MEASUREMENTS: Laptop Audio Survey - Apple MacBook Pros, Acer Aspire, ASUS Taichi...

Now that the asynchronous CM6631A USB to SPDIF device works well on the Mac with full access to sample rates after removing that reset chip, let's have a look at a few of the laptops I have around here - how the internal DAC's measure and compare this to the effect of disparate hardware on adaptive USB and asynchronous USB DAC's. First up, let me introduce you to the laptops I'll be testing:

1. Early-2008 17" MacBook Pro (MBP):

     CPU:  2.6GHz Core 2 Duo
     Memory: 6GB 667MHz DDR2
     Graphics: nVidia GeForce 8600M GT 512MB
     OS: Mac OS X 10.8.2
     HD: Corsair GT 240GB SATA SSD
This has been my work laptop for a few years until a couple months back. Essentially top-of-the-line laptop back in 2008 when I bought it. Still looks good and works well! I updated the hard drive to an SSD and upped the 4GB to 6GB about a year back.

2. Mid-2009 15" MacBook Pro (MBP):

     CPU: 2.26GHz Core 2 Duo
     Memory: 8GB 1333MHz DDR3
     Graphics: nVidia GeForce 9400M 256MB
     OS: Mac OS X 10.8.3
     HD: WD 640GB SATA
My wife's work laptop. One of the Apple UniBody construction machines. The only laptop in this roundup with a hard drive rather than SSD.

3. Acer Aspire 5552-7858 (17"):

This is the least expensive of the laptops; priced at <$600 when I bought it new in 2011. This is also the machine I've been using over the last months for all my "mobile" data collection such as the Oppo tests at my friend's house. Mostly used by my kids and guests if they want to look something up. SSD upgrade about 6 months back.

     CPU: 2.2GHz AMD Phenom II X4 N970

     Memory: 6GB 1333MHz DDR3

     Graphics: ATI Mobility HD4250

     OS: Windows 8 x64

     HD: OCZ Vertex2 120GB SATA SSD

4. ASUS Taichi 21 DH51 (11.6" dual screen):

My newest "ultrabook" laptop (bought February 2013). I know the reviews are mixed but IMO, this is a fantastic mobile laptop / tablet with a high resolution 1080P matte screen in laptop mode and glossy touchscreen as a tablet! Everything in one light and mobile package; fantastic to use in the local coffee shop to update one's audio blog :-). Great for work and although the battery life is at best average in power saving mode, good enough for me with light Office duties.

     CPU: Intel Core i5-3317U (1.7GHz)

     Memory: 4GB 1333MHz DDR3

     Graphics: Intel HD 4000

     OS: Windows 8 x64

     HD: 128GB SSD mSATA

Note that this unit doesn't have USB2 - tested with USB3 ports.

I. NATIVE DAC (RightMark Analogue output analysis):

First, let us have a look at the internal DAC's in these laptops. Measurements were done as follows:
Laptop headphone out --> shielded phono-RCA cable --> E-MU 0404USB --> Aspire AMD laptop

In order to try to ensure "bit-perfect" output, I used Decibel 1.2 on the Macs (turned off all volume adjustment, allowed Decibel to have exclusive access, buffered audio files to memory). On the Windows 8 side, playback was with foobar2000 1.2.4 (latest stable version) playing back with either WASAPI with no dithering or ASIO if driver available.

As you can see, each laptop demonstrated significantly different measurement results. Every one is capable of handling 24/96. The frequency response graph for each machine looks different, especially between the MacBook's vs. Windows machines (perhaps the Acer and ASUS use similar OEM DAC hardware?).

Not surprisingly, the ASUS has the highest noise floor of the 4 machines. It is an "ultrabook" so there's a lot of electronics packed into that case plus the headphone out jack is right underneath the screen hinge (instead of closer up front by the keyboard like the other machines).

I'm impressed by the MacBooks though with >100dB dynamic range! That's fantastic for laptops and certainly as good as lesser "audiophile" DAC's out there.

 II. Adaptive Isochronous USB (RightMark Analogue output analysis):

Using the AUNE X1 (Mark I) as the DAC, let's look at the relative performance of each machine using the X1's USB interface - the old "adaptive isochronous" BB PCM2707 USB1 technology measured at the usual CD sample rate of 16/44.

