Header

Objective Reviews & Commentary - An Engineer's Perspective

November 30, 2011

Objective Desktop Amp (ODA) & DAC

INTRO: Ok, right up front you need to know this article is not one of my usual in-depth write ups. But here’s the deal. The battery powered O2 headphone amp has been more popular than I imagined with 2000+ PC boards already in circulation and more on the way. After thousands of O2 messages and comments I’ve learned many are interested in purely a desktop amp and many O2’s have gone into service sans batteries. Those considering an O2 for desktop use might want to read this “preview” article.

THE ODA: “ODA” simply stands for “Objective Desktop Amp”. The idea is to take the same objective principles behind the O2 and apply them to an amp optimized for desktop-only use (no battery power). Those principles include:

  • Genuinely Transparent Performance – I wrote an article for InnerFidelity that describes what’s required of a headphone amp for transparent performance. I was even more stringent in defining the requirements for the O2. The ODA will meet or exceed the same requirements so you can listen to everything your music has to offer without your being forced to listen to your amplifier’s unwelcome contributions. This means no audible hiss, distortion well below audible levels, and a suitably low output impedance to maintain flat frequency response and proper damping with all headphones.
  • Wide Headphone Compatibility – Music lovers shouldn’t have to worry about trying to figure out what headphones work well with what amp. The O2 has demonstrated it’s entirely possible to have one modestly priced amp that can accurately drive 98% of current headphones on the market.
  • Focus On Design Rather Than Designer Components – The O2 demonstrates design and implementation is often far more important than using trendy expensive parts. The ODA places a similar emphasis on proper implementation to get the most from reasonably priced high quality components. As with the O2, I expect the ODA will outperform many other headphone amps using more exotic components and topologies.
  • DIY Friendly Design – Several first time DIYers have already built an O2 and reported it worked the first time they powered it up. By avoiding surface mount parts, point-to-point wiring, and chassis mounted components, the ODA should be equally simple to build. Because DACs require surface mount components, and for other reasons, the optional DAC board will be available pre-assembled and tested (see below).
  • Real Ground – The ODA will not have third channel masquerading as a virtual ground to mess up the performance. It uses a proper bipolar power supply referenced to true ground and direct coupled outputs.
  • Maximum Value – Like the O2, the ODA is being carefully “value engineered” which takes into account everything from the size and complexity of the PC board to minimizing the number of different components on the parts list. Upgraded components are used where there’s a meaningful benefit or little price penalty. The ODA will use a low cost off-the-shelf enclosure.

ODA VS O2: So what makes the the ODA different and more desirable for desktop use? The main O2 thread on diyAudio is flirting with 1000 posts. And, collectively, there are another 1000+ posts and messages elsewhere. Some clear consensus stood out in what many wanted in a desktop headphone amp:

  • Optional Internal 24/96 High Resolution USB DAC – The number one requested “upgrade” to the O2 is a DAC that’s designed and measured to similar standards of high performance. Just such a DAC is well under development for the ODA (see below).
  • Higher Quality Inputs & Outputs – The O2, for size reasons, has 3.5mm input and output jacks. The ODA adds a 1/4 inch (6.5mm) Neutrik headphone jack for use with high-end full size headphones and RCA input jacks for higher quality connections and even lower crosstalk.
  • Rear Panel Connections – The ODA has the power and input connections in back to keep the cables out of the way for a cleaner look and better ergonomics (the downside being another relatively expensive panel to buy).
  • Wider Source Compatibility – The ODA should work with anything from an iPod Line Out Dock (LOD) to high output home DACs with a wide range of headphones. It has more flexible gain options compared to the O2.
  • Higher Quality Power Supply – The half-wave power supply in the O2 is something of a compromise for size, cost, and battery charging reasons. The ODA power supply improves on the O2’s in several ways.
  • Headphone Protection Relay – The O2, like many amps, produces a “click” in the headphones when you turn it on and a soft “thump” when you turn it off. The ODA uses a headphone relay to eliminate these noises and provide added protection for expensive headphones.
  • Other Possible Upgrades – Depending on how the details sort out, the ODA may have some other upgrades as well such as a possible preamp/line output.

ODA + ODAC: What’s an “ODAC”? It’s a USB DAC designed the objective way—just like the O2 and ODA. It fits inside the ODA turning it into a desktop headphone DAC. Here’s some preliminary information:

  • High Resolution USB – Relatively few reasonably priced commercial DACs, and almost no DIY DACs, support anything beyond 16 bits at 44 or 48 Khz over USB (despite most having 24/192 DAC chips). The problem is, until recently, there were no suitable options for high-resolution audio over USB without needing special drivers or spending lots of money. The ODAC will support up to 24/96 over USB.
  • 111+ dB Documented Dynamic Range – The most dynamic range you can get from a DAC operating in 16 bit mode is around 96 dB with most falling several dB short of that. While that can be enough if you control the volume in the analog domain somewhere after the DAC, it’s often not enough if you want to control the volume at your PC in the digital domain. That’s where those extra high resolution bits show their stuff. Even with the PC volume turned down, you can still get 16+ bits worth of resolution and dynamic range. That can mean the difference between hearing noise or pure silence in your headphones. And, unlike nearly every other DAC out there, the Dynamic Range of the ODA will be fully documented. Many DACs just quote the spec for the chip off the datasheet which is usually nowhere near the entire DAC’s actual performance. See the Tech Section for a preview.
  • USB Audio Class 1 Compliant – Unlike most pro-audio interfaces that support 24/96, the ODAC requires no problematic proprietary drivers for XP, Vista, Windows 7, OS X or Linux. It’s true Plug-And-Play. It also does not require UAC2 drivers, like some DACs, which are not provided in any current version of Windows.
  • ODAC vs DAC1 – I’ve done some preliminary blind testing against my $1600 Benchmark DAC1 Pre with a variety of music and my best headphones, and so far, a least two different people cannot tell them apart. The DAC1 Pre has won a lot of professional accolades for being one of the better DACs money can buy at any price. I can see a formal ODAC NwAvGuy blind listening challenge coming up.
  • Redbook Compliant Output – For many reasons, it’s good to have a DAC that complies with the Redbook standard for digital audio output levels. Many USB DACs, like the AMB gamma, can’t produce the required 2 volts RMS of output and suffer lower dynamic range. And some go way over the specification which risks overloading the inputs of a lot of equipment. The ODAC is Redbook compliant. Among other benefits, this makes it much easier to do level matched A/B comparisons with other DACs and digital sources that are also Redbook compliant.
  • Pre-Assembled Board – Because USB and DAC chips are typically fine pitch surface mount devices they’re not very DIY friendly. To make matters worse, if you want 24 bits over USB with native Windows drivers, I’m not aware of any suitable chips that are available to DIYers. All the current solutions require licensing and/or some sort of contractual agreement. They’re not stocked by DigiKey, Mouser, Farnell, etc. And to further complicate things, they also require custom programming before they will operate correctly. That’s 3 strikes against DIY for high resolution USB DACs. The obvious solution was finding someone to co-develop the board with and handle all the contractual, assembly, programming and financial details so they could offer a pre-assembled board.
  • Line Input Retained – The ODAC won’t defeat the line input on the ODA. You can use both sources or even plug another Redbook compliant DAC in for direct A/B comparisons (even better if a friend does the switching out of the listener’s sight!).
  • Standalone Capable – The ODAC is designed so it’s also usable by itself for other applications while operating entirely from USB power. No ODA required.

NO MONEY FOR ME – There’s been some speculation that an assembled DAC board is a chance for me to finally make some money. But that’s not the case. There’s no money in the ODAC for me. As with the O2, and ODA, I’m leaving that part for others to sort out. This blog, and hence my reviews and commentary, remain entirely non-commercial.

WHAT THEY’RE NOT: A few things the ODA and ODAC are not:

  • No S/PDIF – There’s little need for an S/PDIF output as it seems unlikely anyone would want to use the ODAC as a USB-S/PDIF device for another DAC. And while there are a few applications for an S/PDIF input it’s not something most will ever use. For everyone else, an S/PDIF input would add complexity, size, and substantial cost with zero benefit and may degrade jitter performance. That’s not keeping with the “best performance for the lowest cost” O2/ODA approach.
  • No Balanced Outputs – Balanced audio is great for pro use with long cables and electrically hostile environments. But it’s generally more of a liability than an asset in home audio gear. All else being equal, balanced stages usually have higher overall noise and distortion than their unbalanced counterparts. Headphone drivers don’t magically somehow work better when driven by a balanced source. Most of the stuff you hear and read about balanced audio for home headphone gear is myth, hype and even snake oil.
  • No Arc Welding – If you’re looking for an ultra high output amp for those vintage AKG K1000s, or to weld that broken light fixture, the ODA isn’t it. Just like with S/PDIF, it doesn’t make sense everyone should pay for power they can’t use (and that could also more easily damage their headphones) just to support a tiny minority of potential users with unusual requirements.
  • No Battery or USB Only Power – The ODA isn’t intended to be portable. That’s what the O2 is for. The ODA needs a proper power supply for the best performance and should outperform any USB powered headphone DAC. The ODAC board by itself, however, can be powered from USB for use in other applications.
  • No Recording – The ODAC is playback only. If you want a microphone input for Skype calls, check out the $12 Syba CM-119 I reviewed. If you want to record a string quartet, there are lots of great reasonably priced professional USB audio interfaces optimized for recording rather than playback.

DAC INTERPLAY: I’ve held back on some ODA design details waiting to see if the ODAC was going to be a reality. So in some ways, the ODAC has come first. The ODA PC board layout and other details are subject to change greatly depending on what form a DAC option might take. Because I want to remain non-commercial and not profit from any of this, I had to find someone interested in offering the ODAC for sale in pre-assembled form. It might be a bit premature, but so far the ODAC is looking viable and I’m very pleased with the performance and testing so far. It’s been a joint effort.

HOW MUCH? Many have asked about the cost. I’m guessing the ODA itself will be around $30 more than a complete O2. That would put the total DIY price somewhere around $130 without the DAC option. Just the board would be under $70 complete and fully functional if you build it yourself. And I’m told the optional assembled, programmed and tested ODAC board should be under $100. These are all just estimates and subject to change. The more popular the ODA and ODAC become, the lower the price will probably go.

WHAT’S NEXT & TIMING: My current plan is to publish another ODA/ODAC article sometime in December with more details, more test results, etc. Depending on how things go, the ODA documentation package (including PCB artwork), and even the ODAC board, could be available as early as late January. Being realistic, February is more likely and it might slip into March. It all depends on how many board revision cycles are required, what changes/problems come up, etc. The O2, for example, had a last minute re-design and delay because the main distributors ran out of volume controls and a few other critical parts. For now I just wanted to let everyone know there’s progress, and if you’re thinking about an O2 for desktop use, or perhaps another desktop headphone amp or headphone DAC, you might want to wait for the ODA.

 


TECH SECTION


DYNAMIC RANGE BACKGROUND : Probably the biggest spec thrown around by DAC chip makers for bragging rights is the dynamic range (DNR) performance of their chips. Some sleazy audio DAC marketing types just publish the number from the datasheet as the spec for their completed DAC. But that’s usually cheating in a big way and a bit like saying the tires on your car are rated for 155 MPH so that must be how fast the car can go. Uh, no. But then again some of those designing and selling USB DACs probably have no way to measure the actual DNR so perhaps that’s why they cheat. Getting even close to the chip’s ultimate performance requires great care with the power supply, PC board layout, grounding, and more. For 112 dB of dynamic range with the Redbook standard output of 2 V rms at 0 dBFS, there can be only 5 microvolts of total noise. That means all the noise in the audio band added together has to be less than 5 microvolts! That’s hard to do even if the DAC chip itself were somehow noiseless and that’s hardly the case with lots of noisy digital signals just a few millimeters from the analog pins and noisy USB signals not much farther away.

ODAC DYNAMIC RANGE: Shown below is the performance of an early prototype of the ODAC. The final design could be worse, but hopefully might be even better. Regardless, the results show nearly 112 dB of genuine dynamic range (A-weighted or “dBA” as is the industry standard). Most USB DACs are limited to 16 bits and around 90 dB plus or minus a few dB of dynamic range. I used a –60 dBFS signal instead of my usual –90 dBFS as the former seems to be more of an industry standard for this measurement:

Odac 1 Khz -60 dBFS Noise & Dynamic Range 10K Ohms BW=22Khz (ref 2 Vrms)

 

WHY 24 BITS CAN MATTER: Many find it convenient to leave their headphone amp turned up and use the volume control in your PC’s operating system or player software. For one thing it allows locating the headphone DAC out of reach and/or out of sight. But when you do that with a typical DAC capable of only 16 bits over USB you get less than 16 bit audio delivered to the DAC at anything less than maximum volume on the PC. Turn it down only 6 dB and you have 15 bit audio. Another 6 dB and you’re getting only 14 bits. And so on. Here’s a graph of the ODAC running in 24 bit mode in blue. It’s playing a 6 Khz test track just below maximum at –1 dBFS but the volume control in Foobar was turned down about –45 dB. So the end result is a signal that’s about –46 dBFS being sent to the DAC. In 24 bit mode the THD+N was impressively low at only 0.01% relative to the –46 dB signal. Shown in yellow is the much higher noise floor in 16 bit mode without changing anything else. The dScope can’t show FFT readings from previous sweeps easily, but the 16 bit THD+N was nearly 20 dB worse at about 0.08%—or eight times as much noise and distortion in 16 mode (As has been pointed out in the comments, the readings in this graph may not be correct but the spectrum is correct. I plan to re-run the test):

Odac 6 Khz -1 dBFS Vol = -45 dB THD & THD N 24-44 (blue) 16-44 (yellow)

 

MORE TO COME: Check back in December for more on the ODA and ODAC!

208 comments:

  1. Great initiative with the DAC. Of course, I just got my E10 2 days ago which has the big benefit of being able to do USB power and being really really tiny :P

    ReplyDelete
  2. One feature that is missing from most amps and many DACs is balance control. There is a substantial amount of people who have slightly unbalanced hearing, for one reason or another. It would be great if this amp/dac has this functionality.

    ReplyDelete
  3. "Most of the stuff you hear and read about balanced audio for home headphone gear is myth, hype and even snake oil."

    Please make this the subject of a future post !
    I would especially like to see no-nonsense measurements of so-called 'balanced head-phone amps' (actually 'four-channel amps'), fed from a balanced source vs unbalanced amp .
    I would think that just the fact that you get 'free gain', so to say, is measurable ?

    ReplyDelete
  4. cant wait for the group buy!

    Will the ODAC have its own outputs? (e.g. can I give it its own enclosure?)

    ReplyDelete
  5. Finally, the curtain opens, eh? Very nice article, summarizing the information well in a manageable amount of text.

