Header

Objective Reviews & Commentary - An Engineer's Perspective

May 9, 2012

ODAC May Update

odac production ipodGOOD NEWS! I just completed testing the first two ODAC boards from YoYoDyne. These are known as “first articles” and are built using the same PC boards and parts as the production boards. I’m pleased to report the performance essentially matches the previous prototype I measured in every area and has significantly lower jitter. I still need to publish my final large ODAC article with all the measurements and more technical details, but my involvement is otherwise coming to an end. That’s good news as I can spend more time on the open source DIY Obective Desktop Amplifier (ODA).

AVAILABILITY: I’m not involved with the manufacturing, sales, or, for that matter, receiving any profit. So you need to contact one or more of the following if you want an ODAC. Last I heard, they’re still hoping to ship some before the end of the month:

ODAC CANNOT DRIVE HEADPHONES DIRECTLY: There still seems to some confusion the ODAC is a headphone DAC by itself. It’s not. You need some sort of headphone amp like the O2 (the ODAC fits inside), the upcoming ODA, a FiiO E5/E6/E9/E11, etc.

O2 + ODAC: As previously discussed, when combined with the O2 and wired to the input jack as shown in the Tech Section below, you can still use the line input jack of the O2. When a jack (say from a portable player) is plugged into the jack the O2 will disconnect the ODAC and play the external source. When you unplug the jack the O2 will switch to the ODAC. If you don’t have either source powered up you may hear some hum and/or noise due to the floating input.

SCHEMATIC: YoYoDyne has received approval from Tenor to release the schematic once. More on that in the next ODAC article.

200 COMMENT LIMIT: With some help from friendly blog readers I’ve found a work around for the 200 comment limit. It’s not perfect, and sometimes doesn’t work with Internet Explorer or mobile browsers, but clicking the “Load More…” or “Loading…” link at the end of the comments should load at least some of the newer posts. As long as you see either of those links, there are newer comments not being shown. Hopefully, if I can publish articles more frequently to spread out the comments, you won’t have to worry about it.

WHAT’S NEXT: I’m still working on my Timex vs Rolex article. It’s been my most daunting article yet and I’ve only had limited free time. But it’s hopefully coming Real Soon Now… Really! Honest! There’s also the final large ODAC article with the rest of the measurements and more tech background on the design. And, beyond that, look for the next ODA article. I’m still hoping to have at least a preliminary ODA design published before the end of summer.

 


TECH SECTION


ODAC TO O2 WIRING: Use twisted pair wiring, such as you can find inside any CAT5 ethernet cable or similar. You can also use small shielded wire but it might be harder to “route” within the O2 enclosure. The photo below shows the twisted pairs connected to the ODAC output header. Note the grounds are together (click for larger if desired):

odac to o2 wiring

O2 INPUT JACK CONNECTIONS: There are two options for connecting the ODAC to the O2. If you no longer want to use the input jack, you can just connect to the P1 header alongside the input jack. Left connects to pin 1 (the square pad), the middle pin (2) is ground and pin 3 is the Right input. You must cut the ground traces as shown in the photo with a razor blade or similar. Be careful not to accidentally cut other traces (or yourself!). If you have a DMM you may want to use the Ohms range to verify there’s no longer a short (near 0 ohms) from pin 3 and pin 4 to ground sometimes traces that look cut are not. If you want to still use the input jack, the twisted pairs from the ODAC must be routed to the bottom of the O2 board and connected to the input jack as shown. This can be done through the unused pin pads for the volume control, through an unused mounting hole, or there might be clearance between the front panel and the O2 board—it will depend on the type of wire you use. Keep the wiring as short as possible. The ground wires (and the chassis ground wire) can all attach to the center pin 1 of the jack closest to the front edge of the board or to pin 2 of the P2 header nearby:

o2 odac input jack wiring

odac production boardOVERALL CONSISTENCY: In general the production boards measure very similarly to the prototype. In this case that means YoYoDyne used the right parts and their assembly contractor didn’t mess anything up. It also provides added confidence the ODAC is a solid design. When three different samples all measure very similarly that’s a good sign nothing is on the ragged edge, excessively tolerance dependent, etc.

LOWER JITTER: There were a few minor PCB and other tweaks between my protoype ODAC and the production boards from YoYoDyne. I’m pleased to report the production version has even lower jitter. In the last ODAC article I reported the jitter components adding up to –103.3 dB. In the production version, as you can see below, it’s over 3 dB better at –106.5 dB with far fewer components. In reality both of these are comfortably below the –100 dB threshold I’ve established for transparency, so I’m not suggesting this is an audible benefit, but it’s still nice to see:

ODAC 11025 hz -3 dBFS J-Test Jitter & Phase Error 24-44

382 comments:

  1. Well done!

    Nice to see the unexpected improvement in the already inaudible level of jitter.

    Sorry to hear of the blog enemy BS. Perhaps a little forensic linguistic analysis of a few of the replies here against the writings elsewhere of...well, I'm guessing were we to say the name of a likely suspect at the same time, they would match.

    Rob

    ReplyDelete
  2. Hi NwAvGuy,
    The attacks merely show you're getting famous. Those that hide behind untruth are the first to poke at the real truth. Disregard the nay-sayers, their opinions have ceased to matter. I wonder how much revenue those 'black sheep' companies have lost since your blog came to be? I bet all their supporters haven't been able to continue buying every FOTM they come out with. With so many O2's out in the wild now, they lost those potential sales. That has to hurt the bottom line. Perhaps we will even see the closing of some doors? Those that can't take the heat and compete will fall by the wayside. You have laid down the gauntlet for performance/price and brought the snake oil merchants to task for the crap they peddle on the unawares. Viva La O2 & ODA/C! Keep smilin' to the music.
    ~FLAudioGuy

    ReplyDelete
    Replies
    1. Thanks. If history is any guide, there will always be plenty of FOTM and snake oil customers willing to part with their money for questionable reasons. What I hope changes is consumer demand pressures at least some of the manufacturers to provide better specs and perhaps evidence of real measurements.

      We're already seeing some change. Last February when I wrote my educational article on Output Impedance almost no manufacturer mentioned it and now many are. That's a significant step in the right direction. If company A, B, and C are all touting their low output impedance, it makes it harder for company X to still sell their poorly designed products with high output impedance.

      It's not unlike requiring snack foods display the calorie content or cars being subjected to standard tests for fuel economy. In both cases it puts pressure on the manufacturers to improve their products to remain competitive.

      Delete
  3. I apologize, in my post I forgot to thank you for your time, expertise, candor and herculean effort to bring some sanity to this hobby. I am sure most of us do realize there are countless man-hours you have given away for free here. Let alone, not accepting any pay for your designs. I for one have learned so much poring over each article many times already and still more to come. This blog has become my "go to" source for information. I have directed many of my pro audio friends here to read up as well.

    I don't know if you are an actively employed or retired engineer but there are many better ways to spend your free time. Myself, I would probably go fishing or travel. Point being, your time and effort here is much appreciated by so many. I know I don't stand alone in saying "Thank you". ~FLAudioGuy

    ReplyDelete
  4. Dear NwAvGuy,

    even if this blog is about to close due to "false" claims of violating Blogger's TOS, you can be sure there will be number of volunteers willing to host your blog somewhere else :) Just keep backups of all your articles should this happen some time ;)

    What you've done so far is just incredible - many many thanks for everything! :) You've helped open my eyes and save a lot of $$$!

    And don't be afraid of your "enemies" - they won't threaten/defeat you!

    Best regards,
    Mir (Happy owner of the O2 amp)

    ReplyDelete
  5. It's always a pleasure to read about your work and your interpretation of results. Your objective expertise outweighs any number of ill-informed opinions; no amount of ignorant noise can change that.

    Some food for thought:

    http://www.uwgb.edu/dutchs/pseudosc/SelfApptdExp.htm

    ReplyDelete
  6. Would you be able to ask these vendors to perhaps provide a third party qualification service or an in-house measurement for various metrics? We don't have the testing equipment and would like to know that our pieces meet your thresholds.

    ReplyDelete
    Replies
    1. JDS will be checking all of their pieces. I'm pretty sure

      Delete
    2. @traumerei: All ODACs are manufactured by a single production shop under the engineering direction of Yoyodyne Consulting. Yoyodyne coordinates bulk distribution to all ODAC suppliers, such as us.

      This post from NwAvGuy marks engineering approval of a sample from the first batch of boards.

      Therefore, all boards will measure equally. There's no need for suppliers to repeat NwAvGuys work.

      Delete
  7. Dear NwAvGuy, I'm really sorry to hear people now try to force you off blogger... - but as others have said, there are plenty of other options if it comes to the worst.
    Maybe you should reserver some domains? For example set up an as yet empty wordpress blog to secure your name?

    I also would like to point out that the load more button works on a computer but not in the mobile implementation on my BlackBerry.

    ReplyDelete
  8. Now that is really good news NwAvguy! Congratulations and best regards. I have ordered ODAC and all the parts for "The Wire amp", will post some impressions in near future. Thanks!

    ReplyDelete
  9. Thank you for your contributions to DIY; keep burning the midnight oil! Let reason, objectivity, science, and open technology pave the way forward.

    This blog is an invaluable resource for dispelling audiophile hyperbole. Audio reproduction and technology should not be described in terms of food and wine.

    It would be interesting to read your thoughts on speaker/headphone psychoacoustics, considering that speakers, and room interaction, are the weakest links in the audio chain.

    ReplyDelete
  10. Whoever wrote that email obviously has no idea who he's dealing with here. I seriously doubt that Google will shut down this blog due to wrong claims - after all, the one thing Google's engineers are famous for (and proud about) is their love of measurement data. They can't even settle for a shade of blue without sufficient data to back it up: http://news.cnet.com/google-designer-leaves-blaming-data-centrism/

    But even if they go along with it, the worst case scenario here is that nwavguy.com will be pointing to nwavguy.wordpress.com, or some other site, instead of nwavguy.blogspot.com - so all is still well.

    ReplyDelete
  11. Nwavguy,

    It startles me how small the ODAC is... Is there any "reason" behind all the parts in something like a gamma 2 DAC?

    Also, more specific instructions on how to install the ODAC would be welcome.

    Thanks for everything

    Chris

    ReplyDelete
    Replies
    1. It's much like comparing an AMB beta22 amp to the O2 amp (at least a few people I know have replaced their beta22 with an O2)--there's a big difference in size and complexity but not performance. Once you achieve transparency, a more complex design with more components can't make the device more transparent. So it's a waste of money to include more than needed.

      Many designers lack the ability to properly measure the product(s) they're designing. When you're "designing in the dark" the intuitive solution is to use a shotgun, aim in the general direction of the target, and hope for the best. But if you have access to a good well lit target range, once your rifle can hit the center with every shot, that's as good as it gets. The irony is, a lot of more complex designs perform worse because they were either designed intentionally by ear or not otherwise properly tested. The implementation details are often more important than the components used.

      Other designers include extra components because they mistakenly believe they're required for proper performance. Some audio myths are very pervasive. Or sometimes designs are more complex for marketing reasons to help differentiate the product from the competition. I'll talk more about this in the upcoming Timex vs Rolex article.

      Delete
    2. Yeah, that's what I figured. Excited to get the ODAC, just nervous about installing it. (i imagine I can't be the only one that bought and assembled o2 that is now trying to install the ODAC without shipping stuff off)

      Thanks for the response!

      Delete
    3. Someone on the Head-Fi forums actually used a cheap analyser at his place of work to compare the THD of the O2 and the Beta22.
      http://cdn.head-fi.org/2/29/29a47215_THDNGraph.jpeg
      http://www.head-fi.org/t/588978/look-out-im-using-test-equipment-o2-and-beta22-testing-inside

      Despite the analyser's own THD reaching 0.2% in the lower frequencies (and both curves hence tracking the analyser's own performance) I was a little surprised to see that the Beta22 seemed to have higher LF distortion (above the floor of the analyser), despite the test being conducted directly into the 40k ohm input impedance of the analyser: hardly a difficult load!

      Now, this is hardly conclusive proof that the Beta22 is total crap, but one does wonder...

      As regards the hate mail, I couldn't possibly speculate as to what site has threads in which people do nothing but rage incoherently about you with consistently fruity language. One of the Head-Fi volunteer admins (you'll never guess which) might have popped by on this hypothetical site to register his agreement...the post makes for one of the most hilariously inconsistent and hypocritical things I have ever read. Then again, this is the same site that offered an instant critique of the O2's grounding as abominable (not quite the wording used) because it looked "messy" (Don't you know that REAL star grounds are star shaped?)

      As others have said, when those who claim superiority (we don't really understand how science works, you see, unlike those who are enlightened by their magic ears: we're just mad zealots trying to destroy audio) inevitably resort to the verbal equivalent of hurling feces, it can only benefit you.

      When those new to the so-called "debate" find one side has graphs and logic and the other side is shouting "POO!" from the sidelines, I suspect they will quickly leap to the right conclusion.

      Delete
    4. I believe going beyond transparency is good because in the end the sum noise and distortion of the entire audio chain does add up, so some headroom is always appreciated. I know my chains aren't always as simple as DAC->Amp. I sometimes end up with a source -> converter -> DSP or EQ -> DAC -> Amp -> output.

