slim.a
1000+ Head-Fier
- Joined
- Jan 8, 2009
- Posts
- 1,228
- Likes
- 24
USB to SPDIF converters shoot-out : EMU 0404 USB vs. Musiland Monitor 01 USD vs. Teralink-x vs. M2Tech hiFace
Purpose of this thread 24/03/10:
Since this thread has been assaulted many times by people wanting to discuss theory of jitter, what is possible to hear and what is not I decided to write in plain words, why I started it.
The reason I started this thread was to:
1. To share my experience with the 4 usb converters I have tested in my system
2. exchange any experience that helped you improve the source: whether it is a usb converter, a digital cable, a usb cable, a media player ...
However, if your main goal is to say that:
1. jitter is not audible in most cases
2. cables do not matter
3. all subjective hearing is placebo and only DBT can prove that there is a difference
Please read the following : http://www.head-fi.org/forums/f46/do...-forum-450574/
And in that case, I will ask you politely to not drag this thread off topic by discussing such subjects.
Such discussions are better suited for a separate thread. You can start a new one and post it in this thread. But PLEASE, do not discuss such subjects on this thread as it is not the purpose of this thread.
System Used :
Main Chain :
Foobar 0.8.3 (KS and Otachan ASIO) --> Wireworld USB Ultraviolet --> USB TO SPDIF converter --> Hi-Fi Cables & Cie Sobek or 18 ft. Belden 1694 BNC Cable --> Audio-GD DAC-19MK3 --> Hi-Fi Cables & Cie Khnoum RCA interconnect --> Audio-GD C2C (w/ upgraded pot) --> Moon Audio Blue Dragon V3 cable --> Sennheiser HD-650
Usb to spdif converters :
EMU 0404 USB
Musiland Monitor 01 USD
Teralink-X
M2Tech hiFace
Power related accessories :
Hi-Fi Cables & Cie PowertransPlus Power Cords
Supra Mains Block
Essential Audio Tools Parallel Filter
Vibration Control :
E&T Spider Rack, Vibrapods, Vibracones, Sandboxes, Brass cones, Acrylic and Fiber carbon sheets,
Herbie's Audio Lab Tenderfoot, SuperSonic Component Stabilizer
Other gear :
Sources : Creative Audio 2ZS Notebook, EMU 0404 usb, Zero DAC, Audio-GD DAC-100
Headphone Amps : Little Dot MKIII, Audio-GD ST-3
Introduction:
Over the time I accumulated a few usb to spdif converters commonly discussed on head-fi, so I decided to write a comparative review of those units. Granted it is a bit tricky to describe the sound of a component that exclusively does its work on the digital level, however the differences do exist and I tried my best to describe how they compare to each other.
Jitter - theory :
There has been a lot of things about jitter. I found an interesting article here on TNT-Audio explaining jitter and the difference between transmitter jitter, line induced jitter and interfering-noise induced jitter.
There is a theoretical advantage of using a usb to spdif converters instead of a direct usb connection to the dac is an added layer of isolation and "jitter rejection". Below is a quote from an email exchange with Marco from m2tech : "S/PDIF receiver which, you certainly know, has a feature which is "jitter rejection", a figure that typically ranges 20 to 30dB: this means that the input jitter to an S/PDIF receiver is reduced by a factor of 10 to 30... 970ps become 100ps quite easily."
This might explain while I found that all 4 units performed better than the direct USB inputs of the USB dacs I have.
I even found out that when using the Musiland as a transport (using optical) for the EMU 0404 USB it gives better results than its USB input.
(Edit -- The intent of this thread is not to discuss jitter theory or treshold audibility of jitter. Please read the following post http://www.head-fi.org/forums/f21/do...-forum-227350/ before posting on those topics.)
Test Protocol :
Before writing my review, I did many A/B tests going back and forth between different units. And besides those A/B tests, I also did extended listening with all those units to get more familiar with their sound as sometimes
To make sure I was not adding any unforeseen parameters, for each A/B test I would use the same usb port, the same usb cable and the same digital cable (used in the same direction).
I did try various digital cables but did most my listening with the sobek digital cable as I was very familiar with its sound. For those who might think that digital cables do not matter, you can find Here a link of an old stereophile article that measures the jitter of various digital cables which show that even the direction of the cable can affect the sound. During my test I tried to beconsistent comparing units with the same digital cable (used at the same direction), and then compared again the units with a different digital cable.