    While I'm not currently interested in the ODA, I am very interested in the ODAC for standalone use as a DAC with my amplifier. I'm excited about it, and waiting for the details. I assume the ODAC board will support being powered from a DC wallwart without an isolator, or filter USB power well, since the USB power of two of my motherboards has ripple ranging from 30-50mV, and my desktop, depending on the workload of the southbridge (SATA, Ethernet, PCI) has
    some ugly transients (all measured on a Fluke 196C).

    Just a small request NwAvGuy, when (if) you make another article about the ODAC, it would be interesting to see how it performs from an absolute latency perspective.

    ReplyDelete
  6. Holy crap this is awesome!

    ReplyDelete
  7. I cant wait to get my hands on one of these. The only reason I didnt scoop up an O2 was the lack of USB DAC. Keep up the good work!

    ReplyDelete
  8. Thanks everyone for the comments.

    @jupitreas a balance control is an interesting idea. Not especially cheap to implement, but I agree it could be useful. Any input from others on that?

    @Peter, I do plan an entire article dedicated to balanced, bridged, 4-channel, etc. but the ODA/ODAC is a higher priority for now.

    @MikeyG the ODAC will be useable with its own outputs.

    @Ferongr, when using the ODAC stand-alone no power supply is required and yes there's some carefully designed (and measured) USB power filtering. I hadn't thought about measuring latency, are you concerned about sync with watching a video or ???

    ReplyDelete
  9. I'd actually be interested in latency as well.

    I don't notice any issues when just watching normal videos/movies but I do edit music videos from time to time. I have no way of measuring it myself, but my USB DAC seems to have higher latency than my internal soundcard. Tight sync can be pretty important when I'm trying to time effects down to a single frame.

    Its probably not important to most people, but I remember being impressed by the latency figures given in the DAC1's manual.

    ReplyDelete
  10. I'm especially sensitive to visual/aural timing discrepancies, from my early competitive Quake III days, a twitch shooter where visual and aural feedback matter. Today, I'm still very sensitive to video not being synced well with audio, especially on action material with gunfire and muzzle flashes, and do like to casually game from time to time.

    Some external interfaces use generous buffer sizes to avoid buffer underruns and skipping under heavy CPU usage (USB is very CPU-dependent), and that results in the audio being late by a fixed amount.

    While I doubt that any well executed interface introduces latency, it wouldn't hurt to measure it for documentation's sake. Anything under 40ms should be good on today's digitally-driven LCD panels that themselves have input latency (not panel response time) from 10 to 50ms.

    ReplyDelete
  11. re: latency, I'll have to do some research on if that can be measured reasonably easily. With pro audio interfaces it's done by looping back the output of one channel through the other channel's input and observing the interchannel delay. That yields the total latency of the A/D -> D/A process. But with just a USB D/A, you need some way to know when the signal reaches the DAC (or is sent to the PC's USB driver) vs when it appears at the analog audio output. And the nature of USB is such that's not something even most high-end scopes can probably trigger on. To a scope the USB bus just looks like a bunch of random serial bus activity.

    I have a USB real-time protocol analyzer that can do complex triggering, but it doesn't have a hardware trigger output so there's no way to correlate a given USB event with what's happening at the DAC's analog output.

    You could look at the I2S signal into the DAC chip itself, but that would exclude the latency of the USB interface and delays on the PC side.

    Some gaming hardware will flash the screen and use a microphone to pick up an audio "click" over the speakers and compute the difference between the flash and the onset of the click to compute latency as experienced by the gamer. I'm not aware of any reasonably easy way to pull that off for testing DACs unless such software already exists? But even then I would need to come up with the optical hardware for "reading" the screen flash and feeding the result to a scope.

    If anyone knows of a relatively easy way to measure the latency of just a USB-only DAC, please let me know?

    ReplyDelete
  12. @jupitreas...
    @NwAvGuy...

    I can manually balance the output directly from my Windows 7 PC if I go into the advanced audio panel. I've seen this feature in XP and Vista as well, but I can't comment on other OS's. Since this will have the resolution available, being 24-bit, you shouldn't have a problem tweaking the left/right volume sliders independently.

    To get there on Windows 7, right click the speaker icon in the system tray and select "Playback devices" then right-click the playback device (ADI SoundMax "Speakers" in my case) and select Properties (bolded). From there, go to the Levels tab and select the Balance button to get into it. You can also get to it from Control Panel but the route is just as long. It's not convenient to get to but once it's set, you should be good.

    One interesting thing is I've noted that Windows 7 volume control isn't linear. It's nice that it's not because you don't have to be ultra fine in moving the sliders when down low like I experienced in XP, but I start to get detectable-by-my-ears audio compression once I've dipped below 50% even with my 24-bit DAC (the one at home... not the ADI Soundmax). Just something people should just know about.

    ReplyDelete
  13. About latency, ASIO4ALL will likely support this device and can drive the latency down to ~32ms or less according to my observation with my PG42-USB. If anyone can hear that in an annoying way, then they're part insect because that's 32bps territory and it should just sound like regular imperfect reality and not some actually apparant delay. Anyway, WDM and MME roundtrip latency is about 12x that (from what I've seen) and makes for more time spent with sync in my DAW when I've recorded to a backing track.

    ReplyDelete
  14. A not really scientific (as in not suitable for an AES paper) but pretty accurate method:

    Recording one channel from each of two interfaces playing similtaneously, the one being a low latency one, and the other the ODAC will show the added latency of the slower one. The comparisons are more "apple-to-apple" if both utilize the same connectivity method (e.g. USB), though obviously you don't get the absolute latency of the hardware.

    Can USB audio-class devices ask the OS for a software buffer of a specific size? If that's true, then measuring from the USB outputs and downstream might not show the whole picture.

    ReplyDelete
  15. I'm not sure how to play two interfaces simultaneously in a way where you know how much latency is due to the driver buffers and hardware host controller buffers in the PC? The Host Controller, and perhaps code in the PC side of the USB driver stack, can only do one thing at a time and it will probably start one stream playing before the other. I don't know how much of a difference that can be.

    And what do you use as the reference interface and how do you know you have an accurate latency number for it for playback only? It's an interesting topic!

    In thinking about it further, if video sync is really the greatest concern, it seems playing a video file with a user's favored software player (ideally in whatever format/wrapper/codec the user is most concerned about--VOD, MKV, MP4, etc.) that has a "flash" and "tick" generated in perfect sync is the best way to get the desired measurement with an optical sensor reading the flash and feeding one channel of a high speed scope and the other channel looking at the audio out of the DAC.

    If someone knows of, or wants to produce such a video file, I can whip up the optical sensor? I bet someone already has a patent on this... :)

    ReplyDelete
  16. You can use Virtual Audio Cable for this (essentially it's an audio driver like ASIO4all that communicates further with other drivers) and it has an audiorepeater executable bundled with it that will do the trick).

    Regarding the reference inteface, your Benchmark DAC Pre would suffice. As for the different driver-side latency it would be ideal if it were part of the measurement and not substracted (assuming the DAC Pre operates in USB-class mode too to put both interfaces on equal footing).

    ReplyDelete
  17. I've tried Virtual Audio Cable and the repeater application on two different Win7 PCs and it crashes on both. V.A.C. hasn't been updated in almost 2 years so I'm not sure if something has changed with Win7 it doesn't like? It also has no control over the lowest layers or the USB hardware.

    I use my own custom software player I developed to feed any two sound devices on the PC with the same stream to allow ABX testing of USB DACs (that's how I've already compared the ODAC to my Benchmark DAC1).

    I can try what you suggest with the DAC1 and my own app. I can swap ports around and repeat the test several times to see if the results are consistent or fluctuate due to other variables.

    At 44.1 Khz the DAC1 is spec'd at 2.1 mS but I strongly suspect that's using S/PDIF not USB. It may even be calculated (rather than measured) based on their intimate knowledge of the data path. It's much easier to deal with latency for S/PDIF.

    ReplyDelete
  18. If the ODAC were at my house, I'd first assess if such a deep-level analysis were needed. I'm not as geeky as some of you, so I can sometimes just deal with good enough. I understand that some of you can't settle that easily, though.

    How I'd make the determination is this: I'd use my DAW with my PG42-USB mic via ASIO as the input and have the monitor output through the ODAC into my headphones. I'd then simply speak or play instruments and see if the entire circuit is giving me audible delay. If the A/D from the mic, the PC's processing to reroute the signal, and the D/A of the ODAC together don't give any audible delay from my own jawbone's resonance then I'd call it good enough and move on.

    If there were a detectable delay, then I'd want to go through the lengths you are all attempting to design, in order to determine what the latency is. I'm willing to bet it's far below 20ms, which is absolutely in safe territory. The reason I believe that is because USB interfaces rely on the PC to buffer, so they don't have long buffering built in like a TV would. Am I wrong about that?

    Also, I think Virtual Audio Cable (VAC) can simultaneously stream one source to two ouptut devices so you can do some testing that way. I think.

    ReplyDelete
  19. Hi NwAvGuy,

    I'd like to share some of my experience with latency.

    I play a PC game called "Lunatic Rave 2" or "LR2" where I've come to realise over time there are three kinds of latency you have to address if you want perfectly flawless gameplay.

    1. Input latency into a PC monitor.

    Almost every modern PC monitor has input latency, not to be confused with g2g which is usually advertised as 5ms or 2ms, a typical LCD monitor will have around 33ms or 50ms of input latency.

    You can circumvent this with 1. A CRT, or 2. A very select few LCD monitors like the iiyama Prolite E2210HDS, which you can find at digitalversus.com in their "input lag vs a CRT" testing section.
    ______

    2. Hardware input latency

    Windows 7 is set to a default 125Hz USB polling rate, that means a theoretical low of 8ms hardware input latency via USB.

    The solution is change the USB polling rate to 1000Hz - http://www.ngohq.com/news/15043-how-to-increase-usb-sample-rate-in-windows-vista-7-a.html

    This is done by gamers that want their mouse to be as responsive as possible, or the input from the Konami controller in LR2 which comes down from above 30ms to less than 5ms, the only other way to circumvent this is by using an LPT parallel connection instead of USB.
    _____

    3. Audio latency, since it's a music game about notes falling down a screen and you hit them at the red line ideally within a 33ms window and then they make a sound, any audio latency will throw you off, using ASIO4All and setting the "ASIO Buffer Size" to 64 Samples (lowest) tends to solve this, reducing the audio latency to less than 16ms, at least when it comes to internal sound-cards.
    ____

    This is my experience from an audiovisually interactive PC game based on perfect timing, however it will naturally apply to fast-action sports, music video editing, midi instrument connections and similiar.

    My only advice is to keep the selectable ASIO Buffer Size as low as humanly possible, If I'm not mistaken the Musiland Monitor 01 ASIO driver has a very low selectable buffer size.

    Looking very much forward to your reference quality DAC endeavour!

    /kiteki

    ReplyDelete
  20. @kiteki, thanks for the added info. The LCD issue is absolutely true. Lots of newer monitors and TVs have a "gaming mode" designed to minimize their latency but it still varies a lot from model-to-model.

    The 2nd issue in your list is the HID polling for mouse/keyboard input which is obviously critical to gaming, but not for watching video. I believe all UAC1 audio is polled at 1000 hz (1 mS intervals).

    The 3rd issue, ASIO drivers, is a can of worms. It's often a battle between low latency (small buffers) and avoiding audio glitches (bigger buffers). Glitches are not a big deal in gaming, but they're very unwelcome in high quality music listening. And they're unacceptable when recording high quality audio. The pro-sound forums are full of people complaining about ASIO problems. Either they can't get advertised latencies, or they have drop outs.

    It's my experience native UAC1 USB drivers are much less problematic than third party ASIO and much more widely used for just listening to music and watching video. So what seems to matter most for the ODAC is how does it compare to other UAC1 DACs for latency? It might be the hardware itself is a relatively insignificant portion and switching to ASIO with the right settings might make a much bigger difference.

    The other option, of course, is to use one of the many software video players that lets you tweak the delays to assure near perfect audio sync with your particular monitor/TV and hardware. Some A/V receivers also have adjustable delays.

    In the big picture, for an audio playback DAC such as the ODAC, I think this is a fairly small issue. But the comments here do have my curiosity up so I'm going to do some testing and more research.

    ReplyDelete
  21. I very much like that the components can be used independently. (DAC or AMP)

    I guess we'll be waiting a while but I'm not in a rush. Very exciting.

    "@jupitreas a balance control is an interesting idea. Not especially cheap to implement, but I agree it could be useful. Any input from others on that?"

    You can do this in windows with your DAC so why bother? I have uneven hearing and that's how I do it (this has already been said but I'll say it again anyway).

    ReplyDelete
  22. Love your O2 amp, can't wait to see the ODA amp too.. But im slightly disappointed we cant DIY the DAC portion our self... :(

    ReplyDelete
  23. It's great that the ODA+ODAC will be available soon. Many people have been waiting for it.

    I'm not sure how latency will affect the 98% of the users; which is the same reason why you wouldn't include a recording input. It's better to focus on sound quality rather than making it pro-audio. There are products that are better and cheaper for that purpose.

    For me, I think this project will be way out of budget. On the other hand, I'm still waiting with a lot of anticipation your technical review on E10; how it compares to E7 as a DAC (same chip but with 24/96Hz enabled), and how it compares to E11 as an amp (same opamp without the three channel problem).

    I wonder if a 24bit/96Hz DAC like the E10 is any better than something like that E7's, when handling 16bit/44Hz or 16bit/48Hz recordings. Can audio upsampling improve sound quality like in image upscaling?

    Will E10 be reviewed soon? I'd like to buy it as a standalone, because it's like an upgraded E7 + E11, for only the price of E7 or lower. The headphones have owned so far have been easy to drive like Creative Aurvana Live! and Fischer's FA-006, so I sold the E7 because it didn't make a difference to my laptop onboard. But now that I'm getting Brainwavz HM5 which is 64ohm and should need 0.115 Vrms or 0.19mW to reach 90 dBSPL based on innerfidelity's measurements on FA-003 (same type of drivers in the same housing). What do you think?

    Much appreciated as always. Please don't let the nay-sayers (from head-fi for instance) put you off.

    ReplyDelete
  24. I must say I'm still really happy with the O2 - even in its usually fixed position. If my main amp goes though I know what I'm definitely replacing it with :D

    ReplyDelete
  25. I have all the parts and Board for the o2 sitting on my desk waiting for me to finish the tube headphone project from DIY audio...... AS FAR AS A DAC .... PUT ME DOWN FOR ONE !!!!!!!

    Tim

    ReplyDelete
  26. +1 request pre line out....

    thank you for all the time you are putting on this projects !!!!! i can only imagine how tiering dealing with the projects, documenting, giving support on the foruns -the ones that have not banned you yet :-) - while weareing the flameproff vest against some people...and living your life.... keep it up, you are a good example that should inspire other designers.