      Delete
    5. As someone who's soldered up a Gamma 1/2 from scratch, although i'm not versed enough in audio electronics to discern all the functionality and architecture of that DAC, i *can* say that very little of the complexity has to do with the shotgun approach Nwavguy mentioned.

      A lot of it owes to the fact there is no Gamma 2. Gamma 2 is a daugtherboard for Gamma 1, released after the fact. That caused it to be a bit more complicated than necessary, as well as the fact the whole sandwich was built for different configurations offering completely different functionality.

      Then there are the S/PDIF and coax inputs and outputs and the accompanying buffers, decoupling 1:1 transformers, selector etc.

      Another reason for the complexity is a lot of safeguards that may or may not be necessary, but have nothing to do with actual audio, but in-use performance - if i recall correctly, each and every major IC on there had its own brownout detector and watchdog IC.

      Finally, the output stage of the Gamma actually includes an amp (and drives headphones just fine).

      I wouldn't say Gamma is overly complex, or even complex at all. There's very little there that might be considered over the top or superfluous.

      (NOTE: i am not - in any conceivable way - commenting on its performance, as i don't have the means to test it)

      Delete
    6. You're correct part of the Gamma's complexity is the modular approach. And some of it is optional, like the ASRC, USB interface, etc. Things like the output buffer, wathchdog ICs, etc. however, are not needed with the ODAC. And yet despite the much higher component count, the Gamma is inferior in some significant ways--most notably it doesn't meet the Redbook standard for output level, only supports 16 bits over USB, and has electrolytic coupling capacitors in the audio output.

      Delete
    7. Yeah, the USB limitations are the part i hate the most about the Gamma.
      I was actually considering ditching it, but decided against it as its a convenient way to power the DAC when using a dedicated power supply is too much of a hassle. And i felt leaving a USB receptacle just for power would be stupid, so ended up including the USB support even though it's pretty much just a "backup" option.

      I was actually considering ordering an ODAC as i'd love to do an, admittedly very unscientific and subjective comparative listening test between the two, but decided to wait for the ODA+ODAC.

      Delete
  12. Reagrding the "bad news":

    "First they ignore you, then they laugh at you, then they fight you, then you win."
    ( http://en.wikiquote.org/wiki/Mohandas_Karamchand_Gandhi )

    So you managed to get to 3rd level already! Don't worry, it's just one more :)

    ReplyDelete
  13. Maybe VIA chips solve problem with licensing. They have some nice USB audio chips like http://www.via.com.tw/en/products/audio/usb/vt1620a/index.jsp or even 32 bit http://www.via.com.tw/en/products/audio/usb/vt1731/index.jsp
    They didn't need any drivers and etc. and hopefully VIA have more DYI friendly NDA.
    One of those chips is used in this soundcard - http://audiotrack.net/products/MAYAU5/

    ReplyDelete
    Replies
    1. The VT1731 requires a driver for Windows *and* they will only talk to you if you want 25,000

      Delete
  14. Ah, what a quandary. I will be receiving some Q701s shortly, and had hoped that the ODA/DAC combo would be available in time. As of this moment my sources are a uDAC 2 (ulp!) and a Marantz CD-63SE. I don't doubt the Marantz's ability to drive the AKGs to decent volume, but I'm sceptical as to whether much effort would have been put into the HPO SQ of a dedicated CD player, Special Edition or not.

    Were I to start with an O2 connected to the line-out of my PC's integrated Realtek soundcard, would I incur much of a loss in SQ? It seems like the amping section is most often the culprit for reduction in quality, so perhaps using the line-out would obviate that somewhat.

    Appreciate your posts as always.

    ReplyDelete
    Replies
    1. The performance of built in PC sound hardware varies so much it's impossible to answer your question. The Q701 needs quite a bit more output than most headphones so it's safe to say you'll need an amp regardless. You just need to decide if you want to try an amp first, or just get a headphone DAC that can handle the Q701. Something like the FiiO E10 would almost certainly be an improvement over your PC's output, but might still be a bit marginal for the Q701 in terms of output power.

      Delete
    2. @Brian, a little bit of math indicates that your Q701 can pull over 3V and over 50mA of current if you're a loud listener. If you're a quiet listener like me, then you're usually in the safe territory (below 35mA) in terms of current and most amps handle that alright. If you like it loud, get the O2. You don't need high gain from redbook sources, but you do need its overall power specs for things to really shine.

      Delete
  15. Well I think by this time you realize how many supporters you have garnered. As Belligero points out above there will always be ignorance to combat truth and science.
    Let's face it. Any time you go up against a corrupt, well entrenched system that's lining its pockets or serving its own self interests at the expense of others, you WILL draw fire.
    But when it is all said and done, whose character and integrity would you rather have? Hang in there. Best regards.

    ReplyDelete
  16. A big THANK YOU to everyone for all the supportive comments. I'm not worried about the future of the blog. The internet is full of product reviews, educational articles, and commentary. This blog is only one example among thousands of similar sites. It just happens to be one of the more objective sites on audio.

    As a side note, I couldn't publish two of the otherwise positive comments because of what some might consider offensive language. I don't want to give my critics any ammunition. This isn't an adult blog (something Google allows with restricted access) so I want to avoid any language not allowed on US network television.

    ReplyDelete
  17. Well thanks for your time and work, as always.

    I'm kind of curious: at a high level, what caused the 3 dB improvement in the jitter test performance? (More explanations are on the way in a future article about the ODAC, I'm sure.) It seems like there must have been some small issue (what?) that was knowingly addressed that you were sure wasn't going to screw anything else up, and not some random shot in the dark.

    ReplyDelete
    Replies
    1. I believe it's related to some minor changes in the USB ground scheme. When you're dealing with jitter components below -110 dB, even small changes in the digital grounds can impact jitter. You're correct such changes carry a risk of making things worse instead of better. So, especially where the PC board layout is involved, we tried to play it safe with the final revisions based on everything I had learned along the way.

      Delete
  18. Greetings, NwAvGuy!

    As I began to read this latest blog, I found myself reacting internally in an emotional sense to your reference to whomever was 'throwing rocks.'

    Having lived over the half century mark and experienced, to a certain degree, the slings and arrows of outrageous fortune, I feel marginally qualified as a sentient being.

    Please allow me to say, in that presumed capacity, that under veneers of varying thicknesses and permeabilities, people almost daily provide obvious evidence of our collective animal origins. Who among us does not possess a limbic system (I think the term 'reptile brain' is acutely and accurately descriptive). When emotion takes the reins, Einsteins become dangerous and possibly intolerable idiots.

    Please never surrender. Yours is a force contributing to our struggle to become more human.

    Take heart!

    Jonathan

    ReplyDelete
  19. Just want to add my thanks to NwAvGuy for all your hard work. As a researcher in human factors psychology, a lover of good sound, and a cheapskate, I am strongly in favor more blind testing. Contrary to what seems to be the opinion of many anti-testers, blind testing is not about cold, clinical measurements. It simply removes the biases of expectation and forces us to judge based only on what we hear. Of course, that thought terrifies the audiophile-industrial complex and threatens the mutually profitable relationship between manufacturers and the audiophile press.

    ReplyDelete
  20. I also want to thank you again for your hard work, I am so looking forward to getting an ODAC.

    I have read all of your posts on this blog, and perhaps I am forgetting that you mentioned this in the past, and if so, I apologize, but I would be very interested in reading your thoughts on digital filters. Different filters do measure differently, and in several products, the filter that measures the best is not the one that is preferred by the designer and reviewers for listening. Ayre has settings for filters it calls "measure" and "listen", and the well received (and excellent measuring) Bricasti DAC has 9 different filter settings. I listened as the filters were changed on the Bricasti at the recent NY Audio Show, and there definitely was a perceivable difference, mostly in the high frequency response (which is also what usually changes in the measurements as well).

    Did you have options for the filter that you implemented in the ODAC, or is there just one that is proscribed with the ES9023? I believe that the ES9018 has different filter options built in that can be changed with software via i2c, but that chip is much more feature rich (as it should be) than its far less expensive brother. Are different filters just a kind of EQ, or are they more than that? Because the filter is fundamental to the DAC, and unless you have the option to change it, the one that is implemented seems to be an important decision, and as someone who knows very, very little, I am curious to hear what you think about them.

    Thank you for your time and effort. I have learned a great deal from your writing, as have countless others.

    -Aaron.

    ReplyDelete
    Replies
    1. The short answer is I think all the "filter hype" has been blown way out proportion. Most of those nice graphs and scope shots you see of pre-ringing, post-ringing, etc. were created using completely artificial signals that are not present in real world digital music. It's like driving a car off a cliff to evaluate the braking performance and then complaining the brakes don't slow the car as it falls through the air. It's an unrealistic test.

      Just as cars are designed to be driven on roads, DACs are designed to reproduce real world digital music--not square waves or impulses with zero rise time and infinite bandwidth. In both cases, if you deviate too far from the real world you're likely to uncover some less than ideal behavior.

      I plan an article dedicated to this topic as it gets complex. But, I believe most of the filter options some DACs offer are mostly a marketing gimmick. Some create audible levels of distortion and, as such, may indeed sound different in a blind test. Others don't audibly change the sound of playing music however much they might change the performance reproducing unrealistic waveforms.

      You can insert an A/D and D/A into a signal path and not even tell both devices are there. That's exactly what Meyer & Moran did in their AES high resolution audio study. Over 500+ trials recording engineers, audiophiles, and others, couldn't tell when an extra 16/44 A/D and 16/44 D/A were added to the output of a SACD player. They proved transparent.

      The M&M study speaks volumes as to the transparency of DACs. Not only could none of the listeners hear the rather boring conventional digital filter in the DAC, they also couldn't hear the DAC itself, the op amps, power supply, cables, and even the entire A/D process. Meyer & Moran not only called SACD into question, they provided strong evidence DACs can be audibly transparent despite all the erroneous audiophile claims they all have a sound of their own.

      Your sighted listening at the NY Audio Show isn't surprising. Please see my What We Hear article for a full explanation of why you heard what you heard.

      Ultimately, the ES9023 properly implemented, is an audibly transparent DAC. The window glass is completely clean. There are no perceptible flaws. It's not possible to get the glass perceptibly cleaner. While the ES9018 or ES9012 might even better measurements, I honestly don't believe they can improve on the sound quality when listening blind.

      It's certainly possible to "rig" filter differences in a way that will yield audible differences that survive a blind test. But it's not hard to demonstrate the audible differences are due to lower fidelity. A more valid test is something like Meyer & Moran did. You compare a high quality high resolution pure source to one run through the DAC and filter being evaluated. If even using a high-end system, lots of highly skilled listeners can't reliably hear a difference, it's safe to say the DAC and filter in question are audibly transparent.

      Delete
    2. Thank you for taking the time to respond. I do believe that I am fallible and am prone to hearing differences when there are none (I know that I have in the past), but the filter differences I heard at the audio show were potentially real, supported by measurements in the stereophile review. Are these measured differences potentially audible?

      http://www.stereophile.com/content/bricasti-design-m1-da-converter-measurements

      I might implement an Arduino to control my Buffalo DAC via i2c. Then I could change filters, and perhaps blind test them with an assistant to see if I can reliably hear a difference.

      I am definitely going to blind compare the ODAC with the Buffalo DAC II with matched levels. I'm sure that many people will be doing the same with their own converters.

      I wish that the Meyer and Moran paper that is often used a reference was available to the public for less than $40. I would definitely read it, as would many, many others. Hopefully an AES member friend will let me read the paper.

      Is it possible that the digital volume control built into the 9018 is more transparent than using a digital volume control before the data is sent to the DAC, or an analog control afterwards, as it has to be implemented with the 9023? That alone could be a reason to use the more expensive chip.

      Thanks again for your time.

      -Aaron.

      Delete
    3. If you spend some time with Google you can likely find a free PDF copy of the M&M AES paper. As an AES member myself I don't want to provide any links but they're out there. There's also a follow up by M&M to their critics. They easily put to rest most of the half baked criticisms thrown their way by fantasy-oriented audiophiles.

      Volume controls are a different subject and a possible topic of a future article here. Like digital filters it's not a simple subject. The short answer is I believe the ODAC, or similar DACs, are transparent when used with software volume controls. Once you have around 20 bits ENOB it's really hard to come up with realistic scenarios where that's not enough to be audibly transparent.

      As with digital filters, it's possible to construct unrealistic scenarios where some can claim 20 bits isn't enough, but in the real world, I've yet to see anyone prove otherwise. So that largely renders 32 bit volume controls, etc., irrelevant.

      You have to remember the chip companies are largely under the same pressures as audio gear makers. They're looking for any and every way to differentiate their products however questionable those reasons might be in the real world. If they can show a few dB better performance that might be worth some design wins. But the important thing to focus on is what's likely to make an audible difference.

      I'm all for improved performance--even well past the thresholds of audibility--when it's free. But when you have to spend five times as much on DAC chip, you should really question if it will offer any audible benefits.

      Like with gear, I challenge all the semiconductor companies to publish the results of credible blind listening tests to demonstrate the superiority of their chips. To my knowledge none have. National supposedly did blind tests of their audiophile op amps, but I haven't seen any credible published results.