I also tried different usb cables but I ended preferring the Wireworld Ultraviolet usb cable. I did most of my testing with that usb cable except with the m2tech who doesn't need a usb cable.
Ease of use & Drivers :
EMU 0404 USB :
The EMU 0404 usb is a very versatile unit. It can be used as a DAC, headphone amp, and also a USB to Spdif converter. However, it requires custom drivers and the digital output works only with ASIO.
It has optical and rca spdif output but it lacks a BNC output which might be a drawback for some.
The EMU can do 24/192 thanks to its custom drivers.
Musiland Monitor 01 USD :
The Musiland is a wonderful little unit. It needs custom drivers that work up to 24/192. It works with DS, KS and ASIO so that it can be used with any player.
It has BNC, Spdif and optical out. I am currently using it with the 1.0.3.2 driver and it works flawlessly.
Teralink-X :
The Teralink is the easiest to use. It doesn't need custom drivers and works with Direct Sound and ASIO (through ASIO4ALL).
It has I2S, BNC, Spdif and optical out. This one is a "plug and play" unit that performs flawlessly but is limited to 16/44 and 16/48.
It is possible to install custom drivers (thread here) but there are no sonic benefits in doing it (those drivers introduce many additional processing to the sound which I found to cause a degradation to the sound in my opinion).
M2Tech hiFace :
The M2Tech hiFace is a very minimalistic unit : it has only one output (I ordered mine with a BNC output) and it plugs directly into a usb receptacle without the need of a usb cable. For now, it has only drivers that work with Kernel Stream and it works with Foobar, Media Monkey and WinAmp.
DS and ASIO drivers are expected later down the road.
The unit does 24/192.
Sound :
First, all the usb to spdif converters listed in this review outperformed the usb inputs of the DACs I have on hand. So as long as they are used with a decent spdif cable (such as the affordable 18ft. Belden 1694 digital Cable), there should be sound improvement in many dacs (at least with the ones I tested).
Second, I have noticed big differences between usb cables (Edit -- For those skeptics on why there are differences between usb cables, please read the 23/03/10 follow-up). So I settled for the review on the best one I own (the Wireworld Ultraviolet USB) and used it for all the converters and the DACs except for the m2Tech Hiface. Therefore, it is important to note that the m2tech converter has a slight comparative advantage since it doesn't need the use of a usb cable.
Also, for those who might be interested, I wrote a full length review of the DAC-19mk3 (here) that I used for this review. It should give a good idea of a baseline of the sound in my system. I was using the Musiland back then with the Mode A digital filter for the DAC-19 mk3 and the ST-3 head-amp. Since then I upgraded to DF-1704 digital filter (mode B) and the C2C head-amp which made my system more revealing and transparent.
EMU 0404 USB :
I had the 0404 usb for a long time, so I was very familiar with its sound from its analog outputs. When used as usb DAC, the EMU 0404 USB is "bright" sounding, lacks deep bass and the overall sound can be described as "edgy". The sad thing is that those attributes also reveal themselves in the digital output which sound edgy and congested in the top and light in the bass in comparison with the other converters I have.
However, compared to the usb inputs of the Audio-gd DAC-19mk3 or the DAC-100, it is a noticeable step up in resolution and bass tightness.
Also, it is very sensitive to the latency settings : setting that latency from 2ms to 4ms can definitely improve the sound of the higher latencies. However, the EMU 0404 becomes more sensitive to anything else running in the computer and it shows trough crackles and pops. This is especially true forFLAC files as even 24/96 wav files can play at the lowest latencies without a glitch.
Note on USB cables :
I first discovered the effect of usb cables while using the EMU 0404 usb with the Audio-GD DAC-100.
I was doing a comparison between the spdif input of the dac-100 and its usb input, and for a few trials I preferred the spdif input using the EMU 0404 as a transport. When I did the same experiment a little bit later, I found that I preferred the sound of theusb input of the DAC-100 over that of EMU 0404 as a transport. The only thing that changed between experiments is that I used a Belkin cable with the DAC-100. I repeated the experiment many times and could detect the differences : the culprit was the stock usb cable that comes with the EMU 0404 that it is the worst sounding of all (it is also poorly constructed with thin conductors and ferrite shielding).