    ReplyDelete
  27. Already convinced w/ the ODA, looking forward to it! Looks awesome. Was already planning to build a desktop-only O2 until I heard you mentioning the desktop version in development.

    As for the DAC, was planning on using the Creative X-Fi Go Pro that you reviewed. I plan on using the volume control on the amp. If using the volume control on the amp, how is the ODAC better? All my music is CD-quality at best.

    ReplyDelete
  28. @madcap386, I've been thinking about that same question. I believe most programming interfaces run for $50 - $150, so that's an inhibitor. I don't want to speak for NwAvGuy, but I'm pretty sure we'll have enough detail from him to know exactly how to go about a full DIY. Licensing sometimes gets weird, though.

    ReplyDelete
  29. Thanks all for the positive comments, feedback and encouragement.

    @kingpage, I will review the E10 but I'm not sure exactly when. A good time might be once I submit the first ODA PCB to the vendor while waiting to receive the prototype boards. I'm not sure when that will be.

    "UPSAMPLING" I don't recommend upsampling audio without a really good reason. Most of the benefits of old school upsampling are already built into today's 1 bit DACs. And any sample rate conversions may have negative consequences. There's a lot of current thinking suggesting it's best to play audio at the native sampling rate. It's sort of like using your LCD monitor or TV at its native resolution. If you try to rescale the video to anything other than the native resolution, you get artifacts and it doesn't look as good.

    @Jeff, without doing a decent blind ABX listening test it's hard to say how much difference you would hear between the ODAC and X-Fi Go Pro in your application. I can say the ODAC measures better overall and much better in some ways. For an obvious example, at high volume settings between tracks, you might hear some hiss with the X-Fi but not with the ODAC. But, for some people, that's not a big deal.

    @madcap386 & akgk171, I'm sorry but for contractual (non-disclosure, licensing, etc.) and parts availability reasons you won't be able to do a full DIY of the ODAC unless you can find someone to sell you the parts "under the table" and you get your hands on the required firmware, programming tools, etc. Neither I, nor anyone else, can offer those things legally to DIYers. I think you'll find that's true with most 24/96 UAC1 USB solutions. Perhaps that will change someday with new chip introductions, etc. The only DIY solution I know of is the Atmel MCU-based UAC2 "widget" project which currently lacks driver support in Windows and is more expensive.

    ReplyDelete
  30. I too am interested in latency, because I use my computer as a real time synth, controlled by a MIDI keyboard. I realise there are many USB/Firewire audio/MIDI interfaces, but I'm not sure how many of them have very good quality headphone outputs. Certainly, the two headphone outputs on my M-Audio Fast Track Ultra (FTU) have woefully low ouptut levels. It's a shame, becuase other than this, it's a very nice interface indeed.

    The FTU allows me to use an ASIO buffer size of 128 samples, which at 44.1kHz is 2.9ms. The actual latency is a bit higher than this, due to the time it takes to clock the audio through the DACs & filters, I presume. (my DAW software does correctly report the total latency - not just the ASIO buffer latency). Would this buffer be a hardware buffer, residing on the interface?

    Using a direct USB connected MIDI keyboard, I seem to be able to tolerate about 10ms of ASIO buffer delay, when playing piano. (a MIDI to USB converter may introduce extra delay in the MIDI path, though)

    If something simple can be done to assist with latency, I think that would be great.

    CEntrance provide an ASIO driver for use with their DACport. I've never tried one, but it looks like a very nice product. (although it has insufficient voltage to drive my AKG K601 headphones to the level I seem to need for live piano playing)

    The use of computers for real time piano use, even at home, is increasing. Many digital piano owners are aware that significantly better piano emulation is available by connecting a computer to their digital piano, and all these people need is a simple, but high quality, USB headphone amp. I'm thinking that the ODA/DAC project might be useful for this.

    Greg.

    ReplyDelete
  31. @Greg, there are hardware buffers at both ends of the USB connection--i.e. in your PC's USB port controller and your M-Audio DAC. Both ends are subject to at least somewhat different latency between different hardware.

    I've made some consistent delay measurements with the dScope. It will simultaneously feed the same impulse signal to a USB device (the ODAC in this case) and to its own analog outputs (via the dScope's internal A/D section). You can then measure the delay between the ODAC output and the dScope output. The delay includes the latency of the dScope's own D/A.

    If my current measurements are correct, the delay is under 20 mS. I'll eventually post scope captures and a more exact delay value for the ODAC. I plan to double check the value against the DAC1, etc. But the good news is it appears to at least be comfortably under the threshold for lip sync issues.

    ReplyDelete
  32. Thank's for looking into latency issue NwAvGuy. While it's true you can adjust the timing of audio versus video in capable players, there are many realtime applications where this is not possible. I'd hate to have to switch my amplifier's inputs depending on the application I use, and also assign different applications to different outputs.

    ReplyDelete
  33. For anyone wanting to do a relatively un-scientific, yet very simple test on Audio/Video synchronisation, there are numerous YouTube videos that could do the trick, like: http://www.youtube.com/watch?v=szoOsG9137U&feature=related

    Oh, and regarding balance adjustement, it can be done through Sound Preferences in OS X as well. Not much use for a hardware solution as I see it.

    ReplyDelete
  34. hey..i see you've written an article in innerfidelity.. since you've come in contact with Tyll, have you tried sending him your O2 or ODA for him to try?? I'm just curious what would an "old school" audiophile like Tyll think about your amp..

    ReplyDelete
  35. @NwAvGuy: Sorry if this is a silly question, but how does the dScope send the impulse to the ODAC? Is the dScope just a software app, and because of that, it can send audio to the ODAC over USB? If that's the case, it would be interesting to try the ASIO4ALL driver, and see how low you can get it. (as someone else suggested earlier)

    Greg.

    ReplyDelete
  36. @Aksel, thanks for that. It seems a balance control is off the wishlist.

    @Anon, Tyll has expressed an interest in reviewing the O2 but he seems to have other priorities at the moment. I think the O2 is a bit "political" from his perspective (given his employer, the site's sponsors, etc.).

    @SullivanG, yes the dScope has a software app that manages both the dScope and can communicate directly with USB audio devices. I'm not sure if the ASIO4ALL driver would work or not. I'm still doing more research on latency. It's really not easy with a USB playback-only device to know you're getting a solid result.

    ReplyDelete
  37. Thanks again NwAvGuy. If ASIO4ALL does work, in addition to doing the latency test with dScope, you could also try an audio player that supports ASIO, such as Foobar, and see how small you can configure the ASIO buffer whilst maintaining glitch free audio.

    Thinking about it, you probably won't be able to do the latency test with dScope using ASIO, because it probably doesn't support the ASIO API. (it would be great if it did, though!) \Greg.
    p.s I think I pressed the Post Comment button before adding my second para, but it didn't seem to go through. I decided to take the opportunity to add more stuff. ;^) Feel free to merge if they both come through.

    ReplyDelete
  38. In my experience, while ASIO4All can reduce some of the total latency by bypassing various Windows endpoints applications normally use (DirectSound, shared WASAPI etc), it still can't change the driver's software buffer. In my case, configuring a program to use ASIO4All on my Audigy results in quite a bit of latency compared to configuring it to use the KX Project driver's ASIO endpoint that can manage latencies under 6ms (depending on CPU load).

    ReplyDelete
  39. @Jakejkjaw: I'm curious - what do you think of the iBasso D6 DAC/AMP? It looks like it has plenty of power/voltage. (although, I agree with a comment I saw from NwAvGuy I saw somewhere: their specs are incomplete). \Greg.

    ReplyDelete
  40. Another vote for pre-out, I don't believe it is essenial for it to be switchable (either manually or by inserting the headphone plug) because most users would have a volume control downstream of the PC or DAC output wouldn't they?
    E.g. the chain I use is:
    PC->EMU 0202 DAC->Creek OBH11 HPA->SM Audio passive volume control->Active speakers.
    Headphones are always plugged in, if listening to speakers headphone volume is at zero, if chnage over to headphone listening I just turn the pasisve volume control down to zereo. The creek HPA has a loop through type of connection that allows this.
    Lastly thanks for all the great content you put on this site, it's a real breath of fresh air in the audio world.

    ReplyDelete
  41. @Ferongra: FWIW, ASIO4ALL allows me to play software pianos with subjectively very low latency on my Dell E6500 laptop, using the integrated IDT high def code. I haven't done any proper latency measurements yet though. \Greg.

    ReplyDelete
  42. Thanks for the hard work. I look forward to having one of these built for me as soon as I can.

    ReplyDelete
  43. Wow. 24/96 over USB, with such measurements.

    This begs the question for me: would you rather use USB to S/PDIF converters to allow for 24/96 on a DAC that measures excellently, such as the cheap Musical Fidelity M1 (http://www.stereophile.com/content/musical-fidelity-m1-dac-measurements), or a DAC with inferior meeasurements, but can do 24/96 over USB at a similar price to that of the M1 including converter.

    ReplyDelete
  44. @Tim, you're welcome.

    @Anon, I'm not sure what you mean by "such measurements"?

    In general, USB to S/PDIF is a poor solution as it introduces yet another clock, yet another clock recovery, and yet another source of jitter into the bitstream. It's much better to have a single USB DAC that does it all in one box, with one central clock, with parts proven to play nice together.

    DACs are a lot like amplifiers in that they only have to be good enough to be transparent. Making them even better is nice but generally costs significantly more money and is unlikely to get you any better sound.

    The tricky bit is jitter. But it would seem comparing J-test data with blind listening tests there's a point of diminishing returns there as well. I plan to do more research, and more blind tests, in the future. Having the Benchmark DAC1 Pre is really helpful in that regard (a Class A Stereophile pick many have said is the one to beat).

    ReplyDelete
  45. +1 request pre line out....

    I understand your reluctance to include a S/PDIF input but how about having it as an extra like the USB input?

    ReplyDelete
  46. NwAvGuy, I've been thinking I haven't thanked you nearly enough lately.

    You've put so much of your time and money into your reviews and the O2 project for not much more than good karma for your anonymous handle. I think you're probably the best thing to happen to this hobby since Tyll started measuring headphones.

    ReplyDelete
  47. Built my O2, very happy with it. Quite, clean, powerful. Currently with the UCA222 DAC. Should have used lower gain through (HD565-ovations and HD595s).

    Haven't done any PCB soldering for about 20 years and nothing this complicated - had no problems, worked first time. I did notice that one of the power socket pins (IIRC) overlaps with the case PCB support rail, not sure if anyone else had this problem. Put some tape over it just to be sure.

    Looking forward to building an ODAC and probably a DAC only ODAC to go with my O2!

    Thanks for all the hard work.

    ReplyDelete
  48. The S/PDIF input requests are interesting. I wouldn't want it unless a few things changed around my home. I'd offer it for TV and video system viewers (and gamers). I can't imagine it would cost much to build in a jumper strip that can accomodate it. That way someone could pop the default jumpers off and plug in a ribbon cable that goes to a selector switch. Then all you'd need is some bare contacts on the PCB that they could use to connect their self-built S/PDIF daughter card onto which are routed to the DAC when the switch is thrown. That way the versitility is there but it's all on the builder to take the extra steps.

    I like that idea better than choosing either USB or S/PDIF but if you want to change your source, you either get two amps or you open your amp and swap the daughter card every time. Can it cost more than $0.60? That might be one of the cost options that's worth doing.

    ReplyDelete
  49. Thanks to Maverick and others for the encouragement, suggestions, etc. To Akg, the jumper/daughter board thing is more tricky than you might imagine--especially when the entire ODAC is already a daughter board on top of the ODA board but I'll keep your suggestion in mind.

    The DAC daughter board makes sense as it's an expensive component and some just want a headphone amp while others want an integrated headphone DAC. But beyond that, I'm not a big fan of the "modular" approach unless there are compelling reasons.

    Custom PC boards, especially in low volume, can be expensive. Connectors and longer signal paths are not free and without problems either. Every time you "go vertical" you need to really focus on loop area, ground return paths, etc. on--both of which typically increase jitter.

    I'll keep thinking about it, however. As with the O2, ideally I'd rather walk before running if at all possible. So keeping it simple for now might be best. But I also don't want to be spinning boards in the future if it can be avoided now. Because the DAC board is modular, it can be swapped out for a different one later keeping the same ODA headphone amp, enclosure, power supply, etc.

    ReplyDelete
  50. @SullivanG

    I don't think its *likely* that mods will bring down the efficiency quite that much that the full 7VRMS will be necessary for most people but it's best to stay on the safe side.

    ReplyDelete
  51. NwAvGuy, Jupitreas - I fully agree about pan/balance controls in a DAC. Having worse hearing in one ear, I have to rely on pan/balance controls to 'fix' the sound for me. With most portable players that means using a hacked/bridged inline volume control to act as balance - since they all omit this most basic of audio settings, next to a volume control. Also, Windows 7 seriously messed up the audio mixer, with the balance hidden in some godforsaken sub-sub-sub-menu, and barely usable, since the balance resets when one lowers the volume level to minimum. So, a balance control would be highly appreciated. (AFAIK, only Violectric offer 'hardware' balance in some of their products.)

    ReplyDelete
  52. Thanks for all your hard work you're putting into this, NwAvGuy! I'm new to the hi end audio world, but I was in the market for my first external DAC and amp for my DT 880 to replace my x-fi forte soundcard. I've read that you'll have to spend a pretty penny to get the most of these cans. After stumbling upon your blog by accident, I was sure I was going to get the O2, until I heard news about this project.

    I'll be honest and say that I'm sad to hear it probably won't be in "production" (as I have minimal background in DIY audio projects) until spring, as I'm sure these types of stuff take time. Take your time. I cannot wait until the project is completed. Until then, I'll probably be saving myself for the ODA :) Keep up the good work!

    ReplyDelete
  53. Balance:
    Why not use different gains voor left and right to set a balance. This way it will not degrade the performance and it seems to me, if you use balance, you're gone set on 1 setting anyway. Using the balance in software can degrade the performance. And also would likely not work when you forcing the PC to bitperfect out. The change the level it has to process the stream. Depanding on quality of software used, it can/will lower the bit resolution.

    UcD:
    Heard what Guido Tent could do with the Hypex SMPS en UcD modules. Impressive. (I'm hoping that the ODAC can match the b-Dac in peformance). The "new" UcD should be even better because some Ncore design is also applied to UcD. Also heard 1 occasion of the Ncore module (modded OEM UcD180 module). If you find UcD impressive, wait till you hear these :-) Can't wait till Hypex will sell them to the general public.

    Chipamps:
    Why not use the LM3886/LM3875 so populair in gainclones? Cheap, easy to implement.