      Douglas Self has commented several times on the unproven claims of semiconductor companies regarding supposedly superior sound quality that has no basis in factual reality. If companies don't provide credible information to support their claims, it's just questionable marketing fluff.

      Delete
  21. Congrats NwAvGuy!!! I was getting antsy waiting for this update. I am glad you had so much success designing this dac. I am still eagerly awaiting more ODA details, and really looking forward to getting one.

    Shut up and take my money.jpg

    ReplyDelete
  22. Congratulations! I'm really looking forward to the ODAC when it comes. I'm also looking forward to more product reviews. Really, everything you do is great and I thank you for doing it. (But please pace yourself so you don't get burned out!)

    P.S. Because I know there is only one of you and you can't do everything, can you recommend any other good objectivist review sites? If I wanted, say, some nice desktop speakers to connect to the ODAC once it is available, I'd like to read some objectivist reviews to help me choose.

    ReplyDelete
  23. Do you hear a difference between the ODAC feeding line out to the O2 vs the iPod Touch 3G feeding line out to the O2 via blind testing?

    ReplyDelete
    Replies
    1. I haven't done that test. I also think it would be interesting to compare the Benchmark DAC1 vs the Touch 3G. The problem is my usual solo ABX box switches quickly (which is normally what you want) and it's impossible to get perfect time sync between two completely different sources (hitting "play" on the PC and the Touch and having them start playing within a few milliseconds of each other). Your brain quickly figures out "oh that's the one that's slightly behind so that must be device B" which invalidates the blind testing. So it requires somewhat different blind methodology to do properly.

      Delete
    2. I wonder whether any of the "digital" out docks also have analog out, or if you could fashion a ALO cable/dock extender. Then you could have the Touch as your source and maybe pull simultaneous analog and digital out of it. No need to worry about syncing. Something like: ( http://www.pure.com/products/product.asp?Product=VL-61429 )

      You'd have to pull the analog out before the base because it does it's own DAC.

      Delete
    3. The ODAC won't work with the digital out from an iPod. The Benchmark DAC1 will. But you're still using the iPod as a transport along with whatever jitter issues it might have. So it's different than using a PC-based USB source. But it might still shine some light on how transparent the DAC implementation is in the iPod.

      Delete
  24. Hi Nwavguy,

    I'm a big fan of your blog and your DIY projects like the O2 and ODA, I'm looking forward to ad the ODAC to my collection!
    This may be a bit off topic but are you planning any updates/changes for the O2 amplifier?
    Thanks

    ReplyDelete
    Replies
    1. The power jack for the O2 is no longer available and Mouser has a possible replacement on order that's due in June last time I looked. If the new jack fits the current PC board, I probably won't be making any changes. If it doesn't fit, I may slightly revise the PCB to accommodate a different jack and also make a slight tweak near the gain switch so it doesn't have to be installed with as much care. I'm not planning any changes that will alter the performance or require my doing another prototype run.

      Delete
    2. I'll like to add that the new jack doesn't fit the current PC board because i've tried it, but you can cut the connection to make it fit. Shouldn't affect the sound quality.

      Delete
    3. Wait! In my Mouser Project page, I've got the Kobiconn 163-179PH-EX listed as the one I need. I wonder what I did to end up with that instead of the 163-7620-E. Mouser has ~2600 of the one I have in queue. I'm glad this came up because I would've had an interesting time with that one. It's deeper and the blades are too wide.

      Delete
    4. @TT, you may not have much choice right now but to trim down two of the "blades" to fit (they can also be bent to fit). It's not hard to do with the right tools.

      Delete
  25. Hi NwAvGuy,

    Great work! I've been holding off on any audio related purchases for the ODAC. Just a question, for someone who's going to be powering headphones that are difficult to drive, like the AKG Q701, and perhaps in the future some entry level monitors (Yamaha HS80M, for example), would the ODA offer any advantages over the O2?

    Because it's USB, it's still limited to 2 channels, correct? How would a subwoofer play into this setup? Please excuse my lack of knowledge about audio.

    ReplyDelete
    Replies
    1. If you want to mix speakers and headphones the ODA is a better bet. A subwoofer can be run using "Y" splitters from the pre-out of the ODA but the speakers will still operate full range unless your subwoofer has a line-level crossover built in. You can also get reasonably priced passive line level "in-line" crossovers for the monitors to prevent them having to deal with deep bass.

      Delete
  26. Great stuff, NwAvGuy! My ODAC is ordered - can't wait to try it with the O2!

    One somewhat unrelated question (sorry, I can see you get lots of these... it's just I feel I can only trust your advice):

    I currently own an older Linn Majik I amplifier built in 1997. Recently it's started to develop a fault with unresponsive controls from both the front panel and remote.

    I am trying to decide whether I would be better off selling it (for in the region of $300) and buying a new budget integrated by someone like Denon, or cashing out to have it repaired.

    Do you think an older premium Class A amplifier is superior to a new, budget amplifier? Or has technology moved on to the extent that the latter would be better?

    Also, a popular thing seems to be taking an amp in for a 'service'. Can you see any reason why this is worthwhile, or does it simply consist of wasting a hundred quid to get someone to spend five minutes vacuuming away some dust?

    Finally, what do you make of Linn in general? Do you see them as quality manufacturers, or snake-oil peddlers?

    ReplyDelete
    Replies
    1. Linn makes some good but expensive gear. If your amp is true class A it would run hot. Hot running amps have shorter lifespans. I would suggest replacing it with something like a Denon, Cambridge Audio, etc. Servicing a working amp is only valid for replacing old (usually 15+ years) electrolytic capacitors and servicing moving parts (relays, switches, pots).

      Delete
    2. It's not just the heat, it's the thermal cycling. I bet that, as a group, solid state class A amps left on 24/7 would have a longer lifespan than ones which were turned on and off a few times each day. Of course, I could be completely wrong.

      Delete
  27. Hi NwAvGuy,

    I have been following the O2 adventures for awhile and finally pulled the trigger to order the O2 and standalone ODAC from JDS for my Sennheiser HD580s. This is my first standalone DAC as I'm a bit of a soundcard nut (starting with Gravis, then Gravis Ultrasound, and then high end Creative cards).

    I really appreciate the hard work you've put into the designs as well as the engineering behind the finished boards. I think there is a place for subjective analysis of headphone amps but a flavored/poor performing DAC is just, well, wrong. I also like the fact that you open source as much as you can as I've worked with open source software my entire career. I've been there with NDAs on hardware (hi intel) so I feel your pain.

    Quick question: You've mentioned before that soundcards like the ASUS Xonar and Creative Titanium HD have very poor headphone "amps". Unfortunately I used a Creative to tide me over until the ODAC was released. I previously used an E-Mu 0404 USB but the driver support was awful. Could I have damaged my headphones by using the Titanium HD's headphone out? I haven't noticed any changes other than I have a tiny range of software volume adjustment (~0-20) Win7 x64 before it is too loud.

    ReplyDelete
    Replies
    1. It's not likely you damaged your headphones. The main problems with soundcard and pro audio headphone outputs is the impedance is too high. But that's not harmful to anything but the sound quality.

      Delete
  28. Dera Mr. NwAvGuy,

    having assembled a couple of O2s myself (some of the first examples in Poland) and having ordered my ODAC at JDS Labs I started to think how difficult would it be for such a skilled and determined engineer as yourself, to develop ODAC to a fully-fledged player with its own file storage (e.g. micro SD) and user-friendly UI (e.g. Rockbox)?

    I guess I am not the first one to ask this question, but honestly, I have not read all the comments under your articles, although I read many - maybe an answer to this question (if it is so popular) should appear within the article itself?

    your job is now famous and recognized globally and your informative articles are very much eye-opening to many.

    I think I know one of the reasons subjective perception has been winning with objective percepion - it is people`s (both consumers` and the self-made audio-gurus`) laziness and ignorance + marketing skills of the latter.
    everyone, electronically-educated or not (usually the latter) can write subjective reviews. it only requires marketing skills to sell them. so why bother studying electronics/acoustics? all the more average Joe prefers to read subjective, beefy, colofrul review to "boring", and "brainy" stuff. I am ignorant myself in the area of electronics/acoustics, but in general I am prone to scientific, measurable and objective perception of things, so I am willing to make the effort of raising my level of knowledge, but at the same time I realize not everyone is going to do so, not everyone has time to do it etc. this ignorance has been exploited by hard work of smart marketing people who know more money is in the subjective than the objective world.

    a subjectivist lives in constant uncertainity whether his gear is good enough, he or she is more prone to get convinced a newer, more expensive gear is better than what he/she currently has and will be more inclined to buy it. an objectivist will look into specs (if they are available) and say "well - maybe this new stuff is a bit better in some respects, but I am confident this is far beyond my perception." if he or she still feels tempted to buy the new stuff they will still carry out a blind listening test which can change their mind.

    it is more difficult to sell new stuff to objectivists than to subjectivists, this is why so much marketing and other job has been done to promote subjective perception/reviewing and promoting "no tech-specs standard" in the world of audio.

    again - thank you for being the Robin Hood, Martin Luther and Mahatma Ghandi of audio :-)
    keep doing it!
    you may now have some enemies - as all the three world-changing men had, but you have a great many more supporters and this you should keep always in mind!
    revolutionists never have easy life, but they change the world for better.

    ReplyDelete
  29. Hi again,

    Wondering if you have any plans for discussion of spatial audio? ITDs, ILDs, HRTFs, binaural recording, etc. It's a really fascinating area of audio that seems quite lacking in development and consumer-facing products.

    Cheers.

    ReplyDelete
    Replies
    1. I've messed around with a few demo tracks and understand the basic technology but I'm not an expert in the area so I don't have any plans to cover it. I agree it's kind of fascinating but the lack of content is a big hurdle.

      Delete
    2. Indeed. Considering the growth in headphone popularity in recent times I'm surprised there aren't more binaural recordings out there.

      Delete
    3. Let me chime in on HRTF. It needs some serious revision. What matters is the sound signature as measured at the opening of the ear canal and from which direction. The interpolations that are aimed at displaying how something sounds in your ear are garbage for understanding what actual sound is produced. The concept that leads to subtracting the abnormalities out is solid, but what they aim for in so doing is not useful for headphones.

      Let me put it another way. Have a dozen high-pitched triangles (the musical chime instrument) dinged and then look at the FR chart with HRTF taken into account, and you will see a wobbly curve that doesn't demonstrate the actual SPL at those frequencies. I would believe that if you want something to sound like the chime of a triangle, then you'd want actual SPL/fq and not something tweaked from readings on a HATS.

      It's also worth knowing that the foundation for HRTF is a bunch of subjective hearing studies stitched together.

      Delete
  30. Hi NwAvGuy.
    Thanks for your blog and for your work.
    I am an owner of the HD650 and I dediced to buy the EHP-O2D from Epiphany Acoustics.
    I was looking for some documetation about how much the cable influence audio performance and I would like to know if you can give me any suggestions. Is it worth, in your opinion to change the default HD650 cable with some Mogami or even more exotic stuff or not?
    I know many "SUBJECTIVISTS" that can tell you that they have noticed huge sound changes with their new cables and I am curious to know your idea. Thanks!
    Vittorio

    ReplyDelete
    Replies
    1. It's all snake oil. At best, expensive cables will be more durable, but that's about it. You'll often see people saying how silver cables help the highs while copper help the lows, but it's pretty clear they're attributing the color of the cable to the sound.

      Delete
    2. A properly constructed cable, shielding aside, doesn't change jack. Tests were ran, comparing Monster cables to metal coat hangers: http://consumerist.com/2008/03/do-coat-hangers-sound-as-good-monster-cables.html

      Exotic winding geometries and isolation material are just that, exotic. It's just strands of copper (sometimes a little bit of silver). Cable believers frequently claim various "effects" of different cables, but those things are either purely made-up or does not apply in the audio range. The link between "more expensive cable" and "better/altered sound" is nonexistent. Take "cryogenically treated cables" for example, what is the link between "frozen copper" and "extended high-frequency response"? There is none. I tell ya if these cable companies can make little bits of copper all sound different then they need to apply for the Nobel Prize.

      Of course, that is excluding more shady businesses who slip little resistors or capacitors into the cable body to alter its electrophysical properties. Or those who install little noise generators into the plugs so the added noise output from the user's speakers will be heard as "additional detail". I know, because I've seen these things myself.

      Stuff to read:

      http://home.provide.net/~djcarlst/abx_wire.htm
      http://www.roger-russell.com/wire/wire.htm
      www.nousaine.com/pdfs/Wired%20Wisdom.pdf
      http://www.aes.org/e-lib/browse.cfm?elib=5549

      PS. I've seen a guy using very expensive audiophile cables with his O2 amplifier. Oh the irony.

      Delete
    3. I feel pretty confident calling this one: it's snake oil.

      Delete
    4. @Vittorio, I asked the same question in another blog post's comment and NwAvGuy gave me a detailed answer with alot of calculations. Long story short, the difference between copper and silver is so small that the difference is inaudible. The same goes for your default cable and other aftermarket cable. That's what science says. The reason why subjectivists say that there is a huge sound change is probably due to expectation bias.