That led me to buying different usb cables from Monster, Real cable, Belkin Gold, and finally Wireworld Ultraviolet. All of them have different sounds but the Wireworld is clearly superior sounding and it also allowed me to reach the lowest latency settings with the EMU 0404usb without suffering from crackling and pops.
Musiland Monitor 01 USD :
The Musiland Monitor 01 USD is a definite step up in sound quality compared to the EMU 0404 USB (and incedently compared with the usb input of the dac-19mk3).
The highs are cleaner and the bass tighter. The soundstage is a little bigger and there is better separation of the instruments.
I tried different latency settings with the asio control panel of the musiland but it didn't seem to affect much its sound performance. The good thing about the Musiland is it always operates without glitches whatever the load is on the cpu.
Finally, while it is not harsh sounding like the EMU 0404 usb, it lacks some of the sparkle and life that the better converters (teralink, hiface) seem to inject to music. It is not bad by itself but the music seems to be constrained and tight once it is compared with better units.
Note on usb cables :
While the async protocol should make the kind of usb cable that is used irrelevant, the Musiland is however still affected by the choice of the usb cable. It is probably because it draws its power from the computer but is also possible that is still affected by incoming jitter from
Teralink-X :
When I bought the Teralink-X, I was not expecting much from it. At that time, I already had the Musiland unit and I was just curious to see how it performs because it uses high grade capacitors and a low jitter clock.
Straight out of the box it performed very well. After a few days of burn-in, the sound improved dramatically and it outperformed the Musiland unit easily.
The bass got deeper and more powerful, and the highs became more extended and sweeter at the same time. It is like lighting up a picture and discovering hidden details. But best off all, that increase in overall resolution did not increased the "edginess" of the sound. In fact, the sound became smoother.
The soundstage became bigger, almost limitless. In fact, on most recording I don't feel that I am listening to headphones, the only thing that keeps reminding that I am listening through headphones is the pressure of thesennheiser hd-650 on my head.
The imaging is precise, and hollographic in many recordings thanks to the precise separation of instruments and voices. Every instrument/singer has its own place on the soundstage.
Overall, I find the sound sometimes too good to be true. I got accustomed that only a few "audiophile" recordings sounded great in my system, however with theTeralink -X all recordings sound enjoyable in my system. I can still hear big differences between quality of recordings, the encoding used (mp3,flac, wav), but the Teralink-x, especially when paired with the Belden BNC cable, lets you focus on the positive sides of the recordings which is some might call "musicality".
Note on digital cables :
The Teralink-X works best in my system with the Belden BNC cable.
When paired with the Belden cable it gives a wide soundstage and a very smooth sound, "tubey like".
When paired with the Hifi Cables Sobek BNC cable, I get an increase in resolution, but a smaller soundstage and an increase in brightness "edginess".
Note on burn-in :
The Teralink-X needs a few days to sound its best. It doesn't sound bad at first but after a dew days of continuous play-back, the sound opened up and the bass got deeper. TheTeralink -X uses many capacitors inside that might explain the big change in sound that I did not notice to that extent (or at all) with the other converters
M2Tech Hiface :
The m2tech hiface is my newest usb to spdif converter, I have only owned it for a few days, but I have listened to it for enough hours now to get a sense of how it sounds. This minuscule unit is simply outstanding.
First, the music just makes more sense. The notes flow more easily and it requires less efforts from the listener to understand what the performers were trying to convey. There is also a general increase in resolution that makes the music more real, probably because of its better retrieval of low level information on the recordings which makes reverbs and ambiance hall more audible in Live recordings and Classical music.
Second, the soundstaging is wonderful. It has a better pinpoint and 3D imaging than the other units, and there is a even better separation of instruments and voices. In fact, whether it is because of better imaging or because of better dynamics, the result is that instruments and voices seem to have more acoustic power (while playing at the same volume levels as I did with the other converters), the presence factor of performers and instruments is much higher.
Also, while the soundstage with the Teralink-X is big, there is not as much differentiation between recordings. With the hiface, the soundstage changes a lot (in width, depth and height) between different recordings.