    Airport Express:
    According to MicroMega the dac inside the airport express is just fine, as is the IV/buffer stage. The modded the AE to replace the noisy SMPS and replaced the clock with one with a higher accurecy. And in (subjective) reviews that can measure with 4K cd players....

    Extra digital inputs:
    As the matrix hifi demostrated, even a cheapy cd speler has a decent analog out. So why instead of a complicated digital swithboard, use an analog swithboard? And maybe the seller of ODAC can sell different modules (USB,COAX/TOSLINK) and just a ODAC in (or between) the source and the ODA? Then you can choose to upgrade and your one pace and need. With ±100 USD per ODAC, i wouldn't mind that extra cost.

    ReplyDelete
  54. Hi NwAvGuy! i'm looking forward to the ODA, but was a little disappointed to hear it will not be coming with a S/PDIF. I only say that because i also do a lot of gaming on my computer, so i'd like to still use the DSP in my soundcard via S/PDIF.

    Are there any DACs with S/PDIF that you'd recommend that share the similar feature (price to performance) as the O2, and could also work very well it (thinking about getting an O2 myself)? Something that can be considered a "legitimate improvement" over current computer soundcards?

    ReplyDelete
  55. @Anon, that depends on if you want just a DAC or a headphone DAC and what headphones you plan to use. I've seen enough problems with reasonably priced DACs and a few headphone DACs (like the NuForce uDAC-2) I'm reluctant to recommend anything that hasn't been properly tested.

    The pro audio interfaces universally seem to have awful headphone outputs and nearly all require proprietary drivers. So I can't recommend them.

    DACs with real (not RMAA) test results narrows the choices considerably--especially if you want a DAC with 24 bit USB support. Even if 24 bit USB is not something you might benefit from today it's well worth having for future applications.

    Among 24 bit DACs, the FiiO E10 doesn't have an S/PDIF input nor do the HRT Music Streamers, or Centrance DACport. The Music Fidelity V-Dac II does, but I haven't seen solid test results on it, and it can't drive headphones.

    At the moment, the least expensive headphone DAC I trust with full 24 bit support and an S/PDIF input is probably the Centrance DACMini at around $800. I've been recommending the DACMini since it came out a long time ago and Stereophile just confirmed my pick in their latest issue. It may need a $100 upgrade to a 1 ohm output depending on what headphones you plan to use. The downside is it's relatively expensive.

    Perhaps others here have suggestions for Headphone DACs with S/PDIF inputs capable of 24 bit USB operation that have been properly tested, have low impedance headphone outputs, etc? And perhaps, considering all of the above (and prior comments) we need to re-visit the S/PDIF option for the ODAC.

    ReplyDelete
  56. This is just what I need, many thanks for making the effort!

    I could use S/PDIF, but it's not a showstopper that it's not there, I would buy this anyway. I don't need a balance setting at all, but I didn't spot whether the intention was to have tone controls or not? Tone controls would be kind of nice, but not a critical thing.

    Re latency, Windows isn't really designed for real time apps, neither is USB, so all the super low latency stuff are really afterthoughts, which is why problems can creep in. By real time, I mean the ability to do real time communication, voice chat is quite obviously possible. It all then comes down to how low does the latency really need to be.

    ReplyDelete
  57. @Oskar, thanks for the feedback and comments. I agree Windows isn't a "real time" operating system but neither is OS X or desktop Linux. From what I've seen the performance (including latency) with high-end professional audio interfaces (RME, Motu, etc.) is relatively similar between Windows and OS X and between USB 2.0, Firewire, and PCIe interfaces. Sound on Sound and others have published the numbers.

    The old school days of OS X and Firewire having a big advantage over XP and USB are long gone and Firewire is on its way out. Windows 7 audio is a much closer match to OS X. I'm not really sure where Linux stands, but few (any?) of the popular DAW software packages even run on Linux so its not really in the game.

    But it's also true none of that matters for just audio playback as long as the latency isn't so long as to mess up video sync or gaming. And for those who just want to listen music, latency doesn't matter at all.

    ReplyDelete
  58. NwAvGuy, I have been an avid follower of your blog and your threads and post of the other forums for quite some time. Your objective measurement of devices in this domain is to be applauded, thank you.

    Moreover, your posts are highly educational and being a long-time ham having built much of my own RF equipment, I find this a refreshing introduction to the world of AF.

    I have an O2 kit on order through the Jokener’s GB and will use this to graduate from the success I have had building a few highly effective CMOY amps in tins. I would like to consider next a build of a relatively conservative desktop amp to drive speakers and wonder (hope) you might consider a version of your O2 (wire with gain) amp suitable for this application?

    The same concept as the O2 but with an output of say 25w per channel, it need not be any more in most cases. In fact 12w per channel would be ample for most bookshelf class speakers. This might even (ideally) be in the form of a small (transparent) power amp that takes the out put of the O2?

    I can see an O2 transportable being carried around and a desktop O2 driving phones at the desk, and when one wants to turn on the power amp (or integral higher power O2), to get the same transparent gain through a set of speakers. All nicely packaged in a common case style and appearing and being a well-matched component set.

    Keep up the very informative, highly valued and educational narrative, I know form experience it can be tiring telling the emperor that he is naked, shining a light on darkness of ignorance can be it equally rewarding to those that can see it.

    Regards, Tony

    ReplyDelete
  59. @Roulston, thanks for the encouragement. There's a much greater void for high quality reasonably priced headphone amps and headphone DACs than for amps to drive speakers. There are all sorts of amps for speakers that are sufficiently transparent as long as you keep them within their output limits. It's true most have lousy headphone outputs (or none at all).

    Someone with reasonable DIY skills could morph together one of the better chipamp projects/reference designs, or Hypex UCD modules, into a combination headphone/speaker amp with the O2/ODA. If I were to design what you're suggesting to a reasonable budget, that would be my approach. If you want a serious no compromise speaker amp, get Doug Self's latest power amp book and build one of his "blameless" designs.

    In short, reasonably priced "transparent gain" for speakers has been available for decades in commercial products, reference designs, and proven modules like those from Hypex. But one-size-fits-nearly-all transparent gain for headphones has been mostly an expensive proposition--hence the O2 and ODA.

    ReplyDelete
  60. Lol I noticed someone mentioned the AK4396 / NE5532 DIY DAC in this discussion.

    Here is the diyaudio thread related to it - http://www.diyaudio.com/forums/digital-source/186245-dac-2496-ak4393-dac-kit-cs8416-ak4393-5532-a.html

    I think it sounds sublime, yes, I love it, however I don't have a dScope or the necessary experience, I am only a listener.

    I have faith the ODAC will compete with or surpass it.

    I believe in accuracy, transparency and objective data until the very end of the chain.

    If sonic artwork exists, it's at the very start and at very end of the chain, if anywhere!!

    /kiteki

    ReplyDelete
  61. @kiteki, thanks for the kind words. But just so everyone knows, that DAC doesn't seem to be DIY designed. It seems to be one of many eBay DACs being sold direct out of Asia for profit. I've tested several such eBay DACs and I've found significant problems with every one so far. I was hoping to find a good inexpensive DAC to recommend but I finally stopped wasting my money and gave up.

    That DAC lacks USB support, and does not have coaxial or TOSLINK S/PDIF on board. It is best to have all of those on a single board. Over 20 Mhz of bandwidth is required to properly handle S/PDIF signals. Running wires between boards can degrade the signal, create jitter, create ground loops, etc.

    If you're using a PC as a source, USB is usually the safest way to feed an external DAC. Few PCs have S/PDIF outputs, and those that do, often have poor quality S/PDIF that's prone to jitter. Using a USB -> S/PDIF converter just adds another clock recovery, and clock oscillator, to the digital signal path which generally makes things worse.

    There is much more to DAC design than just using popular chips and dropping them onto a nice looking PC board. But it seems the people who design and sell many eBay DACs, boards, kits, etc. often don't have any way to properly measure their DACs so they're designed "blind". Or perhaps they don't care about the performance and are just looking to make money however they can. But, whatever the reason, I've been really disappointed with eBay DACs. And I strongly recommend a DAC with on-board 24 bit USB support and onboard S/PDIF interfaces (if you need S/PDIF).

    I'll be testing the FiiO E10 soon and I suspect it will outperform the DAC you linked when using an eBay USB -> S/PDIF converter. The E10 is also much less expensive than buying that board, the converter, an enclosure, panels, power supply, etc.

    ReplyDelete
  62. "If you're using a PC as a source, USB is usually the safest way to feed an external DAC. Few PCs have S/PDIF outputs, and those that do, often have poor quality S/PDIF that's prone to jitter. Using a USB -> S/PDIF converter just adds another clock recovery, and clock oscillator, to the digital signal path which generally makes things worse."

    Interesting. . . I got into a fun argument on heavily-funded recently with someone who was going on about how terrible USB is and how using S/PDIF is essential - to reduce jitter no less. Ugh.

    ReplyDelete
  63. @Satellite, early USB audio could be pretty bad so that's likely how what's now a myth got started. The early devices often used the syncronous method of extracting the audio clock and were at the mercy of the PC's USB port timing. But, for quite a while now, most devices use the adaptive method of clock recovery and largely ignore poor quality USB timing. Plus the USB audio chip makers have been getting steadily better at reducing jitter in other ways as well.

    For some proof, and perhaps H-F discussion material, have a look at the current issue of Stereophile. John Atkinson's measured the Centrance DACmini's jitter as being much lower via USB than S/PDIF--the exact opposite of the mythical view.

    The reason I claim USB is safer is simple: With a self contained USB audio device the audio quality can be solely in the hands of whoever designed that device. It's entirely possible to design a USB DAC to be essentially independent of whatever USB port you plug it into. Even the $39 USB Creative X-Fi Go Pro I tested has impressively low jitter.

    S/PDIF, however, is usually different. Unless the device employs some (usually costly) method of completely re-generating the clock (i.e. the ASRC method used by Benchmark in the DAC1), the DAC will be at the mercy of the S/PDIF source it's driven by. And, not surprisingly, PC/Laptop motherboard manufactures spend as little time and money as possible on their digital outputs as they're virtually never objectively tested by anyone.

    Digital outputs on PCs, internal soundcards, and some USB -> S/PDIF converters can also have other problems. Some of the chips used are not bit accurate (transparent) for example. In that case, even the Benchmark DAC1 can't properly reconstruct the butchered audio signal. But with USB Audio you get bit accurate performance with all modern operating systems regardless of the sound hardware in the PC.

    So unless someone has say a $700 Lynx AES16 card in their PC, the quality of the S/PDIF signal is often a big unknown. A decent USB DAC gets around that problem as it bypasses all the audio hardware in the PC.

    ReplyDelete
  64. Seems like the E10 is $76 at Micca, the (finished with enclosure) AK4396 project is around $105 incl. shipping.

    I can't really comment since I haven't heard the E10, yes the AK4396 lacks USB so it's not ideal, I have to agree there, I ordered the Musiland 01 USD, their proprietary drivers look quite advanced - http://hifiduino.wordpress.com/musiland-usb/

    If your ODAC will come with USB, coax and optical, what can I say, that will be fantastic!

    /kiteki

    ReplyDelete
  65. NwAv, any thoughts on linear phase versus minimum phase in this DAC? Pre-ringing versus post-ringing.

    ReplyDelete
  66. @Crystalline. That's a good question and one I hope to better answer after more blind listening tests and future research. It's an often discussed topic, but I've not been able to find any solid blind listening tests that support it makes an audible difference.

    Some are probably getting tired of me bringing it up, but the massive Meyer & Moran blind study inserted an extra A/D and D/A running at a lowly 16/44 after an SACD player and nobody could tell they were there under any realistic conditions. I would be willing to bet that D/A had pre-ringing. And, despite all the supposed evils, nobody could even tell when the D/A was in the signal path.

    The classic arguments go something like: "Post-ringing is OK as it happens naturally but pre-ringing is awful because it's unnatural." But that's a far too simplistic view of what's really going on.

    While there have been some papers published involving blind testing of filter characteristics. They've mostly dealt with speaker crossovers and other areas rather than DAC filters.

    I've read several reviews of high-end DACs with the supposed option to switch between linear and minimum phase filtering but the reviews uncovered that sometimes the ringing characteristics didn't change as expected, and in other cases, several parameters were changed including such things as an audible amount of high frequency roll off. Those DACs are obviously not very useful for conducting a useful blind listening test of filter characteristics.

    There's been enough written on linear/minimum phase at HydrogenAudio to make even a hardcore audio geek's eyes glaze over. Last time I tried to wade through some of it I didn't find anything that was terribly convincing one way or the other. If someone knows of some decent blind studies related to DAC filtering, please post some links here? It's an interesting topic.

    The current ODAC prototype, like most mainstream DACs, is linear phase and does have some pre-ringing. That said, the ringing is above the range of human hearing so it's difficult to worry too much about it from an objective point of view. But the "acid test" is to put the ODAC up against higher-end DACs in blind listening tests as I did against the Benchmark DAC1. I hope to do more blind comparisons with other DACs once the design is finalized. If I find at least some audible differences that offers the challenge of seeing if the right measurements could reveal an explanation for why they sound different.

    ReplyDelete
  67. nwavguy, i appreciate what you are doing here.. and i fully understand the drive behind publishing the original O2, as a portable amp that is well engineered.. However, when you're going into desktop amp territorry, I think there are already plenty of inexpensive amps that measures well.. you mentioned the DIY cost of the ODA would be around $130.. in that price range, there is already Gilmore's dynalo from around 10 years ago that measures extremely well (check the original headwize pages and construction thread), even pure discrete, which is seen as the "better" thing by the old school audiophiles.. there are also other designs like the ck2iii that was redesigned by Dr. Cavalli several years ago, and several others around that price point.. I just dont think you need to push the O2 into a desktop design, if indeed your primary goal is to provide inexpensive well engineered amps to ppl that care.. just my 0.2$ of course..

    ReplyDelete
  68. @Anon, thanks for the feedback. I agree there are a lot of choices, but few are properly and thoroughly measured ("The Wire" project on diyAudio being a notable exception). Kevin Gilmore, for example, has a very old and limited audio analyzer and I've shown where RMAA falls short in many areas. Discrete designs are also more variable in terms of the performance any one person will get with their particular discrete parts.

    I think apples-to-apples, with similar features, specs, headphone drive capability, low output impedance, conventional ground, with a power supply, enclosure, etc. the ODA will have little competition price-wise. And I'll gladly put it up against either of the amps you mentioned--either on a test bench or in blind listening.

    The points above are before you consider the ODAC daughter board option. With it, the ODA becomes a high quality 24/96 USB headphone DAC. I'm not aware of any comparable DIY headphone amps with a similar option. And there may be a future DAC option with even higher performance.