      Delete
    5. Some thoughts. Cable length has something to do with it, but you gotta be getting into some long runs or be in some noisy environments.
      Stereophile ran a cable sound test that I believe is scientifically invalid for a couple reasons, but still a telling example where people who don't know what they're listening to picked a favorite cable MUCH less expensive than the $8K super cable. So long cables in noisy environments put together in bad ways with all the cool technologies doesn't sound as good to people, but they still were only using crazy expensive cables. I would've loved to see them use some simple twisted OFCC for <$50 and see where that landed.

      To avoid the snake-oil traps, I pull my own cable and try to keep everything under 6' per run between my equipment, and I just get the cheapest twisted-pair stuff I can find.

      Also, my AKG headphone cable is stock and I'm leaving it that way... even though it's 10' long and measures 1 solid Ohm of resistance round-trip.

      Delete
    6. I'd like to say thanks to all of you for your useful comments!

      Delete
  31. @anonymous:

    You wrote: "Most likely performance on the same level (no measurements anywhere yet..)"

    That's quite the leap of faith then, isn't it?

    "If you think about budget people, they probably won't need anything better than Fiio E10. It's already transparent enough, even I couldn't detect any differences to my DM100/O2 setup (mind you, just a simple A/B). So one can get good DAC+amp for less price than ODAC itself."

    Your inability to detect a difference in a non-blind, single-subject, single-headphone (just a hunch) test is not a valid basis on which to conclude that others would be satisfied with a FiiO E10.

    "Strangely I find DM100 even more appealing in the "geeky well engineered" realm. Maybe it's all the buttons and leds."

    Don't confuse features with quality of engineering. There are lots of products festooned with switches and lights that don't perform well. Look at the fact that NwAvGuy went through multiple iterations of the ODAC in order to optimize its performance. I'd much rather he spent his time on that rather than on the addition of blinky lights and switches.

    ReplyDelete
    Replies
    1. Extremely well said. Classic example of expectation bias.

      Delete
    2. I fail to see what expectation bias logic is faulty here.

      - A well known and trusted manufacturer can't make a new transparent DAC when it already has done that in the past measuredly?
      - People on a budget (or anyone else for that matter) actually care or can hear differences beyond what Fiio E10 already offers?
      - Leds and extra connections with actual purpose make a product worse engineered?

      Delete
    3. Please see my replies above and below on this topic. Let's not turn this into a E10 vs DacMagic debate. They're both decent products for the price.

      Delete
  32. @Anonymous (May 13, 2012 7:54 AM)
    I love my O2 price/performance ratio also! Here in the US, the Cambridge DacMagic 100 sells for $369, making it 3.7X more expensive and far less of a bargain. Considering that the DM100 specs are not listed for 192k support, I doubt it performs as well as at 44.1k. Also, 192k (using a Windows driver), S/PDIF, Toslink, Asynchronous USB are not needed for audible transparency and may well cause other problems. I do like the LED's though!

    For it's given design use of computer audio, the O2 and ODAC represent a far better value (at least to US customers) than the DM100. For those who simply cannot afford the O2/ODAC combo, use the O2 by itself with an iPod or other good device via LOD or HPA output, as the iPod has sufficient quality for many people.

    I would only recommend a FiiO E10 for someone who is unwilling to save up a bit more for an O2, or simply cannot go above the $80-100 range. By DIY'ing the amp, certainly some money can be saved and spread the project out a bit. The O2 and ODAC, together or separately, represent an almost unbeatable value of performance to price.
    ~FLAudioGuy

    ReplyDelete
    Replies
    1. Seems I poked at the hornets nest with my original ODAC vs DM100 post. That was really not the intention. Or maybe it was unconsciously..

      I'm all about objectivity, but it's funny how seriously people take things. Hey Fred, seriously? I'm a leap of faither that confuses features with quality engineering? So Cambridge Audio is full of snake oil run by dumb engineers then? Even NwAvGuy doesn't think so..

      He also said he hasn't blind tested Fiio E10 yet mentioned that it's probably at the point of diminishing returns. I have no reason to believe that normal people with easy to drive headphones would ever need to pay for ODAC/O2 to be able to listen "free of care". Or need O2 instead of E11 for portable use in similar scenario. Naturally things change with harder to drive headphones etc..

      There is no denying the raw price/performance ratio of ODAC. Personally I just wanted something little more with DM100. I wouldn't even ever pay anything more for a DAC. There's no extraordinary leap of faith here compared to people believing that only ODAC can be "audibly transparent".

      Delete
    2. See my comments about the DacMagic 100 above. I have said the E10 is approaching audible transparency in terms of the measurements if you leave the PC volume all the way up. That's a significant caveat for some. It would be interesting to do a serious blind comparison with many listeners as the E10 is borderline in a few area (so, for that matter, is the E11).

      Delete
  33. I haven't seen it mentioned in this article or comments, so I thought it may be worthwhile to do so. Customers with internally mounted ODAC that are using both outputs (internal 4-pin + 3.5mm external) to keep their O2 powered ON. This will keep the outputs loaded so they don't 'fight' each other. Since the O2 input impedance is 10kΩ, the external device input should be 10kΩ or greater, to keep the ODAC load at 5kΩ and above.

    ReplyDelete
  34. Hi NwAvGuy, I am so grateful for the work that you have done on this blog and for the phenomenal O2/ODAC/ODA. You have made it easy for me to understand the science behind sound rather than having to trust on good faith.

    I have a quick question. I've picked up an AKG K501 which is quite a bit harder to drive than the HD650 I own and supposedly much more demanding than the K701. It has an impedance of 120 ohms and a sensitivity of 94 dB SPL/mW. Considering that the bulk of my music is very dynamic and high quality classical music, would the O2 provide sufficient power and voltage to reach an ideal dynamic range? Would the ODA be any better?

    ReplyDelete
    Replies
    1. If my quick math is correct, the K501 needs about 4 Vrms to exceed 115 dB SPL so the O2 is fine (even on battery power). But 4 Vrms is well beyond what you can get from an E10, E11, and lots of other amps.

      Delete
    2. Thank you very much for the reply. I think I'll still patiently wait for the ODA/ODAC combo and be satisfied with my stereo amp/E10 in the meanwhile.

      Delete
    3. I have two O2`s. One of them is driving AKG K601, and this is a great combination.

      Delete
    4. You are all rolling with better stuff than I am. I've got some alright AKGs fed by a... [wait for it] ...NuForce Icon HD. I found this site because I tried out a uDac2-HP and it wasn't any better than my lousy onboard soundcard. The Icon HD sounds much better (though I don't know what that means other than I heard some new stuff and I subjectively prefer it). But I don't know how it really is because there's only a single measurement chart on the product and that's for it's RCA outputs and not the headamp stage. I do know that it has a 10 Ohm output impedance (ouch).

      I'm waiting patiently for the ODA + ODAC.

      Delete
    5. Hi, I can confirm that AKG K501's can be driven well with the O2. Even at unity gain, it can be driven at a more than sufficient volume! Also tested well - HD600, JH-13pro, HE-400.

      Delete
  35. I was looking at some iPod Classic measurements (http://homepage.mac.com/marc.heijligers/audio/ipod/comparison/measurements/measurements.html) made with Fuzzmeasure (yeah... weak, I know), and the differences in impulse response between the two iPods caught my attention because they're largely due to a combination of chip selection (Wolfson vs. Cirrus Logic) and firmware.

    Are you planning to expand on the ODAC's performance in this regard with similar information and insights?

    ReplyDelete
    Replies
    1. Yes. I've published some thoughts on impulse response already and I'll be discussing more in the next ODAC article. I also plan to do an entire article on the topic.

      The short answer is those impulse and square wave DAC results you often see are nearly always done using what's essentially an impossible signal with zero rise time created in the digital domain. Real music is bandwidth limited (A/D converters require it).

      I'm not aware of any credible blind tests demonstrating audible differences between the filter designs of high performance modern DACs playing real music. AFAIK, when there are audible differences, it's because of other factors besides a properly implemented linear phase or minimum phase filter.

      This is yet another example of what's essentially snake oil and it goes all the way down to the silicon vendors. Most want a way to differentiate their DAC and marketing departments everywhere latched onto filtering in the last few years. But if the conventional filters are already so transparent you can't even tell when an entire A/D and D/A are in the signal path (i.e. Meyer & Moran) how do you "improve" on that?

      Delete
  36. Thanks for your hard work and contributions. It is astonishing to me that anyone would comment negatively when you have devoted hundreds of hours to a project you receive no profit from. It makes you wonder what they've ever done with their lives?

    I've always been curious about burn-in. You hear it mentioned a lot in the audiophile community, even in regards to DACs, amps and (lol) cables. Is there any validity to burning in components? Just wondering what is scientifically is supposed to be happening?

    Not that in an audibly transparent DAC or amp you should hear a difference, but have you ever seen anything change in your measurements?

    How about with headphones and speakers? Does the suspension and diaphragm need to become more malleable like the leather of a baseball glove?

    John Krutke/Zaph recommends burning in his diy speaker designs before listening to them. He's a proponent of objective measurements, so I wonder if there's any validity to the argument.

    ReplyDelete
    Replies
    1. Yup, speaker burn-in is real (although tweeters are a stretch and midrange drivers are up for debate). Woofer suspensions do break in just like a new pair of athletic shoes--and in fact are sometimes made from similar materials. You can easily document the change in performance.

      As for electronics burn-in is nearly always a myth driven example of expectation bias. That said, Doug Self has documented certain capacitors in relatively rare applications (phono preamps being perhaps the most common example) do perform better after they've been in use a while. The solution is to use the right caps.

      The other aspect is what's known as the "operating point" which can vary with temperature. While op amps, as used in the O2, tend to deliver very similar performance stone cold and hot, discrete circuits are often not nearly as well behaved. Discrete circuits, especially "exotic" or poorly designed ones, can have much higher distortion when cold (or, for that matter, when unusually warm).

      In both exceptions above it's usually just matter of about 30 minutes of use before the performance settles down to whatever performance level the gear is ultimately capable of. The idea of burning in gear for days or weeks is a total unproven myth (speakers excluded). And with cables it's even more of a bad joke.

      Neither the O2 nor ODAC

      Delete
    2. I saw a datasheet from either Linear or National on one of their opamps showing a warmup time of greater than 4 minutes for full R2R swing power. I was shocked until I did the math and realized that I'd never get anywhere close to the envelope and everything was wathin spec in the first few seconds. That's not burn-in. I'm pretty sure that as long as heat buildup isn't a problem, the components shouldn't change any from the first milliseconds 'til it begins to fail. If you get a bit of distortion from voltage scatter at the transister gate, it shows up right away and stays there for the life of the IC.

      On headphones, I once had a blog and did a writeup on burn-in. Since hearing adapts, and does so quickly, I had a well burned-in set of headphones (K171) that I would listen to for a couple minutes before testing out the new headphones for changes. This established the consistent standard in my ear. I pulled out a new AKG K272HD and listened for very short periods of time (returning always to the K171 for recalibration) and found it very sloppy in sound positioning and bass performance. I then let it burn in and checked it again after 8 hours, 24 hours, 36 hours, 56 hours, and finally after 72 hours.

      The internet folks are wrong about burn-in times required for headphones because my notes indicated that after 8 hours, bass was improved and never changed again. I don't know if that change took place in the first 10 minutes or closer to 8 hours, or how long it took to change. But, the soundfield (instrument positioning) took a lot longer to tighten up. By the 56 hours, are hardly picked up any difference (assuming I could interpret my notes well and recall things clearly... yeah right!). Up to 72 hours made no difference at all.

      Now, contrast that with my same experimentation with a new set of K142HD's. The sound never changed from minute 1... ever. It's like they were already burned-in or some other material in the headphone is absent compared to the other because they use the same capsule in both models to produce sound. So I'm totally baffled and I'm thinking that it may not be the driver that takes a lot of time to burn-in with the K272HD's.

      There are many unexplored facets to this that may actually be understood by some manufacturers as trade secrets, or not. I'd like to dig more into it. If I were rich, I'd be doing audio experiments all day.

      Delete
    3. Oh, if you read my previous comment on headphone burn-in and thought my self-experimentation had scientific or even any form of difinative credibility, you're mistaken. I've made some conclusions myself but primarily a better approach needs to be taken on these things. Tyll Hertsens has done some work on the matter, but not the kind of stuff I would've done.

      Delete
    4. I conducted impedance plots of my HD650 headphones brand new out of the box and after 48 hours of break-in. There was a noticeable shift. The test set up was completely undisturbed to make sure if wasn't related to any other variable. The bass resonance dropped about 5% from around 90hz to 85hz. While I'm going to say that's certainly audible, it could be. There's at least some objective evidence to support headphone break-in. It's entirely possible the same shift would have happened after 8 hours or even less.

      And to my earlier comment above which got cut short, the last sentence was supposed to read:

      Neither the O2 nor ODAC change significantly from right after power up versus after running for days. The O2 does degrade slightly if you get the output op amps seriously hot (which I've only managed to do with full power sine wave testing into punishing loads) but it quickly recovers when used normally to play music.

      Delete
  37. I have been following your progress on the ODAC on and off for some time and although I do not need another source/dac I will be buying one to try it. I feel this may be a step towards audio moving away from the world of "magic" and back to a world of performance.