Test laptop --> Shielded USB cable --> AUNE X1 --> Shielded RCA --> E-MU 0404USB --> Aspire laptop

Notice how close the results are except for the Aspire-X1 results. Obviously since we're now utilizing the same DAC for each machine, the sound output is dictated by the properties of the DAC. The difference with the Aspire-X1 setup is that the Aspire is both the test machine and the measurement device (loopback setup). This means the Aspire needed more CPU processing and had much higher USB traffic than the other machines (maybe the mixed USB1 & USB2 modes have something to do with this). What we're seeing is likely electrical noise generated by the busy machine, and as I'll show later, this is NOT likely due to significant timing jitter in the digital domain.

Otherwise, I see no significant difference in measured "sound" coming out of the DAC whether the machine is Mac or Windows despite the significantly different underlying hardware, OS, playback software, etc... 

 III. Asynchronous USB (RightMark Analogue output analysis):

Test laptop --> shielded USB2 --> CM6631A --> inexpensive decent 3' TosLink --> AUNE X1 --> E-MU 0404USB --> Aspire laptop

Look at what happened with the asynchronous and very inexpensive ($50) CM6631A performing USB to SPDIF duties! Audio output from the AUNE X1 DAC is now IDENTICAL from every laptop. This shows that compared to the USB1 input to the X1, the CM6631A converter, in using TosLink provides excellent electrical noise rejection from the USB port (note that I'm running the converter USB powered; NOT separate power supply even). I think this is a nice demonstration of how the TosLink interface has its place despite measurably worse jitter in some cases.

IV. Jitter - Native DAC:

The Dunn J-Test was created to look at jitter in SPDIF interfaces, therefore there really should not be any issue with built-in DAC's for each laptop. Reminder, "bit-perfect" settings for Decibel with the Mac's, foobar2000 WASAPI 16-bit undithered for the Windows machines used. Here's how the 16/44 J-Test result looks for each machine:

2008 17" MacBook Pro:

2009 15" MacBook Pro:

Acer Aspire:

ASUS Taichi:

Quite different spectra as you can see. What is learned from the graphs is NOT jitter but rather the loss of the jitter modulation pattern with the 2008 MacBook, Acer Aspire, and ASUS Taichi. This means that despite using bit-perfect settings with the players, these computer DAC's are incapable of "bit-perfection". Other than the 2009 MacBook Pro, all the other machines either truncate or perhaps have some kind of 'forced' dithering in place such that the LSB jitter modulation has been corrupted.

V. Jitter - Adaptive Isochronous USB AUNE X1:

Using the same method, lets look at the jitter pattern with the USB1 interface built into the AUNE X1 (Mark I). Again either using Decibel or foobar2000 with bit-perfect settings in place. USB cable is generic shielded good quality 6' cable:

2008 MacBook Pro to AUNE X1:

2009 MacBook Pro to AUNE X1:

Acer Aspire to AUNE X1:

ASUS Taichi to AUNE X1:

As you can see, the jitter modulation tone can be see in each situation. This confirms that indeed bit-perfect data is being sent to the DAC containing the LSB data. If anything, it looks like the Mac's may have a few more jitter sidebands than the Windows machines. If you recall from the RightMark results above, the Acer Aspire is performing both playback and recording duties in this situation so you can see that the noise floor is higher than for the others (noise increases significantly in the lower frequencies as shown in the RightMark graphs) likely due to the increased CPU and USB activity causing noise to spill over into the X1 DAC through the USB electrical interface.

VI. Jitter - Asynchronous CM6631A USB to TosLink to AUNE X1:

What about when the CM6631A is used? Two things are happening here - USB communication is now asynchronous allowing better clocking (ie. lower jitter), and galvanic isolation is also happening by way of the TosLink SPDIF intermediary. As I measured in a previous blog post, the CM6631A appears to have just as low jitter whether with coaxial or TosLink.