    As for discrete design being preferred by some, they're welcome to read Op Amp Myths & Facts, and why I believe ICs are better for most headphone amps in O2 Headphone Amp and the O2 Design Process articles. Those who prefer audiophile mythology, or intentional distortion, to a more objective approach are not likely to be interested in the O2 or ODA. And that's OK. There still seem to be plenty interested in properly measured transparent amplifiers. And ODA is a small fraction of the price of even the least expensive Violectric.

    ReplyDelete
  69. Thanks for the tech-brief, any thoughts on this article? http://www.cirrus.com/en/pubs/whitePaper/DS668WP1.pdf

    ReplyDelete
  70. That's an interesting paper but it's mostly re-hashing the same conflicts that have been well debated elsewhere (although I didn't see a date--it may have been more timely when it was first published). Cirrus is also going to promote their most profitable solutions and differentiate their products capable of IIR filters however they can. So they're not exactly an unbiased source of information.

    The ODAC has less than 0.01 degree of inter-channel phase shift at any audio frequency so it's free from the most audible phase shift mentioned in the article. But things are much more fuzzy when it comes to intra-channel shift of the sort created by IIR filters. In an analog amplifier, or even most analog EQ circuits, many would object to the relatively massive amounts of phase shift Cirrus is arguing are inaudible.

    When debating the greater phase shift of IIR filters vs pre-echo from typical FIR filters, I've not seen anything conclusive in terms of credible listening tests. Among the high-end digital audio manufactures both methods have their loyal fans.

    If you look through the current Stereophile Class A "best of the best" digital components you'll probably find a roughly even mix of approaches favoring linear phase, minimum phase, and some offering the supposed ability to select either one. So even the highly subjective golden ears can't seem to agree on a clear winner.

    What we really need is a well run blind study where the only significant variable is the filter characteristic. If someone is aware of even something close, please post a link here?

    ReplyDelete
  71. I know it could be too much to ask when your hands probably are still tied with your ODA project, but is it possible to see a review on E10 some time this week?

    It's just that I might be able to get it cheaper now than afterwards. I hope I didn't come across too demanding. Whenever you're able to...

    ReplyDelete
  72. @Kingpage, if I had to guess, with the holidays and all, the E10 review might be the first Wednesday in January. The earliest would be December 28th. I try to publish articles on Wednesdays. There will be a new review tomorrow but not the E10.

    ReplyDelete
  73. You should try the Numark Stereo-IO (under $40) and Numark DJ-IO (under $80), the former being 16-bit and phono input capability, the latter being 24-bit ASIO with dual line outs, headphone out, and no inputs. Numark/Alesis 24bit stuff tends to sound good.

    Did you ever pick up a 0204 when they were really cheap on Black Friday?

    ReplyDelete
  74. I have tested some Numark gear, and perhaps I've just been unlucky, but it's been mostly junk. They are several big notches below say Behringer in the quality department. From what I've seen Numark shaves every last penny out of their designs including the actual R&D. They also have obscene profit margins. Dealer cost on a $99 MSRP Numark item is often around $25 and might have a "street price" of $50. And it likely cost less than $10 to build.

    ReplyDelete
  75. what about the traktor audio interfaces?

    ReplyDelete
  76. Traktor interfaces are not widely available in the USA. I also expect the FiiO E10 is a better product for headphone listening at a lower price.

    ReplyDelete
  77. Came across some nice articles:
    http://www.benchmarkmedia.com/discuss/feedback/newsletter/2011/12/1/do-specifications-lie-do-our-ears-lie-where-does-truth-lie-examination
    and
    http://www.benchmarkmedia.com/discuss/feedback/newsletter/2011/12/2/0-ohm-headphone-amplifier-sonic-advantages-low-impedance-headphone-amp
    The first article is particularly interesting. Would love to see some measurements of the ODA done with headphone load (like your HD650) in addition to standard ideal resistive load.

    ReplyDelete
  78. @Anon, thank you. I was already aware of John Siau's latest articles. I've had the pleasure of discussing high-end audio with him a few times in person. I have great respect for him. Not many in the industry have walked the objective/subjective line as artfully as John has. He genuinely understands what matters, what doesn't, and why.

    John has not only applied his impressive talents to Benchmark products, but unlike most commercial designers, he's also shared a lot of his knowledge in the form of articles like the ones you linked, detailed measurements of Benchmark gear, etc. Benchmark almost stands alone in that regard. Their approach starkly contrasts with most of the marketing-hype nonsense, unsupported claims, weak specs, and/or half-baked products being promoted by competitors.

    I didn't think of showing graphically how higher output impedances create much higher distortion at low frequencies when using real headphones. But John did and it's brilliant. It's yet another blow to the tired "higher output impedances are better myth". Kudos to John and Benchmark!

    I discussed some of the resistive-vs-real load issues in my Music vs Sine Waves article. But an article comparing resistive loads to real loads in detail has been on my to do list for a while now.

    The output impedance of the O2 and ODA is low enough they perform very much like Benchmark's HPA2/DAC1 headphone amp--the load makes very little difference. The only thing you can't usually test using real headphones is maximum output. You would likely cause damage with full power sine wave testing from a high power amp like the O2/ODA.

    ReplyDelete
  79. wondering just how much an improvement something like the sotm tx usb for a usb async dac, if any? although you can already get isolators/batterypoweredusbhubs, it seems somewhat unique in that it gives cleaner power to the usb clock -- akin to running a laptop off the battery.

    ReplyDelete
  80. Hi NwAvGuy, I got a question for you. Hope you might answer it as I'm relatively clueless about it. I got an Balanced Stereo Amplifier at home for me stereo setup, It got balanced XLR Inputs and Dual mono Design internally. It also accepts Normal unbalanced signals just fine, I've always being thinking about getting an Balanced source for it. So now I've been looking for a nice FW/USB DAC with Balanced outputs to match and also feed my headphones with. (AKG 701)

    If I understand you correctly, using a balanced source for my amp won't do any good, will just be a waste of effort? or might it even make things worse? Because I were just considering a Focusrite Scarlet USB Audio Interface with Balanced outputs for my audio needs. Should I just wait for the ODA instead? I'm perfectly comfortable in building an ODA for myself:)

    ReplyDelete
  81. @Anon, I'm sure some DACs benefit from cleaning up the USB power and/or timing. But there are several ways to make the DAC itself fairly immune to USB issues as long as the USB port meets the required specs (and nearly all should--they're certified to).

    Ultimately the proof is in the measurements. If I can plug a DAC into multiple ports on multiple PCs and fully loaded hubs/controllers with no significant change in the audio performance. It's expensive overkill to try and "improve" the USB power and/or timing further. Adding hardware between the USB port and the DAC might even make the performance worse.

    It's like using a AC line power conditioner when your gear isn't suffering from any AC line power related problems. Lots of audiophiles do that but it's a big waste of money. And sometimes the power boxes make things worse--either by creating L-C resonances and/or restricting peak current to power amps.

    True (galvanic) isolation is a different issue. It can have benefits for preventing ground loops between the PC's ground and your audio system's ground. But that's usually only applicable when something in your audio (video) system is grounded--i.e. a cable TV connection or a big power amp with a three prong grounded AC plug. Headphone gear is rarely grounded.

    It's expensive to isolate 480 Mbit USB 2.0 signals externally via the USB connectors. It's much easier to build the isolation into the DAC as HRT does with their Music Streamer line.

    @Daniel, you are correct there's no benefit to a balanced source unless you're going to run really long cables or use it next to an arc welder. If you just need a good DAC, and not a headphone amp, an HRT Music Streamer II would be a good choice. They even have a balanced model but it's a lot more money.

    Pro audio interfaces (like those from Focusrite) are mainly optimized for recording, not playback. You're mostly paying for mic preamps, phantom power, metering, input level controls, and the rest of the recording hardware. A dedicated DAC designed for high quality audio playback is usually a better choice as 100% of the design budget went into playing music.

    ReplyDelete
  82. @Justin, I'll be publishing "updates" as I make significant progress, but there won't be ODA boards, or even a final design, available over the holiday break. Sorry. If you have much use for a battery powered amp, go with the O2 if you want something sooner rather than later. If it will rarely leave your desk, especially if you can use some of the extra features, you might want to wait for the ODA.

    ReplyDelete
  83. OK NwAvGuy, no waste of money on Balanced cabling then. I do however need a pretty decent line input for some conversion of analogue material. I'm currently using a decent soundcard but due to space limitations it won't fit in the computer anymore. Also, would be nice with a decent headphone output on the DAC as well.

    ReplyDelete
  84. Sorry for doing 2 posts, it seems the HRT Music Streamer II is just way overpriced here in sweden. Close to 300 dollars för the non + model, Ouch.

    ReplyDelete
  85. @Daniel, I will be publishing some tests of various pro audio USB interfaces. I have access to at least four or five typical current models but none are from Focusrite. I need to do more research, but it's interesting most of them cannot do bit-accurate digital S/PDIF via USB. I suspect some may use an inexpensive combination CODEC chip rather than high quality discrete A/D and D/A chips.

    I expect the ODAC, despite its lower cost, may outperform most of the interfaces purely as a DAC. And the ODA should easily outperform any of the interface headphone outputs. I think most interfaces, including the Scarlet series, will have a hard time with your K701 headphones which need a low output impedance and quite a bit of voltage and current.

    So, if you want to wait, the ODAC/ODA might be a very good choice. If you want even lower cost, wait for my FiiO E10 review coming soon, but it will struggle with the K701 (it is the same headphone amp as the FiiO E11).

    ReplyDelete
  86. Thank you for your time to respond and give feedback on these topics. And yes, it's kind of strange/interesting 'feature'. I'll wait for your upcoming reviews, and hold my breath about buying a DAC for a bit.

    I'm very much looking forward to your reviews and the ODAC design. :)

    ReplyDelete
  87. Hey Nwavguy,

    Its been said so many times that its almost cliche, but thanks for the blog, it's been very helpful and informative, as well as allowing a layman like myself to understand some of the more technical aspects of the hobby. I was wondering if you could take a look at Teac UD H01 DAC amplifier for £299, it seems like it has good specs and is not too expensive. I would appreciate your opinion as you're much more knowledgeable about these matters than me. Thanx.

    ReplyDelete
  88. @Anon, glad you like the blog. The Teac UD H01 is a hard call as I can't find any useful specs beyond what bit rates it supports. Notably missing are any specs for the DAC performance, headphone output power, output impedance, noise, etc.

    What Hi-Fi did a really brief "review" of it but they didn't have anything very useful to say either. I can say most Asian-designed headphone outputs have far too high of an impedance. See: Output Impedance

    I'm not sure what to make of Teac these days. I think they've changed ownership a few times and the once famous Japanese maker of tape decks is probably long gone. What scares me about the current Teac is they slap their name on everything from $49 plastic all-in-one boom boxes at Walmart to their supposed "reference" gear. A lot of old audio brands have been bought up by no-name Chinese companies looking for more credibility. But it's the same questionable no-name gear with a different front panel. I don't know if Teac is now one of those.

    Even once legendary brands like RCA in America and Thompson in Europe are now owned by Chinese companies (both of those are owned by TCL in China). The CEO of TCL was famously asked at a press conference what would become of Nipper, the famous RCA dog, and his reply was something like: "What is a nipper?". So you can't put much weight in a lot of brand names these days.

    ReplyDelete
  89. I'm Currently using a Xonar STX soundcard, I believe it has an output impedance of roughly 11 ohms and adjustable gain. It's pretty good for a $160 consumer soundcard I guess.

    The headphone output as I understand it, is decent but not good. The line out is supposed to be pretty good however. I don't know if you head about it. Any idea how it fairs against other products you've tried or the ODAC as most reviewers are either biased, not very technical or both.

    ReplyDelete
  90. What are your thoughts on these two articles?

    http://www.nutshellhifi.com/library/FindingCG.html
    http://www.nutshellhifi.com/library/tinyamps.html

    ReplyDelete
  91. @Anon, The Xonar STX was tested by Stereophile. You're correct, at least in some PC's, it has a respectable line output. But the ODA/ODAC has several advantages over the STX:

    - No problematic proprietary drivers required
    - Much better headphone output
    - Consistent performance with most any PC
    - Works with laptops, all-in-one PCs, small HTPCs, etc.
    - Portable and can be easily moved from PC to PC or used as a stand-alone headphone amp.

    The STX is really at the mercy of the PC you put it in for noise. Look at how one channel of the Apple MacBook 5G I just reviewed has 20 dB more noise than the other channel. That's entirely due to being close to noisy circuits. Similar things can happen to the STX in some PCs. And there are lots of PCs the STX won't work with. The trend is strongly towards smaller PC's without PCI slots.

    @Kingpage, one of those articles is mostly about tube distortion. Which, because there's relatively a lot of it, is a rather different topic than an ultra low distortion amp like the ODA. The author clearly has a large tube/analog bias. In my opinion he's trying to make a case for tubes, which is difficult to do in an objective way.

    He is correct about a few things. A lot of early solid state amps where indeed pretty bad. But a lot of tube amps commit even bigger sins.

    The other article states "resolution that simply isn't there in the data storage medium" with respect to 16/44 CD audio. He really needs to read Meyer & Moran's extensive AES study and explain how adding a 16/44 ADC and DAC to the signal path was inaudible to a large group of listeners over 500+ trials. The same has been done to vinyl playback in otherwise all analog systems (including those with tubes). Even the owners of the systems cannot tell when the 16/44 ADC and DAC have been added to the signal chain if they don't know it's there. There are dozens of trials and studies that support 16/44 really can be 100% transparent. He's got nothing but some psuedo-science and personal opinions to support his argument.

    I believe the explanations for his views can be found in Subjective vs Objective. It's fine to prefer the sound of tube gear, but it's not OK to try and justify it with erroneous pseudo-science.

    ReplyDelete
  92. Hello NwAvGuy, I've been following your work, and I would need an information:
    I have a few headphones, one of them is a Beyer Dynamic DT48A 5ohms for audiometric applications.

    Would the O2 or the ODA (I am more interested in that one) drive my Beyer?
    Or what about the ERGO AMT (4 ohms, it goes even to 3 at some frequencies, I don't own it but when a friend of mine lent it to me, I couldn't get good results with my Denon Amplifier and a few other)?

    If your Amplifier can't do the work, I'm thinking about a Violectric V181 or V200, but it's quite expensive...

    Can you help me with that?

    Congratulations for you work!

    ReplyDelete
  93. @Robin, the "below 16 ohm" issue came up a few weeks ago in the diyAudio thread and there was considerable discussion there. I plan to test it down to 8 ohms one of these days.

    Those DT48As look like seriously UNcomfortable headphones with their rubber earpads, etc. I'm curious why you're not more interested in the 25 ohm DT48 E (or lots of other options) for music listening? Or do you want to use them for hearing testing?