    Recently I asked the gentleman at Audio Poutine if this would work with the ipad camera kit and he posted a video of it working. http://www.facebook.com/photo.php?v=351915454862820. This is awesome as there is a growing market for products that work this way. If you could get your unit to work with android tablets, then you would open up an even larger market for yourself. I do wonder if there is a way to power the odac from something other then the USB though.

    IMO there is going to be a large demand for usb 5.1 devices as devices like the raspberry PI and MK802 start flooding the marketplace. When will you be making your 5.1 ODAC? hint, hint ;)

    I do have one question though. Is there any way to grab a signal and make a spdif output for the odac? I know it may sound a bit odd but it is something that would allow me to use it in my car with my bitone processor.

    Keep up the great work
    James Zatopa

    ReplyDelete
    Replies
    1. No 5.1 plans here, and in theory, the Tenor chip supports S/PDIF output. It's not something we ever tried and it may require different programming (which requires a proprietary tool).

      Delete
  38. In regard to TT and the HRTF comments. I'm not quite sure what is being argued there, because HRTF's measured for a given individual do in fact show lots of complexity in frequency response, as they should.

    The work on HRTF's is mainly made difficult by several things, all of which are related to the fact that each individual's HRTF's are different, and that there is no "one size fits all" HRTF to be had.

    In particular, externalization (i.e. not in the head sound) for directly in front is quite difficult because at that position, one is accustomed to the DIFFERENCES in the HRTF's to the two ears, and taking the differences is even more subject (as it must be) to individuals than the HRTF's themselves.

    It is a complex subject. Unfortunately I don't have an article up strictly on HRTF's at the present, although I did a talk in that regard a few weeks ago in Turkey, it isn't available as a recording, sorry.

    ReplyDelete
    Replies
    1. This point comes up in other online platforms a lot too, but it is important to see the whole picture here. The alternative to HRTF'ed sound over stereo headphones is a 5.1/7.1 setup, and that is not just expensive, but needs to also be properly set up. Both these conditions are deal-breakers most of the time for most of us. If you look at the average PC user having his $100 Logitech/Creative/etc. 5.1 setup in his room - and maybe even something like monitor (as in LCD-display) right between his head and one of the satellites - there's a good chance that opting for HRTF will actually be an upgrade, despite the flaws.

      Something like Beyerdynamic's Headzone or that Smyth Realiser may be better solutions than the standard Dolby Headphone or CMSS-3D implementation, but they also are 10-100x more expensive.

      Delete
    2. Really, I'm calling for a simpler method of testing FR for headphones and such than by using a system that requires HRTF compensation. I'm a fan of simpler. I became one with my technical background in plastic colorants and learning of the ins and outs of reception, perception and transmission. Surprisingly, a lot translates over into audio in terms of the theories of human perception and an instrument's usefulness in gathering data.

      Delete
    3. Testing headphones via coupler is not a particularly arcane art, and it is a method that has a great deal of perceptual justification, as well as analytic reliablity. What do you mean "HRTF compensation" for a headphone?

      Delete
    4. The O2 measures flat/neutral/transparent. The ODAC is largely achieving the feat as well. Some speakers get really close to having a flat FR. Headphones, not so much... or do we even know?

      HRTF is about perception and therefore illusrates how one would perceive a device with a flat FR playing white noise. It doesn't illustrate if the device measures flat. Because a HATS system is designed to measure how something would sound in the ear, it has its own canal resonances that need to be compensated for in order to show what should be a flat FR.

      So I do think that a measurement system which relies on the extra variables of sound measurements through a synthetic ear canal and then compensates for the ireegularities is prone to more mistakes than just measuring a headphone's earpiece directly.

      Delete
    5. @audioskeptic, I like your blog, BTW. I admit that my comments have not been as clear as my thinking on the matter. I do believe that HRTF is "sound" science. But since one's perception of audio is based on the signal generated external to the ear (excepting IEMs), I'm calling for a system that measures the sound at that point to determine the nature of the FR in a way that all users should be able to interpret for themselves, and hopefully has less of a margin of error (see Tyll's article on his own system's accuracy). It's the use of a HATS for these things that I believe needs to be revised or rethought... not the actual science itself. There are different battles to pick with my suggestion, but I'd like it to be explored moe. IEMs are a bit trickier though because the sound is produced beyond some absorbtive and/or resonant material in the ear canal. I can't think of a finer use for a HATS than with IEMs (when it comes to audio quality). Calibrating the HRTF for IEMs at a certain insertion depth is necessary, but I don't know if it's been done yet. B&K don't have calibrated IEMs for such a purpose, so I don't know how to interpret plots I read online.

      @NwAvGuy, I'm hoping your ODA has room for three gains positions, one of which can optionally be a voltage divider. I'm planning to set one up with 0.5x, 1.7x, and 6x gain settings. The first two will be perfect for my main listening devices when fed by the ODAC, and the highest gain for my hardest headphones when fed from iDevices.

      Delete
    6. I agree with you TT on HATS and free-air measurements being a suboptimal solution. I think a much more reliable way would be to measure the magnetic field fluctuations caused by the coil. Though I'm not sure how that would properly take the driver's diaphragm-stiffness into account.

      Delete
  39. I'd like to receive some comments about the balanced input/output. I read that many audiophiles are using the balanced cable and I was wondering if it can really make a difference. I am using the popular HD650 with a Fiio E10 and I am moving to the EHP-O2D.

    ReplyDelete
    Replies
    1. I was on the balanced bandwagon for a while until I realized that it makes no quality difference in the sound at all from the perspective of how much push-pull you get on the earspeaker. The differential signal requires about twice the equipment to get the signal from A to B and usually results in slightly poorer performance in terms of distortion. You can still get transparency but it takes a lot more to get you there and your options component selection can start to get restricted.

      I thought it would be cool because most DACs output in differential mode and I thought it would be more purist to go differential all the way. It turns out I was wrong and it took a while to figure that out.

      Balanced output for a preamp can make sense in some places because the overall noise rejection of balanced cable runs is bettter than unbalanced but those differences don't show up until you have long cable runs. Anything under a couple meters doesn't matter. For headphone use, you're likely to have no need except to impress some other audiophile friends.

      Delete
    2. Is there any advantage to balanced amplification for transient response? (though maybe it doesn't matter because it would be limited far before the transient limits of the amplifier.)

      And/or more available power? Though I guess that if you could just have a single ended design that was twice as powerful w/o twice the complexity of a balanced amp.

      Maybe I missed the article/post about the lack of merits of balanced designs... just curious as it's always been one of those "maybe I should get a balanced amp" nagging feelings.

      Delete
    3. There's nothing in digital music that can even come close to challenging the transient response of even modest headphone amps (excluding electrostatic amps). It's also not hard, as even the O2 demonstrates, to get enough power to drive nearly any headphone well past the threshold of pain and/or its dynamic limits with a conventional amp.

      Bridged/balanced output stages offer benefits only in extremely high power amps for speakers. In that case they reduce the peak current demands on the power supply which can be significant when you exceed a few hundred watts per channel.

      And balanced line-level gear only offers benefits, as already stated, for very long cable runs and/or operating in electrically hostile environments.

      In general, all the extra circuitry in balanced gear degrades the overall performance. Noise is typically worse, and some forms of distortion are typically worse.

      Delete
  40. Balanced cable runs between equipment are detrimental in a consumer environment: you end up with higher noise and no advantages unless you have stupidly long cable runs.

    Balanced headphone amping is simply detrimental in all possible environments!

    ReplyDelete
    Replies
    1. Except for electrostats, since they're push-pull transducers.

      I wonder if a DAC with balanced output would be useful there. I suppose it would probably depend on the quality of the amp's phase splitter but I'm not sure.

      Delete
  41. Educational reading here, NwAvGuy :)

    But I must ask you, have listened Nuforces Icon HD(P)?
    I'm looking for a neutral-sounding DAC with above average soundstage. I've tested uDAC-2 and concluded, that it's not for me, not so good sound-wise and it had problems with playback at 24/96. Then I found out about Icon HDPs little brother HD. After a month of researching and smoking some forums, I think, that Icon HD is very good toy for 300$. But in a buying day, I found myself reading your blog and thinking, what is better? Nuforce Icon HDs DAC or ODAC? Can you help me with that :)

    With a basket of thanks,
    Eugene

    ReplyDelete
    Replies
    1. I really don't trust NuForce. Their response to the uDAC-2 issues causes many reasons for concern. In my opinion there are many better options than NuForce. The uDAC-2 is far from their only product that has been found to have significant problems.

      The ODAC is entirely audibly transparent. So it's impossible for NuForce, or anyone else, to audibly improve on that. Given NuForce's track record, their products will likely perform worse.

      Delete
    2. Do not get NuForce or anyone who designs products by ear and without any credible measurements. I learned my lesson, never again Nuforce.

      Delete
  42. Hi NwAvGuy, I'm glad you're here separating the wheat from the chaff.

    I've had a mail from Epiphany Acoustics today saying the EHP-O2D (AMP and DAC) is slightly delayed but they will start distributing at the beginning of June. I think it was the O2 amp boards they were waiting on. Anyway I'm patient enough.

    In the meantime I'm looking for a pair of cans that would complement the kit.
    IEM's are my usual staples (Hifiman RE-262, UE triple fi 10's), but I'm looking to get a bit more comfortable whilst at my desk at home.
    I've been reading up from various sites on a selection but as of my initial statement it's difficult to work out what's what from the subjective comments on them.
    I've nearly pulled the trigger on one of this lot: AKG Q701's, Beyerdynamic DT 990 and Sennheiser HD650.
    Am I on the right track with these, should I consider something else or does it really matter above a certain price level?
    I'm grateful for any pointers and/or comments.

    Keep up the great work

    Will...

    ReplyDelete
    Replies
    1. Have you read my HD650 review? Headphones are very subjective but of those three I would personally choose the HD650. The Q701 (assuming they sound the same as the K701) are too bass-shy (but otherwise wonderful), and the DT990 is too bright and even somewhat harsh. But you may not like what I like.

      Delete
    2. I'll second what NwAvGuy has to say about these. Of the bunch, I'd give a word of caution on the Beyerdynamics concerning the higher impedance models (above 32 Ohm). They have higher distortion levels compared to the lower impedance models as seen on charts at HeadRoom's Learning Center. I'm thinking that they use the same magnet types but find some other way to introduce resistance which hurts the end effect of damping factor. Sennheisers HD650's seem to do it right and have a real higher impedance model that is legit for good sound reproduction and not just to try to compensate for overly powerful amps like Beyer seems to be doing with the variety in their lineup.

      Delete
    3. You should check out the Hifiman HE-400. Its current price is $100 less than what the 650s are retailing for these days, and is really cheap for a planar magnetic. There seem to be build quality issues, however.

      Delete
    4. "Of the bunch, I'd give a word of caution on the Beyerdynamics concerning the higher impedance models (above 32 Ohm). They have higher distortion levels compared to the lower impedance models as seen on charts at HeadRoom's Learning Center."

      InnerFidelity graphs suggest it is the other way around, with the 600Ω model having the least distortion. I guess random sample variation is significant, and there is no conclusive evidence either way. However, your assumption that the higher impedance models are the same as the 32Ω one with serial resistance added is not correct, since it would mean the efficiency decreases significantly with the impedance, but that is not the case.

      Delete
    5. I haven't heard the DT990's, but own DT880's and haven't found them at all harsh (quite laid back and pleasing, actually). My buying choices came down to roughly this same list but comfort was a big factor, which led me to the Beyers... so while they are down two votes, I have to plug them for what it's worth. :)

      Delete
    6. I've got the Q701s. NwAvGuy, some say they're different, but frequency graphs from InnerFidelity make them look pretty similar: http://www.innerfidelity.com/content/akg-quincy-jones-q701-sound-and-measurements

      I feel like the bass quantity is just right. Mostly though, people looking to buy in this price range already know if they prefer balanced, mid-centric, or analytical sound. I fall in between mid and analytical, leaning towards the latter. I feel that the bass quantity is just right, and that the quality is much better than my Senn HD25 1-IIs.

      Delete
    7. Thriller doesn't sound bass-shy at all on any of my AKG's .
      Maybe it's just me, but some mixes sound spectacular on AKG-cans,
      almost as if they where created on them ..

      HD650's are nice, but IMO they are not 'monitors',
      they clearly 'pretty up' the sound .
      Beyer ? GREAT cans, you just have to listen to all the models before choosing the one with a sound-signature you like, the difference is quite significant .
      They are also extremely comfortable to wear for extended time IMO ..
      You really should listen to some different headphones, playing music you like and know, before choosing .
      Personally I would rather have 2 different 250$ pairs than
      one 500$.

      Delete
    8. to Va1z4rDMay 23, 2012 1:01 PM

      The thing with AKG's, at least for me, is that they seem to give the most correct bass-representation possible compared to switching to front-located speakers .
      Considering the fact that real BASS is at least 50% nasal-cavity and abdominal PHYSICAL 'impact'-perceived, it always astonishes me how close the AKG's got the bass, when switching to speakers at 'realistic' (harmful) levels ..
      Seemingly AKG's don't try to over-compensate for the missing physical properties of real felt bass .

      This is btw also where the CD-format gets it wrong :
      Just because you can't HEAR something, it doesn't mean you can't FEEL it !
      If it's loud enough ..sub-sonic bass will make your guts vibrate in unison and you will vomit! and maybe even need a change of underwear !!
      There IS such a thing as 'The Brown Note', but not @ 16/44.2
      and not with 99,9% of the gear out there ..