2008 MBP to CM6631A to AUNE X1 (TosLink):

2009 MBP to CM6631A to AUNE X1 (TosLink):

Acer Aspire to CM6631A to AUNE X1 (TosLink):

ASUS Taichi to CM6631A to AUNE X1 (TosLink):

Notice how the jitter spectrum looks almost identical now just as the RightMark measurements above look essentially identical in each case with the CM6631A. With TosLink galvanic isolation, the Aspire's noise floor is in line with the others and in each case, the jitter modulation signal is well defined, again proving that Decibel (sending to the Mac's native USB 2 Audio driver) and foobar2000 (using the C-Media ASIO driver now) are capable of bit-perfect output when the hardware allows.

VII. SUMMARY:

1. Even though most laptops/motherboards these days advertise their audio hardware as "HD Audio", they are generally nothing of the sort. Although the MacBook Pro's showed good dynamic range in the RightMark tests at 24/96, the 2008 model was incapable of true 16-bit processing. Likewise, neither of the two Windows 8 laptops were capable of preserving the 16-bit LSB jitter modulation pattern. Perhaps obvious, but the bottom line is that you cannot trust the built-in DAC's for even 16/44 resolution audio.

2. Watch out for noise creeping into the DAC from the USB interface as shown with the Aspire-X1 Adaptive USB example. This can originate from internal electrical components as in this case or other times ground loops. Galvanic isolation with the TosLink interface is one way of dealing with this.

3. Overall, the asynchronous USB interface is better than adaptive isochronous USB for jitter in this example (obviously I cannot vouch for all asynchronous DAC / interfaces). This has been demonstrated before already and seen again here. These days it's a moot point since asynchronous USB interfaces are readily available and if your DAC doesn't have one, just grab an inexpensive converter like the CM6631(A) I used here. There are of course some very expensive USB-SPDIF adapters but given the CM6631A results, why spend more from a sonic perspective? As I have also stated before, I still have no good evidence to say that jitter at this low level is ever audible with real music despite what subjective audiophile reviewers or Internet posters generally claim! In ABX tests and even non-blinded flipping between measured jittery vs. clean audio sources, I cannot tell the difference at least up to ~2ns peak-to-peak jitter; nevermind the usual few hundred picosecond jitter with decent modern gear [Ed. apologize for the error...  I previously had a typo of "2ms"]. It's possible that my ears aren't sensitive to this but to be honest, unless one has superhuman hearing abilities, I can't imagine how it would be physiologically possible given these test results.

4. If you keep electrical noise/interference at bay, and use appropriate software player & drivers to ensure bit-perfection, there is no reason why Macs vs. Windows vs. ANY computer would 'sound' any different paired to a good DAC. As you can see above, when the CM6631A asynchronous USB-to-SPDIF converter is used and galvanic isolation is in place, there's essentially no difference in the analogue RightMark measurements and the Dunn J-Test spectra irrespective of CPU, speed, memory, HD/SSD, OS, USB, etc. (obviously as long as the specs are good enough for bit-perfect output - this is the beauty of digital). (Here's another interesting article showing identical Mac vs. Windows output using JRiver and different methodology.)

I hope this article has been useful in demonstrating a few things about computer audio. Measurements allow an opportunity to verify claimed improvements and demonstrate technical progress in hardware performance - amazing that few audiophile magazines / web sites actually include independent objective measurements and reviews come out looking like infomercials and PR. Realize that none of what I've shown is all that mysterious or requires audio "voodoo" whatsoever - science is all you need; let the musicians provide the passion and art. The cables are all generic, reasonably high quality (ie. nothing looking like they're going to break!), and universally costs <$20 for standard 3-6' lengths. The AUNE X1 DAC can be bought off eBay for ~$200. The CM6631A USB to SPDIF costs $50. The analogue and jitter measurements are on par with much more expensive gear I've come across. No need to further obsess over details like jitter or bit-perfection because the objective data has set the mind at ease...  If this were my main listening system, this then is where I would start more subjective evaluation with the amp and speakers asking myself whether I prefer the sound of the technically accurate setup or not. Realize that not everyone wants technical accuracy - terms like "clinical" or "sterile" impart this pejorative connotation. IMO, there is no such thing as a system being "too resolving", there's only source material that's "not good enough"!