    On paper even the 5 ohm DT48A is easy enough for the O2 or ODA. The 48A is rated at 110 dB SPL at 1 mW. That's 70 mV of signal and a peak current under 20 mA which is 1/10th the O2's 200 mA.

    I can't find any specs (in English at least) on the Heil Air Motion AMTs. They seem designed to work mainly with their own dedicated ERGO amp (much like electrostatics). Without knowing more about their specs I couldn't say if the O2/ODA could drive them well.

    You're correct the Violectric amps are expensive. The ODA is intended to perform much like a Violectric with 98% of popular headphones for a small fraction of the cost.

    ReplyDelete
  94. Well, I was told the DT48A was a little bit better than the DT48E 25&200ohms.
    Indeed, he's not very confortable. The best I did was about 1h listening to him, after that, I didn't even feel my ears anymore, there was to much pressure on them... but I'm gonna buy the DT48E earpads, I heard it was more confortable...
    For now, I use my integrated amplifier, a Denon PMA700AE with all my headphones: Hifiman HE-6 (on speackers system), Beyer DT48A on the amplifier headphone socket, HD650 both, but there is a noise on the speacker system with the HD650 unless I listen to music at a good level... I tried this way since the HD650 impedance varies form 300 to ~500ohms and the impedance of Denon headphone socket must be about 330ohms or something this kind. I do the same thing with my T1 and a few other.
    That's also why I want a new amp, so I can be sure I get the best from my Headphones... without noise or anything else...

    Tkanks for your quick answer!

    ReplyDelete
  95. RE: latency, I've received my iBasso D6, and I am very happy to report that I can use the generic ASIO4ALL driver with an ASIO buffer size as low as 128 (@44.1kHz) and it is absolutely fine for real time piano playing.
    This looks promising for the ODA/DAC, because the iBasso, as we know, does not come with an ASIO driver and hasn't been optimised for ASIO.

    I lose the multi-client capability, though, which is a shame. I.e - I cannot use normal (non-ASIO) Windows applications at the same time, for example, to play music while I am playing live piano. My M-Audio Fast Track Ultra allows me to do this, and it is a very nice feature.

    There's one serious problem with the D6 so far though - the output drive is NOTHING like I expected. It is not even as loud as my V-Can, yet it should be even louder. I need to take some measurements, because it's going back if it does not meet the specs. \Greg.

    ReplyDelete
  96. I know this is very off topic, but just re: this D6 amp, as far as I can tell, it is clipping at 2.8V RMS (unloaded), when fed with a full scale sine wave over the USB. This isn't enough - it should be 4.56V RMS in order to achieve 650mW into 32 ohms. I don't think I am doing anything wrong, but of course I'll contact iBasso for assistance. \Greg.

    ReplyDelete
  97. AWESOME!! Im so happy the Desktop Amp project is going fowards , expecially the DAC wich is what matters to me.

    What abuot your outside US fans and folowers , international shipping of the ODAC will be a reality?
    I live in Argentina.
    You are awesome Nwavguy

    ReplyDelete
  98. @Stefano, thanks for encouragement.

    Assuming there is enough interest, I think International shipping for the ODAC will be possible one way or another but that will be up to the ODAC vendor. I'm not sure what's involved to ship to South America but at least eBay makes International sales somewhat easier.

    @SullivanG, yeah, this article is about the ODA/ODAC not iBasso products. To be honest I would not be surprised if your D6 simply doesn't live up to the claims. Companies like iBasso provide very sketchy and incomplete specs. I suspect many don't even have the equipment to properly measure their own products. Nearly every Asian-designed DAC/Headphone product I've tested from small vendors like iBasso has been seriously flawed.

    iBasso may have a nicer website than many similar companies, but they seem to follow the same FOTM design philosophy of constantly releasing new models as quickly as possible. It seems they're mostly interested in making products that look nice and use popular chips rather than taking time to focus on getting the performance correct.

    Besides using popular DAC and op amp chips, companies like iBasso, Matrix, Audinst, HA Info, Yulong, etc. tend to use the cheapest parts possible, poorly designed PC boards, cheap cloned versions of audiophile parts, etc. For example, thick film SMT resistors are much cheaper than thin film versions but look exactly the same. The cheaper thick film resistors have 30 to 40 dB more noise and distortion. Care to guess what kind pretty much every off-brand semi anonymous Chinese, Taiwan and Hong Kong vendor uses in their amps and DACs?

    ReplyDelete
  99. NwAvGuy, I have a Yulong U100 which uses a Cirrus Logic CS4398 D/A chip and a ADA4075-2 opamp. What can you say about these components, and would the ODAC be an improvement in any way?

    ReplyDelete
  100. @Daniel, Given the large number of O2 faithful in Europe I suspect the ODAC will absolutely be available in Europe. Ideally there will be someone there (like Jokener on diyAudio for example) who can work with the ODAC co-developer to distribute them directly to EU customers.

    As long as the STX is quiet in your computer its main weakness is the on-board headphone amp and its high output impedance. So you're more likely to need the ODA rather than the ODAC.

    ReplyDelete
  101. Thanks NwAvGuy. Just for closure on the D6 - iBasso say that the DAC output is 1.2V, and given that the maximum output I'm getting is 2.8V RMS, that means the maximum DAC output is 1.2V peak/0.85V RMS. This is despite their specs saying "Line Output: 1.5V RMS". Their specs do not mention the DAC output, so I assumed the DAC output would be 1.5V RMS! This means the true power spec when using the DAC is 225mW, which is a far cry from the specified 650mW. They have allowed me to return the unit, which I will of course now do. At least I now have good confidence in these kinds of amps being suitable for low latency ASIO work \Greg.

    ReplyDelete
  102. Thanks NwAvGuy. The lack of measurements/specs make decisions very difficult... not to mention that subjective opinions are rarely shared unanimously. How can people make informed decisions then?

    ReplyDelete
  103. @Anon, the lack of meaningful specs, and proper measurements, does indeed make it very difficult to make informed objective decisions. It's my belief that's how many manufactures want it.

    A lot of companies know they won't win many spec contests so most just publish a few vague numbers and depend on subjective marketing hype, touting the chips/parts they use, and buying ads from those who might review their products. Schiit Audio is a textbook example. They do an amazing job at marketing without sharing hardly any technical data anyone can actually verify (because it's all too vague).

    Those who advertise on Head-Fi often end up with a small army of shills plugging their products. And some of the more outspoken shills react very defensively if anyone tries to bring objective performance into the discussion. Just look at what happened with NuForce, Audio-GD, and Schiit Audio.

    It takes a lot more time (hours of labor), genuine audio engineering knowledge, and expensive test equipment, to properly design audio gear that not only sounds good but measures well. A lot of the small companies selling products on eBay, and/or through very limited distribution, take shortcuts more of those things.

    That's why I believe there's a real need for more people like Tyll Hertsens at InnerFidelity, John Atkinson at Stereophile, and myself, to measure more gear in a credible way and publish the results. It helps those buying stuff make more informed decisions and also helps keep the manufactures honest. Objective reviews provide incentive for manufactures to genuinely improve their products instead of just throwing together yet another half-baked design with a new front panel and the latest DAC chip.

    Anyone who says measurements don't matter is either misinformed or they're trying to protect their own interests. Measurements don't tell the whole story, but when combined with blind listening tests, build quality, features, etc., you can make a solid objective case for what's worth spending money on and what isn't.

    ReplyDelete
  104. Hi NwAvGuy,
    How much current will the ODAC draw when operating? I plan on using it with an iPad using the Camera Connector Kit, and apparently anything drawing more than 100mA will require a powered hub. Also, when used (installed in the same case) along with the ODA, will the power supply for the ODA also supply the DAC?

    Thanks,

    SAB.

    ReplyDelete
  105. @SAB, the ODAC will be over 100 mA and all bets are off with it working on an iPad even with a powered hub. Are you aware of anyone doing 24 bit UAC1 audio on an iPAD? It's possible iOS has the full driver stack required but it may not. It's not something I've researched. The ODA will have a power header for the ODAC board.

    Also, as an aside, I'm not aware of any 24 bit USB audio chips that can be used with even a remotely decent DAC and stay under 100 mA for the total power requirement. There might be some all-in-one chips, like perhaps whatever is in the Creative X-Fi Go Pro, but I doubt they'll deliver even full 16 bit performance. So even if the driver and hardware support is in iOS, I'm not sure and external device under 100 mA could perform much better than the built-in audio.

    ReplyDelete
  106. Thanks NwAvGuy,
    I found the following link that contained developer questions with some answers re. iPad + CCK:

    http://lists.apple.com/archives/coreaudio-api/2010/Apr/msg00124.html

    It appears that it fully supports Class 1 compliant devices, but, back in 2010 it was limited to 16 bit (that may have changed with OS5)

    I believe there are threads on Head-Fi and such discussing which DACs are capable of working with the iPad. I will investigate further.

    SAB

    ReplyDelete
  107. Hello NwAvGuy, I've got another question for you.
    For christmas, I might get an AKG K1000 (almost in new condition).
    Would your ODAC drive the K1000? (120 ohms and 74db)

    Thanks for your help!

    robin.

    ReplyDelete
  108. @Robin, the K1000 was mainly designed to be driven from a speaker output not a headphone output. Perhaps that's why AKG called them "earspeakers". The ODA, like most headphone amps, can't properly drive the K1000. It's the most power hungry "headphone" I know. I have heard the K1000 driven from a tube speaker amp and they're interesting. They're so different it's hard to even consider them headphones but I enjoyed listening to them for a short time at least.

    ReplyDelete
  109. Benchmark definitely supports iPad input..they've tested and I use myself.

    http://www.benchmarkmedia.com/discuss/feedback/newsletter/2011/06/1/ipad-streams-high-resolution-audio-dac1

    ReplyDelete
  110. Hi NwAvGuy!

    Thank you for your good work and generosity. I can't follow this high level discussion but I understand that you are doing good stuff. I don't own any amplifiers and speakers, I like to keep it simple and my question is: How about two phone jacks? Then I could enjoy the music together with my friend who owns some expensive audio machines. Certainly would I not be ashamed.

    I am presenting myself as anonymous just because I don't know how to create an ID.

    Kemal

    ReplyDelete
  111. Not to stray too far OT, but further to the iPad as a transport concept (it works well in my opinion), what might be the sonic pros/cons vs. computer USB? I assume we can measure the performance...

    I think it's the future; no reason to use a full blown computer to stream music anymore.

    By the way, Innerfidelity has a timely post today.

    ReplyDelete
  112. Yeah, I can understand the desire to use the iPad as a "transport" although it's still somewhat limited in some ways. It has relatively modest amounts of local storage, no SD slot, and I'm not sure what the options are for streaming content from a NAS in any non-Apple format or way (i.e. FLAC, CFS shares, etc.). And if you have sketchy WiFi you may experience drop outs.

    I guess it depends on how much you want a given device to perform double-duty. There are several network media players (wifi and wired Ethernet) that cost far less than an iPad and can be controlled from an iPhone, iPod Touch, iPad, Android phone/tablet, any web browser, etc.

    That way the "remote" can stay on your person and the player is just a small low power box that can be tucked away anywhere. Some of the better known examples are the Squeezebox and Sonos system but there are many others. Such systems play nearly any format, from nearly any source, instead of being largely constrained to Apple's formats and media sources.

    ReplyDelete
  113. First off, thank you.

    Second, I ask that you reconsider the includion of an optical SPDIF input. As you may know, almost all,MacBooks can output optical SPDIF if one plugs an optical cable into the headphone jack. The Mac Pro has standard optical SPDIF outputs, which I feed into a decent AV receiver. This is the preferred way to get high quality sound out of a Mac, no driver required. I know that some listeners with Cd players would also welcome the opportunity to use SPDIF from their CD player in conjunction with the desktop DAC/amp that you are building.

    ReplyDelete
  114. @Fred, for reasons already explained in the comments above, it's not trivial to add S/PDIF to the ODAC. It would only be used by a minority of users and the rest would be paying for something they had no use for.

    As for Apple's current MacBooks, only the Pros have optical out. And the ODAC will work on any MacBook via USB with no driver required.

    Most external USB DACs only support 16 bits over USB. Some of those DACs can run at 24 bits using S/PDIF. That's probably at least partly where the preference for S/PDIF comes from. But with a 24 bit USB DAC, like the ODAC, that argument is no longer valid.

    24 bit USB is, in many ways, superior to S/PDIF for computer audio. With S/PDIF, the source provides the digital audio clock and the DAC is at the mercy of the source for jitter and bit accuracy. But with USB and native drivers you're assured of bit accurate performance and the jitter is managed by the DAC as it generates its own audio clock.

    ReplyDelete
  115. I'd just like to thank you for an amazingly informative and honest blog. I'm glad someone cut through all the marketing propaganda and gave us a solid scientific approach.

    I've been looking for a new amp (to replace my old Creek OBH-11) and look forward to getting a ODA when available.

    ReplyDelete
  116. Another well considered and valuable narrative and I for one will be requiring two ODACs so once you sort out the partnering arrangements they can be assured on my order regardless of kit concept. (pure DIY kits, Pre-mounted SMDs, or fully assembled).

    Like some other commentators I too would like to untether from a computer source. With my library growing mostly with .FLAC files that are massively oversampled the last thing I want to do is have to down sample them, or touch them at all for that matter.

    Great work. I hope to have my O2 kit soon, and the ODAC will be a perfect desk side complement.

    Thank you for your enthusiastic contribution to we poor souls.

    Tony

    ReplyDelete
  117. I'm still unconvinced by the "minority" of users wanting digital optical input, and even if it is a minority - some who want it actually need it.

    Increasingly TVs offer ONLY optical audio output - no RCA - so you are really alienating a lot of potential users with that decision.

    What's more for applications like gaming consoles and cheap CD players, it offers a way to bypass the often inferior DACs.

    If you buy a DAC without digital input then its basically denying you using it with anything but a computer.

    ReplyDelete
  118. If UAC1 can support 24bit audio, why does UAC2 even exist and why is it used? Earlier I thought it's a requirement for 24bit audio over USB, as every 24bit DAC seems to use UAC2.

    You seem to have an amazing amount of patience for questions, thanks for that.

    ReplyDelete
  119. @Anon Dec 27 3 AM, you're free to design your own open source DAC with S/PDIF and share it with everyone.

    @Anon Dec 27 5 AM, UAC1 is limited to 24/96 with only 2 channels in one direction. UAC2 supports up to 24/192, multi-channel, and simultaneous hi-res recording and playback.

    But none of the above are an advantage for 2 channel audio playback. So, indeed, it's hard to argue how UAC2 is better for headphone or two channel listening with speakers. For those who might think 24/192 sounds better than 24/96, you need to do some research. It's been well established 24/192 offers no benefit as a playback format.