      Delete
    9. Bass is very subjective. In my opinion most serious full size speakers have subjectively more deep bass than the K701 does. And if a system has a powered subwoofer, as most do these days, it's all about how the sub is set up.

      I have no idea why you feel 16/44 cannot reproduce subsonic bass properly. Higher sampling rates and/or bit depths do nothing for improving bass performance. The 16/44 format has flat response down to DC. Meyer and Moran, and others, have shown how 16/44 can be entirely audibly transparent.

      Delete
  43. Hi there NwAvGuy,

    I have an off topic question regarding damping and damping factor. Is there any audible difference between a damping factor of 400 v/s 800? Violectric's V100 has a damping factor of 400, while their V200 has a damping factor of 800, which is the double. Does that implicate an audible difference.
    The difference in specs between the V100 and the V200 appear to be minimal. Are there real audible differences between the two amps?

    Sorry for the selfish and off topic question.

    ReplyDelete
    Replies
    1. While it's a measurable difference, I'm not sure it's an audible one. My rule of thumb for headphones is a worst-case damping factor of 8 (the so-called "1/8th rule"). That means the output impedance should be no more than 1/8th of the headphone impedance. That said, Benchmark does show some added measurable benefits to even lower output impedances. See:

      Sonic Advantages of Low Impedance Headphone Amps

      and:

      Output Impedance

      In my opinion you're not going to hear any differences between the Violectric amps (unless you need the higher output power of the more expensive ones) as they all should meet the criteria for transparency.

      Delete
    2. The funny thing about the Violectrics is that they do look nice - everything seems to be neatly arranged into rows - and they advertise the fact of a fully discreet output stage. These usually would have been hints to avoid the product, judging from your prior articles.

      The other weird thing about them is the 60V internal voltage. Does that help in any way? Even the DAC1's rails are at just +/-18, right?

      Delete
    3. Only the higher-end Violectrics have a discrete output stage. At least the V90 uses an IC output. But, again, that's hardly a problem if the entire amp is audibly transparent.

      You're correct about the DAC1's rails. Even +/- 15V rails will drive 99% of dynamic headphones to levels you probably shouldn't be listening to.

      I've had some private correspondence with Violectric and, while I respect most of their approach, they seem unable to resist at least a bit of snake oil with some of their products. The reality is their least expensive V90 should be audibly indistinguishable from any of their more expensive amps with virtually any headphones. And, for that matter, the O2 (and upcoming ODA and DAC1), should be equally indistinguishable. As much as some would like you to believe they all sound different, until you listen blind, it's all just very biased opinion with little basis in objective reality.

      Delete
    4. Thanks All,

      NwAvGuy, I've read your HD650 article now and am tempted by them more so.

      1st Anon They they would be a nice complement to my 262's which work well for me. Any word on there HE-300 which is a dynamic I believe?
      2nd Anon I was thinking of the 600ohm ones as well but can't find them listed in stock anywhere in the UK at the moment so they'll likely drop off the list :(
      Hotsoup I've not read up on the 880's yet but I was thinking the 990's would be a step up until I read the HD650 article and now realise how things can go backwards as well as left and right with any make/family of cans...phew.
      Va1z4rD I'm leaning towards Q701's as I'm not too worried if they need a bit of EQ for a touch of bass but then again I'm not a bass-head so maybe nothing at all.
      Peter, in the UK they are quite similarly priced approx £230 with the HD650's seen for about £270. I think AKG's and Beyer's are quite a bit cheaper in the US than over here but the Senns are about right. You're right though why not go for 2, so looks like the AKG's this month and the Senns next unless anyone else has other ideas for me?

      Thanks again

      Will...

      Delete
  44. It would be great if you could address all these differences between the DAC/Amps of the different manufacturers that make audio transparent components, but arrive at similar results through different designs, in your upcoming article "Times vs Rolex".

    Violectric believes that 60 Volts make sense, but that obviously goes against several issues you've discussed earlier. Like you said, Violectric has some weakness for snake oil, but I suspect that HeadAmp, CEntrance, Anedio, etc., also have a minimum amount of snake oil arguments for this or that design choice.

    These little details may make little difference if like you say, even the V90 is audio transparent, but the plus or minus dollars sure do. It's not easy to develop a sufficient shopping criteria, but it would be a great thing to make something like a buyers guide. Obviously very much opposing the common ones, which are only more publicity.

    Cheers!

    ReplyDelete
  45. NwAvGuy,
    What a fantastic blog you have! Very informative and rock solid information in this snake oil driven audio world.

    Do you know if there are any (serious) measurements available on the Violectrics products (like they are e.g. on the O2 and the DAC-1)?

    ReplyDelete
    Replies
    1. I have seen a few measurements here and there and, from I've seen, they indicate Violectric is accurate with their specs.

      Delete
  46. The argument Violectric offers to explain and justify the high voltage of their headphone amplifiers, is this:

    A headphone doesn't really require high power, but from the equation P = U2 / R we can see that the square of the supply voltage determines the power into a given load resistance. The higher the headphone's impedance, the more voltage will be needed. But this deals with the achievable loudness to a limited extent only: Technically spoken, music lives on fast transients which put high demands on signal processing. And thus a fast transient can easily push an average amplifier with +/-15 volts supply to its limits. Due to VIOLECTRIC´s high supply voltage you will benefit from doubled output swing capability."

    Any thoughts on how accurate that really is?

    ReplyDelete
    Replies
    1. Please see my More Power article. It explains, in a lot of detail, just how much power you need. An amp either clips or it doesn't. The right kind of oscilloscope can readily capture clipping playing real music into real headphones. And your ears decide, even with high dynamic range music, what's "loud enough."

      There's nothing magic about "transients"--they either are reproduced normally or the amp clips. Having more headroom above the required clipping point doesn't help the sound at all.

      Additionally any headphone driver will overload at some voltage--even on a brief transient. At lower frequencies, where the greatest transients usually can be found, drivers are excursion limited. Adding more voltage doesn't help at all and risks damaging the driver. At higher frequencies drivers overload in other ways. And very few drivers can handle more than about 25 volts peak-to-peak (9 Vrms)--an easily obtainable output with +/- 15 volt power supply rails.

      Unless you own very rare headphones that need unusually high amounts of voltage I just don't see any real-world need for more than +/- 15 volt rails except as marketing hype to try and justify more expensive amps.

      Delete
  47. Just an update... I've had some other obligations lately--hence the lack of articles. The ODACs are starting to ship to customers and we're looking forward to more feedback.

    We have discovered one minor surprise related to a late change of a filter inductor. For those wanting to put the ODAC into an O2 there's a ferrite inductor that might contact one of the battery terminals. Electrically it's not a problem but mechanically it might be if the battery terminal wasn't installed fully vertical. The solution is to heat up the solder pads on the battery terminal and make sure it's fully upgright and not tilted forward. As long as the ODAC PCB contacts the base of the battery terminal everything is fine. I'll have more info in the next ODAC article.

    ReplyDelete
  48. NwAvGuy,

    Thanks for the effort you are putting into designing, measuring and explaining your DAC.

    You said

    "I’ve even tried several different PCs and the ODAC’s performance is relatively consistent between them."

    I'd be interested in seeing measurements using different PCs so that I understand how large real world differences might be.

    Two questions: If two ODACs are connected to the same PC, will they work independently? How will their names appear in the Windows control panel applet for audio devices? (I have a dedicated MusicPC that provides sound to 3 rooms via 2 or 3 audio devices. I might want to use 2 ODACs in that PC.)

    Bill

    ReplyDelete
    Replies
    1. Yes they will work independently. The names depend on which version of Windows you're running and where you look. It should be either "ODAC" or "TE7022 Audio" and, I believe, Windows adds a number after a second identical device when both are plugged in--i.e. "ODAC 2"

      Delete
    2. A bit of technical info in this scenario. Advanced software (like a DAW) can bus both devices together for output but programs like Windows Media Player and iTunes will only play the sound through the default playback device... in other words, only one of the ODACs will get audio. Does anybody know how to get this to work through traditional software?

      Delete
    3. For the record in Linux the ODAC identifies itself as "UAC1 DAC" ....

      I had some problem switching between my on board sound and the odac. .... I found the pavucontrol if your using pulse audio the easiest way to temporarily switch between sound cards

      Delete
    4. @Me, there are not many reasons why someone would want the same audio stream out of two devices at once (besides blind A/B testing). It's not an easy thing to do. There is a freeware virtual driver that's supposed to do what you ask, but in my experience it was unstable and unusable.

      In general if you need to drive say speakers and headphones from the same software source, just split the output of a single DAC. If you want to use two different sources, that works without any extra software. You can, for example, play Windows Media Player on the ODAC and all other Windows audio on your built-in audio.

      Delete
  49. Newb quesetions about ODA upgrades:

    Wider Source Compatibility: you said it has more flexible gain options. can you more specific like a selectable 3-gain modes?

    Higher Quality Power Supply: is a new bigger panel/case needed to fit in the new ODA power supply? how much would it generally cost more in the price range?

    ReplyDelete
  50. To answer a few questions at once, I don't have an ETA on the ODA beside I hope to have something published before the end of the summer. I'm still working on some of the other details. The gain options, for example, are partly a mechanical/ergonomic decision.

    The idea with the ODA was to keep the price of the board + power supply within about $10 of the O2 board + batteries + wall transformer. The larger enclosure and front/rear panels, however, will add additional cost but that will depend on the type of panels, manufacturing volume, etc. I really can't yet predict what the assembled price will be from someone like JDS.

    ReplyDelete
    Replies
    1. Just received my ODAC in the mail today! It arrived from JDS Labs without a line out jack. I noticed that if you flip the ODAC board to the bottom of the O2 board and mounted it on the opposite side it, with some filing of the ODAC board perhaps could easily be made to fit in the O2 and still retain the batteries.....Physically, If there are electrical issues then.....

      Also, it would seem that all the O2 functions, inputs and outputs could be fully retained.

      "I think" to quote the great Carl Spackler(Bill Murray's character from Caddyshack.

      Delete
    2. I think a lot of the questions on the ODA are not so much impatience as they are curiosity and excitement - I can't imagine anyone here would want you to rush the ODA design.

      Maybe you could publish an article on the various options you are considering or the design decisions you are facing in the mean time?

      I read in the original ODA & ODAC article about the improved power supply, but I always assumed there would still be some form of a wall-transformer for the ODA too. After reading your comment above I'm not so sure any more - will the ODA have a fully internal power supply? Is that a safe choice for a DIY design? Also, isn't the internal power supply the DAC1's weakest link? (I'm not challenging your decisions here of course - what little I now about amps, I've learned on this blog.)

      One more thing - I wouldn't worry too much about the price. Since the ODA will be a DIY-design again, the various vendors will be somewhat constrained in their pricing range. Additionally there finally seems to be some competition for the objective products taking off with that new Swiss manufacturer.

      Delete
    3. Thanks for your comments. The power supply issue is all about trade offs. An internal supply does indeed present safety issues for DIY but the O2's half-wave supply creates other problems--especially for those who live outside North America.

      I'm always concerned about price. One of the main points of this blog is to show you don't need to spend a lot of money to get the best sound possible. I don't want to fall into the trap of adding expensive snake oil just because some mistakenly believe it helps the sound. A $0.50 op amp used properly really can be just as transparent, and sound just as good, as a $7.00 op amp.

      Delete
  51. What do you think of the EHP-O2D + HD800 combo?

    ReplyDelete
    Replies
    1. The HD800 is well within the O2's capability. The comment below is from someone with the somewhat similar HD700. Transparent is transparent.

      Delete
  52. Amazingly, I received both HD700s and the ODAC today. Feels like Christmas! Anyway, I soldered up a 1/8 jack to the ODAC and plugged it in to the O2 and am up and running. I have just started listening, but everything seems blissfully free of any sounds that are not part of the music. Lots of smiles over here.

    I'm still getting used to the HD700s, as I've spent many hours listening to the HD595s, and these are definitely a different beast. They seem to have a brighter top end, which is a good thing, as I felt the 595s were a little rolled off. Bass is also a bit emphasized vs. the 595s, which I also feel is a good thing for the same reason. The more I listen, the more impressed I am.

    In short, I'm one happy camper! Thanks NWAVGuy for getting me as close as possible to headphone bliss.

    ReplyDelete
  53. Hi NwAvGuy, you've done some sterling work on the O2/ODAC and I'm looking forward to hearing the dac in action coupled to The Wire Se-Se and O2.

    Just wondering, whenever you’ve compared the O2 or O2/ODAC against the Benchmark you normally cite ‘can't reliably tell the difference’. The only area that looks to deviate away from the requirements for total transparency are the distortion figures. I understand you set a higher threshold, <80/90dB compared to the <100/110dB shown to be needed. What do think it'd take to bring the O2 and [less so] the ODAC into that area, so 'can't reliably tell the difference' becomes 'can't tell the difference'?

    ReplyDelete
    Replies
    1. There is no black and white answer to what you're asking. On the one hand I've been criticized by some for setting the bar at 80/90 dB as that excludes a lot of gear. And there are others who want to believe they can hear the difference between 80/90 dB and 100/110 dB. In the What We Hear article I tried to explain my justification and, in the Tech Section, provided links to some sound files used in the Myths Video so people can listen for themselves to "worst case crud" that disappears even at -75 dB.

