgrades” expecting better performance—and they hear what they expect to hear even when it’s not true. But the ODAC demonstrated those upgrades often make things worse. So instead of getting even 98 dB DNR like the first ODAC revision, those following audiophile myths and designing-by-ear probably would have ended up with something even worse. Unless you’re making the right measurements, you really have no idea what you’re getting.

THE POWER SUPPLY: For reasons explained in the earlier ODAC articles, the ODAC is USB powered. This allows it to work standalone, as an internal add-on to the O2, and in the upcoming ODA. There are many obvious advantages to USB power but it often degrades performance due to noise. To get around this, the ODAC uses split digital and analog power supplies each with their own filtering and regulator. The analog supply has additional filtering and the critical reference voltages, and negative supply for the DAC chip, are further optimized. I literally tested more than 100 variations of components, including different brands of capacitors, to get the most out of the ES9023. This level of refinement would be impossible without a serious audio analyzer.

PUMP YOU UP: The ESS chip has the huge advantage of a built-in low noise charge pump. It generates its own regulated negative power supply allowing a Redbook standard 2 Vrms output from a single 5 volt USB power supply and a direct coupled output. This is an important distinction compared to a lot of USB powered DACs. Without the charge pump, or some other negative power supply, USB DACs can’t produce the Redbook standard 2 Vrms which reduces their effective dynamic range, lowers their ENOB, and creates level matching problems. It also requires an undesirable output coupling capacitor and usually results in loud transients on power up. The AMB Gamma, and most USB powered DACs I’ve tested, don’t meet the Redbook standard. The ODAC does. And it produces only a soft click on power up.

ADAPTIVE USB INTERFACE & LOCAL CLOCK: I’ve talked about this before, but just to be clear, the ODAC is NOT clocked by the USB port. So the quality of the audio clock, and any resulting jitter, is largely independent of the PC’s USB timing. It has its own low phase noise 12 Mhz crystal controlled oscillator that’s used to generate the MCLK and SCLK audio clocks.

A NOTE FOR 24/88 FANS: Some have asked about 24/88 high resolution audio support (popular for SACD rips). While the ODAC doesn’t support 24/88, it does support the audibly identical 24/44. It’s trivial to re-sample 24/88 audio to 24/44 with no artifacts as it’s a simple divide-by-two operation (and one the operating system will perform for you automatically). I know many audiophiles probably think they’re losing something, but nobody has proven they are. Meyer & Moran demonstrated in a very in-depth study that even 16/44 audio sounded identical to SACD. Another good read is 24/192 Music Downloads. And if you refuse to believe all that, try resampling some 24/88 audio to 24/44 and compare them yourself with Foobar and the ABX add-on. It’s been done at HydrogenAudio and elsewhere always with the same result: Unless you mess up the resampling somehow, or change the levels, you can’t tell them apart.

TRANSPARENCY GUIDELINES: The What We Hear article offers information and references outlining guidelines as to what’s required for a piece of audio gear to genuinely disappear from the signal path and not alter the sound in any audible way. Here are what I believe to be relatively conservative criteria for audible transparency and the ODAC passes all of them:

  • Frequency Response 20hz – 19 Khz within +/- 0.1 dB (Most DACs, due to the Nyquist limit of 22 Khz, start to roll off past 19 Khz when operating at 44 Khz sampling rate—the ODAC is down about 0.4 dB at 20 Khz). The widely accepted, but less conservative standard is +/- 0.5 dB (1 dB total variation) from 20 hz to 20 Khz.
  • All Harmonic, IMD, Alias, Modulation, & Crosstalk Components Below –90 dBFS and total sum below –80 dBFS (0.01%)
  • All Noise Components below –110 dB and total sum below –100 dBFS
  • All Jitter Components below –110 dB and total sum below -100 dBFS

GREEN GUIDE LINES: A few months ago I introduced green guide lines on several of my measurement graphs to help show the worst case ideal performance. Some of these are slightly more lenient than the above criteria or take into account more detailed thresholds (i.e. that power line hum can be slightly higher in level than midrange noise). For now I’m keeping the green guide lines consistent with earlier reviews. But please note the ODAC meets even the tougher criteria above.

PARTIAL MEASUREMENTS: I’ve made LOTS of ODAC measurements including some things I’ve never measured before—such as true latency. For this article I’ve only shown some of the more common measurements. In a future article I’ll cover additional measurements, 16 bit operation, 24/96, as well as several comparisons to the DAC1, FiiO E10, and an HA-Info eBay headphone DAC with a well respected DAC chip. So, in the interest of getting this article done sooner rather than later, and keeping it to a manageable size, only a sampling of 24/44 measurements are shown below.

DYNAMIC RANGE: A DAC’s noise floor impacts Dynamic Range (DNR), audible noise, THD+N, and can even exceed jitter-induced distortion. If you have to pick a single number to evaluate real world DAC performance –60 dBFS dynamic range (DNR) is one of the most revealing. The guys in the white lab coats have determined DNR greater than 100 dB results in transparency under realistic conditions. And, if you want to adjust the volume in software, it’s best to have at least 110 dB DNR to keep the noise floor inaudible even if the downstream gain is left cranked way up. Anything beyond 110 dB is past the point of diminishing returns—it looks nice on paper but doesn’t help the sound quality. The ODAC is a very substantial 14 dB better than the FiiO E10.  Here’s both channels of the ODAC referenced to the 2.03 Vrms at 0 dBFS. Note the channels are very symmetrical indicating a careful PCB layout:

ODAC -60 dBFS Dynamic Range & Noise Both Ch Line Out 100K Ref 2 Vrms 24-44

ODAC VS ESS: As explained earlier, the DNR quoted on datasheets is often something of a Holy Grail. The chip specs are typically from a very high quality AES/EBU or I2S laboratory quality signal (as output by high-end audio analyzers like a Prism dScope or Audio Precision). And they’re typically running from expensive ultra low noise bench power supplies costing thousands of dollars. It’s safe to assume the datasheet numbers were not made with USB data while running on USB power. ESS rates the ES9023 DNR at –112 dB A Weighted. The ODAC delivers –111.1 dB A-Weighted under the same conditions. In other words, even using USB data and power, the ODAC comes within a fraction of a dB of achieving the datasheet spec! I’m fairly proud of this aspect of the ODAC. It wasn’t easy.

CCIF IMD: This 19+20 Khz twin tone is a difficult test for many DACs running at 44 Khz. Old style (NOS) non-oversampling DACs especially struggle due to aliasing problems. In addition the output buffer (or I-V stage) in many DACs contributes high frequency distortion because the RC filter can be a challenging reactive load at these frequencies. If you look back through my reviews, you’ll find lots of products struggle on this test. Even the E10 turned in a marginal result. The ODAC, however, due to careful optimization of the output filter, and the superior digital filtering of the ESS DAC, does very well here with everything in the audio band well below 100 dB (both channels shown). Note also the 19 and 20 Khz tones are visibly equal in level which is not the case for many DACs:

ODAC -3 dBFS CCIF IMD Both Ch 100K Ref 2 Vrms 24-44

 

SMPTE IMD: This twin tone test is more revealing of low frequency problems including power supply interaction. Again, the ODAC does very well with everything well below 100 dB (both channels shown):

ODAC -2 dBFS SMPTE IMD Both Ch 100K ref 2 Vrms 24-44

 

100hz THD+N @ 0 dBFS: This test checks for clipping of the DAC at 0 dBFS and also shows the maximum output and channel balance error. You can see the ODAC produces 2.03 Vrms which is within an in significant 0.03 volts of the Redbook standard for digital audio. And even at 0 dBFS, the distortion is still 3 times less than what’s required for 100% transparency. The channels are perfectly balanced to within 0.001 dB. This spectrum is shown all the way out to 96 Khz and you can see the ESS DAC is well behaved even above 20 Khz with all noise still below about –110 dBFS. This is excellent performance:

ODAC 100hz 0 dBFS Max Out Both Ch 100K Wideband & Channel Balance Ref 2.03 Vrms 24-44

 

THD+N VS FREQUENCY: Here’s the distortion performance at –1 dBFS from 20 hz to 20 Khz into a more challenging 10K load with a measurement bandwidth of 22 Khz. At 1 Khz the distortion is only 0.0027% and it remains around 0.003% over most of the audio band with only a slight rise up to 0.0048% at 9 Khz before the harmonics fall above the audible range. This is excellent performance and both channels are very closely matched (yellow vs blue):

ODAC -1 dBFS THD N vs Freq Both Ch 10K Load BW 22 Khz 24-44

 

NEW JITTER FINDINGS: I did quite a bit more research on jitter during the ODAC’s development. I’m also using a new dScope method that shows the same spectrum as before but now the symmetrical jitter components are marked (with a white “X”) and summed to obtain a total numerical value (previously the dScope was just showing the total residual noise floor). Having the single number (-103.3 dB below) made it easier to optimize the ODAC for the lowest jitter. The objective evidence conservatively indicates  if you keep all related components below -110 dB, and the total below -100 dB, the jitter will be entirely inaudible. Jitter creates dissonant distortion products in the audible band. It’s reasonable to assume if the audible effects of jitter are kept at or below the inaudible noise floor, they too will be inaudible. So the same levels of –100 dB and –110 dB that apply to noise also apply to jitter contributions. This is also consistent with various professional reviewers and their anecdotal opinions on jitter performance as well as my blind testing against the Benchmark DAC1 which has even lower jitter.

ODAC JITTER: The ODAC passes the conservative criteria with several dB to spare even on the worst-case J-Test signal. And the spread at the base of the signal (very low frequency jitter) is extremely minimal being entirely below –130 dB. It’s also worth noting the jitter here looks subjectively worse because the noise floor is much lower than most of my jitter measurements which are done with a 16 bit test signal. A 16 bit noise floors masks most of the “spikes” seen below. The ODAC also has negligible inter-channel phase error and essentially perfect pitch accuracy:

ODAC -3 dBFS 11025hz J-Test Sum of Marked Jitter Components Ref 2 Vrms 24-44 ref

 

CONCLUSION: Hopefully the above provides some good evidence the ODAC delivers transparent performance. The next ODAC article will compare the 16 and 24 bit performance and I’ll be comparing it to several other DACs. I’ll be publishing many more measurements such as modulation noise, channel separation, square wave/impulse response, latency, frequency response, absolute noise, and more.

316 comments:

  1. I do believe you've done it again! And with 2V RMS from USB power too! And it's going for £60 in the UK! And it doesn't need any drivers! The only conceivable blemish to my mind is the lack of 24/88.2 operation, but as you said you're limited by what chips are available at a given pricepoint.

    That, and the link to the YoYoDyne announcment appears to be broken (address that appears when you hover over is right, address when you click is not)

    I'm really nitpicking now...the stuff you deliver is delivered at a virtually unprecedented price/performance ratio, and considering the performance is transparent...

    ReplyDelete
    Replies
    1. Thanks. I fixed the link. Everyone might be wise to refresh the article before they read the whole thing as I often find lots of things to fix in during the first few hours after publishing.

      You're correct about 24/88 and it being a Tenor limitation. It's been well proven 24/88 doesn't sound any better than 24/44. So an easy solution is to simple convert those rare 24/88 tracks to 24/44 which should not produce any nasty artifacts as it's a trivial divide-by-2 operation. Anyone can verify such a conversion is inaudible with Foobar ABX and a DAC that supports both formats.

      Delete
  2. Thanks for your contributions. It is fairly significant accomplishment to develop a DAC with a ESS Sabre ES9023 while keeping it under $100.

    ReplyDelete
  3. Hello,

    Could you please explain the discrepancy between the numbers published in the current article and your previous "Early April" update?
    http://nwavguy.blogspot.com/2012/04/early-april-update.html

    I thought you had managed to get even better numbers than the DAC1.

    Keep up the good work! ;)

    ReplyDelete
    Replies
    1. You must be from a country where April Fools Day (April 1st) is never celebrated?

      Delete
  4. Congratulation NwAvGuy! This is truly a milestone for the audio community, especially for the DIY crowd.

    What do you think about combining ODAC and O2 into one unit to avoid using interconnects, and also having a simplified/shorter output stage between the DAC section and the amp section? Would that improve performance if done right (in theory)?

    ReplyDelete
    Replies
    1. Please see the linked ODAC Update article. It shows a picture of the ODAC inside an O2 as a single unit. What you describe was a major design goal of the ODAC.

      Delete
  5. I swore you were going to use the CS4351 because it has similar specs in terms of voltage output and such, but I see now that set the bar a little low in terms of cost and overall performance.

    Anyway, I'm planning to hook this up to some H/K Soundsticks... and to have one in an ODA.

    ReplyDelete
  6. Looks like a great bang for your buck DAC as one would expect. I sorta wish the USB and 3.5mm jacks were on opposite sides of the board for using it as a standalone DAC but I suppose if you put it into the O2 you don't have this issue.

    ReplyDelete
  7. Wow. Great stuff.

    I'm amazed that it's so small!

    I look forward to comparing it to the MSII. One thing that will be interesting to hear is if the dead flat ODAC sounds darker than the MSII since I have always thought the MSII sounded bright, although the specs of the MSII say otherwise (so chances are they'll just sound the same I suppose).

    Faulty ears or faulty specs? Hmmmm.

    ReplyDelete
  8. The ODAC uses a properly implemented ESS DAC :o

    No wonder us regular DIY folk couldn't access this kind of stuff...

    That said, the (properly measured) high end Oppo Blu-Ray/DVD players use properly implemented ESS DACs as well, so it would be great if someone could compare ODAC, a high end Oppo and the Benchmark DAC Pre....

    ReplyDelete
  9. How do you connect to speakers? 3.5mm to RCA adapter? Do you know if there is any plans for a version with RCA connectors?

    ReplyDelete
    Replies
    1. you would want to route the line signal to a preamp and amp, not directly into the speakers, there are 3.5mm to RCA adapters

      Delete
    2. Because of the excellent dynamic range and 24 bit support you can run the ODAC straight into powered speakers or any other amp and control the volume from the PC. You don't need a preamp. If you try that with a typical 16 bit DAC you're likely to hear at least some noise from the speakers (unless they have a volume control and you keep it set relatively low).

      Delete
  10. Do you plan on releasing test results of the O2+ODAC combo (that is, when they are put together?)

    ReplyDelete
    Replies
    1. Yes, the next article should have some tests of the combination. My most recent blind test also used the O2+ODAC. It should be a lot like the DAC1 where the headphone output performs a lot like the line output but with more current and/or voltage as needed.

      Delete
    2. does the o2+odac have the same "issue" as the dac1 where you can measure the noise from the switching digital circuitry in the headphone out of the o2? how would this affect the sound?

      Delete
    3. The O2+ODAC noise floor is, visually, a bit quieter than the DAC1. I suspect that's because there's a lot less digital circuitry in the ODAC. The DAC1 has an FPGA, ASRC, and the S/PDIF hardware. The O2+ODAC also has the advantage of being on a different board and run from a completely different power source. But in terms of total noise, they're both comparable and both sufficiently transparent. So it's a non-issue for both of them.

      Delete
  11. If I use ODAC+O2 for music production on the go (mobile desktop musician with MIDI controller), how is the latency measurement in DAW setup (i.e.-steinberg cubase) when additional driver like 'asio4all' apply?