    ReplyDelete
  120. Hello NwAvGuy, wow, lots of improvements and progress. would like to say I'm getting the O2 and thank you in advance for your massive contributions. I look forward to more of your projects in future.
    It's nice to see someone who is willing to do this for the audio community.

    ReplyDelete
  121. So USB audio is the way to go, when done properly?

    I was interested to read of your disdain for USB to S/PDIF converters. I understand that the ideal is a USB cable straight to the DAC, but when S/PDIF is the only option for connecting to your DAC, is a decent asynchronous USB to S/PDIF converter really so bad? My (limited) understanding of such converters is that they minimise their dependance on the jitter of the computer's USB bus: surely they are superior (when reasonably designed) to simply using the S/PDIF output of your computer itself? When there are models available with around 130ps of jitter, can they really be deemed "poor" solutions? Suboptimal, perhaps, versus a direct USB connection to the DAC (assuming the DAC boasts a decent USB implementation), but surely not anything approaching an audible compromise?

    ReplyDelete
  122. @Willakan, my dislike of the "two box" solution is mainly for people looking to go out and buy both boxes rather than a single USB DAC.

    If someone already has a nicely performing DAC, it's possible some of the better USB-to-S/PDIF converters will do a good job. But there are only a few reasonably priced ones I know of. And perhaps the most popular is an Asian designed unit mostly sold on eBay under a variety of names. I wouldn't automatically trust it to be "reasonably designed" unless you've seen full credible measurements.

    The Mobile Fidelity V-Link seems to be a somewhat more solid product. But it also exceeds the price of say a HRT Music Streamer II. So, given the choice, it's hard not to recommend the async 24/96 HRT DAC as the better overall solution.

    The audibility of jitter is a can of worms I hope to try to shine more light on in the future. It's far from black and white but I do tend to agree that once all forms of jitter are below a certain threshold it's not an issue.

    ReplyDelete
  123. This talk of USB vs S/PDIF is key to me at the moment because my MacBook (older non-pro version) is my primary computer source for my music but my NAD M2 doesn't have USB input, only S/PDIF.

    As it's a true digital amp (rather than just an amp with a DAC built in) I'd prefer to feed it a digital signal/

    Have you any experience of the M2Tech HiFace USB to S/PDIF adapter because I was considering one.

    ReplyDelete
  124. @twelve, I don't have any experience with the HiFace but you may have seen my other comments on the subject. In short, I don't trust it's a properly executed design as it's from a no-name company and, to my knowledge, has never been properly measured. I would have more faith in the Music Fidelity V-Link. If your Macbook has S/PDIF out, and you're happy with it, just stay with that.

    ReplyDelete
  125. Hey NwAvGuy,

    First off, I want to say the O2 is an excellent benchmark amp and a great baseline amp for my headphone reviews. Excellent work!

    But on a more serious note, I think a safe price range for a fully functional ODA (with DAC capabilities) should be no more then $300 as of now.

    Could you clear something up about the DAC itself? In the article you never really stated what DAC will be used, so will the DAC be a custom made DAC or am I missing something?

    Also, it will be safe to assume this will be a project for those confident in their DIY skills like the O2 (so no step by step) as I'm curious if I should wait for a preassembled version or try to build one myself (novice amp DIYer right here)

    ReplyDelete
  126. @jr kong, Thanks. The design of the ODAC board has not been finalized yet. Once it is reasonably final, I'll publish a lot more details.

    The DAC board will be only available pre-assembled. So it doesn't make the ODA any more difficult to assemble. The ODA board itself is similar to the O2 in difficulty.

    I anticipate the price of a fully assembled ODA/ODAC to be comfortably under $300 but that's up to those who choose to sell them.

    ReplyDelete
  127. NwAvGuy,

    I was reading about your Odac and I'm wondering if it will also be compatible with the O2 amplifier as well as the upcoming ODA? Many people -- I'm going to go out on a limb and say 99 percent -- won't need the extra grunt of the desktop version, and might like the option of going portable even if the amp was primarily desktop based. Still, they might crave a high quality Dac option. Making the ODac modular would avoid shutting out those who have bought/built O2 amps. Maybe this is your plan all along, or maybe it's impractical; just wondering.

    Cheers,
    Tim

    ReplyDelete
  128. line me up for a pre-assembled ODA/DAC! i like your objective approach to audio... while music is highly subjective, audio should be purely objective

    ReplyDelete
  129. I'm just fascinated, really. I wonder how this ODA/ODAC will fare against something like the CA DacMagic (which is fairly affordable, compared to other DACs). I'm so into the market for a good "PC Soundcard". if this ODA/ODAC proves to be great, I'm so into buying it - in a neat case like the CA :P

    ReplyDelete
  130. A pass-through to my (powered) speakers is ESSENTIAL for me.

    ReplyDelete
  131. Do you have a *rough* estimate of the board size for the ODA? Me and a friend of mine are planning something a bit more motivated for a desktop amp and I'd love a board size estimate. It doesn't need to be too precise, just a rough idea.

    ReplyDelete
  132. @Atilla, the board size is still subject to change. One of the more likely possibilities is 160mm wide (along the panel sides) by 80mm deep.

    ReplyDelete
  133. Super interesting project. Would be great to develop a cable that grabbed digital output from iPod dock to feed the DAC. My understanding is that this is difficult, but not sure why.

    ReplyDelete
  134. Hey NwAvGuy,
    Regarding the DAC, who exactly is the one creating the DAC board? Can you provide any details?

    You mention that the ODAC will not support recording, so can we assume that the ODAC won't "overrule" our existing soundcard when it is connected?

    ReplyDelete
  135. The DAC in at least the iPod Touch 3G and 4G is quite good and likely hard to audibly improve on. And some rumors suggest Apple may raise the game further with 24 bit audio support in the future 5G Touch.

    Apple chose not to use typical S/PDIF signal levels/impedances/etc for the digital output in their dock connector. That does make it a bit more challenging but you can buy lots of docks that offer S/PDIF output and/or have a supposedly better DAC built into them.

    But I can argue the internal iPod DAC may have an advantage for at least jitter. It uses an I2S interface which is vastly superior to S/PDIF as it doesn't require embedding the data clock into the audio signal. Any external DAC has to use an embedded clock which is always a source of added jitter.

    So, at least with the iTouch 3G/4G, I think using the line out (LOD cable) and not bothering with an external DAC is the best plan. The current Classic is also likely respectable. If you have an older iPod, or a Nano, your money might be better spent on a current Touch (or the new 5G) rather than an external DAC.

    ReplyDelete
    Replies
    1. Thanks. Very good info regarding the Classic, which is my front end at the moment.

      Delete
  136. @kong, AFAIK, what happens when you plug in a driverless USB audio device is up to the operating system, not the device. In many cases the device will become the current default playback device as that's usually the behavior most people want. So I'm not sure if that's what you mean by "overrule"?

    You can leave the device connected and easily switch back to internal audio or another device if you want. If you're using a software player (like Foobar) that lets you choose your output device, it will remember your preference. For other ODAC details, please be patient.

    ReplyDelete
  137. pardon my poor wording but what I meant by "overrule" is:

    Can I use this alongside my other soundcard? As I tend to record audio from time to time I was wondering if the DAC in the ODAC might disable the default sound card when connected.

    ReplyDelete
  138. @kong, One of the advantages of using the native UAC1 sound drivers is multiple audio devices are cleanly supported. You don't have to worry about ill behaved third party drivers messing up your other audio devices or taking over your system.

    So you can easily use the ODAC alongside other sound devices. If they're left connected all the time, and you're using Win7, your playback (and recording) preferences should be remembered. If you only plug in your recording interface once in a while, it may be come the default audio device when you first plug it in, but it's easy to change that if you want.

    In general, Win7 is quite a bit better behaved than XP in terms of both flexibility of audio routing and remembering preferences by device. But even XP shouldn't be a problem for what you want to do. In theory OS X and Linux behave similarly and should also allow multiple audio sources with the ODAC being one of them.

    ReplyDelete
  139. This is super exciting and I find myself checking everyday just to see if there are any updates. I do have a few quick questions though. I know you've given an estimated of the board size in an earlier comment but I was wondering if that included the ODAC components. Also, what your thoughts on using the ODAC/ODA combination as a transportable DAC/amp combo? Something that can be easily brought back and forth between work/home/hotel room. I was hoping that the size and design would be conducive to bringing it around.
    Thanks!

    ReplyDelete
  140. @Kye, the ODAC is a "daughter" board that will sit on top of the ODA board. So it doesn't change the "footprint" but it does change the required height of the enclosure. For a transportable, the O2 with the ODAC would be a better and smaller option and would also work standalone from an iPod, phone, Clip, or other battery powered player without needing any AC power.

    ReplyDelete
    Replies
    1. Thanks for the reply. For my purposes the ODAC/ODA would definitely serve as a desktop amp as I'd prefer the benefits of the ODA while at home/work.
      Just to clarify; I was just wondering if, rather than making one for home and one for the office, it would easier to get an extra set of power and USB cables and simply throw the combination ODAC/ODA unit in my office bag with my laptop. From the discussion thus far, it certainly sounds like this should definitely be possible. A standalone ODAC unit would be wonderful for portable use though.

      Delete
  141. Wait, so that means the ODAC is small enough for portable use as well?

    ReplyDelete
  142. @kong, the ODAC, by itself, can't drive headphones. But the plan is to allow it to fit a small (smaller than the O2) Box Enclosures case for stand-alone USB powered use. Combined with an O2, the ODAC would make a great (trans)portable headphone DAC.

    The ODA/ODAC is designed for home/office/etc. use only from AC power and will be much larger.

    ReplyDelete
  143. That's really good to know! I'm probably going to build one of each cuz I can see places where I can use both.

    ReplyDelete
  144. So it sounds like the there will be an ODA/ODAC product as well as a standalone ODAC for those that already own the ODA.

    If true, that seems to be the "better way to go" as it will at least allow for portable use. Other than size, will there be any performance improvements for the ODA portion of desktop version?

    ReplyDelete
  145. @Tim, the benefits of the ODA vs O2 are explained in the article and comments above. How much the improvements matter will depend on your individual needs. And, just to be clear, the ODAC can be added easily to the ODA at any time. So you can start with an ODA amp and use your existing DAC, or something really inexpensive like the the Creative X-Fi Go Pro, Asus U3, or Behringer UCA202. And, later, you can easily add the ODAC board and have a "single box" high performance headphone DAC.

    ReplyDelete
  146. Thanks! I assume you can add the ODAC to the O2 as well (as separate boxes)?

    ReplyDelete
  147. @Tim... stay tuned for more... :)

    ReplyDelete
  148. Waiting for the odac... I already can't wait.

    ReplyDelete
  149. First of all I'd like to thank you for all the time and effort you put into this blog and these projects. Really great stuff going on here. I thought about picking up an O2 but to be honest the ODA/ODAC sounds like it would much better suit my needs. Could you give a rough estimate of when we might see the ODA/ODAC for sale? Thanks in advance!

    ReplyDelete
  150. @Anon, the plan is to release the design before the end of March. That's only a goal, no promises. It will take at least a few weeks beyond the documentation release before there are completed boards/amps for sale. But it should be well before summer.

    ReplyDelete
  151. Hmmm... that looks like the CS4351.

    I may be wrong. I'm prepared to hear it.

    ReplyDelete
  152. I have a question about dual headphone use (i.e. me and my girlfriend) with the ODA or O2 amps. Hopefully this comment thread is an ok place to ask.

    How much does it degrade performance when you split the output to 2 sets of headphones? They would be (most likely) identical models, Fostex T50rp orthodynamic headphones. So the impedance curve is basically flat over the whole frequency response.

    I would prefer not to build multiple amps if one will be sufficient.

    ReplyDelete
  153. @Sam, if the source has a low output impedance (under 1 ohm) and plenty of current capability, as with the O2 and ODA, using two headphones at once won't change the sound quality. But if the output impedance is higher and/or the source has insufficient current capability it may well degrade the sound quality to connect two headphones at once. The other common problem is headphones with significantly different sensitivities will play at different volumes.

    The T50 requires a fairly high output and will suffer using many sources. I would strongly recommend a high quality amp like the O2, ODA, a Violectric, etc.

    ReplyDelete
    Replies
    1. Very good news. I'm patiently waiting on my O2 kit from the latest group buy. Can't wait! I forgot that the loads of 2 headphones in parallel shouldn't really notice one another, if the amp can handle the current. I had in mind the (potential?) problem of wiring speakers in series and having their complex impedance interact.

      This means that we can both use the high quality front output of my Xonar D1 until the ODAC is ready.

      Delete
  154. I'm wondering if you further justify your decision to not include optical input to the ODAC. There are many applications of it like being able to use legacy sources, and from PC, allowing you to use the sound card for DSP effects.

    I understand that it does not exactly fit in with the project's motive, but will it necessarily add a whole bunch of extra circuitry and degrade jitter performance? Well, I guess it would completely depend on the ODAC design, but how hard would it be to go down the DIY route to add an optical input if I need one?

    Cheers

    ReplyDelete
  155. @Anon, I think it will be relatively difficult to add an optical input to the ODAC. As previously discussed in the comments, it would require a new and larger PC board and entirely different topology to switch between the I2S USB data and I2S S/PDIF data. It's possible a more expensive larger board may be offered someday with S/PDIF and it would still hopefully be "plug compatible" with the ODA as a daughter board.

    ReplyDelete
  156. Just wondering if you would release the gerber file for the ODA, like you did with the O2?

    ReplyDelete
  157. @Anon, yes there will be gerbers for the ODA (not sure about ODAC).

    ReplyDelete
  158. I just dug into what it would take to put S/PDIF into the ODAC and it looks like a huge undertaking to add ~35 additional componenets and a way to resync clocks when the input is switched. I'm sure some DIY guru will whip up a seperate DAC for those who want S/PDIF instead of USB, but it'll be quite the feat to have both in the same box.

    Anyway, I see merit in it but it would just get out of hand in terms of your design objectives. I think you're spot in in assuming it's the computer listeners who will use your designs most and will also be best served by the cost:quality ratio you're after. This will definately be somewhat of a shocker in terms of total cost compared to the O2, but it seems like it'll be great.

    ReplyDelete
  159. I'm not sure exactly where to ask this, but what are the issues with Audio GD's stuff? I ask because I recently got an NFB-5 and I want to make sure I'm not drinking their kool aid. That's the exact kind of thing I've been trying to avoid, in fact.

    In their defense though, I will say that most of their designs seem to be based on previous iterations. The NFB-5, for example, appears to be an NFB-12 with a better user interface.

    ReplyDelete
  160. @Phos, In my opinion (and I know many others agree) Audio GD does indeed want you to drink their kool aid. You don't have to read very far on their website to encounter total nonsense that has no basis in fact.