      I used the qualifier "reliably" because blind ABX testing is based on a statistical result. Listeners often won't score a "perfectly random" result even when listening to the same source as both A and B. There's more information on blind testing in the What We Hear article.

      I don't believe anyone can manage a statistically valid score comparing the ODAC/O2 to the Benchmark but a few might randomly get say 7 out of 10 correct and (incorrectly) claim that means they can hear a difference. If they don't want to believe in the math, the best solution is to conduct more testing with many more trials.

      If you do want a DAC/amp with performance well beyond transparent I would encourage you to buy a well measured commercial product. For me to design such a product is somewhat in conflict with the message I'm trying to send. I can't honestly tell anyone it will sound better.

      Delete
    2. I am a little confused by the question from Paul Blythe. Which chart/measurement is he referring to? Weren't the distortion components, that were approaching the -80dB threshold, well within the ultrasonic spectrum?

      Delete
    3. Thanks for the reply, reading back I made quite a mess of asking two separate questions. Q.1, should have been a qualification of what "reliably" meant in this instance.

      The second hypothetical question should have been; if the principles that have led to the O2, ODAC and ODA were to be set aside slightly, a relaxation of cost vs. performance. What would it take to produce a headphone amp or DAC that not only hit the transparency mark, but surpassed it by say 10% [a safety margin against detractors]... Would it incur a 25% increase or a 100% increase in the cost of parts? Plus extra development time etc. Is it possible to quantify such an increase even if it’s not truly justifiable?

      Anyhow, I took delivery of an ODAC via JDS Labs on Monday and WOW! Sounds identical to my old Pioneer DV-757Ai (£800 when new), and is a few steps better than my Oppo BDP-83. Fantastic stuff :)

      Delete
    4. Glad you like your ODAC. I think it's difficult to quantify increased cost and development times for XX% increased performance. Datasheet performance numbers can be elusive (the next ODAC article will discuss this more). And, it's worth pointing out, the ODAC already exceeds many criteria, and the criteria itself arguable has a substantal "safety margin".

      Delete
  54. Hi nwavguy

    Recently stumbled upon your Blog and first off, a BIG thank you. Not just headphonia needs people like you, the whole audio industry does. I have been so far down the hole I stopped trusting anything but my own ears. Too much money for uncertain results. Am I really getting a product that technically solid or does it simply use tricks to make my ears believe it sounds good? I have been to a lot of audio dealers and get the feeling they are actually believing the shit that is going on. One guy even told me you need different mainline filters for different devices because a CDP with too clean power will sound boring. I wish I was making this up.
    For a guy living on a small budget I have already dumped about 3k into my audio hobby. So yeah, saying delusional audiophiles are not harming anyone... they certainly did to me. And it still started so harmless, with me being an engineer with a sense of quality and eyeing a pretty fine sounding compact Yamaha Stereo set for 500 Euros. I should never have listened to the voices saying seperates will get you more for you money.
    At least it has sped up the process of trusting my own judgement. So thanks a big lot for the thing you are doing :-)

    As for the ODA, I have a question: will it be possible to split Dac and amp so I could hook up stuff like a tube buffer to tune the sound to my liking? I think that also would go a long way of showinh people that what they thought was better was just different.

    Oh yeah, you know what to do next: design or suggest some speaker amps. ;-)
    Oh, what's your opinion on the preamp, how much difference does that part make?

    Best wishes, Fox

    P.S: tried to send you this message via contact form twice, said it went through but no answer. Could be broken.

    ReplyDelete
    Replies
    1. A DIYer could split the DAC from the amp in the ODA if they wanted but it's not something that will be built-in as few are likely to use such a feature. If you really want the sound of a tube amp, I would suggest going with a tube amp otherwise it's sort of like putting tractor tires on a Porsche.

      Delete
  55. I pre-ordered the ODAC + O2 from JDS Labs but I didn't opt for the power adaptor since i'm in Australia and it won't work here. My question is, what power adaptor should I be looking for? I have searched the internet and i'm struggling to find an answer. I'm using the HD650 and HD25-1 II headphones.

    ReplyDelete
    Replies
    1. Jaycar Electronics has a suitable 240 VAC to 12 VAC power adapter. You can find more in the O2 Details article under the Components section and/or in the Bill Of Materials (linked under O2 Resources). Also search the O2 Details article page for "adapter" as there are many comments about power adapters and more in the O2 Summary article.

      Delete
  56. GOt mine yesterday ... mounted it in a stand alone box w/ nice RCA connectors .... SOUNDS OUT OF THIS WORLD!!!!!!!!!!!!

    It lifted a blanket from between me and my awesome AKG's. THey sound so clear in a way I've never heard.

    I bought my AKG 701'because they were listed as a flat neutral set of cans, matched with the ODAC and the O2 amp .... I belive I'm hearing pretty close to whatever the studio mix's sounded like.


    Thanks NWAUDIOGUY ..

    TIm
    Indianapolis

    ReplyDelete
  57. I'm using a DT 770 pro-80 on Fiio E11. The Fiio E11 has low gain mode (6db or 2X) and high gain mode (12db or 4X). On low gain mode, it gets pretty loud at volume 2.5 to 3.0. If I were to build on O2 unit for myself, what gain settings should I set for use with this headphone? I'm thinking 1X-1.5X for low gain and 2X-2.5X for high gain. Does this seem right?

    ReplyDelete
    Replies
    1. I own the DT770 Pro 80 and it matters a lot what your input source is as to how you set the gain. With an iPod's 0.5 volt line out (LOD) 3X gain works well. With a 2 volt "home" source, however, you might be happy with 1X. There's more info in my More Power and All About Gain articles.

      Delete
    2. Thanks for the links. Along with a couple other pages I shoudl set the minimum gain to 1.4X and the maximum gain to 2.0X

      Here's wat i got,

      #1 Beyer Dt 770 pro-80 is 2.8 volts RMS

      #2 my 'home' source ASUS - Multimedia- ASUS Xonar DX says its 2 volts if i'm correct. FYI, the headphone isn't loud enough with this input source alone. I need an amp to make it louder for my taste.

      #3 not tottaly sure ow much excess gain i want but 3 db is reasonable maybe?

      but anyway I did your gain calculations to get 1.4X (or 2.9 dB) minimin gain. Another 3 dB excess gain is 5.9 dB (1.98X) of maximumg gain.

      is it ok to set it to 1.5X minimum gain to 2.0X maixmum gain?

      Delete
    3. "my 'home' source ASUS - Multimedia- ASUS Xonar DX says its 2 volts if i'm correct. FYI, the headphone isn't loud enough with this input source alone. I need an amp to make it louder for my taste."

      Note also that this sound card has an output impedance of 100 Ω, which, among other unwanted effects, reduces the maximum voltage at most frequencies to less than 1 Vrms with the DT770 Pro 80 as the load.

      Regarding the gain, you may want to keep the 2.5x option if you plan to buy higher impedance headphones in the future, while 1.5x should be fine for the DT770-80.

      Delete
    4. alright thanks!..i'm' going to keep my other options open for gain settings..

      i.e. i'm starting off w/ 1.5X/2.5X, then other options would be 3X, 3.5X 6.5X 8X... shoudl ineed to change them out...

      as for what you said:
      "Note also that this sound card has an output impedance of 100 Ω, which, among other unwanted effects, reduces the maximum voltage at most frequencies to less than 1 Vrms with the DT770 Pro 80 as the load"
      what unwanted effects are you talking about?
      and.....
      why does an output impedance of 100 Ω reduce the maximum voltage to 1 VRMs with the DT770 Pro 80 as the load, while the website says 2 Volts?

      REgarding the gain, i agree with you that a higher impedance headphone should warrant a higher gain setting. a higher impendenace headhone would have higher ohms which needs more volts so headphone volts divided by xonar dx volts = higher gain setting... i would say the xonar dx, on higher impedeance / ohm headphones' load, ouputs more than 1 volts and up to 2 volts.

      Delete
    5. The "Headphone & Amp Impedance" (February 2011) article on this blog explains why you want the output impedance of an amplifier to be the lowest possible when driving headphones or speakers.
      The voltage is reduced to less than 1 Vrms because the output impedance with the load forms a voltage divider, therefore, with a 80Ω load, the voltage drops to 2 * 80 / (80 + 100) = 0.889 Vrms. Since the headphone impedance varies as a function of frequency (see "Beyerdynamic DT 770 Pro 80 Review" from April 2011), the actual voltage will depend on the frequency, i.e. the frequency response will not be flat. That is one of the problems caused by a headphone amplifier with high output impedance.
      When the sound card is connected to the O2 instead, the output impedance is not an issue, and there is no significant voltage drop as the O2 has an input impedance of about 10 kΩ.

      Delete
  58. FYI, the ODAC will in fact fit in the B2-080 case with the batteries still in tact. pics can be seen here: http://www.diyaudio.com/forums/vendors-bazaar/211184-odac-offerings-epiphany-acoustics-20.html#post3040133

    ReplyDelete
  59. I apologize if this question has been asked or answered already:

    Presuming Murphy's Law, is the ODAC adequately protected in the event someone plugged headphones into the 3.5mm jack?

    ReplyDelete
    Replies
    1. I don't recommend it but it's not likely to cause harm. There's a series resistor and the output is inherently current limited.

      Delete
  60. Still waiting for Timex vs. Rolex. Keep up the good work!
    -B

    ReplyDelete
  61. got mine and it sounds clean and beautiful. compared to the uca202 (dac part only) there is a notably clarity in the upper ranges and seems tighter. granted, this was not a blind test so take it as a whim.

    thankyou nwavguy :]

    ReplyDelete
  62. With the ODAC installed in the O2 does the O2 still allow for input in the O2 line in jack when the ODAC is not in use?

    ReplyDelete
    Replies
    1. Yes. See the revisions to the article above.

      Delete
  63. UPDATE: I've posted photos and hopefully enough details to help DIYers properly connect their ODAC to their O2. See the Tech Section in this article above.

    For those wanting to solder on their own 3.5mm jack to the ODAC board, the one I used is the CUI SJ-3523-SMT from Digi Key. I haven't verified the footprints and dimensions, but the Switchcraft 35RASMT2BHNTRX and Kobiconn 161-4034-E from Mouser may also work but check the datasheets. To enable the jack you need to jumper across both the pads leading to the jack on the bottom of the ODAC board.

    We were also at 190-some comments here nearing the magic 200 mark so I went back and cleaned out some of the redundant, and far off topic comments to make more room to try and avoid the "Load More" situation.

    I feel bad I don't have the Timex vs Rolex article up yet along with more progress to report on the ODA but I've had several higher priorities taking most of my time. I'm still hoping to have the ODA design more or less finalized this summer. Being able to buy boards or finished amps will likely be in the fall.

    The good news is the ODAC feedback so far has been overwhelmingly positive which is encouraging. Sales have exceeded YoYoDyne's expectations.

    ReplyDelete
    Replies
    1. I believe that the Kobiconn equivalent is the 161-4033-E, which is "stereo" and has no internal switch.

      Delete
  64. Thank You for the detailed pics and explanation on how to wire this bad boy up! Will cutting the traces on the input jack so that its use may be retained adversely affect the noise level/sound quality of the amp?

    ReplyDelete
  65. Is the ODAC power regulated well enough so there aren't any problems if I use a USB y-splitter for power boost with an iPad but let my kids use it too and maybe they plug in everything and get double the power?

    ReplyDelete
    Replies
    1. I have no explored iPad some use. Some early reports indicate it can be done but you're on your own. Apple has gone on record saying the iPad does not have a enough power to run external DACs like the ODAC that require over 100 mA of current.

      Delete
    2. Let me rephrase: if I plug the cable into two full power USB ports (thought this is not my intent... just a possible accident) and then into the ODAC, will it fry? From what I can tell, that's going to be in the range of 6.5 - 10V.

      Delete
    3. Double USB cables connect the power in parallel. That's still a nominal 5 volts but with twice the available current. I'm not aware of any way to fry the ODAC with USB power. If you do somehow manage to feed it more than 5.5 volts, yes, that may harm it.

      Delete
  66. My ODAC arrived two days ago here in New Zealand. I didn't get it to use with an O2 but have it wired up straight to my power amp. Software volume controls on the pc work fine. The sound is fantastic!
    Thanks for making possible such a great and affordable piece of hi-fi gear.
    B.B.

    ReplyDelete
  67. diyAudio ODAC - European and Worldwide pre-orders. #124 mabur

    [ Disclaimer: This brief ODAC review is based on my own limited listening sessions to date with the ODAC + O2 combo during last weekend. The opinion is my own and based on my own impressions.

    Setup: I started by soldering the supplied 3.5mm stereo jack onto the ODAC board and bridging the two pads, which connect the stereo jack to the DAC chip output. I then connected the ODAC to a Windows 7 64 notebook with the enclosed mini USB cable. The device was immediately recognized as a sound device and I enabled 96KHz with 24 bit resolution in device settings. I connected the ODAC output via a short, shielded, self-made connector cable to the O2 input and the O2 output to AKG K550 headphones.