    ReplyDelete
    Replies
    1. The total hardware latency of the ODAC is approximately 1.6 mS (it should be under 2 mS worst case). But that's just the hardware. You have to add whatever latency the driver (ASIO or native UAC1) requires for buffering, etc. I'll have more information in the next ODAC article. I haven't yet tried it with DAW software.

      Delete
  12. The ESS chip has the huge advantage of a built-in low noise charge pump.

    I was curious as to how you'd solve this, and now i have to read you essentially cheated your way arround it, by letting the IC do the "dirty work" xD

    Seriously though, great job! And thank you very much for doing it.

    I am a bit puzzled as to why you left something as important as the frequency response off of the article. My guess is there are some non-quality-degrading oddities that require more in-depth explanation? Either way, I'm excited to read your further articles on the ODAC.

    Another Issue you didn't adress here, that I'm curious about, is the crosstalk-performance over the 3,5mm jack. That should be notably worse, somewhere around -60dB or so, right? I think it might be quite important to mention this to the buyers thinking about getting the ODAC in the separate enclosure JDS/Epiphany are offering.

    My Other question is about the full O2+ODAC package. Is that still capable of an amp-only mode so as to plug in another analog source into it? Or will soddering the ODAC board in ultimately mean "sacrificing" the pure O2 functionality? I guess a DAC-only mode is out of the question since neither JDS nor Epiphany will be offering the O2 with RCA jacks for line out. But then again this might be a topic worth adressing for the ODA+ODAC project rather than the O2 itself.

    ReplyDelete
    Replies
    1. The frequency response is given in the table. It's flat +/- 0.1 dB from 10 to 19 Khz and down 0.4 dB at 20 Khz due to the normal Nyquist Limit at 44 Khz sampling. I didn't show it because it's boring and very similar to the E10, Benchmark DAC1, etc.

      Crosstalk from a 3.5mm jack is really only an issue with the relatively low impedance of headphones. Feeding a 10,000 ohm or higher line input, the resistance of the shared ground is negligible. That said, my crosstalk measurement was taken from the pin header with RCA output jacks. It might degrade very slightly from the 3.5mm jack but still should be better than 90 dB.

      Yes, the O2 (as was explained I think in the ODAC Update article) retains its line input jack. When something is plugged into the jack the ODAC is disconnected. Remove the plug, and you're listening to the ODAC. The only think you sacrifice are the batteries.

      The use of the ES9023 isn't cheating at all. It's minimalism. I don't believe in adding parts when a piece of gear doesn't need them to be entirely transparent. I talk about that more in the O2 design sections. A lot of extra money and complexity goes into gear either for marketing reasons or for mythological reasons. It's like sacrificing a goat to the magic spirits and just as unnecessary.

      Delete
    2. Hi NwAvGy,

      I decided to put my question here because it concerns the O2, ODAC and ODA, and it's also related to the last part of Grunty's comment which you didn't answer.

      1) Grunty asked if it's possible for a ODAC-only mode in the ODAC+O2 package and i'm guessing no because a) There is no output I can see that will serve this purpose and b) likely the analog out of the ODAC will be fed directly to the O2 amp.

      2) Can you give a more in-depth explanation of how to solder the ODAC onto the O2 amp? JDS Labs covered it only briefly and I was wondering how would you suggest I hold the ODAC board in place so that it won't move about when I move the O2 amp and so that the USB port will line up with the back panel opening.

      3) The O2 amp is able to accurately drive 98% of all current headphones. That is respectable considering that the O2 amp is battery-powered. Is it possible to modify the ODA such that it is able to accurately drive 100% of all headphones? (Since the ODA will be run off AC)

      4) Is it possible to have a ODAC-only mode in a ODAC+ODA package?

      Thanks!

      Delete
  13. you mention a "relative consistency" when using different PCs, can the PC used cause performance variations? is there any way we could optimize the pc for the ODAC?

    thank you

    ReplyDelete
    Replies
    1. I plan to cover this more in the next article. There is some measurable improvement from using a USB cable with a ferrite "choke" at the ODAC end of the cable. If you don't have a cable that has one, you can buy such a cable for a few bucks or get clamp-on ferrites in the appropriate size from electronics suppliers for a few dollars.

      With any USB audio device it never hurts to use a USB port with either nothing else sharing that USB controller (cluster of ports) or only something benign like a mouse. High bandwidth devices like USB WiFi adapters, Bluetooth adapters, video hardware, etc. create more bus contention. While I haven't heard or seen any problems with the ODAC on shared controllers (in fact I have routinely shared a controller with the dScope which generates lots of traffic) it's still not a bad idea with any USB audio device.

      If your PC has front and rear panel USB connections, try to split any high bandwidth hardware between the two sets. Likewise some laptops have two groups of USB connections. As for other variations, the ODAC's extra power filtering helps make it more immune to noise issues which can vary from PC to PC.

      Delete
    2. It would be great if you could include some testing with one of those 25$ usb isolators with linear ps found on ebay..they cost a quarter of your dac, and I wonder is the benefit is worth it specially when using the ODA+c as preamp on power amps

      On ebay you can find the isolators by searching for "ADuM4160"

      Thank you for the time and energy you are putting on this projects, I hope you are having as fun as I am waching the reactions to your blog and designs

      Delete
  14. Congratulations on finishing this, cannot believe your not getting paid for this, which in turn is helping to keep costs down for us. I hope karma comes back around on you big time. Few questions for you:

    Do you know if the ODAC will be built inside the casing of the future ODA or will it dock in, similar to a e9+e7?

    Any idea on when the ODA might be completed? I kinda want to hold out buying the ODAC if I can resist.

    Any recommendations on a power amp to use with the ODA to drive speakers that are good bang for buck? Behringer a500? (saw it mentioned on a previous post)

    ReplyDelete
    Replies
    1. See the previous ODA/ODAC articles. The ODAC was created mainly as a "daughter board" for the ODA. Both boards fit in the same enclosure.

      Yeah, the A500 is a lot of bang for the buck. The new Emotiva A-100 for $220 doesn't have as much power but is much more desktop/space friendly. If you're on a tight budget there are even some reasonably priced TriPath and other "Class D" desktop amps that are semi-respectable. Or perhaps a used Emotive UPA-2 (they sadly stopped making it).

      Delete
  15. Very impressive! I can't wait to get my hands on the upcoming ODA.

    [URL="http://www.kaptainklak.com/gfx/gallery/133191134044.gif"]http://www.kaptainklak.com/gfx/gallery/133191134044.gif[/URL]

    ReplyDelete
  16. Great job!
    I have some probably obvious/dumb questions to ask!
    1) To retro fit the ODAC into the O2, will require removal of batteries, and a new backplate for the USB port.
    Q: Does JDS provide the new backplate?
    Q: How would one connect the ODAC to the O2? (internal connections? or an external interconnect?
    Q: Does this mean a required USB connection to power the ODAC plus the AC adapter to power the O2?

    ReplyDelete
    Replies
    1. @'Nick^ We'll offer the new O2+ODAC rear plate previewed in our announcement as soon as it's in production. Check the link to the JDS announcement at the top of this article for pictures.

      Yes, a retrofit requires removal of the O2's 9V batteries. Yes, you will need an AC adapter for the O2 and mini-USB cable to the ODAC.

      Delete
    2. Nick,

      This is mainly handled in the previous ODAC article from a few weeks ago.
      1) yes + you need to soler three wires from the output of the ODAC to the underside of the O2's input jack
      2) I assume they will, but you better ask them
      3) See answer 1 for internal. For external, just get a short interconnect cable. FiiO sells some nice short ones.
      4) From a standard computer, you need a USB cable + AC power for the O2 if you're listening to the ODAC as the audio source. From something like an iPad which doesn't supply enough power for the ODAC, you will need a power boost Y-cable for the USB cable + AC power for the O2. If you're only listening to the O2 as an amp stage from something other than the ODAC (audio source plugged into the O2's input jack externally), then you only need the AC power while it's performing that function.

      Delete
  17. Hi nwavguy,

    As usual, its another great updates on the odac.

    One small question on the ODAC enclosure. Given the pcb which measures 49 x 58mm, Its actually a pretty odd size..
    I was originally thinking it should fit a b1-080 box,(b1-080 box only have a internal width of 51.5mm)

    For Diyers like us who want to use the odac as a standalone unit with customized faceplates, is there any recommendation on the enclosure available on the market that we can use (aside from the jdslab options)?

    ReplyDelete
  18. Is there a limit to how much volume can be attenuated from the computer before you start noticing noise? Thanks.

    ReplyDelete
  19. Several answer in one comment...

    @Anon, no estimate on the ODA time table. I'm aiming before the end of summer but it could, worst case, drag into fall.

    @Nick, JDS plans to sell O2s with the ODAC and will have a suitable back panel. I'm not sure if they'll sell just the plates but they should be happy to sell them with the ODAC board. There's also a guy on diyAudio considering offering laser cut acrylic panels. The ODAC connects internally with 3 soldered wires to the input jack of the O2. And yes, you still need AC power connected to the O2 to use it. USB can't supply enough power for a serious headphone amp.

    @madcap, JDS should have at least one enclosure and panel available for a standalone version. I'm not sure about the other ODAC board sources. It is an odd size, but that's because of the O2 compatibility (it's basically the size of two nine volt batteries).

    @anon, it's not so much how far you turn down the computer volume, but how far you crank up the "downstream" volume (gain) after the ODAC. Generally if you were to set your PC volume to max and then set your headphone amp, powered speakers, etc. to the maximum level you ever want to listen at, you should never hear any noise from the the ODAC even with the PC volume reduced to a whisper. You can just leave the downstream volume set to that high level and control the volume exclusively from the PC if you want with no audible degradation.

    ReplyDelete
    Replies
    1. Any chance we'll learn how, if feasable, to hook up an I2C volume controller to this thing?

      Delete
  20. Nice, I've been reading your articles for a while.

    Also looking forward to the Timex vs Rolex article.

    You have gained quite a cult-like following on 4chan's /g/ (technology) forum.

    ReplyDelete
    Replies
    1. Thanks. A following is good, but I'm not sure about "cult-like" ;) I'm trying to offer information to help others make better choices on their own rather than being spoon fed the usual hype and myths. The Timex vs Rolex article will hopefully further help the information cause and should be up by next Wed or sooner.

      Delete
    2. Well, /g/ is a cult indeed. Now they have at least two gods to worship, RMS and you.

      Delete
  21. I like your science based point of view NwAvGuy, i hope that the odac will be a success for you.

    Are there any chances to see a 32bit AD converter in the future? ESS has those 32bit ADC chips, and i think the one who developes an mastering grade ADC with lots of headroom for a reasonable price would change the bedroom music production scene completely.

    Using AD converters with lots of headroom is really useful if you want to track out as loud as possible without loosing dynamics and running into clipping. Opposed to highend DAC's for music listening, highend ADC's are a really useful tool for beeing as competitive as possible!

    Cheers

    Ramón

    ReplyDelete
  22. Is it April first again? Damn that's RESULT! It still amazes me how you can put so much effort in.

    ReplyDelete
  23. The real story here is going from 98 dB dynamic range to 111 dB dynamic range in four major design iterations, with 112 dB being more or less the limit. It's kind of surprising that, even with you using the reference design as a guide and (I'm assuming) following pretty good design practices, the first iteration turned out so much worse than what's actually possible. It makes me wonder how many iterations some of the boutique audio companies are going through, and how much improvement they get.

    It sounds like a lot of hard work paid off. Congratulations.

    Or maybe you intentionally screwed up on purpose originally, to make a point about something or another and somehow further your shady ulterior motives?! ;)

    P.S. I still consider anybody who can solder those tiny SMT parts, even aided by the right tools, to be a wizard—-or at least with some magic combination of more manual dexterity and patience than I probably have. I don't even want to think about all the part swapping that was required to test the combinations on the dScope.

    ReplyDelete
  24. I have preordered the ODAC-board. Can I just put the ODAC-board into a case of wood, and then connect it to my O2 without missing out on the audio quality? I don`t have the skills to build it into my O2.

    ReplyDelete
  25. I´m willing to see the objective2/odac assembled together. Congratulations!
    Latter you could measure them against fiio e17.

    ReplyDelete
  26. I own a Dacport LX and an O2 amplifier. Will the ODAC measure significantly better than the dacport?

    ReplyDelete
    Replies
    1. Based on the numbers up here and the measurements of the DACPort at Stereophile, it seems they'll both sound the same (transparent).

      Delete
  27. First of all, great job! If you don't mind I have a few questions/request that can be interesting for we, your readers:

    - Could it be possible to include in the future comparation between other DACs and ODAC a few "recpectable"/know "music oriented" soundcards like Creative, Auzentech, Xonars? They can be in the same price range or higher (like the DACs). They have different features than the ODAC like recording, gaming effects vs portability of ODAC, but just to compare comsumer's options.

    - Talking about "digital atenuation" (soft vol control) explained above to an anon, could you like to explain it further in a cuantitative way?
    E.g. 24bit DAC chip "X" gives you... let's say DR 126dB, and a DAC+AMP design including "X" gives you DR 110dB. If people like to use only their computer software volume, how far can they push down software volume before loosing transpacency versus the DAC+AMP analog volume? It's true that if output stage noise is "worse" than DAC's, all measurements better than 96dB and 24bit DAC, that allows you to use soft volume without audible loss?

    ReplyDelete
  28. Why do you use different units in your comparison chart?
    like dB vs. dBA and dBuA/dBuC/dBu

    ReplyDelete
  29. Hi NwAvGuy... as allways an impressive work. Congratulations.
    I'm planning to buy your ODAC without amp (all mounted with an enclosure) because I have a decent amp now, but my problem is that I need an optical input at least. I know you won't provide this input in your DAC but (I don't know if it is a silly question) Are there any option to combine a optical source with your ODAC?... I know that exists some good products to convert USB to coaxial like the M2Tech hiFace, but I don't know if it's possible to make the inverse or do you have any idea to get it.
    Thanks in advance!

    ReplyDelete
    Replies
    1. @Zine, you should be able to get an inexpensive USB audio device that will accept optical S/PDIF in. Here is a device that offers S/PDIF optical in and out:

      http://www.siig.com/it-products/multimedia/sound/usb/usb-soundwave-7-1-digital.html

      Now, if you plug headphones in to that device and listen to it, it will not sound anywhere near as good as an ODAC+O2. However, I would expect it to do a good job of importing bits through S/PDIF. I'm not an expert on this stuff, but I would expect it to do a perfect job, since S/PDIF is digital. Or maybe there could be jitter issues or something? (I'm a software guy, not hardware.)

      P.S. @NwAvGuy, I am listening to music with an HRT Music Streamer II plugged in to an O2, and I am enjoying the clean sound very much. I look forward to someday getting an ODA. Thanks for doing the things you do.

      Delete
    2. @SteveHA & Zine, Glad you like your O2, but the SIIG device doesn't help. The ODAC only accepts digital input via USB. For a device to convert S/PDIF to USB it would have to be a full USB HOST with the entire USB Audio driver stack implemented in it. It would essentially have to be a small computer that implements the entire audio subsystem and USB host functionality. You can use a PC to do the conversion, but at that point, why not just play the music on the computer?

      Delete
  30. @NwAvGuy:

    So the DAC1 Pre and the ODAC cannot be reliably differentiated in blind testing. What about the FiiO E10? Are you sure you can reliably tell the difference between the FiiO E10 and the DAC1 Pre in blind testing?