After all this, you might ask: "So how does it sound?"
     Great!  (No need for superfluous adjectives.) Now go enjoy some music...


-------------------
For posterity, a response I put up on Audio Asylum questioning the results (April 21, 2013)...

http://www.audioasylum.com/forums/pcaudio/messages/12/122768.html

Gordon,
I appreciate your feedback and critique. I used to have the QB9 and indeed it is a nice DAC so was quite happy that Stereophile's measurements concurs.
For this article, I believe the data was "bit perfect" since the LSB jitter modulation was evident in the 16-bit tests. Due to the hardware limitations, this was of course not seen in the 24-bit graphs. I did discuss this fact and the issue that the built-in computer DAC's had problems with bit-perfection.

Thank you for bringing up the "skirt" issue. I'm going to have to update my ASUS Essence One measurements because this was worse with the USB input but not the coaxial or TosLink. However, most of what I've seen suggests that this low frequency jitter is masked by the primary signal. Please correct me if this is wrong so that I may pay more attention to it especially if there's a point of audibility.

As for "at some point you have to listen to the damn thing" (I'm directing this to the general viewership, not just to your comments). It is precisely because I've been listening AND reading forums like this for the last 15-20 years that I started doing the measurements and putting up the blog! I don't think it's at all a stretch to say that people are notoriously unreliable in their experience... I have never found an audiophile who could reliably differentiate high bitrate MP3 for example, yet when I ask, everyone seems to say they can (sure there are people who can ABX 320kbps but that's very few). For those who have a copy, listen to track 26 on Stereophile Test CD2 simulating 10ns jitter. I would estimate that half of the audiophiles I've tried this on over 50 years old have difficulty hearing this simulation on a test tone yet they universally claim that a few hundred picosecond jitter is a "problem" audible in actual music.

There's precious little objectivity out there anymore for audiophiles - look at that bizarre 3-part article on computer audio in TAS early 2012; if "everything's possible" (bit-perfect rips sound different if ripped at different CD speeds according to that article) then nothing is known anymore. Just look at the numerous comments when those new to computer audio asks straight forward questions here. IMO, some of the responses border on delusional.

Time is limited but I have spent many many hours listening to cheap to very expensive gear - even owning a few of the more expensive items over the years. It is my belief that "good" sound can be independent of factors like price, or any of the external factors like workmanship, etc. In fact, I've come across situations where high priced gear are just plain inaccurate. For me, accuracy is all that matters, and this can be quantified objectively.

Speaking of "misinformation is what is sending most people into shock when they start CA". Realize that over these months of testing, I have not advocated anything I would consider as "fringe". I've shown that asynchronous DAC's are better with less anomalous J-Test spectra, coaxial and good USB is better than TosLink in the gear I have, shown 'good' DAC's like the Oppo BDP-105 (ESS Sabre) measures superbly. I advocate for "bit perfect" and showed that both Windows and Mac are capable of accurate output, and have interspersed these tests with listening to make sure nothing appears awry - usually spending a couple hours at a time in the evenings listening (eg. when I was doing some cable measurements recently). As I've said in some posts, there is no voodoo or magic here, and the tests are showing me such. How in the world is this shocking anyone out of computer audio!? If anything, it's reassuring that measurements line up with a rational empirical approach that is reproducible.

Now as for the test gear I use and software like RightMark. You don't need the Hubble telescope to identify Jupiter in the sky. Likewise, why do I need expensive gear like an AP when I just want to make sure the frequency response is relatively flat, or that the dynamic range of a DAC can exceed 16-bit CD quality, or that the Dunn test isn't atrocious? All these things are within reach of what I have and the beauty of computer audio and technological advancement is that stuff that can do this is easily within the consumer's grasp. However, I'm not saying that anyone should do this since it has taken me countless hours to learn how to get the calibration right and make sure the software setup works for me in order to achieve a high level of reliability in the measurements.

When my simple setup can easily demonstrate analogue cable differences, XLR vs. RCA differences, show me the spectral smearing between 320kbps MP3 vs. 400kbps AAC, demonstrate J-Test differences between coaxial vs. TosLink vs. AES/EBU, when I post about this, in what way is this "misinformation"? In fact, my ASUS Essence One measurements showed a disturbing anomaly when upsampling is used resulting in attenuated frequency response which has since been confirmed by ASUS (and hopefully to be remedied)... How come subjective Essence One reviewers missed this when it was so obvious with just a little testing?