    Sound: What immediately struck me, when I listened to selection of tracks from itunes, which were either lossless or compressed, is how much “more” of the music you can hear i.e. my brain focused on clarity of the sound and additional details that have been veiled before. Needless to say, all compressed music is pretty much unlistenable. I could easily hear distortions in AAC compressed music content purchased from Apple (256kbit) which was a new and unexpected thing for me. Since this clarity is amazing, what seems to happen, is that it initially kind of overshadows the lower frequencies (the brain focuses on the things in the sound that were not hearable before) and seems to let you think that they are somehow missing from the content. At one point I was afraid, thinking to myself, where the bass has gone, however, after a while, the brain re-adjusted and then all I heard was music, perfectly balanced and clear. The ODAC has dead silent background on my AKG K550 headphones, which are quite sensitive. I have not tried listening with other headphones or earphones yet since now my son is the happy owner of the ODAC (I also built an O2 amplifier before for him) and he confirmed the amazing clarity (I have not warned him) and that he could also hear compression distortion (that has not made him very happy), even on his Koss Porta Pros, although these headphones are not best known for analytical or discerning sound... ]

    ReplyDelete
  68. Anonymous Stat guy:

    Thank you NwAvGuy, the Benchmark dac 1 has always been my favorite dac listening to, even before the NwAvGuy development/s. Thank you for such hard work, there are many who appreciate it. Thank you also to your codevelopers or guides who helped out along the way.

    Their work does not go unappreciated.

    I've got 60 minutes on the thing so far, over the last day. Never before have I thought I would get something so close to the Benchmark dac 1 for such a cost. There are many of us who this technology would not be available to without this effort. Keep writing and keep up the good work.

    Thank you.

    The machine work from JDS Labs was also beffiting of the 150$ price tag. The StandAlone ODAC's are beautiful, anodized black case with similar but higher quality (improved) panels. It's about the size as a pack of gum.

    ReplyDelete
  69. The Stat guy:

    Back off to school so I can hope to contribute to the world in quite the same way you have!

    I'm sure the work was tireless, and was probably exhausting, at times. Thanks NwAvGuy.

    There is a Chinese saying: "May you live, in interesting times." The times have changed, and they are definitely interesting these days. Thank you for your moral courage. Congratulations.

    ReplyDelete
  70. This is kind of off-topic, but I was wondering what your thoughts are on the latest Headfi squabble. There are people, mostly modders, who are saying that headphone companies - Beyer, AKG, Audio Technica - among others are releasing incomplete headphones because they have no dampening which causes resonance which makes certain frequencies sound wonky. Seeing as dampening materials are pretty cheap relative to the cost of a headphone, I would think they would have reasons not to put dampening, but these people also argued that even the cheapest speakers have dampening materials. What are your thoughts?

    ReplyDelete
    Replies
    1. If the $25 Senn HD201 lacked damping material I'd say it may have been a cost issue. If $300 AKGs lack it, I would suspect it's intentional. I'm not a headphone design expert, but in speakers, you often want a certain amount of damping that can be obtained in different ways. Over-damping can be just as bad as under-damping. I haven't seen the HF comments, but they may at least partly be fueled by the common audiophile myth that more damping is always better.

      There are at least three kinds of damping in headphones. Damping of the driver itself, damping of the air space in the enclosure, and damping of the enclosure. And there are multiple ways to accomplish all three, some of which may not be obvious just from visual inspection.

      Delete
    2. Even really expensive headphones can benefit from this. If you look up the "HD800 "Anaxilus" Mod" you can see some before and after measurements. They show a show some changes to the frequency response which may be a matter of taste or design but they also show a reduction in stored energy around the 5-6khz region on the waterfall/CSD plot which isn't something you can easily argue as being better, more accurate, or probably even pleasant to most people. Tyll did an article on IF about this mod as well with extensive measurements.

      That second part is really what this whole issue is about. Some people, like me as well as HF user purrin who did those CSDs, appear to be much more sensitive to those kinds of issues than others and consider them to be pretty glaring flaws.

      Delete
    3. You don't even have to look as far as head-fi, or limit yourself to $300 models - just check Tyll's page: http://www.innerfidelity.com/content/diy-modification-sennheiser-hd-800-anaxilus-mod

      At some point I wonder if these people - given the chance - wouldn't even go as far as encouraging the recording engineer to have a glass or two before he sat down to do his work, so as to bring out that final smoothness in the recording.

      Delete
    4. Taking an objective approach like the waterfall plots could have merit. But what may appear to be a flaw could be part of the headphone's "sonic signature" and was intentional. I would think, as with speaker enclosures, you want the earcup material itself to be well damped so it makes as little contribution to the sound as possible. But the damping of the driver, which depends on many factors, is very different. Just changing earpads from leather to suede will change the sound of headphones because they damp the semi-enclosed chamber between the driver and ear differently. But ultimately, the change in sound comes down to subjective preferences.

      Delete
  71. Thankfully, Headfonia have reported that they're not going to grace the ODAC with another of their "reviews". I suppose this is mildly regrettable from one perspective - no such thing as bad publicity - but then again this is a site which writes things like the following with the utmost seriousness:

    "At the multi thousand dollars price bracket, there is no bad amplifiers."

    This could perhaps be amended to "At the multi thousand dollar price bracket, there are no bad reviews."

    I don't mean that in a conspiratorial sense, just an-obvious-result-of-expectation-bias sense!

    ReplyDelete
    Replies
    1. They're also comparing open design headphones with closed design headphones (see recent K550 review) giving courageous statements that a very specific open headphone is grainy, has vague soundstage and there are tons of other totally subjective statements, when compared to that closed headphone they're reviewing...

      So after reading that article I've come to conclusion that I must be probably deaf with my K702s and the reviewers @ headfonia are probably very serious sound engineers with many decades of practice and experience, and with golden (y)ears of course, because they are able to pick even the slightest difference between the black background and a much blacker background :-D Also statements about the graininess made me ROFL :-D Who knows, maybe they've been using some broken tubes when A/B'ing those headphones :-D

      It's very funny reading for "objective" people, but it's sad that many headphone novices out there are reading and trusting headfonia's reviews... :-(

      Delete
  72. I've wired up my DAC to my O2 amp but the left channel is very soft. I've done a continuity test and it suggests that the connection is okay. Anyone have any ideas? Could it be a grounding problem?

    ReplyDelete
    Replies
    1. Perhaps you swapped the ground and signal in one channel (the "striped" and "unstriped" wires in my photos)? Or you may have a solder bridge somewhere. Have you confirmed the O2 still works properly with another source?

      Delete
    2. The O2 works with my Bifrost DAC. So I know its not the amp section. I'll rewire the ground wires.

      Delete
  73. Love your work NwAvGuy!

    I have an O2 amp and will soon have an ODAC.

    My setup is currently a Corsair SP2500 speakers and the O2 amp for headphones. The thing that I don't know enough about and that concerns me is the split to both of these things.

    I'm currently doing a 3.5mm split out of my Xonar DX2, when then go to each amp. Volume is fine on both, but someone had mentioned to me something about this perhaps causing imbalanced impedance.

    They didn't go beyond that, but, through a little Googling on the issue, apparently this impedance imbalance can lead to some kind of reflecting in the cable, causing some degradation in the signal quality and can also cause the lower impedance side to short out and damage something.

    Is this current setup safe? I've been using it for about a week without an issue, and everything appears to sound great. I know there is a lot of misinformation and are many snake oil theories out there, and what I said above is pieced together from a few sources. So I figure you're the perfect person to set me straight on all of this.

    Also, once the ODAC gets here, assuming the above setup is safe and of acceptable signal quality, would the ODAC support the same split? If not, would I need to just use the ODAC for my headphones and I guess run the speakers off the sound card?

    Thank you much!

    ReplyDelete
    Replies
    1. The "reflection" issue of impedance mismatch only is a concern at frequencies much higher than audio. So you don't have to worry about that. You do need to make sure both pieces of equipment receiving the signal are always turned on. If one is powered off while you're listening to the other one it may present an improper load (the input impedance of most devices drops dramatically when they're turned off).

      The other possible concern is the total impedance should be about 5K ohms. If both inputs are 10K or higher each, everything should be OK.

      Delete
  74. NwAvGuy, Thanks for your dedication and work! I have the O2 and love it and have an ODAC on the way! I have a quick question that hopefully hasn't been answered elswhere. I built and then installed my ODAC in the larger B3-080 case (I think that is the model number) and I did this with the intention of being able to install the ODAC above the O2 without removing the batteries from the O2. Now, thinking about it, is this going to be a workable configuration? Will I still be able to have my battery powered O2 and USB powered ODAC? It seems like kind of a silly question, but one I thought I would ask before I hook everything up and try to run on battery and screw something up.
    Thanks again for your great work! Looking forward to the ODA!
    Bruce

    ReplyDelete
    Replies
    1. Yeah, the batteries can stay as long as you can figure out how to mount the ODAC board.

      Delete
  75. Does ODAC support bitperfect playback? ASIO4ALL, WASAPI etc.

    ReplyDelete
    Replies
    1. It supports bit perfect playback without any proprietary drivers at 16 or 24 bits and 44/48/96 Khz. I don't yet know if any of the ASIO drivers will support the ODAC but I'm guessing someone has, or will, try a few and hopefully report their results. Normally the only reason to use ASIO would be for lower software latency but that's a non-issue unless you're doing live recording which the ODAC doesn't support.

      Delete
    2. I noticed that the Win7 endpoint volume control for the ODAC works in WASAPI exclusive mode.

      Am I correct in assuming that's the OS using the TE7022L's volume control?

      Delete
    3. While I believe the TE7022 can support adjusting the PC's volume (much like the FiiO D5 does with volume up/down buttons), the ODAC has no volume control of any sort.

      Delete
  76. i got a fiio e10. How will the this Odac combo be compare to fiio e10 ? listening mainly to c-pop, kpop, jpop, ballad, female voice.

    and which one should i buy HD598 or HD650 for this ?

    ReplyDelete
    Replies
    1. Did you read my E10 review? The E10 is a decent DAC if you leave the volume all the way up on the PC side. But the headphone amp in the E10 is nowhere close to the O2. I consider it a bit marginal for the HD650 but some might be happy with it. If you like the sound and it gets loud enough, keep using your E10. As for the 598 vs 650 I can't comment on that one as I'm only familiar with the 650. Did you read my HD650 review?

      Delete
  77. i need to ground the input jack to the case...i looked a t a couple of pictures

    ..
    Ground Lead Picture
    O2 ODAC INput Wiring Jack picture



    can i use a thin wire from a resistor cap thingy to ground the case from the white 'ground' that is labeled in the second picture?

    ReplyDelete
    Replies
    1. I'm not sure what resistor/cap you're talking about. But the case ground wire I use is shown the picture (it's the bare wire). You can connect your case ground there or to the center ground pin of the 3 pin header.

      Delete
  78. Well I sent a message to Corsair tech support yesterday, asking if they could give me the input impedance when on and off. Nothing back yet. I may not get anything back, since it's such a weird question.

    What is at risk of being damaged if one side is on while the other is off? I did this often for about a week before contacting you. Could something possibly be damaged already?

    I'll assume by your post that the input impedance of the O2 amp is 10K or higher. Something tells me the SP2500 amp is 10K or higher too, but hopefully Corsair still gets back to me.

    To try and solve this, I'm actually looking for a mini receiver thing that can switch the audio source so I don't have to make sure they're both powered on all the time. I don't want whatever I use to introduce any signal quality issues, so do you have a recommendation?

    Although a preamp typically isn't use for this, a preamp can have an offterm meaning in home audio to describe something like what I'm looking for. These seem to entail volume controls, which I don't really want.

    I found this. Seems pretty cheap, but I don't really care if it serves the purpose. Again, I would like to preserve signal quality, so I'll be a bit more if you know of something of quality that will serve this purpose.

    Sorry for all the trouble. Thanks!

    ReplyDelete
    Replies
    1. It's very unlikely you harmed anything. And yes, the O2's input impedance is 10,000 ohms when it's powered on. When it's off, it's more complicated, but it's as low as a few hundred ohms. That's true of many devices.

      If you're worried about having both powered on, you just need a passive switch of some sort. The ODA (if I can ever get it done!) may also solve your problem.

      Delete
  79. The only USB DAC amps I've tried produce terrible results in Windows. Static when sounds start and stop... static when the computer is busy... terrible. Will your design do better?

    ReplyDelete
    Replies
    1. I've never had any problems on any computer I've used the ODAC on. The only time I have problems with USB audio devices is when they need proprietary drivers--the ODAC does not. It uses the native Windows USB Audio support. If you're having audio problems using native drivers, something is likely wrong with your Windows installation or your PC's hardware.

      Delete

PLEASE NOTE! If you see "LOAD MORE..." OR "LOADING..." just above this paragraph, you need to click that link to load the newest comments. It may take several seconds. When all the comments are displayed the Load More/Loading link disappears.

COMMENT APPROVAL: Comments may take anywhere from a few minutes to a few days to be approved. You might want to save a copy (highlight Ctrl-C) of your comments before submitting in case something goes wrong. The approval process is necessary to prevent obscenity, etc. Please keep comments on topic. Comments promoting commercial interests will not be approved and may be reported as SPAM.