    As for the O2, would it be possible to make a SMD version to make it more portable? (I'm strongly suspecting the answer is in the negative, but it can't hurt to ask.) I don't know how much effort and money goes into designing/producing a new board and if you could easily just switch out the components to SMD counterparts and have it work? (I don't agree that SMD soldering is DIY unfriendly. I think most people simply don't know the correct techniques.)

    ReplyDelete
  31. Does the ODAC have USB isolation like the HRT Music Streamer II ?

    ReplyDelete
  32. WHat needs to happen to get a pre order ???? Are you running the it??? or do we have to go through headfi or other forum??

    I can't believe it's time !!! I thought it would take at least another year to make this happen .... PLEASE let me know what I need to do to reserve one. I am VERY excited!!! My O2 amp needs a friend,

    Tim
    Coacharnold@gmail.com

    ReplyDelete
  33. Here come the dumb questions:

    1) Does the type or length of the USB cable matter at all to the ODAC?
    2) My E10 is 44/16 on Windows shared mode because of your results. Should that be the same for the ODAC with Pandora, Spotify, MOG?
    3) In the exclusive mode, given your readings of the E10, should the foobar WASAPI be 16 or 24 bits resolution to play files some of which are FLAC and 96/24 WMA?

    ReplyDelete
  34. IT IS DONE

    GIVE THIS MAN A MEDAL

    ReplyDelete
  35. Wow, thanks everyone. We're already up to 70 comments so here's a bunch of answers:

    @Kite & Nick (multiple), The ODAC's 3.5mm output could work while it's installed in the O2 as long as the O2 is either powered on, or you plug something into the O2's input jack. I'll be providing the soldering details soon. It's just a few twisted wires from the 4 pin header on the ODAC to the O2 input jack and you have to cut the two "jumper traces". The board mounts with a single screw and the side groove. The ODA will have a preamp output that can feed speakers, etc. More soon.

    @Anon & Anon (USB isolation), The ODAC doesn't need USB isolation for transparent performance. I'll look into the eBay USB isolators. It would be interesting to test one and show the "before and after" results.

    @Anon (32 bit ADC). 32 bits is 99% marketing hype. There isn't a 24 bit audio ADC with true 24 bit (144 dB DNR) performance. There also isn't 24 bits of dynamic range in music (jackhammer at 1 inch away to absolute silence orbiting pluto). I understand about headroom, clipping, etc. but the best microphones you can buy are inferior to even 24 bit ADCs.

    @Anon (miniaturization). More cramped circuits usually have less optimal layouts. Smaller resistors perform worse than bigger ones. There's less room for power supply filtering, etc. Another factor in portable gear is often power consumption. Ultra low power circuits tend to be inferior. That said, the proof is in the measurements and Apple has done a very respectable job.

    @Mikeaj (98 to 111 dB). George and I both would have been ecstatic if the first board would have come close to the ESS chip specs. Having to spin three more iterations was a lot of time and money. The O2 also went through four board revisions and many changes. Based on the measurements I've seen, a lot of others designers don't bother because they think the first version sounds fine--but they're listening with heavy Expectation Bias.

    @Anon (amp needed?). The ODAC cannot directly drive headphones (or speakers). It can directly drive most any line/aux input (powered speakers, desktop audio system, receiver, etc.).

    @Anon (wood case). Yes, a wood case is fine for the ODAC.

    @Anon (ODAC vs Dacport). The ODAC won't outperform the Dacport LX. Both are transparent DACs.

    @Marcelo (software volume), The dynamic range and software volume issue gets complicated and I'll cover it more in the next ODAC article. As I said above, it's related to how much gain you have after the DAC. If you have a FiiO E10 with 97 dB of DNR, and you plug it directly into a typical power amp with 26 dB of gain, you get 97-26=71 dB of likely audible noise at the speakers.

    @Anon (dB units), Sorry for the confusion. The letters in the measurement table are rating "grades" A - F with "A" being the best. I'll add a note explaining that.

    @Zine (USB to S/PDIF), Sorry I'm not aware of anything that converts S/PDIF to USB.

    @Anon (E10 blind testing). I don't have the E10 anymore. It would be an interesting test.

    @Anon (SMT O2). The main reason to switch the O2 to SMT would be mass production using automated assembly. It won't make it much smaller as the batteries, jacks, controls, electrolytic caps, and regulators can't shrink much without compromising the performance. We have talked about putting the ODAC and O2 on one board. Stay tuned.

    @nostradamus (multiple), My ODAC has worked fine with 10 foot cables but, with all USB devices, shorter is better. If possible, just leave the ODAC at 24 bits even playing 16 bit material.

    ReplyDelete
  36. Hey, what kind of connector does the USB cable need on the ODAC side? It does not look like the std. one used for peripherals like printers, or the MSII. Is it the same kind that comes with DAP's, because that would be incredibly convenient as I have like 10.

    ReplyDelete
  37. I'll buy two the second schematics are available. Being fearful of other manufacturers copying and cost-reducing your product is not the way to make a great product. Since you have the knowledge and are first-to-move, other manufacturers will always lag behind. Plus, people like to support the creator directly, even if it does cost a few bucks more.

    ReplyDelete
  38. @Satellite, Good question. I just added an "OTHER DETAILS" paragraph to the article and included the type of USB cable. It's a "Mini B" which indeed is used by a ton of other devices. The even thinner one is the "Micro B" which is on many of the newest phones.

    @Jonathan, That's your choice. Next time we do a new design, we'll come to you to fund the large production run and see if you want to take the risk yourself not knowing how many others you'll be competing against. My bet is you'll not fund the project.

    @Maty, I did not approve your post with the giant cartoon image and the repost from diyAudio. I'm not sure it's legal to repost content from there, and it only clutters up the comments here with redundant information. We're going to have a difficult time keeping this article under 200 comments as it is.

    And, to everyone: So far I've published nearly every comment I've received. The only exceptions were either redundant, already answered, or corrections to the article which I made. Please be aware I'll probably be forced to go through and clean out some comments to make room for new ones unless Google comes up with a workable fix to the 200 comment limit.

    ReplyDelete
  39. How does the ODAC differ from the Calyx Coffee from a design standpoint? It seems like both are using the same DAC and USB interface as well.

    ReplyDelete
    Replies
    1. We've not seen the Calyx in the flesh. But, on paper, it does indeed look very similar. If they sweated the details properly it should perform similarly. Interestingly, however, their performance specs are not as good. It would be fun to measure one. I don't know much about them, but they seem to be yet another "boutique" manufacturer.

      Delete
    2. @Marcelo Apr 19, 2012 01:44 AM:

      "Could it be possible to include in the future comparation between other DACs and ODAC a few "recpectable"/know "music oriented" soundcards like Creative, Auzentech, Xonars?"

      More tests and measurements would be interesting, there is unfortunately very limited objective information available on the large majority of devices, even the most commonly used ones. From the specifications and loopback based tests, it looks like the better sound cards such as the Titanium HD and Xonar D1 and above may be capable of quite competent DAC performance (possibly even better than the ODAC in some aspects, but not in others); however, the usual caveat with these is that the (mainly noise) performance can vary more between machines. The X-Fi HD USB looks good, and has a decent set of features for about $100, but it apparently does not support 44100 Hz sample rate in hardware, which is a major limitation.

      Delete
    3. The other major limitation is the headphone outputs often have an unacceptably high output impedance--even on the expensive products. So the cards that can play bit accurate native CD audio (which seems to exclude a lot of Creative products including the X-Fi HD), and have acceptable measured audio performance (which excludes a lot of the cheaper options), might be suitable as a line output source. Just don't plug headphones into them.

      Delete
  40. Another amazing effort by NwAvGuy. You Sir are simply the best.

    ReplyDelete
  41. Very impressive indeed NwAvGuy! My congratulations and my deep hearted thanks for your generous contribution(s) to the DIY and/or audio communities!
    It seems that you managed to filter PC supplied USB power differential noise quite effectively. But I am concerned (like other posters before me) about common mode noise (the noise that is the same on both signal polarities in respect to real ground), which is, probably, the main reason behind the need of ferrites in the USB cable. Common ground with a PC will inevitably lead to excessive common mode noise, especially in some cases. Do you believe that adding some ferrites in the USB cable is sufficient enough for eliminating this? Maybe a USB isolator would be a much more potent solution? The use of USB isolators (as others have already noted) maybe is, in some cases, essential. Please do check some USB isolators!
    I suspect that the presence of common mode noise is one of the main reasons behind some weird audiophile myths (like elevating the cables off the ground with wooden bases, or plugging the mains voltage socket with a ‘correct’ polarity).

    ReplyDelete
    Replies
    1. If you're talking about common mode noise degrading the USB data I haven't seen that to be an issue. If you're talking about analog ground loops with other connected equipment, that's a potential problem with any non-isolated USB DAC including the Benchmark DAC1 and lots of other high-end gear with USB connections.

      The ferrites do suppress common mode noise and also radiated noise. That lowers noise on the ODAC's power supplies, which in turn, slightly improves the overall performance. The ferrites are also a good idea for RFI suppression from cell phones, etc. The ODAC has ferrite filtering on board, but it helps to keep as much noise as possible from even reaching the ODAC.

      For headphone use, because headphone amps usually are not grounded to the mains ground, and headphones "float" with respect to ground, it should not be a problem. The entire signal chain will have only a single external ground--the USB ground. But if you connect to the ODAC to downstream equipment with an AC line/mains ground, there is some chance of ground loops resulting in audible hum but even that's relatively rare.

      In terms of the digital signals, and USB power bus noise, I tried a total of four different PCs and a couple different external USB hubs. I didn't see any cause for concern with any of the USB sources. But it's conceivable there might be some especially nasty PCs that could significantly benefit from an isolated connection.

      I will try to test one of the cheap eBay devices and report what I find.

      Delete
    2. Did you consider or test any opto or GMR isolators for I2S? I think the jitter added may be worse than the problem they solve for a DAC without an ASRC, however.

      Delete
    3. We were happy with the results without. I'm not sure there would be any benefit without additional power supply complexity, a 4 layer board, and as you suggest, the isolator itself could make jitter worse.

      The more parts you throw on a board, however high-end they might be, the longer the development time will generally be to really get it all right. And the higher the performance the parts, the more time you'll spend chasing out the last bits of reducible noise, distortion, and jitter to extract what they're capable of.

      While it would be interesting to go down that path as an academic experiment it would also be somewhat like trying to build a high performance car in your garage from scratch. If someone wants that level of total overkill I'd suggest a Benchmark DAC1 or something that measures even better like an Anedio or dCS. At least that way the multiple man years of development time are amortized over a larger number of units sold and you can have your great DAC tomorrow instead of 2 years from now.

      Delete
  42. Another laudable step forward towards great sound, Fantastic!
    I'm already lining up to get an ODAC and some true transparency, that much is certain.

    I would like to make a couple of suggestions regarding your anticipations for your future article with further information and data on the ODAC. I believe it would be highly useful -and healthy- to compare the results of the ODAC with some real test results of the more emblematic design-by-ear DACs you mentioned (Schiit, Burson, etc.), as well as with the highly priced crowd like, Ayre, Wavelenth, Bel Canto, Wyred 4 Sound and dCS, etc. (I don't know if the measurements that Stereophile has made of the latter DACs could fit and expand the chart you've already made with the DAC1, the FiiO E10 and the ODAC, but that would certainly put things in perspective.)

    It would be sobering and healthy to see if the incredibly priced DACs I mentioned above, are actually worth their performances -and design concepts-, and it would be equally healthy to see how far off design-by-ear can really be, and of course the particular design errors, mistakes and flaws of each case.

    This might be just as ideal as unrealistic, and perhaps as plausible as getting George from YoYoDyne to lead a blind test for Headfonia's reviewers. It's probably easier to expand your FiiO/Benchmark/ODAC with already available measurements than getting all the design-by-ear DACs and get down to testing them, and I also suspect your confidence in numbers, the lessons learned with the ODAC, and the logic argumentation that excludes the design-by-ear manufacturers from hitting gold by mere chance, probably makes it more a waste of time than anything interesting to you. Nevertheless, I believe it would be ideal to have the facts, the numbers, and thus a definitive argument. (I believe there are people who really WANT to sell us snake oil, and that there's people with good intentions who really don't know they are... I could be wrong of course.)

    I hope I haven't repeated suggestions someone else already might have made, but my enthusiasm got me straight past the comment thread.

    Fabulous job as usal; looking forward to (not) hearing the ODAC ;-)

    ReplyDelete
    Replies
    1. Thanks for the kind words and suggestions. I agree that would be helpful and perhaps enlightening. But, as you suspect, it's easier said than done.

      The Benchmark DAC1 is reasonably representative of the "highly priced crowd". It's a Stereophile Class A component that's been compared with many other expensive DACs. I happen to have one which makes it an obvious (and cost/time saving) choice.

      I'll also be comparing the ODAC to an HA-Info DAC, and while it's not a Burson, it likely was designed by ear (at least I hope it was because some of the measurements are inexcusably bad). Again, I happen to have this DAC so it's a convenient comparison.

      While some high-end DACs are an interesting exercise in chasing perfection, what ultimately matters is what's really needed for the best sound? Those want to spend more than necessary for transparency on higher end gear may have some good reasons for doing so. But believing they're getting audibly better sound shouldn't be one of those reasons. This is covered more in the soon to be published Timex vs Rolex article.

      It's not possible to do a "differencing" (null) test on a DAC or player without some significant compromises. But it is possible to do correctly with amplifiers, other analog electronics, and even cables. What I hope to demonstrate soon is high-end gear, beyond the requirements for transparency, doesn't offer any further benefit even using music rather than test signals. An amplifier either lowers the fidelity in audible ways, or it's audibly transparent. Those are the only choices.

      If I can show some of the "believers" are wrong about the sound of amplifiers, hopefully some will realize they also may be wrong about the sound of DACs and players. Differencing nicely fits between blind listening and conventional measurements. And it's virtually impossible to credibly dispute.

      Delete
    2. Differencing the ODAC would be possible by recording its output with a high quality ADC, such as the Benchmark or even the dScope (if it can be used for this purpose), and then using software for the test against the original file. Of course, the extra conversion and any software processing needed does reduce the accuracy somewhat, but it could still be an interesting experiment.

      Delete
    3. That's a great idea, and it's been tried. But the problem with differencing digital signals made with different clocks is you get "sample drift". In the example you give the samples in the file made with the ADC1 would not match up exactly with the samples in the original file. Bill Waslo explains this in detail in his AES paper, and he tried to correct for it in his software, but in practice the software can only partly correct for it. To make a long story short, the limitations of the software's ability to correct for sample drift masks the extremely low level distortion produced by a transparent DAC.

      It's a complex problem when you look at it from a mathematical perspective. As the music plays the samples drift further apart. So you're having to apply a constantly changing correction factor to compute the difference. And you're having to interpolate between samples to do it. The result, at least in Bill's software, is only perhaps 99% effective. To expose the ODAC's 0.003% distortion, it needs to be better than 99.9997% effective.