Look guys, ultimately either what I say and write about makes sense, or it doesn't. I mention a new post up here once awhile that could be of interest like this one with the laptop tests to outboard DAC's. I have ZERO financial interests or otherwise. I do it for fun, for my own education, and enjoy sharing what I've found with others. Anyone can freely share their thoughts in the comments section (haven't had to censor anyone at this point for nonsense), and I've invited people to give me evidence/reason if they think something I say is wrong. Of course, with my objectivist mindset, "Level 4 or 5" evidence (ie. individual or series of reports) doesn't really impress compared to experimental results or controlled trials.

Cheers,
Arch

MUSINGS: On SACD & DSD audio...

SONY multi-page spread (with hybrid SACD 'RS500' sampler!) - Rolling Stones magazine, December 2003.

Okay, now that I have posted a number of measurements over the last months, permit me a moment to play "Speakers Corner" and talk a little about the state of affairs around DSD now in early 2013...

Firstly, remember that DSD isn't anything new. It's basically a digital format based on Delta-Sigma modulation which has been used in consumer digital audio since the late-80's and early 90's - remember those old Sony 1-bit and Panasonic MASH players from around that era? It was said that as the CD patents were set to expire, Sony and Philips decided to create a new disk format for the 21st century - hence the birth of SACD around 2000 using the "1-bit" encoding method instead of multibit-PCM carried on DVD-like disks capable of higher data capacity (multi-bit ended up being DVD-A of course). Sound quality was said to be better because of things like "100kHz frequency response", better noise floor and in time, multichannel. Of course in creating this new standard, copy protection was job one and they did a great job. I remember in 2004 visiting China and seeing pirated DVD-A's which were easily "cracked" soon after release, but not so SACD's.

MEASUREMENTS: Oppo BDP-105 does DSD.

Well, it's out...  The March 26, 2013 BETA firmware for the BDP-105 that allows native DSD playback from this unit's USB ports as DFF and DSF files on a USB stick. As far as I know, there are no plans currently to allow computer playback connected to the USB port as a DSD DAC.

With this recent development, I headed back to my friend's place to run a few more tests...

Here's the basic premise of what I did...

Using the freely available KORG AudioGate 2.3.1, synthetic 24/192 test signals from RightMark 6.2.5 were converted over to DSD64 (2.8MHz sampling) and DSD128 (5.6MHz) for testing. We soon found out that the DSD128 files created could not be played back properly by the Oppo (it would play but the DSD128 files had timing issues - played too slow). Presumably there is a bug here somewhere with the beta firmware or the AudioGate converter. As such, I was only able to test DSD64 playback.

The first thing done was to go through the Oppo's settings menu making sure there were no volume settings, bass management, etc. active. I believe if any of these are turned on, the DSD will get converted to PCM.

The hardware setup is similar to what I did with the original BDP-105 tests (only difference being use of the front USB connector):
Patriot Rage XT USB2 memory stick 16GB with test files --> front USB port of BDP-105 --> shielded 6' RCA --> E-MU 0404USB --> Win8 AMD X4 laptop

RightMark results:

The first 3 columns are the tests done in PCM mode at various hi-res sampling rates. These essentially measure <1 dB different compared to my original tests using the Oppo's USB asynchronous DAC. Nice confirmation of inter-test reliability.

The last column is with the 24/192 test signal converted to DSD with the KORG software. As you can see, it's almost the same. Note however that RightMark is calculating these parameters just within the audible spectrum between 20Hz - 20kHz (AES17 standard).

Frequency Response:

First hint/reminder of the DSD effect. There's some high frequency noise breaking through up in the 70kHz range. Otherwise, the curves are relatively comparable with -5dB extension out to around 40kHz for DSD and 50kHz for PCM 24/192.

Noise:

Demonstration of the DSD noise shaping through the Oppo. From 20kHz onwards, the noise level rises quite remarkably as you can see. It's all ultrasonic of course so unlikely to cause an audible problem and would only matter if this creates any strain on your amp/speaker system or if nonlinearities cause distortion in the audible spectrum.