Finally, the frequency extremes seem to have been extended with a more shimmering and extended highs and more articulate and tuneful bass. Compared to theTeralink -X, there is also a shift in the tonal balance to a more neutral and faithful balance. There is a better differentiation between the quality of the recordings and it doesn't impose a sound signature on the recordings : Warm recordings come out as such and over-processed recordings are highlighted as such but are still enjoyable to listen to.
Note on digital cables :
Differences between digital cables are more easily spotted : the Belden is muffled, there is also haze and distance to its sound which gave me the impression of a big and wide soundstage.
The Sobek has a better resolution, is razor sharp, highs are more extended, and the overall perceived volume is louder probably because there is more information coming through to the dac.
Once I bought the 18 ft. Belden cable, I didn't like the sobek as much as I used to before when I used it with the other converters because it was more honest and showed the shortcomings of those units.
With the hiface I can enjoy the high resolution sound provided with the Sobek.
Note on media players :
Since I have had the hiface only for a few days, I did most of my testing with foobar 0.8.3 and foobar 0.9.6.4 with KS.
I preferred the sound using the old foobar 0.8.3 but the current version of the driver has a bug with 0.8.3, the time slider of track doesn't move when playing.
I tested briefly MediaMonkey and I made the Hiface work with ASIO4ALL. Then I tried ASIO4ALL in foobar with the hiface and it worked but the sound quality was not as good as with KS.
Final note on the Hiface :
Each time I was doing A/B comparisons with one of the other units, I would end up listening to the M2tech and forget about my review. Unplugging it to put another converter was a tough experience each time because I had to be pulled out of the music.
Beyond the criteria I described above (soundstage, resolution, ...), the M2TECH hiface was the most satisfying unit from a sonic point of view as it was the closest to real representation and offered the most believable representation of music.
Conclusion :
Getting a clean spdif signal from a usb port of a computer seems like a simple task but appears to be a rather complicated endeavour. There are many usb to spdif converters today in the market and it is hard to predict how they perform. Their performance is tied both by the quality of the drivers and the quality of the components (clocks,spdif transmitters, ...)
Out of the 4 converters I tested in my system, the m2tech hiface is the one that did the best job. I have indeed been very impressed by the performance this extremely small unit and it doesn't even require a usb cable. The designer of the hiface paid great attention not only to the proprietary drivers (like EMU and Musiland) but also to the quality of components and clocks that are used.
Granted it is the less flexible unit to use as it is limited for now to KS (Kernel Streaming) with limited media players but its superior sound quality makes you forget about its limitations.
The Teralink is the second best performing unit that is leaning a little bit on the "warm" side. However, it was only when I listened to theHiface converter that I became of that coloration.
If your library contains only CD files and mp3s, the Teralink-X is very good as it combines both a "high resolution" sound with a slight warmth that makes it more forgiving toward less than perfect recordings.
However, when you add the cost of a high quality usb cable, the overall cost exceeds that of the m2tech hiface. So it is hard to recommend it because it is limited to 16/48 and doesn't sound as good as the m2tech. Any "audiophile" who is willing to buy such a converter would likely either purchase theMusiland for its convenience or the m2tech for its performance.
The Musiland Monitor 01 USD is a very nice little unit that can do 24/192 and allows for high flexibility (ASIO, KS, DS). Sound wise, it is not a match (in my system) for the Hiface or the Teralink but I am pretty sure that if it used on other DACs that have better reclocking, the performance gap with the two converters might not be as noticeable.
The EMU 0404 usb comes last (old drivers, components of average quality, and requires a high quality usb cable). However, it is the only unit that has analog inputs and outputs. In my opinion, it should not be bought to serve only as a USB to Spdif converter. There are many units out there that are cheaper and that perform better.
Follow-up (26/10/09)
When I initially wrote my conclusion, I didn't intend to make it sound like the Musiland was a bad sounding unit. But I guess I didn't choose the appropriate words. What I meant to say is that in my system, and with my ears I preferred the Teralink-x and m2tech Hiface. The Musiland is still a very good unit.
I did some other testing since I first wrote the review with another good digital cable (the stereovox xv2) and ended up with this conclusion : I would rather use the Musiland with a good digital cable (Stereovox, Sobek) than using the m2tech hiface (my favorite of the group) with the Canare or the Belkin coaxial cable. This means that while there are clear and audible differences between those units, I was talking about the last few percents of performance in my system.