      Delete
    4. I actually tried correcting the drift in practice, recording some test signals from onboard audio with a sound card. The main issue is not the quality of the sample rate conversion (which I can easily make good enough that the degradation caused by it is not significant compared to that of the analog hardware, especially since the recording sample rate is 96000 Hz, and most of the resampling artifacts are near the Nyquist frequency), but rather finding the correct delay and pitch correction factor. For that purpose, I added a few seconds of 12 kHz tone before and after the music; from the phase difference compared to the original signal, and its change over time, the correct parameters for the resampler can be determined with good accuracy.

      Delete
    5. I'm sure he probably has better things to do, but you might try contacting Bill Waslo (if you have't already) and at least sharing your ideas with him? He's a great guy and I admire his efforts to promote additional objectivity in high-end audio.

      If someone really can sufficiently solve the differencing problem when comparing asynchronously clocked files, that would help strongly dispute the "UAC2 asynch (or insert another mythological technology here) sounds way better" theories.

      Of course, I'm sure the same fanboys would then claim even a high-end ADC was completely masking the differences--never mind most music is recorded with lesser ADCs to begin with. And all you're looking for are relative differences which, if audible in the first place, should not be significantly masked by a high quality ADC. Meyer & Moran demonstrated an ADC can be entirely transparent in the signal path.

      Delete
    6. Well, in this case, one of the samples used for differencing or ABX testing would not go through any ADC or DAC at all. Therefore, potential problems with the method that can be subject to criticism are limited to the ADC improving the sound quality compared to the typical usage with an amplifier, or the original signal being degraded by any software processing (level adjustment, delay, sample rate conversion) applied, although the latter can be made more transparent than even a high end D/A-A/D loop.

      Delete
  43. So this ODAC beats easy to Benchmark DAC1 a superior price dac.
    I´m waiting for your ODAC realized as I wish to buy a Cambridge audio Dacmagic,but in place I´ll order your odacMagic.

    ReplyDelete
    Replies
    1. The new DACmagic 100 or DACmagic Plus should be be similarly transparent to the ODAC. They have more features but they also cost a lot more. The older original DACmagic doesn't support 24 bit operation over USB.

      Delete
  44. Great work! I'm buying one and putting it in a standalone case!

    My power amp works best if fed by a balanced line.
    How would I go about implementing the Jensen recommendation found below?
    http://www.jensen-transformers.com/an/an003.pdf

    (See page 3, paragraph 2.4)

    Thanks in advance!

    ReplyDelete
    Replies
    1. Jensen is a great company but those techniques mainly apply to long cable runs, studio environments, etc. For home use, you can simply use an inexpensive unbalanced-to-balanced adapter that grounds one of the input pins of your amp (unless you plan on running the cable several meters or more). If you end up with ground loop problems (unlikely--see my answer to the common noise question above) then some form of ground isolation may be appropriate.

      Delete
  45. I have to agree that the commercial interests of your partner NwAvGuy are in a bit of a contrast to your prior "preachings". Since essentially it makes "peer review" impossible and the only way to even verify the measurements is to obtain a sample of the first production batch, which will already be shipped to end users. Essentially meaning that if it turns out that the ODAC doesn't live up to its promises, or that if your partner can't or won't deliver the promised quality, then buyers will have fallen prey to some very elaborate marketing.

    Still though, it's hard to believe that anyone would have put up with the effort you have made so far, just to con a few thousand(?) people into buying a moderately priced DAC board. Besides, there have already been ample opportunities to do so before.

    @Jonathan with regard to "Since you have the knowledge and are first-to-move, other manufacturers will always lag behind." - I don't think competition is really that much of an issue here as NwAvGuy has been led to believe by his partner. If it was, we would be seeing a lot more companies selling the O2 today. Thus given the additional licensing/NDA hurdles of the DAC chip, it's doubtful to assume that the ODAC would receive any more attention than the O2.

    The real motive behind the non-public design I believe is instead some creative marketing - trying to create artificial demand by putting "thumbscrews" on the buyers and creating this notion that "if the first batch doesn't sell well enough, then there won't be a second batch, and you will have missed out on the ODAC - so you better hurry up with that preorder". It's not pretty, but that's the price for getting in bed with .. uhm .. "entrepreneurs".

    On a side note, the posts in the diyAudio forum seem a bit concerning too: "no objections to the project, though i'm not looking forward to a lot of objectivity thugs now armed with an article on digital, going around telling people their diy not commercial fun is a fruitless/pointless pursuit". While I clearly dislike the commercial aspect of the ODAC as much as anyone else, it still should be clear that being an "objectivity thug" is a good thing, regardless of the surrounding circumstances. Link to the forum post: http://www.diyaudio.com/forums/digital-line-level/211046-odac-questions-answers-related-device-2.html#post2991598

    Love it (because of the brilliant engineering) or hate it (because of the marketing), the ODAC is here now either ways – not much point in crying over spilled milk. But if there is one thing to take away from this “ODAC affair” NwAvGuy, it is that compromising on your ideals, even if it is to make the otherwise impossible happen, may not seem so clever any more in hindsight. Maybe a (DIY-friendly) fully open source ODAC would have been the better way, even if this meant that it would only exist at some point in the future and not today. I myself am certainly looking forward to that day and holding off any ODAC purchases and recommendations up until then.

    Full disclosure and peer review are among the cornerstones of science. I can see how people in business suits may not like agreeing with that, but an engineer should know better.

    Please remember why you started this blog. :)

    ReplyDelete
    Replies
    1. In our defense, George and I both believe very much in open source when it's possible. We have both demonstrated as much with prior efforts and both have future open source projects on the table. It's unfair to lump us with marketing-oriented entrepreneurs.

      We would have preferred a DIY friendly open source DAC but I don't think an open source ODAC would have worked. It either would have offered little benefit, or nobody would have wanted to take the financial risk. Yes we could have just opted for no DAC at all, but that would have left everyone out in the cold.

      George and I still would like to offer a DIY-friendly open source DAC that will fit into the ODA as a "plug and play" alternative to the current ODAC. I have commented on this in the previous articles. If we were only interested in "putting thumbscrews on buyers" we certainly wouldn't be letting that cat out of the bag early on.

      The ODAC isn't some elaborate marketing scheme as you suggest. I was receiving requests to design a DAC before the O2 even came along. If you go back and read my comments, I said then it wasn't practical to offer a DIY design significantly better than what was already available.

      The early ODA feedback had many wanting a DAC option. I even talked to Violectric about using their USB 24/96 board but, especially shipped to the USA, it's more expensive than the ODA itself. But many liked the idea. So the ODAC was born as something similar at a more realistic cost. To my surprise the ODAC has largely overshadowed the ODA. That's not my doing, that's just what the majority chose to focus on.

      The fact is most O2s are purchased assembled. So having the ODAC sold by the same vendors isn't much different. You're making it sound like the ODAC was some radical departure from the O2 but it's not. It's just that, for entirely valid reasons, it can't be built by DIYers.

      I view the small number of O2 manufacturers as evidence nobody else wants to take the risk. Once a few are doing a good job, others are much more reluctant to take the risk they can compete.

      If we had made the ODAC open source from the start it would have resulted in a "chicken and egg" problem. It's very difficult to find anyone willing to take the risk to have hundreds of something made not knowing who else might be doing the same thing. I ran into exactly that problem with the O2. Several expressed interest but wanted to wait to see who else was going to produce it.Because it couldn't be built in small quantities, an open source ODAC would have been a non-starter and nothing more than an expensive (for me) academic exercise.

      If you accept the above as reality, we've fully disclosed everything we could and still get the first batch of ODACs produced. And anyone is free to "peer review" and verify my measurements.

      And speaking of "peer review"... Rumor has it at least one of my critics arranged for an O2 to be tested on a AP analyzer. The results, apparently, were close enough to agreeing with mine he never made them public (to do so would have only helped the O2's cause). That puts me in the unenviable position where the only people with real incentive to verify my measurements are the ones who want to see me fail.

      If some want to label me an "objectivity thug" that's nothing new. It's hard to have much sympathy for people who object to factual information. Wine lovers don't like hearing $2 wines win blind taste tests either.

      It's true the first batch could have a problem. George will receive samples before the entire run is approved. With any production run things can go wrong. If there are significant problems we would either correct them or tell everyone. That's hardly unique to the ODAC and is also true of the O2.

      So, I've only tried to give people what they asked for the best way I could come up with. I don't expect everyone to agree with the end result.

      Delete
    2. It's further worth noting Tenor requires a signed non-disclosure agreement to even get the documentation for the TE7022 USB chip. As such, "full disclosure" is legally prohibited to anyone who doesn't have a signed agreement with Tenor in place. So not only can DIYers not buy the chip, they can't even see the datasheet.

      We have explained why we chose the TE7022 and I'm not aware of any better options that could have been used in this application. There are some very good reasons why the TE7022 is used in many similar designs. There just are not many choices for 24 bit USB interfaces. Hopefully there will be some better options in the near future.

      Delete
    3. If its worth anything I have also been combing the semi sites for a suitable diy usb part and agree with your choice to use the 7022. Who knows when something better will be available?

      I think those making a stink about the odac not being fully open sourced are mostly the usual audiophools who probably see the odac as a threat to all thats holy. If they were half as good at finding the problems with all the overpriced toys they build n buy as they are at finding ways to criticize your efforts, the world of audio would be way different. I don't see any of them offering a better dac for 100 bucks open source or not. Whats it cost for an assembled usb gamma or twistedpear 16 bit dac?

      Delete
    4. I do biotech research and, at least in my world, there are roughly at least as many colleagues trying to authenticate and further my work as there are trying to prove it wrong. Your comment about only your critics wanting to verify what you publish, and failing to share the results if you turn out to be right, really sucks. That’s all the downside of peer review without any upside.

      Perhaps someone will eventually take the baton and not only help verify but further what you have started? You obviously have your fans; you just need a suitably eager one with access to an audio engineering lab. In the meantime, my advice is to stay focused on the science. Thanks for all your work.

      Delete
    5. You misunderstood me. I didn’t say open source the ODAC. You explained very well why that is not possible and why it wouldn’t make any sense. However I am saying that you should have created an open source DAC, if that was somehow possible – or might have been possible in the future. You are of course free to pursue whatever projects you deem worthy, and I am very grateful that you have chosen to spend so much time helping us. But I am afraid that the time spent on the ODAC, as it is today, might have been put to better use elsewhere.

      If you think of the impact the O2 has had, then it certainly was a monumental achievement. If you think 10 years from now, then the O2 may very well carry the same significance as CMoy amps.

      How does the ODAC compare to this? It is essentially, despite its great performance, a magical black box of which only 2 people in the world know the inner workings. So 10 .. or even just 5 .. years from now it will “just” be yet another DAC, that may or may not still be in production. It may still have a following, but no one will really be confident about its true performance, whether the ODAC they bought is the same revision some reviewer might have had, and whether the ODAC may have generally been outdated. Not really that different any more from other ancient audiophile gear that is surrounded by mysticisms. If things go utterly wrong then the ODAC could even result in creating further audiophile myths instead of dispelling them. Just imagine someone looking at your design and trying to replicate parts of it, failing to do so properly .. and afterwards insisting that it has to be great because it was designed by NwAvGuy.

      This is why I think that another compromise, be it on sound quality, price or timeframes, might have been so much more acceptable than a compromise on your ideals. I understand that you have been “pressured” into doing the ODAC by readers. And while creating and producing the ODAC will certainly make quite a few people happy today, I am afraid that in the long run the ODAC’s impact probably will be negligible – at least when compared to the O2.

      A word on DIY-friendliness, since you mentioned that most O2s today are assembled commercially. DIY-friendliness should be a criterion for your designs mostly because it allows you to be plausibly decoupled from manufacturing, and thus from the guys handling the money. As for a DIY-friendly DAC being offered later on, you only mentioned it in your April-1st article, so I didn’t think too much of it.

      Delete
    6. Once the ODAC ships you won't need magical powers to convert the board into a schematic. I've already disclosed the basic design and I'll soon be disclosing more details in the next ODAC article. It's hardly a "magical black box". A 10 year old could trace out the PC board. But why does it matter if they can't get the parts to build one?

      My hope is the ODAC is a catalyst for plenty of similar products to come. And that probably won't take long. Even FiiO could have made the E10 into something entirely transparent if they just would have spent a little more time on it. So yeah, 5 or 10 years from now I hope we're swimming in affordable transparent DACs and the ODAC might be long forgotten. That will be a good thing.

      I've said before, the bigger problem is the state of headphone amps. There are a lot of poorly designed amps on the market. In contrast there are plenty of decent DACs. That's why I started with the O2. The ODA was the next logical step. And many made it clear a headphone DAC would be an even better next step.

      Nobody is likely to try to replicate the ODAC on their own as a basement DIY project as they can't get the parts. They should wait for a decent 24 bit DIY DAC, build the Atmel-based "widget", or buy a commercially made DAC.

      My original vision was to marry an enhanced desktop version of the O2 with a transparent DAC to create a genuinely transparent headphone DAC at an affordable price. The ODAC helps accomplish that goal. I don't really care if the ODAC itself has its own Wikipedia page 10 years from now. The main idea is to raise the bar and encourage others to follow.

      Delete
    7. My concern with the entrepreneurial aspects of the ODAC design is that the other USB interface chips seem to have been passed over without a great deal of testing. This is at least partly because one of the primary objectives of the design was to allow the ODAC to be retrofitted to the O2 enclosure and expand its marketability, especially with the followers of this blog who are obviously quite likely to have a O2 amp.

      I'd also like to know if there was any testing to verify the idea that the XMOS implementation provided no audible benefit over the TE7022L that was used? It seems (based on looking at other designs and this one) that the TE7022 restricts you to using only a 12MHz clock and as a result you are limited to a synthetic 44.1kHz clock and rely on the ESS DAC jitter immunity to remove any jitter, sure the jitter tests used indicate that this has a small impact. I'm more interested in knowing was it tested or just disregarded on grounds of the licensing issues.

      Is there anything that indicates a device managing to measure even better than ODAC+O2 or DAC1 Pre wouldn't be in fact more audibly transparent? Has any testing been done to verify that the DAC1 Pre is as good as it gets for sound and that exceeding its measured ability is indeed futile?

      Delete
    8. @Chris, we have explained many times why the XMOS solution wasn’t suitable for this design. It’s more expensive, it requires a proprietary driver for Windows (70% of the traffic to this blog is from Windows), and it requires an expensive license to distribute the proprietary driver.

      The jitter of the ODAC is very similar at different sampling rates. I’m not sure what you mean by “synthetic 44.1 Khz clock” but using a PLL to derive the audio clock from the on board 12 Mhz crystal is a good thing as it decouples the audio clock from the USB timing. The Tenor and ESS chips share in keeping jitter low.

      If you’ve read Meyer & Moran, you know it’s possible for a DAC to be so transparent you can’t even tell when it’s in the signal path. As this article and others explain, a lot of smart people have figured out what sort of performance it takes to make that happen. I’ve presented a lot of evidence supporting this and it’s the basis for when I say the ODAC and DAC1 are transparent.

      The ODAC and DAC1 both measure sufficiently well to be transparent. But they do measure differently and the DAC1 has a performance advantage in several areas, especially jitter. Yet they sound the same. This is further evidence you’re not going to get better sound by trying to further improve the ODAC, use the XMOS interface, etc.