Although also not a problem, notice the noise floor from about 12kHz to 20kHz is not as flat with DSD.

THD:

Another view of the ultrasonic noise.

Jitter:
The J-Test cannot be used with DSD of course. This is just for completeness. I've already shown previously that jitter isn't an issue with the Oppo...  Here's just what the 24-bit Dunn J-Test looks like going through DSD transcoding.

If we compare it to the PCM:

Note the loss of the regular modulation pattern. Basically this is telling us that the LSB in the 24-bit signal has been affected and effectively dithered over by the conversion process to DSD.

Analogue Output:
Lets now have a look at what a 1kHz -6dB sample looks like after going through the DSD process. What I did here was record a few seconds of a pure 1kHz test tone in 24/192 to have a look at the waveform zoomed in.

PCM 24/192 FLAC played back:

DSD64 KORG transcoded DFF file played back:

The high frequency noise in the DSD signal can be seen (you might have to click on the images to get a good look). Not a big deal in that this is not audible but a reminder that DSD64 cannot reproduce a simple 1kHz sine wave as smoothly as that produced by the reconstruction filter in the PCM domain.

Impression:
As I had hoped when I wrote that piece on the Pioneer DV-588A last month, here are some results from a device that performs "pure DSD" decoding.

Within the audible spectrum, DSD64 produced by the KORG AudioGate software looks good. Standard measurements like dynamic range, noise floor, distortion are all looking great and reminds us of the high level of performance the Oppo BDP-105 is capable of. I was disappointed that I could not get the KORG-encoded DSD128 test signals to play properly. I don't know if this is due to the KORG software or the beta firmware. Maybe I'd have better luck with Weiss Saracon if I had access to this conversion software... Oh well, maybe next time :-)

As a reminder, all the tests I've shown were converted from the 24/192 PCM domain into DSD64 and therefore will be subject to the limitations of the conversion software and PCM source (note that at 24/192, this is not likely a technical issue).

An interesting observation; even though the encoded DSD128 could not play, free downloads of demo material from 2L worked just fine on the Oppo! They sounded great with a wonderful sense of space, timbre, and dynamics.

Bottom line: The Oppo did a great job with DSD playback just like it did with PCM. Limitations of DSD are clearly seen (ultrasonic noise pollution mainly). From a purely technical perspective, within the 20Hz-20kHz audio spectrum, there's really nothing to differentiate all these hi-res formats. However, if you include ultrasonic characteristics, PCM is definitely cleaner.

Given the frequency response curve demonstrated, this KORG DSD64 conversion + Oppo playback system can likely be encapsulated within the parameters of a good 24/88 system.

MEASUREMENTS: [UPDATE] Adaptive AUNE X1, Asynchronous "Breeze Audio" CM6631A USB, and Jitter...

In late February, I received an interesting USB2 --> SPDIF box from Asia. It's based on the C-Media CM6631A chipset (it says "Breeze Audio" on the metal case and "CM6631-V1.4" on the PCB) which communicates with an asynchronous USB2 protocol (recall that the Asus Essence One uses the CM6631).

Total price ~$50USD shipped.

Although it's said to be plug-and-play compatible with Mac OS X, I've run into a few snags which I won't discuss here - just be aware if you're planning to use this on the Mac (SEE ADDENDUM - ISSUE FIXED!)...  However, I'll restrict these measurements to the Windows 8 environment only where the custom drivers work well.

The idea I wanted to explore was whether using an asynchronous converter like this connected to the AUNE X1 would be better in terms of jitter than the adaptive USB port of the X1 itself. Also, how sensitive is jitter to computer load?

Note that beyond jitter, there are some other obvious reasons to do this... Firstly, using the CM6631A box allows hi-res TosLink/coaxial output up to 24/192 if you don't have a USB2 DAC. Secondly, it might be better to keep peripherals all running at high speed if you're going to be plugging this into a USB2 hub.