Also, there are people who found it curious that the order of purchase coincided with the order of prefrence. First, as I explained in the test protocol, I went back many times to each unit and did not do my review from memory, I gave a fair listen to all units. I also kept all 4 units on hand and won't be selling them in the near future.
Second, a little explaining about why I was "lucky" enought to have a better unit each time I upgraded. I owned the EMU 0404 usb for more than 2 years and used it mainly as a dac/headphone amp before using it as a transport for my dacs. So it was never my intention to use it as a usb to spdif converter at the time I bought it.
However, the spdif output of the EMU 0404 usb worked only with asio, which meant that I had to keep switching between the usb input of my DAC (for watching movies) and its spdif input (for music). It is at that time that I found a very promising unit the Musiland. When I had it I was very happy with it but I was a little bit disappointed about the marginal improvement it brought upon the EMU 0404 usb. I started reading about the Musiland unit and I learned that while the async protocol and use of a FPGA were good things, however it didn't use audiophile grade components (clocks, capacitors). It was when I was looking for low jitter clocks on ebay that I stumbled on the Teralink-X converter. On theory, it was supposed to be inferior to the Musiland (it uses an old cm-108 usb adaptive chip) but it used a low jitter clock, Oscon capacitors, ... and since it was pretty cheap (less than my Wireworld usb cable), I took the chance and ordered it. When I got it, the sound was superior to the Musiland unit which was rather surprising. This lead me to the quest of another usb converter but this time I was looking for an async usb converters with audiophile grade components. Then someone mentioned this great new usb to spdif converter on the Audio-gd 19 mk3 dac (I started). I went to their web site and I was positively surprised to read on their white paper about the use of 2 low jitter clocks and an async usb protocol. So this is how I ended up buying the m2tech. If I didn't learn about the m2tech I would probably have had my musiland modded with a low jitter clock. But for now I am happy with the m2tech hiface and hopefully won't be looking for other converters in the next few months
I hope this explanation can shed the light on my upgrade path and maybe also avoid skepticism about why I ended up prefering my newest unit each time
Jitter measurements:
While jitter measurements should be interpreted with great care (different results with different equipment and test protocol), it is sometimes a nice tool to make jitter comparisons in the same group of transports done with the same equipment.
Here you can find jitter measurements of the m2tech, a LynxTwo pci sound card, a Nagra DII and a Marantz CD player. In that test the m2tech was slightly better than the LynxTwo which is a professional grade PCI card, and better than the Nagra and Marantz cd player.
On a stereophile test, there were measurements of the EMU 0404 usb and other converters. The EMU had the worst jitter of the bench, and its jitter was 8 times that of the one measured with the m2tech (Granted the tests were done with different measuring equipment).
So, it seems that at least, I must have gotten the order of those 2 units right
Follow-up 2 (17/11/2009) : Digital Cables - Which length to choose ?
For those who are interested in any usb to spdif converter they are likely to use a spdif cable and wonder about the optimal length.
There are many people claiming that longer cables are better. There is even an article written by Steve Nugent on digital cables (here) which suggests that longer cables are better because they avoid reflections. While I have nothing against Steve Nugent, you have to be keep in mind that Empirical Audio also happens to be selling digital cables.
Dan Lavry, from Lavry Engineering, corrected some misconceptions about longer digital cables being better and explains why a shorter cable is better than a longer cable.
While I am no engineer, I believe that the facts stated by Dan Lavry are accurate for two reasons : Lavry has a high reputation in the pro audio world and since Lavry Enginnering doesn't sell digital cables, Dan Lavry has (in my opinion) no agenda or no interest in stating that a shorter cable is better than a longer one.
Below is an extract from the conversation and if you want to read the whole conversation, which goes on for a few pages, it starts here :
Quote:
The intent of this follow-up is not to start up the debate about jitter and cable lengths, however, I felt that it was necessary for any reader to learn what Dan Lavry had to say on the subject.
Also, there is something that everybody seems to agree with. To get the best performing digital transmission, it is best to respect the 75 ohms characteristic impedance throughout the digital chain from the transport, the connectors, the digital cable to the digital receiver.
One more thing, Dan Lavry said the discussed issue is similar with usb : "what I stated would be good information for ALL cables including USB".