      Put another way, I doubt the XMOS interface beats the DAC1 for jitter. Yet the ODAC and DAC1 sound the same. And it’s not surprising. All the audible ODAC jitter components, using a worst case signal designed to expose the maximum amount of jitter, are below -110 dB and the total is below -103 dB. Consider Ethan Winer’s distortion audibility test. The distortion is entirely masked by the music around -75 dB. A level of -103 dB is massively lower. Ethan very conservatively sets the bar at -100 dB for assured total transparency.

      So yeah, I believe it is indeed “futile” in terms of audible sound quality and fidelity to try for better sound than you can get from the ODAC or DAC1. I don’t expect everyone to believe that, but I really hope some of the more vocal disbelievers will agree to a credible public blind listening test so they can see for themselves and others can learn from the result.

      The other option are DACs with lower fidelity that have audible distortion and/or problems. If some prefer viewing their world through a dirty lens, that’s their choice. But that’s not what the ODAC is all about.

      Delete
  46. Hello NWAVGuy, this could be too much to ask especially since your partner want's to recoup investment, but could you elaborate on the lesson's learned? for example, what is the engineering/physics reason that more PS filtering results in higher jitter? I see a large inductor in front of the regulator for the ESS DAC (I think), was this a large factor in getting the additional DR?
    Thanks in advance...

    ReplyDelete
    Replies
    1. I plan to cover more of that in the next article which will be much more of a "geekfest". To be honest, some of the improvements were more about trial and error than physics. Some things, that should have helped, made the performance worse. When you're dealing with a noise floor down around -140 dB that's only 200 nanovolts. At such miniscule levels it's much more difficult to model all the potential sources of noise.

      Delete
  47. Hi,
    your USB licencing discussion above is a bit controversial to me. Or at least a bit out of the spirit of previous articles.
    Earlier article referenced Meyer & Moran study which found that 44.1k/16 resolution is good enough for audio transparency. That means there is no real point in using fancy 96k/24 USB solution!
    The majority of our stored music is anyway 44.1k/16 and the minority rest can be easily down converted to 44.1k or 48k.
    ODAC could use as well old but off the shelf TI or CM USB SPDIF/I2S chip @ 44.1k/16 or 48k/16. Than an ASRC chip from TI or ADI can remove any USB related jitters and feed a good enough DAC chip. Look at Matrix Cube structure. That solution could be completely open sourced !
    So if M&M is right than why do we need fashionable 96k/24 USB for music playback?

    ReplyDelete
    Replies
    1. I've explained this many times in various ways including on this page. It's all about where in the signal path you control the volume. Meyer & Moran tested with the volume control after the 16 bit DAC. If you put the volume control before the 16 bit DAC you can easily have an audible problem.

      A typical power amp, for example, has around 26 dB of gain. If you plug your 16 bit DAC, with a -96 dB best case noise level, into such a power amp it amplifies that noise floor to -70 dB. In a quiet room, -70 dB of noise is plainly audible. But if you use the ODAC, in the same circumstance, the noise is now around -85 dB and likely inaudible.

      M&M also observed if they cranked up the volume control after the DAC to a level well above their test listening level you could hear the noise floor of the 16 bit D/A.

      I've said all along 16 bits is usually enough if you only want to control the volume after the DAC. But with a USB DAC that plugs into a PC, many want the freedom to control the volume from their PC as well. A good 24 bit DAC gives them that freedom.

      Further, some want to play bit accurate hi-res content in the native format. With the exception of 24/88, the ODAC lets them do that as well.

      Delete
  48. Hi,
    thanks, now I got your point about the PC volume control. That really might need 20+ bits even for 16 bit files.
    BTW what feature I miss on ODAC/ODA is and SPDIF out. I could use ODAC/ODA box on my desk next to my PC. And I could also hook my power amp via SPDIF which is few meters away.
    Is there an easy heck on TE7022 to make a transformer coupled SPDIF out? Some of the old TI chips could do that.

    ReplyDelete
    Replies
    1. The TE7022 has an S/PDIF output that other designs have implemented. Because everyone seemed to only want an S/PDIF input we didn't bother with the output. It may require different programming to enable the output. I'm not sure if it shows up as a different USB endpoint or if it's enabled in parallel with the I2S output.

      Delete
  49. That is good news! Maybe once you are over the big tasks you can launch an app.note how to heck ODAC for SPDIF out :)
    I have any application of an SPDIF in. Have no CD player etc since years. All CDs are ripped and put on NAS or mobile phone.

    ReplyDelete
  50. Have you seen this guy's reviews? He uses a Rohde and Schwarz UPL to do measurements. I thought you might like his site if you haven't seen it before. I stumbled on it looking for measurements of the Dacmagic Plus, which he posted last month. Perhaps he would be willing to verify measurements for you, he seems to have some pretty good equipment and is through with his measurements in his reviews.

    http://kenrockwell.com/audio/cambridge/dacmagic-plus.htm#meas

    Thank you again for all of your hard work, I look forward to getting an ODAC!

    Aaron.

    ReplyDelete
    Replies
    1. Thanks. I was aware of his website. I think he's mainly a camera guy but he does some good audio reviews as well. I wasn't aware he'd reviewed the DacMagic Plus. It's good to see it performs well as I expected.

      I'm all for more objective measurements with credible equipment. And he appears to be doing it right. So it's a good resource for more objective reviews.

      The Rohde & Schwarz UPL is a very respectable analyzer for what it does. It's worth noting, however, the UPL can't do a lot of things the dScope can. It's more of a stand-alone instrument and, at least last time I played with one at a trade show, their PC software was much more limited than the dScope or Audio Precision software. Still, the basic analog performance of the UPL is at least the equal of the dScope and light years ahead of anything RMAA can do.

      Delete
  51. Greetings NwAvGuy! All I can say is WOW and be grateful that you have provided so much useful education material and practical application to a field that sorely needed it. I don't care one whit that I can't buy a Tenor chip or even that I cannot solder it most likely. I find much of the discussion here trite, pointless and derogatory to your stalwart efforts. People, why does it bother you so much that you can't see a schematic diagram when you can't even get the part, nor can any of you design, test or build one? Will you critique the traces or say a different should have been used? Do you consider yourself digital audio designers that can actually "peer review" this device and make any constructive comments towards it's design? I think not. Some of you need to seriously check your ego at the comment box. All I see coming from several comments is bruised vanity. How many of you stepped up to the plate to make this dream come true? I see one. George. Who has every right to recoup his investment made on everyone's behalf. Don't forget, this ODAC was wanted by so many people posting for it, so doing more than they had to, NwAvGuy & George made it a reality. You all asked for it, you got it. Now grow up and quit your griping. Eat your peas and be thankful you have them :)

    P.S. I have messaged NwAvGuy several times for assistance and he has been nothing but courteous and professional. This man, I can tell, has integrity not commonly found.

    ReplyDelete
  52. Congratulations on the birth. Well done.

    Rob

    PS - for @zine - S/PDIF to USB: Edirol UA1D and others available.

    ReplyDelete
  53. Hi everyone,
    i wanted to know if this ODAC could be a good choice between my ONKYO 875 amp and my squeezebox duet music player.
    The squeezebox has as outputs Coax, SPDIF and RCA.
    thanks for your info.

    ReplyDelete
    Replies
    1. No the ODAC won't work. I would just connect your Squeezebox directly to your Onkyo. The Squeezebox players generally measure very well and are likely transparent on their own (or very close). The head engineer at Slim Devices, Sean Adams, was a fanatic about audio performance and a smart guy. With the exception of their very first players, he seems to have gotten the audio hardware right.

      Delete
  54. I'm not sure if this has been asked before but, can you plug any source to the 3.5mm O2's input having the ODAC already soldered and powered up?

    -What would happen if both the ODAC and the 3.5mm O2's input feed signal at the same time?

    -Thank you!

    ReplyDelete
    Replies
    1. Yes, the jack itself acts as a switch. When you plug in another source, the ODAC is disconnected. If something is playing over USB you won't hear it until you unplug the other source.

      Delete
  55. Several SACD users have dismissed the Meyer and Moran experiment because they claim "SACDs support beyond 20 kHz and the ultrasonic transients affect how we perceive sound".

    Does this have any scientific basis? According to physics teacher, digital audio preserves transients up to the fifth one and that no one can hear anything higher (sixth and beyond) in the first place.

    ReplyDelete
    Replies
    1. Those dismissing Meyer and Moran have done so with myth, hearsay, armchair pseudoscience, psychobabble and, in some cases, blatant lies. It was all bad enough Meyer and Moran published a Follow Up Paper to address all the erroneous claims and rock throwing.

      Basically, it would be trivial for the multimillion dollar SACD industry to have arranged their own blind test showing the benefits of the SACD format to salvage their reputation and business investment. But they didn't and SACD Was Deemed A Failure.

      The same holds true for all those outspoken critics of M&M. Where's their blind test and AES paper demonstrating what they claim is true?

      It's stunning to me so many believe all the babble over a peer reviewed well conducted scientific study. And there's plenty more evidence. HydrogenAudio has documented several blind studies comparing 16/44 to high resolution audio with very similar results to M&M.

      So to answer your question, no, the M&M critics don't have any applicable science on their side. They only have empty beliefs and re-cycled myths.

      Delete
    2. Thank you very much for your answer. Would you please recommend some AES papers (or other academically certified) regarding audio cables, specifically ABX testing of such cables?

      Thank you again!

      Delete
    3. @anon, Sure, check out http://www.aes.org/e-lib/browse.cfm?elib=5549 which includes references to several cable tests including the conclusion "special audio cables do not make a sufficient difference in the sound to be audible in double-blind tests". And for those who like to blame the blind test conditions, check out http://www.nousaine.com/pdfs/Wired%20Wisdom.pdf which used Audiophile's own systems and homes to conduct the testing. The score? Audiophiles: zero The cheap cables could not be distinguished from the expensive ones.

      Delete
    4. While I agree it is probably the case no one can really hear anything better than 16/44.1 provides, I would still like to see something higher in sampling rate at the very least become standard.

      It has taken years to overcome the problem of creating good reconstruction filters because people and mfgs are hellbent on having the passband extend to 20k.

      It's still not easy enough, 96k is overkill but would be helpful.

      As for 24-bit, I say why not? Hell, make it 32 to satisfy the software guys and DSP people. Storage and bandwidth are cheap.

      Delete
    5. The ODAC supports 24/44, 24/48, and 24/96. And the reconstruction filters, even at 16/44, have been demonstrated inaudible in many different published tests. But I agree some (especially old ones) do contribute potentially audible problems.

      In some ways 16/48 is a more elegant solution if only on paper. And the filesize/streaming bandwidth is barely higher. Had CD audio been designed several years later it's far more likely that's what we would have ended up with.

      24/48 means significantly bigger files and greater streaming bandwidth (which is especially an issue for wireless connections). As I've said, 24 bit hardware can have some audible advantages, but any advantage of 24 bit file formats is extremely slim to non-existent. Moving up to 24/88 or 24/96 means roughly triple the file size/bandwidth requirements of 16/44 or possibly even worse due to packing issues (32/64 bit word size).

      Unlike storage, wireless bandwidth (and the spectrum it requires) are finite and not nearly as scalable. There's a reason nearly all of it uses lossy compression at fairly modest data rates. As users are increasingly moving towards subscription, streaming, and cloud-based music, that's important. While that doesn't itself prohibit a higher-resolution audio standard it would likely help marginalize it to being a niche product much as happened with SACD.

      The last figures I saw at an AES presentation indicated hi-res music sales (including from sites like HDTracks) was something like 0.1% of total music sales and they've been at it for many years. There's been quite a bit of speculation about what Apple is going to do with 24 bit music with several hoping for lossless 24 bit.

      At least Apple (an 800 pound gorilla in music sales) is raising awareness of the relatively poor quality of a lot of downloaded music. But the quality problems have nothing to do with it being in 16/44 format. If they do start selling "premium" 24 bit music it may end up being an expensive "fix" for the wrong problem. But if it results in music with less/no lossy compression that's more carefully mastered and encoded, it will still be of value to some people.

      Delete
  56. First para: " I’ve run a second blind listening test and can report the O2+ODAC held its own against the $1600 Benchmark DAC1. "

    Do you have a video of this floating around on Youtube or somewhere else? Would be keen on seeing how you conducted the blind test. That one line sounds like typical sales talk but you're not making any money off this, etc etc.

    ReplyDelete
    Replies
    1. No, it wouldn't help silence the critics even if there was such a video. As has already been pointed out, they would just claim I (or whoever was wearing the headphones) was "sandbagging" and pretending not to hear any difference.

      What WILL help silence the critics is doing such a YouTube video with one of the critics wearing the headphones. But not one has come forward and volunteered to do so. Go find me some critics who are not camera shy and I'll do my best to arrange a credible properly supervised blind test for them to take.

      Delete
    2. Not going to happen.

      I've offered six-figure "bets" to audiophools, including some of the big-mouths at the magazines du jour.

      With odds, or where all they had to do if they could not pass a blind test was to publish the results, while I would put up serious coin if they could to be paid to them or the charity of their choice.

      Not a single taker.

      We know why. So does everyone else with the intellectual capacity and honesty required to see through their bizarre world.

      Rob

      Delete
  57. Epiphany in the UK sell the EHP-O2 amplifier, which they acknowledge on their web-page as designed by NwAvGuy. They also sell something called the E-DAC. Looking at the specs this would appear to be your design but, if so, I think it's rather unfortunate they chose to rename it. Your name is not mentioned on that web-page. What does this mean? Did they modify your design? Are they not using the exact components you specified? I would be much happier if it said something like "certified by NwAvGuy", "implements NwAvGuys reference design" or perhaps "NwAvGuy Inside". Can you confirm this is the ODAC?

    Secondly, I am interested in the EHP-O2D, which integrates the (E-)ODAC and O2 amp, for use with a laptop. If I understand correctly, this looses the battery and, while the DAC will run off USB power, the amp requires a power supply? I don't quite see the point of that. It would be more attractive, if the DAC/AMP combo retained a battery or if it could run entirely off USB. Perhaps by taking power from a second USB socket?

    ReplyDelete
    Replies
    1. Yes the E-DAC is the ODAC. I agree it's a bit confusing. I'll mention the issue to Epiphany.

      The point of integrating the ODAC and O2 is a lot of people use their O2 only for desktop use where AC power is available. They don't need the batteries and would love to transform their O2 in a high quality headphone DAC.

      The other point is USB power isn't sufficient for a headphone amp that can come even close to the O2 in terms of the headphones it can properly drive. At full power the O2 alone consumes about 5 watts of power and ODAC uses a bit under 1 watt for 6 watts total. A USB port can only provide 2.5 watts of power and you lose a good chunk of that as waste heat generating the voltages you need to run the DAC and a high output amplifier.

      So that's a long way of saying USB powered headphone DAC/amps are inherently limited in their output. Many struggle even with the HD600/650.

      Delete
    2. Sorry about the confusion - we rushed that page and so it wasn't very clear. The necessary ammendments will be made to fully reference NwAvGuy & the ODAC, as well as clear up a couple of other confusing bits.

      NwAvGuy seems to have answered your other queries fairly comprehensively :)

      Delete
  58. Please. Can you add to your post a pdf with the schematic and, if not the pcb layout, at least the mechanical dradwing of the situation of the parts and their values? Yes, you say 'no schematic', but this does not serve to construct a ODAC but to repair if damaged, many people are equipped to work with SMD parts.