Lets take a look at the adaptive USB jitter spectrum directly from the AUNE X1 (remember, only up to 16 bits, so I'll be using the 16/44 Dunn J-Test signal) [UPDATED April 1, 2013 - SEE BELOW]:
(Setup: i7 computer --> shielded USB --> AUNE X1 --> shielded RCA --> E-MU 0404USB)

Numerous sidebands with spurious noise. I guess this is pretty typical of a very pedestrian adaptive USB1 interface. It doesn't "look" good but for the most part, these spikes are down below 100dB from the 11kHz primary tone.

Not only is adaptive USB said to be bad for jitter, but some believe it's especially prone to timing errors if you're multitasking...  Lets see what happens when the CPU is under 100% load (Prime 95, 8 threads), and to make it worse, lets also run the GPU (nVidia GTX 570) 100% with FurMark:

Essentially no difference to the spectrum whether the computer is fully loaded or not! You see a bit of noise showing up between 8-9kHz but it's all very low amplitude (again >100dB below the 11kHz peak).

Let's now put the CM6631A asynchronous USB --> SPDIF in the chain. Here's the setup now starting with a proper coaxial SPDIF 'digital' cable (Acoustic Research brand):
i7 computer --> shielded USB --> CM6631A --> shielded digital coaxial --> AUNE X1 --> shielded RCA --> E-MU 0404USB

This is a very nice 16/44 J-Test graph with minimal data correlated sidebands and most of the peaks are just due to the 16-bit J-Test modulation tone.

What happens with the CPU and GPU running at 100%?

No difference!

What about we use the TosLink output from the CM6631A instead?
(Setup: i7 computer --> shielded USB --> CM6631A --> decent plastic TosLink --> AUNE X1 --> shielded RCA --> E-MU 0404USB)

In this setup, the TosLink interface is almost exactly the same as coaxial (I'm actually impressed by this since TosLink often tends to be significantly worse than coaxial).

What about CPU & GPU running full tilt at 100% using TosLink?

Again - no difference! (Ignore that FurMark banner... Just got in the way of the screen capture.)

OK. We've also heard that a poor coaxial cable could be bad for jitter...  That is, if there is severe impedance mismatch between connectors (supposed to be 75ohms), the theory goes that there could be reflections which could damage the digital signal transitions. Furthermore, it is said that a short 3' cable is worse than a longer cable due to the transition time for these reflections.

What then would jitter look like if I replaced a proper shielded coaxial with cheap "freebie" 3' RCA connector that came with an old Pioneer DVD player (I used the red connector)?

Eh? That's all? Really no different than with the "proper" coaxial cable!

How about with CPU and GPU running 100% using the cheap RCA cable?

Again, that's all!?

Realize that in the analogue domain, I can measurably prove that the cheapo RCA is worse than the shielded proper coaxial digital cable based on noise floor differences. But in terms of jitter (digital phenomenon) which we're stimulating here with the Dunn J-Test, in this "real world" setup, I cannot show any significant difference despite the likely poor impedance match and 3' length! (Note that I'm not saying a cheap RCA cable is as good as proper coaxial in general, just that in this inexpensive setup, it works fine... I have measured in other situations like with my ASUS Essence One paired up to a Squeezebox Touch where a cheap RCA cable actually resulted in very severe jitter issues.)

Summary:

1. The asynchronous USB2 interface works to lower jitter! Having said this, I think we have to also recognize that apart from some noise spikes, the actual jitter sidebands with the adaptive USB interface wasn't terrible and likely not audible given how low in amplitude they were.

2. Whether the computer was running full-tilt at 100% or not, I could not demonstrate any change in the jitter characteristic! This applied to adaptive USB as well as asynchronous USB --> SPDIF. IMO, this makes me question the belief in "software-induced" jitter. I suppose if your computer is totally bogged down such that the audio buffer cannot be filled on time, you'll hear severe audio degradation. However, some people believe even light multitasking worsens sound quality, or the importance of shutting down background OS processes or running "stripped down" OS'es. By extension, some people swear by various playback programs having an effect on the audio quality (ie. Decibel vs. Amarra vs. JPlay vs. Audirvana vs. iTunes vs. foobar...) even if all that's promised is "bit perfect" output. It almost always comes back to "jitter" as the explanation some how. Really? Is there any objective proof for such beliefs or even a testable hypothesis?