    Try Googling "Samuel Groener" AND "op amp" and you will hopefully find his massive paper on op amps. He tested a couple Audio GD discrete op amps and they performed horribly. That same zero feedback design philosophy is applied to many other Audio GD products. It just doesn't work very well if you're interested in accuracy rather than audible distortion.

    If you're happy with your NFB-5, that's fine. But if accuracy and transparency are the goal, you should be shopping elsewhere.

    ReplyDelete
  161. Thanks for the reply, I'm fine with my NFB-5 for now, but if it isn't accurate it's not really what I was looking for. I'll probably sell it on Head-Fi, they love kool aid there. I'll probably get a hold of an ODA set up. It'd be nice if the DAC would support Toslink for my set up so I could run it out of my X-Fi for games, but I suppose I could substitute an analog connection to the ODA's line in. If I want a discrete DAC that does have Toslink, I'm not sure where to start after reading about how few of these manufactures do proper tests.

    Though one thing I don't get is the bit on it running in 24 bit mode which you also say you aren't planning to support over USB, or perhaps I am misunderstanding the purpose or circumstance if its use. If you're only sending 16-bit audio, turning it down in windows is just going to send less.

    Speaking of Head-Fi, what are they going to do when Tyll gets around to measuring amps made by sponsors, I see that going badly, but no idea exactly how.

    ReplyDelete
  162. @Phos, I'm not sure what you mean about 24 bit vs 16 bit. it can be a confusing topic. Just to be clear, I'm talking about the data mode used between the PC and the DAC. With 95+% of USB DACs, the PC can only send 16 bit data to the DAC. That's all the USB interface of most DACs allows. But some newer and higher-end DACs will accept 24 bit data via USB. It's important to note a lot of DACs that advertise 24 bit support only support 24 bits via S/PDIF not USB.

    The ODAC supports 24 bit mode via USB. For any DAC you'll never get a full 24 bits of effective resolution (24 ENOB). The FiiO E10, for example, only manages about 16.5 ENOB. But even just a few bits over 16 is enough to allow use of software volume controls with minimal or no penalty. Put another way, a DAC that delivers close to 16 ENOB is "good enough" if you always leave the software volume control(s) at max. But if you turn down the volume on the PC side, you ideally want a 24 bit USB DAC with 18+ ENOB.

    ReplyDelete
  163. NwAvGuy, greetings and hope all is well with you and yours. Thank you again for all you hard work and effort to bring sanity to this "hobby". In regards to the iPad (just for clarity), IIRC it is specifically the 'iPad2' that supports 24bit audio output via the camera dongle. Also, some may have missed in the Benchmark Media paper this quote: "In this graph, there are three measurements of the same 30 kHz digital tone going through the DAC1 USB. The red measurement is the 30 kHz tone coming from the iPad. The cyan measurement is from the Macbook via optical. The blue measurement is from the Macbook via USB. As you can see, they each perform identically, achieving full 96-kHz, 24-bit performance." So for the majority there is no performance benefit offering optical input on the ODAC since the USB performs equally well.

    ReplyDelete
  164. Hi NwAvGuy,

    Is anything gained over 16 bit with 24, if the PC's volume settings are left at maximum? I ask because I have an Emotiva XDA-1 on the way and I'm wondering now if I bit the bullet too quickly.

    If I decide I've gotta have 24 bit output from the computer in the future, are there any inexpensive sound cards with S/PDIF output quality worthy of the XDA-1 and o2? I would be grateful for any advice.

    That said, I want to thank you for all the effort you have put forth. I've learned a lot, and relearned plenty, too.

    Thanks,
    Rob

    ReplyDelete
    Replies
    1. If you leave the PC/software volume all the way up, 16 bits is fine. I haven't tested the XDA-1, but like most gear from Emotiva, it seems like a bargain. I've had an Emotiva power amp on my test bench and it was impressive. Emotiva is sort of the home audio version of Behringer--well designed gear built inexpensively.

      Delete
  165. NwAvGuy, Thank you for taking the time to research, develop, test, and produce this device. It's not often you come across an individual, let alone a corporation, that is willing to invest the time, research, and money to properly develop AND TEST a product on his/her own time and at personal expense. It's refreshing to say the least. And I'll be eagerly anticipating this device's release. :)

    Now a question or two. ;) Do you think that the ODAC will work with the iPad with the USB/Camera Connection Kit or any of the newer Android Tablet devices that support USB Host mode? I have a friend who is using the iPad 2 with the new Pure i-20 dock to get digital output, and he seems happy with it, but I am interested in 7" tablets and Android seems to be the best option. I'd like to bypass the tablet's headphone amp, and preferably the internal DAC as well.

    Bill B.
    bbfoto AT hotmail DOT com

    ReplyDelete
    Replies
    1. I have a first generation iPad but I've never tried to use it as a USB host as I don't have the cable/adapter. Supposedly the iPad (mostly) supports UAC1 audio devices. If that's true it should work with the ODAC. But don't count on it until it's been tested. I'm not clear on what, if any, differences there are between the iPad 1 and iPad 2 regarding USB audio.

      As for Android tablets acting as USB audio hosts I know even less. I do have a Samsung Galaxy Tab 8.9 (way lighter, slimmer, and arguably faster, than even an iPad 2) but I haven't heard if it can send audio to a USB DAC.

      Delete
    2. According to Android specifications, the USB host mode is supported in Android 3.1 or newer platforms. Android devices and USB_CLASS_AUDIO compliant devices like standard USB DAC can interoperate:
      http://developer.android.com/guide/topics/usb/index.html
      http://developer.android.com/reference/android/hardware/usb/UsbConstants.html

      The USB audio driver for ALSA (Advanced Linux Sound Architecture), present in the Google Galaxy Nexus source code, is compliant with USB audio class 2 devices allowing for 24/192.
      https://bitbucket.org/paulobrien/android_kernel_galaxynexus/src/bc4f9b72a51d/sound/usb/card.c

      Currently at least four Android tablets are modded to enable USB audio out:

      . B&N Nook Color
      http://forum.xda-developers.com/showthread.php?t=1459892

      . WitsTech A81G
      http://android.modaco.com/topic/343858-external-usb-devices/page__st__20

      . Ainol Novo 7 Advanced Allwinner with modding instructions
      http://www.slatedroid.com/topic/29289-usb-audio-cards-support-drivers-and-libs/

      . Archos G9
      http://update.archos.com/9/gen9/changes_firmware_archos_android_gen9.htm

      The Google issue "Enable USB audio", requesting Google to activate the feature "USB audio out" on Android smartphones/tablets, is currently ranked 24 of 18252.
      http://code.google.com/p/android/issues/detail?id=24614&sort=-stars&colspec=ID%20Type%20Status%20Owner%20Summary%20Stars

      Delete
  166. Hi NwAguy,

    Thanks for all your contributions. Always a great read. I know that you say that the ODA/ODAC will not come out probably until May or so but I am in the process of upgrading(budgeting - headphones/amp) and would your input for a ballpark figure for this amp/dac. Does $400 sound about right?

    Also, would Grado's be powered well from this amp?

    Thanks...Dave

    ReplyDelete
    Replies
    1. The ODAC board, fully assembled, will be under $100 and a fully assembled ODA, with enclosure, will hopefully be available at least some places for under $200. So I would expect under $300 to be possible for a fully assembled ODA/ODAC pair and considerably less for those willing to build the ODA themselves. If China Inc ever gets on the bandwagon a complete ODA/ODAC for under $150 should be possible.

      Delete
  167. May I also recommend the inclusion of test points on the board? It's a lot more convenient to say 'Test the voltage between TP1 and TP3' in the initial test phase, than to say 'Test the voltage between pin 1 of X and 3 of Y'.

    The best ones are like seen in this photo - http://frankeleveld.com/picture_library/monctrl-psu2.jpg - the little red things with a metal loop on top. Very easy to hook up crocodile clips or probe leads to.

    ReplyDelete
  168. Hi NwAvGuy,

    I submitted a question here a few days ago and the comments count has jumped from 200 to 205 since then, so I'm assuming you responded (thanks for that). I'm writing this time to let you know there may be something wrong with your blog software.

    Clicking the "Load more..." link beneath this field doesn't seem to work. I've tried it with Firefox, IE, Safari via an iPhone 3gs (both the mobile and the web versions of the site), and an Amazon Kindle. The best behavior I've gotten from it is a "Loading..." busy message that seems to generate no network activity. IE complains about the script impacting the PC's performance, but I couldn't seem to get any specifics out of it (I'm far from being a web dev.).

    I just wanted to let you know since, if the issue is server-side, a lot of other people won't be able to see comments past number 200 either.

    Thanks again,
    Rob

    ReplyDelete
    Replies
    1. As I posted on my "Feb 1 Update" article, Google is having some real problems with the comments portion of Blogger. I've had a variety of issues the last few days including exactly what you describe. I've gone back and deleted some less-than-relevant comments to get the total back under 200 for now.

      Delete
  169. Heck yes this answers my previous question, cannot wait to see this come up. Looking forward to this ever so eagerly

    ReplyDelete
  170. Will the ODA + ODAC combination have extra RCA outputs from the ODAC so that it can be used simultaneously with active desktop monitors or an electrostatic headphone amp, for example?
    Thanks, I'm really looking forward to this project, by the way. I've been reading your blog since this past summer, I'm a huge fan.

    ReplyDelete
    Replies
    1. No promises but that's the current plan. Although if you have only electrostatic headphones, you don't need the ODA. :)

      Delete
    2. Hi NwAvGuy,
      Really appreciate all the great work you're doing on your blog, breaking snake oil bottles with science :)

      I'm planning on buying the Hifiman HE-500
      http://www.innerfidelity.com/content/comparing-world-class-headphones-hifiman-he-500

      And I thought I'll buy the ODA as a desktop amp for the HE-500, since all I have is my Yamaha receiver 150mV/100Ohm headphone output.

      Can you please explain why I don't need the ODA, and maybe give me a recommendation?

      Thank you very much!

      Delete
    3. The HE-500 are not electrostatic, they're planars. According to InnerFidelity they have a 47 ohm impedance and need over 3 Vrms to play reasonably loud on peaks. Driving 47 ohm headphones from a 100 ohm source is a bad idea. See Headphone Amp Impedance and Sonic Advantages of Low Impedance Headphone Amps.

      The O2 or ODA will drive the HE-500 flawlessly. Your receiver won't.

      Delete
    4. Thanks, my bad for mixing the two up...

      I'm trying to find a pair of HP which will be reliable\flat\natural enough for me to use in my production studio, making mixing\mastering decisions with them. Any suggestions?

      And when should we expect the ODA? the idea of having an O2 as a desktop amp with extra RCA outputs is great.

      Thanks again for your help, I really appreciate that.

      Delete
  171. I don't know if this already answered, I tried read all this discussion but lot of them are technical so I skipped them, a lot. Sorry again. So I'm wondering, would it possible to convert O2 into ODA? I like the sound of O2 and since I'm not using them for portable it's kinda hassle to re-charging it every time...

    ReplyDelete
    Replies
    1. First of all, you can just remove the batteries from your O2 and run it from AC power all the time just like a desktop amp. If you want a bigger case, 1/4 inch headphone jack, RCA input jacks, etc. there are panels available for using the B3-080 enclosure (which is taller). See the O2 Details article in the construction section and also the diyAudio thread. Many have posted pictures of their desktop O2 amps there.

      Delete
    2. Thank you very much for your explanation sir! Will do. I'll ask my friend who good on electronic to do it :)
      I just want to remove the batteries...
      All those electronic term made me dizzy! (really, I'm not one of those electronic savvy guy)

      Delete
  172. ODA and Home Theater Bypass. I'm wondering whether or not you've thought about having a switch to allow for home theater bypass with the ODA? If it has a preamp out, I'd love to just hook it directly up to my main speakers and get as much out of my signal path as possible. Then when I need to play movies, I can just hit the HTBypass button and presto the fronts are just getting driven from my Preprocessor.

    I guess this might not be in the ODA design and rather the packaging, since HT Bypass might simply be a good passive physical switch.

    ReplyDelete
    Replies
    1. The ODA+ODAC can only accept a 2 channel line input or a USB audio signal from a PC. You can run the front L/R pre outs from your A/V receiver into the ODA and wire the ODA pre outs to your power amp. Then you could listen to either the A/V system or USB computer audio over your front speakers. I think you're asking for a switch that would connect the ODA line-in to the line-out to bypass the ODA entirely when you were not using it? That could be done with external switching, or as a DIY modification inside the ODA enclosure.

      Delete
  173. One thing that hasn't come up so far was the power supply. I'm guessing you'll go for an AC-AC one agin, so as to use a real ground? Since these don't seem to be as common in Europe as they are in the US, please consider, if at all possible, using the same power supply specs for the ODA as for the O2, so as to allow the reuse. Epiphany has shipped their O2s with a 12V 500mA power supply, so it's mostly just the voltage and connector requirement that is critical.

    Also is there any chance of having an additional 1x-gain setting available on the front/back-panel switch, without having to cut connectors inside the casing? The 2.5-setting really is too much for most earbuds/in-ears, and opening up a pre-assembled unit may isn't ideal with regards to warranty, etc. I can see how this might complicate the PCB layout though.

    ReplyDelete
  174. more i/o beyond the basics is good, but not if it turns the whole thing into a big mess. one step at a time. you can always do another design.

    analog and usb in, analog headphone with volume control out, analog line out. bring me your little ones

    ReplyDelete
  175. Is it going to be possible to select the output (1/4 and 3.5mm jacks) path with the ODA+ODAC, either in software or hardware?

    I ask this, as I would like to connect a 2.0 Multimedia Speaker system (3.5mm jack) and my IEMS at the same time, obviously without both playing.

    Many thanks and good to see a serious engineering approach.

    ReplyDelete
    Replies
    1. You're welcome. The ODA will have both headphone jacks always active. It will also have an RCA line output that may be switchable. I haven't sorted out that detail yet but I can understand why some might want it. For most, however, it's fairly easy to just shut off your powered speakers or whatever amp/receiver is driving your speakers when you only want to listen to headphones.

      Delete
  176. Will this ODA+ODAC run 24/96 in Mac OS X?

    ReplyDelete
  177. Hi NwAguy,

    Is there a way to add a separate line out to a subwoofer.

    ReplyDelete
    Replies
    1. You can just use a "Y" splitter off the main line-outs.

      Delete
  178. Any update on when the ODA will be available?

    ReplyDelete
  179. Just wondering, does this have an ETA? Sorry I'm not sure because the site is a bit hard to navigate.

    ReplyDelete
  180. I cant wait for the desktop version with the dac and powersupply!!

    Any updates on the timeframe??

    Thanks again
    Alex

    ReplyDelete

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.