    Also. Can you recommend one or two USB cables, good and cheap? You have done much work in designing the layout and reducing the jitter. This article discusses the importance of layout:

    http://www.eetimes.com/design/medical-design/4214331/Don-t-let-analog-digi-tal-misinterpretations-put-your-design-in-jeopardy?cid=NL_PowerManagement&Ecosystem=po-wer-management-design

    But it also includes a link to this application note:

    http://www.maxim-ic.com/app-notes/index.mvp/id/4345

    Where can see that electro magnetic noise can cause jitter. If external jitter enters can ruin your efforts, the cable must be sufficiently well designed to avoid that enter the electro magnetic noise. Maybe you can make some measures. If your measurement system serves to this, perhaps may need a measurement system for RF, I do not know.

    ReplyDelete
    Replies
    1. Yes, the PC layout was optimized for lower jitter. And your point is a valid one. The manufacturers and DIYers not conducting proper jitter measurements have no idea what they're ending up with even when they're using all the latest FOTM chips. The devil is in all the little details, not what chips you choose.

      We'll look into USB cables. The ODAC is still many weeks away from shipping to customers. I can try to buy samples and see if there are any measurable differences. So far I've just been using several out of my drawer full of USB cables but I have no idea where the came from.

      You won't find schematics published for most audio gear. Many manufacturers only provide them to their authorized service personnel who work under signed agreements. So if you want to hold the lack of a schematic against the ODAC, it's in really good company. I don't see Schiit Audio, Musical Fidelity, or Burson publishing their schematics either.

      We're under a non-disclosure agreement with Tenor and it's not clear to me if we're even allowed to legally publish the schematic as the design is based on Tenor documentation marked Confidential. If repair becomes an issue we'll address it at that time. It's really no different than buying most any other DAC.

      Delete
  59. monoprice.com is pretty amazing. I've bought RCA cables from them before.

    http://www.monoprice.com/products/product.asp?c_id=103&cp_id=10303&cs_id=1030302&p_id=5446&seq=1&format=2

    (I hope you will allow to post)

    ReplyDelete
  60. I wanted to again say "thank you" for your efforts in designing the ODAC. I can't wait to add that to my O2. I have both the ODAC and a pair of HD700's on preorder--they may actually arrive around the same time. Speaking of the HD700's, have you had a chance to listen to them yet? If so, what was your opinion of them as compared to HD650s and/or HD800s?

    ReplyDelete
  61. I see that the odac and O2 are designed mostly for headphone use.

    My question however is how usable are these advancements you have made for things like 5.1 audio. I really wonder how a 5.1 channel version of the odac connected to a standalone home cinema reciever, or even connected to a 5.1 O2 would compare.

    Currently i have connected the pc over hdmi to a reciever which does all the dac and amp work towards the speakers.

    ReplyDelete
  62. @Satellite, I've ordered from monoprice before and agree they seem to be one of the better sources. Thanks for the reminder and letting everyone know. I'll check out their USB Mini-B options.

    @Paul, I haven't heard the HD700. I'm curious as well. So far I've heard mixed reviews but I haven't done that much research. For me, the big question will be if they're worth more than twice the street price of the HD650, and how they compare to the similarly priced LCD-2 and a Stax/Amp combo.

    @Anon, you're correct about the O2. The ODAC can be used for any 2 channel USB DAC application. For a variety of reasons I think it's best to leave surround duties to an A/V receiver or an A/V preamp/processor. If you have reason to believe your receiver isn't doing a very good job, the best bet is to upgrade the receiver.

    One reason is if you use an external 5.1 DAC and run that into the 5.1 analog input on your receiver, many receivers re-digitize that analog signal--especially if you want to have the subwoofer crossover and many other functions be available. Some receivers have a true "direct" mode, but on the newer ones I've tested the signal is still re-digitized. It's easy to tell. Just use a splitter and feed the output of an iPod (or similar) into your receiver while you're listening with headphones directly connected to the iPod at the same time. If you hear an "echo" or delay from the speakers compared to the headphones that means the signal is being re-digitized and going through the receivers DAC regardless.

    ReplyDelete
  63. Great news!!! I really liked the O2 and I'm looking for a new DAC, the price and measurements make a great combination and since the ODAC has the same transparency aimed approach as the O2 I'm expecting a superb combination. Must free up my American Express ASAP... LOL!!!

    ReplyDelete
  64. Nice work there. I wish a success for it, although I'm not happy with closed design. Is there something for us - users to do about it? E.g. writing Tenor and ESS to release their datasheets publicly?

    ReplyDelete
    Replies
    1. Unfortunately, it's the entire distribution model for Tenor so I wouldn't expect them to change it based on some DIYers. Their methods are not that uncommon. Lots of semiconductor companies (or subdivisisions within them) only want to deal with reasonably sized OEMs. And lots of them want to protect their intellectual property so their lawyers make them implement Non Disclosure Agreements before releasing proprietary information.

      Some companies are making a bigger effort to appeal to DIYers. But it's mainly to lure college students into becoming familiar with their offerings, tools, etc. much in the same way Microsoft offers cheap versions of Visual Studio to students. It's usually for more general purpose products like microcontrollers, WiFi hardware, op amps, power controllers, etc. Those products have wide applicability across lots of different industries so the "seeding" among students pays off.

      But the more specialized the product, the more likely they're not generally available in small quantities through the usual distributors. And the more likely you need to sign an NDA to even get all the data. USB audio interfaces are fairly specialized. I would guess Tenor probably has fewer than 100 customers using the TE7022. And DIYers just aren't worth the trouble for the comparatively small quantities.

      Delete
  65. Hi great work! I am a bit concerned with the synchronous USB port.
    It is generally known that synchronous USB could result in serious jitter problem. Is the Benchmark DAC1 pre using asynchronous USB?

    ReplyDelete
    Replies
    1. Actually, both the ODAC and DAC1 use the Adaptive USB Audio interface. Synchronous can indeed perform poorly but it's not applicable here. And the jitter results published in this article speak for themselves. The ODAC's jitter is well below the threshold of audibility. Once something is genuinely inaudible, it's hard to improve on that. That makes Asynchronous USB a waste of money in this case.

      Delete
  66. Is there any reason why someone couldn't take the board out of a standalone odac and put it in the o2?

    Also, anyone experience any noise with the pot on the o2? I'm getting some. Prolly the pots fault. It comes and goes away with some movement. Any suggestions on what I can do, apart from replacing it?

    ReplyDelete
  67. My EE and audio knowledge is limited, so I may be overlooking the obvious, but why do your dynamic range (THD+N?) plots look so different from the ones on Stereophile? Yours seems to go "down" with higher frequencies, while on Stereophile it rises signifficantly.

    Is there any way to compare them to each other?

    ReplyDelete
  68. can you give us an update on how the ODA design is going? I assume you are overloaded with work, life and writing for the blog...but hey, you teased us first :-)

    thank you.

    ReplyDelete
  69. 3 answers:

    So far there's only one kind of ODAC board. So yes, you can move the board from a standalone ODAC to an O2 if you want (some soldering required). If you're hearing noise in just one channel when you move the pot it's very likely either the pot's solder connections or the pot itself. If you hear noise in both channels, make sure it's not a jack problem, ground problem, etc. If your board has the Alps pot, it's the same pot used in the $1600 Benchmark DAC1 Pre and lots of other high-end gear but it's still possible it could become noisy. If you don't have desoldering equipment, the pot can be difficult to remove.

    Stereophile does -90 dBFS 1/3 octave swept noise graphs which show much less detail (note how "wide" the 1 Khz tone is) and essentially are averaging some of the noise together and create the upward slop. It's like putting a 1/3 octave filter in front of an AC volt meter and "sweeping" the frequency of the filter. I show high resolution FFTs which reveal much more detail. The rise to the left in the ODAC graph is a characteristic of the ESS DAC. And the noise plot in this article is at -60 dBFS which is the industry standard for measuring dynamic range. I also often do -90 dBFS plots as well for absolute noise. Interestingly, with DACs, their dynamic range is often significantly different when measured at -60 vs -90 dBFS. The idea behind -60 dBFS is it's loud enough to still be audible yet low enough for any distortion products from the 1 Khz tone to alter the measurement.

    The ODA is moving ahead a bit at a time here and there. I'll publish an update once I have enough new info to share.

    ReplyDelete
    Replies
    1. Hi NwAvGuy,

      Have you definitively decided on the features to be included in the ODA (eagerly awaiting it)?
      In late November last year you listed a series of features that responded to what the majority of comments suggested; are you considering all of them, or only the most essential?

      I think your design of the O2 and the ODAC have fulfilled the expectations that motivated them; you have clearly exposed the hifi industry, the myths they have spread and thrived on, and certainly taught them a valuable lesson. In this sense, I believe that keeping the internal components at the necessary and sufficient performance level vs overkill is economically healthy, but I also believe that planned obsolescence is an important industry element that has to be addressed.

      I remember you pointed out that if we wanted to hand our headamp to the next generation, we should get a Violectric. I nevertheless believe that the ODA + ODAC could be yet another milestone and achievement, if it matched or beat the average features and build quality of the best HeadAmp/(PreAmp?)/DACs available, and still managed to keep costs well below the competence.

      I hope I'm not suggesting anything that you already ruled out with the impeccable argumentation that so well characterizes you, as I have still much reading ground to cover on your blog.

      Cheers!

      Delete
    2. I've tried to share the basic outline of the ODA and, so far at least, I'm sticking to that outline. But product design can be fluid. Sometimes people suggest things I haven't thought of. Sometimes it's not until you build a prototype that one discovers new ways to improve it.

      One thing the ODA offers that Violectric doesn't is the ability to customize the end result. If a DIYer wants an ODA in a vault-like enclosure with massive aluminum panels, they're free to make that happen. If someone wants to use ultra high end $50 headphone jacks that might outlive their children, they can. I'm just hoping to provide a solid audio design and PC board layout. What others do with it, is up to them.

      Delete
  70. Can averaging really account for that much difference? Those look almost like different measurements. I took some liberties with your charts, and if you compare the ODAC to the DACmini, the ODAC outperforms the DACmini by a rather large margin after 4kHz: http://i44.tinypic.com/300bgqb.jpg

    Granted, the DACmini didn't do very well otherwise either (despite being beyond the barrier of transparency) - and its price is probably more the result of the bling-factor rather then true objective audio quality - but also Asus' Xonar STX, that should have had a somewhat similar performance, judging by the numbers, seems beaten by quite a margin at higher frequencies - when looking at the charts: http://i44.tinypic.com/fymkxw.jpg

    Even the V-DAC (that has supposedly >120dB SNR / 19 bits ENOB and the noise floor at -150dB; despite its other flaws) is vastly outperformed by the 111.1 dB SNR ODAC in the higher frequencies - according to the charts - and has simmilar noise/distortion for the rest of the spectrum: http://i40.tinypic.com/35cfu43.jpg

    If these indeed are different measurements, what other approach can be used to compare the results?

    ReplyDelete
    Replies
    1. I would suggest using the overall number--i.e. 111.1 dB SNR--for comparison purposes and consider FFT spectrums to be far more detailed than the swept 1/3 octave result John Atkinson publishes. It might be Stereophile is trying to make some products, like tube gear, look better by using "smoothed" graphs.

      It could also be he's using the old school method for historical reasons to be consistent with how Stereophile measured things decades ago. While they have a relatively new Audio Precision analyzer on loan from AP, I think the one they actually own is more of an ancient boat anchor. Their older analyzer might not be able to conveniently do 256,000 point high resolution FFTs.

      Regardless of why they do it the old the way, you should be able to compare the summed result if they provide it. I put the readings right on the graphs, but Stereophile typically mentions them in the text. In both cases, the noise and dynamic range is the sum of the non-signal energy from 20 hz to 22 Khz. I also provide absolute (rather than relative) noise measurements. For the ODAC that's -102.8 dBu A-Weighted.

      As a further point of interest, I try on many graphs to include enough readings to make it easy to reproduce my result. The extra information also makes it difficult to "cheat" without resorting to Photoshop or similar. In contrast, a lot of other published results lack the test conditions. And particularly results given in dBr (relative units) could be made to look much better by changing the unknown reference. And I also typically include additional information in the graph caption. The whole point is to make it easy for others to reproduce my measurements.

      I should someday do an article comparing the various measurement techniques by different reviewers including J.A. at Stereophile. I've seen far worse measurements than his. Sound & Vision, for example, seems to get their THD+N vs output wrong in nearly nearly every review. And Noel Keywood at Hi-Fi World also often does things in a rather misleading and/or non-standard way.

      Delete
  71. Hi Nwavguy, I just purchased a set of nearfield speakers. Does balanced output make a difference more so with speakers than with headphones?

    Thanks

    ReplyDelete
  72. Pardon my ignorance, but why couldn't you just plug headphones into this and control volume with the operating system? I thought I read in an earlier blog post that was part of the reason for 24-bit over 16-bit.

    ReplyDelete
  73. To answer the three questions above, the ODAC needs a headphone amp, or other amplifier, receiver, music system, powered speakers, etc. It's has a line-level output and can only drive loads of 5000 ohms or higher (most line inputs are 10,000 to 100,000 ohms). Headphones, however, are 16 - 600 ohms. The ODAC will be very unhappy trying to drive headphones directly. The same is true for most any line output.

    Balanced outputs/inputs are only useful if you plan to run very long cables or have an usually electrically noisy environment. Balanced audio isn't about better sound, it's about better noise rejection. Besides external noise, balanced connections typically perform worse than unbalanced connections.

    Assuming your connections are not more than a few meters/yards, you can just use an unbalanced to balanced cable or adapter with the ODAC.

    ReplyDelete
    Replies
    1. Hi NwAvGuy!

      First of all - thank you so much for your effort in designing the O2 and the ODAC! I'm building my own O2 atm an will surely buy the ODAC when it's released.

      Anyways, the main benefit I've had from balanced interconnects in PA and Studio use is its rejection of ground loop noise (assuming the equipment is correctly grounded). Wouldn't equipment for home use benefit from that as well?

      Thanks

      Delete
    2. I addressed potential ground loops in answering comments about isolated USB. Generally with headphone listening ground loops are not a problem. And even with home A/V gear they're still usually not a problem. But you're right they can be under some circumstances and balanced interconnects are one potential fix. But there are other ways including USB isolation which should break any ODAC ground loops.

      Delete
  74. Hi Nwavguy,

    Nice design! While I can understand not sharing the schematics due to the TE7022 chip, what about sharing what your learnt from your prototype testing?

    Looking at the board picture, I don't see any special techniques applied. Comments:
    1) Good - proper use and placement of MLCC decoupling caps.
    2) Good - top layer fill with ground net and via stiching to bottom layer ground plane.
    3) Good - cutout separating digital return from analog return.
    4) Weak - lack of 85 ohm controlled impedance layout for USB traces.
    5) Weak - analog stage has lots of room yet L/R traces are very close. Lineout conn does not use GND pin between L/R signals.

    It'll be interesting to know what good EE best practices might have missed.

    You did mention oversizing the digital caps isn't always a good thing. Good lesson for the "buy-a-kit-roll-the-parts" DIY types. It might be interesting to compare against the Peachtree DAC.iT. It uses the same chips, but has a separate DC power feed.

    Thanks,
    Arius from diyaudio

    ReplyDelete
  75. I bought a Maverick Audio D2/A1 combo for around $520 *sobs*. Is that what you call an eBay boutique audio company?