As I have hinted in previous posts, even though I (obviously!) believe that jitter exists and is measurable, I do not believe it is audible with music unless extreme like maybe >10ns. It's also worth remembering that research into the threshold of jitter audibility decreases with higher frequencies; but this also correlates with human hearing losing sensitivity the higher up we go - especially as we get older (therefore, sensitivity to jitter should decrease with age). These days, well engineered "high fidelity" gear should not be even close to showing 10ns jitter. Likewise, software player programs should be capable of "bit-perfect" output without difficulty or cause timing errors.

Over the years, I have also tried many software players like Amarra, Decibel, JPlay, etc... Ultimately I've never been convinced of significant differences in sound between them (so long as EQ and plugins not active). Foobar + ASIO driver sounds great to me on the PC side and I use Decibel for the Mac simply because it's inexpensive and works well to switch sampling rates.

Bottom Line: Don't worry about jitter! It's more than likely inaudible in a modern computer system and with decent (not necessarily expensive) audio gear. I see no evidence that high CPU/GPU load makes any difference to jitter. Isolating your DAC from electrical noise polluting the analogue output seems much more important.

ADDEDDUM (April 1, 2013):

I found out why there was the 15kHz peak in the adaptive USB1 tests with the AUNE X1. Had to do with some old ASIO4All drivers I had installed on the i7 computer...  A good reminder of how easy it can be to mess up settings when using a PC for audio playback!

Anyhow, for completeness, here are some updated graphs of the X1 using adaptive USB measured off my Win8 AMD Phenom X4 laptop. Note that even though the graphs look different, the conclusions are the same - asynchronous USB2 is less jittery, and when the computer is at high load, the noise floor worsens for the adaptive USB case only.

Adaptive USB1, AUNE X1 using WASAPI for 16/44 J-Test:

Now with 4 threads running Prime95:

Notice that slight noise floor elevation from 8-9kHz. Not as extreme as my i7 testbed with the powerful GPU running.

Asynchronous CM6631A --> shielded coaxial --> AUNE X1, WASAPI, 16/44 J-Test:

Nice and clean...  Looks just like with the i7 testbed. What happens with Prime95 running?

No change.

Asynchronous CM6631A --> TosLink --> AUNE X1, WASAPI, 16/44 J-Test:

Now again with Prim95 running on all 4 cores:

No difference...

OK. So I have now shown that the jitter results with the CM6631A device is reproducible on 2 platforms (i7 & AMD Phenom X4) as being low for both coaxial and TosLink. Also, the CM6631A device seems quite immune to noise from the main computer (compared to the adaptive USB interface). Both ASIO and WASAPI work well for the CM6631A. Still no evidence that the jitter pattern itself changes with increased CPU load.

ADDENDUM (March 31, 2013):

Noticed this very useful post on diyaudio.com from 'bvs':

I've recently found a solution for issue that caused when CM6631A module is connected to any Mac USB 2.0 ports. When you connect module to Mac it is detected as USB Audio Class 1.0 device and you have no ability to use anything more than 48/16. Module has a reset chip LM810M3-2.93 (http://www.nscrus.ru/content/catalog/pdf/LM810.pdf), but as we can see from the CM6631/32 datasheet, the CM6631A has a power-on self-reset. So, reset chip has been removed from a board and now the module is always correctly detected as USB Audio Class 2.0 device on any Mac USB port. This method has been tested on MacBook Pro, Mac Mini and this CM6631A module: New CM6631 USB Module Assembled Board for DAC3 AD1955 DAC7 WM8741 by Weiliang | eBay LM810M3-2.93 is a 3-pin chip located on the top of CM6631A.
So, with a needle-nosed plier, I went to work on removing that little 3-pin chip (labelled "SA B" on the board)...  Here's the result (chip removed): The unit works as expected on my MacBook Pro's now with full playback up to 24/192, no need for an external power supply, and no need for custom drivers. Remember, this modification is for the CM6631A only. Earth to Microsoft - isn't it about time we got native UAC2 driver support in the OS!? Especially considering that it's been available for OS X and Linux since 2010!