    Other than the lack of digital feedback (hissing/popping) I can't hear a damned difference between the combo (with the fricking upgrade tubes) and the headphone out of my PC.

    Would you say that your ODAC/ODA would be audibly superior?

    ReplyDelete
  76. @Anon, thanks for the feedback. I will be covering more of that in the next ODAC article. The TE7022 is only 12 Mbit so the very short traces from the USB jack to the chip are a non-issue. With a 480 Mbit interface it would be much more important. The cross talk of the ODAC is better than 90 dB so the L/R traces are not a problem.

    The ground is on the "right" to provide a direct path to the "star ground". if we had put it in the middle of the connector the ground would have been cut off from the "star" creating a longer and less desirable ground path adding noise to the output. Having the ground in the middle is exactly the sort of thing you see all the time as it intuitively makes sense on the surface. But, in reality, in this case it would result in lower performance (unless it was a 4 layer board). All the analog ground return currents are on the back layer and below the line out header on both layers. So it matters how you ground the header.

    It would be interesting to test the Peachtree. I know their tube gear hasn't measured very well. But I haven't seen any good measurements for their non-tube gear. The Calyx Coffee, as previously mentioned, is another commercial version that appears similar to the ODAC

    @Tony, I'm not familiar with the Maverick so I really can't say if the ODAC would sound better. I can say the ODAC very likely it will at least measure better than anything with vacuum tubes in the signal path. DACs with tubes, even from well regarded companies such as Peachtree, tend to measure rather poorly and many fail the criteria for audible transparency. But that doesn't mean they will necessarily sound bad depending on how they're designed and how close they are to being transparent. Some, like a Little Dot tube amp I briefly tested, are train wrecks and some are borderline respectable.

    ReplyDelete
    Replies
    1. This is grand news, I only wish you had a paypal button so that admirers could donate to either show appreciation or help with R&D

      Delete
  77. I already answered the same question elsewhere. You're correct about 24/88 but the other "benefits" of the TE8802 would have required proprietary drivers and that would have opened an unnecessary can of worms, created confusion, and more.

    When this project was started we were not aware of the TE8802 being used in any shipping products. So it was a total unknown (and in terms of measurements still is AFAIK). Even now I'm only aware of a couple DACs using the TE8802 and they're Asian designed from companies that are known for snake oil and substantially more expensive than the ODAC. The TE8802 is a more expensive chip.

    But, above all, the ODAC is about what's required for audible transparency and the TE8802 isn't required. We could have used all sorts of more expensive parts, including the TE8802, but if they don't make it sound any better, that defeats one of the ODAC's primary design goals.

    ReplyDelete
  78. Should I take it from autumn release projection that the differences between the ODA and O2 are more significant than previously thought?

    ReplyDelete
    Replies
    1. Possibly. But mainly the ODAC needed to come first to make sure the ODA could accommodate it. And it was just released about a week ago. There are also some blog/testing/etc. projects that will be going on in parallel (ideally while waiting for ODA PC boards, etc).

      Delete
  79. Funny enougth. Looks almost like a uDAC2 clone ) with local MCk generator (same as in SE version on uDAC2)
    Without headphone amp thougth..

    ReplyDelete
    Replies
    1. Yeah with several important differences. First the uDAC uses the older ES9022 chip. And, as I keep saying, implementation is everything. And NuForce implemented the ESS chip wrong and the DAC clips badly at 0 dBFS. It also doesn't have the real world dynamic range and it has significantly higher distortion--especially at high frequencies--even when it's not clipping. And when you combine the ODAC with the O2 or ODA, the headphone amp is vastly better than the highly compromised one in the uDAC-2 in many regards (much higher output power, lower impedance, much better volume tracking, adjustable gain, etc.). There are other benefits as well--especially with the ODA.

      But yeah, given there are so few choices for 24 bit USB interfaces, the TE7022 is an obvious choice for many designs. And for a USB powered DAC like the uDAC and ODAC, the ESS chips are also an obvious choice. So, ultimately, we're all picking from the same very limited selection of chips. The difference is in how they're implemented and refining the entire design.

      Delete
    2. So in other words Nuforce has been pwned by one guy in his basement because that guy actually happens to know what he's doing and has the equipment to verify his results.

      Go NwAvGuy!

      I've got my ODAC preordered from JDS and can't wait. Hopefully I'll get it before a meet coming up in the middle of June.

      Delete
  80. On that note, has there been any measurements of the TE7022, with regards to its intrinsic jitter?

    ReplyDelete
  81. I messed around with trying to convert the TE7022's I2S output to S/PDIF to allow it to be measured in the digital domain by the dSCope. Prism sells what they call a VSIO adapter that allows such measurements with high accuracy but it's an expensive piece of hardware. It's something I'll consider purchasing if I can justify for future DAC development. You can read more here: http://www.prismsound.com/test_measure/support_subs/resource.php?rid=54

    Without the VSIO Adapter you're at the mercy of whatever hardware is doing the conversion and the connections, grounding, etc. involved. So you're not measuring just the intrinsic jitter of the USB interface, but also the jitter of your improvised adapter.

    Because the ESS9023 has significant jitter reduction built-in, what really matters is the final jitter performance of the entire DAC from PC to analog. And I've shown that result above in this article. It's hard for anyone to credibly argue jitter components that are under -110 dB and total jitter of -103 dB is somehow audible.

    ReplyDelete
  82. Let me ask you (yet again, if you don't mind =P) an unrelated question: what are your thoughts about bi-wiring? Is it pure snake oil, or is there some truth to it? And what about bi-amping, does it make sense?

    Thanks a lot!

    Oh, and feel free to delete this when you need more space (200 comments).

    ReplyDelete
    Replies
    1. Unlike a lot of audiophile myths and snake oil, bi-wiring and bi-amping both have credible reasons they can make a difference. Speaker cables add to the output impedance of the amplifier. That means the signal at each end will tend to be at least measurably different. If it's different enough to be audible depends on the length and resistance of the speaker wire and what sort of load the speakers present.

      A speaker that might dip to 2 ohms at some point, that's at the opposite end of a big room from the amp, might sound audibly different than using just a few feet of cable. With the long cables under those circumstances, bi-wiring could make an audible difference as it not only cuts the resistance in half, but it decouples the relatively high current bass losses from the rest of the audio spectrum.

      But with better behaved 8 ohm speakers, and a few meters or less of heavy gauge cable, it's unlikely you'll hear any difference with bi-wiring.

      Bi-amping would make a similar difference with lousy amps that have a relatively high output impedance (poor damping factor) such as many tube amps. But with a properly designed amp, even a cheap one like the Behringer A500, it's not likely to make any audible difference as the output impedance is so low to begin with.

      Bi-amping does allow using a far more advanced and precise crossover--either using analog op amps or digital DSP. And that can have very audible advantages. For one thing the amps can be directly connected to the woofers allowing increased damping (no passive inductor in series with the woofer). You also eliminate all the losses in the passive crossover. But, above all, you can have more precise control over the crossover points, slopes, and phase because you're not trying to compensate for the non-linear reactive impedance of the drivers themselves (which can only, at best, approximate with a passive crossover).

      Delete
    2. @Anon, if that headphones output has too high of an impedance, or not enough power, to properly drive headphones there can be big improvements in using the O2 from the DAP's headphone jack. But if the player already has a suitably low impedance (< 2 ohms) and enough power for your headphones, like say the Sansa Clip+/Zip, the O2 isn't going to help any.

      Delete
  83. NwAvGuy, your blog is a very interesting read and I'll be buying an ODAC, when it's available. Also I want to ask: would you review/compare those portable hi-fi digital players, like the Colorfly C4 and HiFiMAN HM-801, to see if snake oil has crept into the portable hi-fi scene as well?

    ReplyDelete
    Replies
    1. I can't speak for Colorfly but HiFiMAN has repeatedly been caught with their pants around their ankles. The HM-602 portable player was claimed to be 24 bit but wasn't, had audibly flawed frequency response, did very poorly on high frequency distortion tests, and had the worst jitter I've ever seen when connected via USB. See:

      http://www.stereophile.com/content/head-direct-hifiman-hm-602-digital-audio-player-measurements

      In short it's a train wreck and an iPod completely shames the HiFiMAN and is in a vastly nicer, more elegant, and even cheaper package. HiFiMAN players seem to mostly be a massive Head-Fi infomercial phenomena.

      Delete
  84. The Peachtree DACit is reviewed in the most recent Stereophile, and more importantly, measured. It also uses the TE7022L/ES9023, but they weren't able to get the same performance as you were with the ODAC. Like the NuForce, it has problems with 0db. It might be interesting for you to take a look at the measurements section of the review. It isn't available online yet. When the DACiT was measured with a 50hz sine wave at 0dBFS with a 600ohm load, it fully clipped, and into a 100Kohm load, it had distortion products at -73db L and -70db R. When he lowered the signal level to -10dBFS, it decreased the distortion to -94db and -96db.

    Interestingly, Atkinson goes on to question whether the 9 volt power supply was inefficient and that cause the measurement problems. He should see what you were able to achieve with USB bus power.

    Thanks again for all of your hard work. I'm very much looking forward to listening to the ODAC, the ODA, and whatever you set your sights on designing after that!

    -Aaron.

    ReplyDelete
    Replies
    1. Thanks Aaron. I haven't seen that review yet. It's likely that Peachtree got it wrong as it's a delicate balance with the ESS DAC's internal operating point. The nine volt power supply comment by J.A. is rather misplaced as the maximum supply voltage for the ES9023 is only 3.6 volts.

      The ODAC does show slightly higher (but still transparent) distortion at 0 dBFS but that's true of a lot of DACs. It might vary slightly from sample to sample due to tolerances within the chip, power supply regulator, and the 1% resistor that determines the reference operating point of the DAC. We specified a tighter tolerance regulator specifically for that reason. This seems to be more of a potential problem with the ES9023 (and apparently ES9022) than with some other DACs. But, in short, our implementation does not have audible levels of distortion at 0 dBFS into 100K as Stereophile measured.

      I'm guessing Peachtree got sloppy or was otherwise unaware of the issue. If you read the Stereophile review of the Peachtree iDecco you'll see it also has some rather severe problems which J.A. mostly attributed to the tube. It would seem Peachtree and NuForce might both have similar "design by ear and don't worry too much about the measurements" philosophies?

      http://www.stereophile.com/content/peachtree-idecco-da-integrated-amplifier-measurements

      Delete
  85. why is it so common for companies to produce under performing electronic products -specially affordable audio-? I like to think that the engineers in charge of the design stage are educated enough to produce, while enjoying the challenge, good performing designs and products. Is it maybe the need for cost cutting the production that really causes this large amount of under performing products? is it the tolerance of the consumer to crappy products? is it maybe because of the intrusion of non qualified people in the audio world?

    either way, I think you are doing a good service on educating people, keep it up!!

    ReplyDelete
    Replies
    1. In my experience it's usually one or more of the following:

      Private Label - There's a big section of CES with small Asian booths seeking US companies to distribute their products. It's like walking through the dogs at the Humane Society with the products screaming "adopt me". The online version is Alibaba. Basically UberAudio in California finds an existing DAC made by Cheng Moon Super Good Sounds (CMSGS) in Shenzhen China. CMSGS will make it look like the rest of UberAudio's products and deliver them ready to ship in nice box with a big gold embossed UberAudio logo. UberAudio had nothing to do with the design of the circuit board inside and might not even bother to fully test it--or the sample they test may not be representative of the product they'll ultimately be shipping. It's basically a no-name product designed by people often trying to make quick buck any way they can. And they either dupe UberAudio into believing it's high quality or UberAudio doesn't really care. You would be shocked how often this happens and some of the big names in audio who do it.

      Design By Ear - A lot of high end gear is designed mainly by ear complete with all the unavoidable problems outlined in my What We Hear article. The company either lacks the equipment and/or knowledge to conduct proper measurements and/or they don't really believe in them. NuForce has admitted to this approach on multiple occasions, including in a letter to Stereophile, as have other companies.

      Contract it Out - A lot of companies are mainly just sales and marketing organizations. They don't even have engineers. They contract out their design work to consultants, design houses, or ODMs--Original Design Manufactures. And that gets expensive--especially to do multiple iterations. So they often have a lot of incentive to go with the first prototype that seems to play music and not catch on fire. Someone might be saying "but it clips pretty bad at 0 dBFS", and management responds with "it sounds fine to me and they want $8000 to do another prototype, so just ship it!".

      Business Management - Even in companies that have their own engineers they can only do what they're told. Just look at all the broken software that gets shipped and is virtually unusable until 6 months worth of updates and fixes have been applied. It's the same sorts of questionable management decisions that result in shipping marginal hardware. But, unlike software, you can't download free updates.

      Marketing Knows Best - Marketing decides they want a tube sticking out of the top of an otherwise perfectly nice DAC. And to avoid being labeled poseurs, the tube has to unfortunately be part of the audio circuit complete with all its non-linear distortion, etc. Even the best engineer can only make a Ford Model T go so fast.

      Maximize Profit - The best examples of this are on eBay where mostly Asian vendors are selling audio gear using trendy popular chips but some of it is very marginal in one or more ways. They don't care because they know it will cost too much to bother returning. It's often sold by semi-anonymous eBay vendors with multiple identities. When one gets too much negative feedback they just adopt a new identity. The magic of Expectation Bias works very much in their favor and most of their customers probably never even realize how half baked some of the gear is. They bought a DAC with a fancy Wolfson chip someone recommended on Head-Fi and expect it to sound great so that's what they hear.

      Dilbert Effect - When the big companies get it wrong it's probably because people hate their job, the last merger didn't go well, and/or the audio analyzer broke last year but there's still no money in the budget to get a new one.

      All of the above typically have one thing in common: Zero or minimal/vague useful measurements

      Delete
  86. What happens after O2, ODAC and ODA? I mean, where do you go once you've achieved true audible transparency of those components?

    Does this mean I'll only have to listen to music now?

    ReplyDelete
    Replies
    1. I've got seemingly years worth of article ideas, product reviews, blind tests, etc on the list already. But yeah, once I find gear I know is letting me listen to the music, not the gear, I trust it and keep it around. Once something represents a visually spotless window into the music, you can't get the glass any cleaner.

      Delete
    2. This is a good question for all of us I think. I am really tired of looking the gears I wanna buy someday for years now, or reading the well known audiophile sites for the best snake oil treatment. These articles and gear here can help us all to decide if we want to stop looking and just focus on music listening with a glass of fine wine.
      I hope in a short time I can lean back and contentedly open a bottle.

      Delete
  87. Thanks for all your work nwavguy! I've been lurking your blog lately, and I have learned a lot. I am going to order an O2 as soon as a get some cash, and I am really looking forwards to getting an odac in the near future too!

    ReplyDelete
  88. I'm curious if you've had a chance to test the Emotiva XDA-1 DAC. They claim to be very involved in the design and implementation of the system.
    http://shop.emotiva.com/collections/processors/products/xda1

    If you're interested and others are interested, I'd be happy to send my extra one up to you (and/or find someone who owns it locally).

    I've really enjoyed reading all these posts, it's refreshing to see real tests and numbers. I'm very much looking forward to getting the ODAC for home and one for the office.

    Thanks,
    Adam

    p.s. Reading this reminded me of this article (http://www.stereophile.com/content/carver-challenge)

    ReplyDelete
  89. Sorry if this has been asked, but when the ODA(+ODAC) comes out, will it be possible to build it with a (variable RCA) line out?

    ReplyDelete
  90. Subjectively speaking, does the ODAC provide any increase or changes to the sound quality compared to the iPod Touch 3G? Both devices feeding analog via line out to the O2.

    ReplyDelete
  91. @Anon, you're welcome

    @Greenwiki, Rumor has it the XDA-1 will be replaced soon so I haven't wanted to test one as I'd probably get the review done and they'd no longer be available. It only supports 16 bits over USB but it otherwise is probably a bargain. Emotiva generally does a good job.

    @Mkubota1, Yes it will have variable line out.

    @Sonus, I haven't done a blind test as it's a bit more complicated to do as an ABX (you have to account for the inevitable time sync differences when playing from two entirely different sources). My plan was to go even further and compare the iTouch 3G + O2 to a PC + Benchmark DAC1. My guess is they'll so sound so similar they'll be difficult or impossible to tell apart. The iPod's biggest flaw is the headphone amp and the O2, obviously, solves that. So the pair makes for a hard to beat combination.

    ReplyDelete
    Replies
    1. Yeah... it will be interesting to see what the follow on product from Emotiva is. I guess my wondering was whether they'd actually done their implementation correctly and whether that had any reflection on their next implementations. But I guess with everything, you might get one implementation right and then mess something up the next time around.

      Their USB support could have been a bit better, I'm guessing the next one will be 24/96.

      I've definitely been enjoying mine.

      Delete
  92. Thanks for the honesty NwAv. There's very little people like you in business. Is this just a marketing trick used by ALO Audio to boost the sales of the "iMod" or do iPods really don't have true line outs, but rather just a volume locked "glorified" headphone out?

    Here's what I'm talking about.

    "But the fact is, most if not all portable player on the market doesn't really have a real line-out. Instead, they deploy a pseudo line-out by re-routing an signal from the amplifying circuit in maximum volume, mimicking the stronger, higher output power of a real line-out. For those who don't know about it, one of those player is your regular iPod. This is why a few years back, a group of people start to modify iPod and sell it as 'iMod', an iPod that has its real line-out signal re-routed to its port so the user can use it with an external amplifier to avoid the what they believe to be an unacceptable internal amplification circuit."

    ReplyDelete
  93. I was wondering if this works with Android. I've been going back to my CD collection, ripping at higher quality and would like to have a solid, portable head phone amp.
    thx!

    ReplyDelete
  94. @Sonus, at least with the iPod Touch 3G the LOD appears to be taken upstream from the headphone amp. For one thing, contrary to ALO's claim, the LOD has LOWER output than the headphone jack. It also measures a bit better. I can't speak for the other flavors of iPods, but it's essentially free for Apple to tap off the line out upstream in the circuit so I don't know why they wouldn't.

    As for "unacceptable internal amplification circuit" the iPod's headphone out and line out flat blow away the outputs of much more highly regarded gear including HiFiMan Players, many DACs, etc. ALO seems to sell a lot of snake oil so I'm not surprised they're spreading F.U.D. about the stock iPod.

    @ShelbyDZ, you don't want an ODAC, you want an O2 (headphone amp). The ODAC only works with PCs.

    ReplyDelete
  95. http://lifehacker.com/5903575/unleash-your-headphones-full-potential-with-a-usb-dac-and-amplifier

    So bad. The replies, even worse. *sigh*

    ReplyDelete
    Replies
    1. Perhaps some who are more objectively educated should add some comments there? There are much worse products to recommend than the E10 but it's interesting it's such a FiiO-centric article.

      Delete
    2. Well,

      He does mention why it's FiiO-centric.. It's what he had to review (they probably provided samples). However, he doesn't once mention impedance as a factor with amps. He simply claims they provide more "power" and "volume". Also, he claims all dacs sound radically different, which is BS.. It's just a shame to see all these people going along with it.

      I thought I had to be a member of the site to comment.. I will try, and update on whether I'm censored out or not. Seems some links to you have not been blocked so, I'm hopeful.

      The discussion about WASAPI on the comments is laughable at best.

      Delete
  96. Any chance of an early reveal of which Ha Info DAC you've got in the review pipeline? I was thinking of buying one as I live & work in their backyard. I'm assuming you've got one of their designs that uses the TE7022 for obviously relevant comparison.

    ReplyDelete
    Replies
    1. It's an NG98-II which has a loyal following on Head-Fi and elsewhere but doesn't use the TE7022. The AC adapter it came with also violates all international and US safety laws despite being labeled as compliant. Based on how badly they screwed up the DAC design in a really obvious way, I personally wouldn't trust anything they sell. It's not just a case of a subtle PCB layout problem or using some substandard parts, it's fundamentally designed wrong.

      Delete
    2. Wow. And I remember Audio-GD with their ground-swapped-with-zero power cords.
      Can't wait to see its performance!

      Delete
    3. Thanks! Will be waiting for the smackdown. All of their newer designs appear to make use of the TE7022 and other perfectly respectable parts at very attractive prices...thus my interest. China is certainly the new epicenter of boutique audio, warts, snake oil and all. Really, it's just so easy here to DIY up your design with attractive-looking results and then sell it. Wish the entry barriers were as low in the good ol' USA so we'd see an unleashing of more projects like the O2 and ODAC. I've been schlepping the odd piece of equipment back home during breaks in work...this year I'm thinking it's going to be the reverse! :)

      Delete
    4. I was kind of hoping someone in China might offer the O2 at a more attractive price while keeping the PCB layout, using the correct components, etc. They could put in their own enclosure if they wanted. But I haven't heard of that happening yet.

      Delete
  97. Now that I've preordered an 02+ODAC, I'm looking for a transparent power amplifier to hook up some speakers. This could turn into the ultimate price-performance ratio hi-fi system!

    Do you have any experiences/recommendations/reviews of power amplifiers or speakers?

    ReplyDelete
  98. About blind testing, how do you solve the problem that you expect to NOT hear any differences before you've even heard the sources? Don't you need two separate people with opposing expectations to verify the result (one person who DOES expect to hear a difference, and one person who does not)?

    ReplyDelete
    Replies
    1. Ideally you're correct IF THE LISTENERS KNOW THE GEAR BEING COMPARED. When they know what they're comparing it's best to have blind listeners that cover the extremes of expecting differences and not expecting them. In reality, however, it's like showing someone a red paint chip and a reddish-pink paint chip and asking them if they're the same or different. Even if they're not expecting a color difference they generally will still realize when there is a difference.

      But the real magic of proper blind testing is when the listener really doesn't know what to expect. For all they know you're playing the same source twice or one is a $10 MP3 player and other is a $1600 Benchmark DAC1. If they don't know what either of the sources are, it's impossible for them to expect a difference or not.

      For many of the Audio Engineering Society blind tests the above is exactly the situation. The listeners didn't know anything about "A" or "B". They were just tested to see if they could detect any reliable difference between two anonymous sources.

      Delete
  99. Nwavguy,

    I can buy the standalone ODAC from jds labs and add a line out jack to it right?

    That way I can use the odac to the o2 or the o2 as a line out. Correct?

    ReplyDelete
  100. EVERYONE PLEASE NOTE: We keep bumping into the 200 count limit here and it's getting harder to decide what to delete to make room. I should have the new article up soon (it's taking much longer than I hoped) providing a fresh canvas for new comments. I'll keep looking into other solutions to the problem as well. In the meantime, if your comments are not showing up, or older comments have been removed, that's why.

    The point I was trying to make about the TE8802 is I'm not aware of any known reputable DACs using it that are available today let alone back when we started this project last year. I'm not saying it's a bad chip and only suitable for snake oil DACs.

    The RCA output issue is an interesting point. I'll pass that along. The ODA+ODAC, however, will have RCA preamp outputs. If you want to replace your desktop amp, the ODA+ODAC should solve your problem. And it should be available fully assembled with the ODAC and RCA outputs already installed.

    For the record, 3.5mm connections work surprisingly well for line outputs. There are lots of high quality 3.5mm to RCA cables available. Because of the impedances involved they work far better for line outs than for headphones.

    ReplyDelete
  101. Congrarulations on the release of the ODAC! It should go very far towards proving that great performance can come cheap! However, I am even more intrigued by the more expensive version you plan to release. What areas does it aim to improve on? What kind of measurements is your target for the more expensive build. Are we looking at an attempt to match truly balls-to-the-wall designs such as the Weiss dac202, the DAC1, and dcs stuff, without the complexity of such designs?

    ReplyDelete
    Replies
    1. A higher end version really depends on how the ODAC does once it's been out a while. If the ODAC proves transparent in more blind tests against much more expensive DACs, and there's not massive demand for S/PDIF, a better DAC doesn't make a lot of sense--at least if it has to be built in similarly large quantities.

      If a new USB interface chip is released that's genuinely DIY-friendly, however, that would publishing the design with no required to make a huge batch of them. I'll be explaining in more detail in the next ODAC article why there are no compelling DIY-friendly USB interface solutions.

      But no, as I explained earlier in a comment reply, it would be a man-year or more of work to match some of the best DACs out there. I have no interest investing thousands of hours in something that likely won't sound any better than the ODAC. If you really want that level of performance for some reason, buy an existing commercial product that has the measurements to documents its performance.

      Delete
  102. Hi NwAvGuy,
    For the 200 comment issue, have you tried this:
    http://www.ravisaive.in/2012/02/fix-200-comments-pagination-problem-in.html

    ReplyDelete
  103. While I understand the reasons behind not releasing the schematic etc., is there a reason for not releasing a front plate design file or enclosure options for standalone ODAC. I am okay with $100 for the ODAC board but I would like to run it standalone and for that JDSLabs enclosure seems somewhat steeply priced at $51. It would be nice to see people coming up with some custom enclosures and faceplates just like they did for O2.

    ReplyDelete
    Replies
    1. Good question. I haven't designed a panel to fit a standard enclosure and I'm not sure YoYoDyne has either. We did include the jack locations on the drawing in this article, and I know JDS supposedly plans to sell ODAC panels for $12 which is less than you would pay getting your own from Front Panel Express. There's also someone who's talked about offering one or more laser cut acrylic panels.

      If someone comes up with one or more proven "open source" panel designs I'll be happy to host the files/drawings at Google Docs and link to them from the next ODAC article.

      One problem is the ODAC was designed to fit in the O2 battery space which makes it not a "slip in size" for say the Box Enclosures cases like the O2 uses. So if you want an off-the-shelf enclosure you need to get one slightly larger and figure out the mounting details on your own. Getting a custom enclosure made in low volumes is expensive. By the time someone covers their overhead, some profit to make it worthwhile, etc. $51 isn't unreasonable. But perhaps there will be some less expensive options down the road--especially if the standalone ODAC proves popular.

      Delete
    2. just use your DIY skills and use a 2´5inch external HD case...they cost 7€ and look very nice.

      Delete
  104. For the 200 comments issue, have you tried this out:

    http://www.ravisaive.in/2012/02/fix-200-comments-pagination-problem-in.html

    ReplyDelete
  105. Tyll goes on record with a self-proclaimed "subjectivist"

    link: http://www.innerfidelity.com/content/are-objective-headphone-measurements-relevant-audiophiles-subjective-experience-page-2

    [quote]
    SG: What really bugs me are the people who insist on ABX testing. I'm no scientist, but it's weird that blind testing mostly proves there are no differences even when two products are very different. Blind testers can't reliably tell the difference between coffee and tea. The audio guys who steadfastly believe in blind testing have it easy, they can buy the cheapest crap and live happily ever after!
    [/quote]

    how respondest thou?

    p.s. love your blog, you helped me with a hard pick a year ago!

    ReplyDelete
    Replies
    1. My response: Blind Testing has been widely used for many decades for all sorts of things. It's used to determine matters of life and death in medicine (every single drug for which medical benefits are claimed has to go through blind testing). It's used (and often required) in all sorts of sciences to remove subjective bias. And it's used in wine tasting.

      In wine tasting blind testing doesn't mask differences as Guttenberg (SG) suggests (i.e. coffee and tea taste the same) it simply removes expectation bias. The wines still taste different you're just more likely to pick the one you really like best instead of the expensive one or the one from a winery you really like.

      In my opinion it's simple. Guttenberg has a strong interest to discredit blind testing for his previous positions to even make sense. It's also in his best interest for his various paid audio gigs. His comments serve himself and the industry that pays him making him (and Tyll) hardly paragons of unbiased journalism in this case.

      Blind testing doesn't magically remove real differences. It only removes subjective bias. And ABX testing, because the switching is done rapidly, has proven much more sensitive at detecting differences than listening to gear for weeks at a time (I reference an AES paper on that in my articles).

      So Guttenberg is correct a lot of people don't hear differences in blind audio testing, but that's not the fault of blind testing, IT'S BECAUSE THE DIFFERENCES DO NOT EXIST. Sorry Steve. If you want to make wild claims, please give us some credible unbiased references that support your wild opinions?

      Further, Tyll is being rather misleading using headphone (vs electronics) measurements for the objective side of the discussion. Any time acoustics (microphones) are involved in measurements you layer on immense amounts of measurement complexity including reflections, comb filtering, the microphone itself, the 3D space everything is in, absorption of surfaces, and more. Showing a few square wave graphs is over-simplifying to the extreme.

      Nobody is claiming two different headphones sound the same. There's no such thing as an audibly transparent headphone. They're all audibly flawed. So it's a bit silly to show headphone measurements and then talk about blind testing and things sounding the same. Tyll, frankly, should know better and probably does.

      Delete
    2. Thanks for the response! I'm slowly making my way through the comments under that article, and it's proving fascinating reading:

      [quote]
      Accuracy, schmaccuracy
      Submitted by Steve Guttenberg on April 26, 2012 - 6:23am.
      So what do measurements provide that my ears can't detect? If the problem is so small I can't hear it, then it wouldn't show up in a blind test, would it? We're back to where we started, the truly worrisome measurements merely provide visual proof of what you can hear with your ears. If you can't, what's the problem?
      [/quote]

      good times had by all

      Delete
  106. Hello NwAvGuy

    I'm interested in a simple DIY mod to add RCA connectors. The spot for those is clear in your circuit board pic, above. My question is if RCA connectors are added, what happens to the signal, is it then split b/n them and the 3.5mmm jack, or does the signal to jack kick in only when inserted, or? Could you summarise the implications and practicalities of DIY adding RCA connections? [BTW my intention would be to use the ODAC to connect both to a headphone amp, as well as to desktop powered nearfield monitors, without needing to unlpug everything depending on what output I needed].

    Regards

    Reg

    ReplyDelete
  107. Amazing work.I wish there was a way to use the same approach to transducers ,imo thats where most of the problems and differences in sound lies.Linkwitz has a good site for speakers but i haven't found anything similar for headphones.
    Alex.

    ReplyDelete
  108. I am looking for the comments on the cross channel talk with 3.5mm jacks. Used to be an issue of poor isolation between channels vs. RCA. Not sure if this is an issue with Standalone ODAC that JDS Labs is selling with 3.5mm LINE OUT. ODAC specs an excellent noise floor and I hate to lose it due to poor 3.5mm-to-RCA cable. Any recommendations for 3.5mm-to-RCA cables? Nothing esoteric just good noise rejection. Thanks.

    ReplyDelete