As a result the usb cable should be kept as short as short as possible (no surprise here).
This is good news for our wallets
Follow-up 3 - 16/03/10:
I have been using the Hiface for months now and I am still amazed about how good it sounds. Since I wrote my usb to spdif shoot out, it seems that the Hiface has gained popularity.
Below are the link to 2 reviews by EnjoytheMusic.com and 6 moons.
http://www.enjoythemusic.com/magazin...hiface_usb.htm
http://www.sixmoons.com/audioreviews/m2tech/hiface.html
More on the USB to SPDIF converters ...
Comparison between usb to spdif converters:
So far, I have compared the Hiface to 4 other usb to spdif converters. I have testes each of those converters with different drivers and usb cables (Wireworld Ultraviolet, Monster, Belkin Gold, Real Cable...). And so far, the Hiface seems the most neutral in comparison.
Here is a quick recap of how they sound:
-Emu 0404 USB: thin and edgy sounding, small soundstage, limited extension at the frequency extremes
-Musiland 01 USD: relatively neutral balance, better extension at the frequency extremes than the EMU 0404
-Teralink-X: relatively warm sounding, better extension at the frequency extremes than the Musiland
-Purepiper usb to spdif converter: relatively warm sounding, limited extension at the frequency extremes.
-Hiface: The most transparent of the group
Using the Hiface with different digital cables:
I tried the Hiface with many digital cables. Here are some quick impressions using the dac19mk3 as a DAC.
-Hiface + Canare cable: The sound is constricted, edgy, the soundstage is small
-Hiface + Belden cable: The sound is warm, the soundstage bigger than life (but not very precise), the transients are slowed and muffled in comparison with higher end cables
-Hiface + Sobek (modded): The sound is slightly on the warm side of neutral. The soundstage is not as big as with the Belden but is more defined and varies from one recording to another. There is a big increase in details.
-Hiface + Stereovox XV2: The sound is slight on the thin side of neutral, a little bit bright/metallic timbre. However, it is more detailed and spacious than the Sobek.
-Hiface + Oyaide DB-510: The sound is neutral and transparent. It the most detailed but also the most natural combination.
Using the Hiface and Oyaide with different DACs:
Since I have settled for the Hiface + Oyaide as my reference front end, I have used it to test different DACs and here is how they sound
-Hiface + Oyaide + Purepiper DAC A-1: It has a relatively neutral tonal balance with a slight emphasis on the upper midrange. It is a little bit dry sounding. With the wrong associated equipment it can sound harsh.
-Hiface + Oyaide + FUN version A (with the AD1852 dac chip): The sound is darker than the Purepiper. It is smoother; it has a lot less apparent details (different voicing) but has more low level details. It is also richer sounding.
-Hiface + Oyaide + FUN version A (with the WM8740 dac chip): The sound is warmer than with the AD1852 but there is a definite loss of details and a little bit of mid bass bloat.
-Hiface + Oyaide + FUN version B (with AD1852): It has a more limited bandwidth than version A and has dirtier highs. The sound is constricted and edgy in comparison with version A.
-Hiface + Oyaide + DAC19mk3 (with the DF1704 digital filter): The soundstage is huge and well defined, with holographic imaging. The transients are faster. There is a great sense of transparency. The bass is deep and accurate. The highs are cleaner than any other combination above.
-Hiface + Oyaide + DAC19mk3 (with the PMD100 digital filter): While the PMD100 is not as fast as the DF1704 filter, it has a purer tone with greater tonal density. It is “analog heaven” not because it adds pleasing distortion but because it has purer highs.
Conclusion
Overall, I am not saying that the Hiface + Oyaide should be used in any system as component matching is important to get a musically satisfying result.
But both the Hiface as well as the Oyaide digital cable are very transparent components. It is possible to achieve a similar tonal balance with other combination, but by using lesser components, you loose at the same time low level details, frequency extension, soundstage and imaging clues, and timbre subtleties that cannot be recovered elsewhere in the chain.
So while I understand that a specific system can sound thin and edgy with Hiface, I believe it is not because of the Hiface but because of other components in the chain (digital cable, DAC) that are more likely to be the culprit.
Side Note: I suspect that in highly resolving systems when using DACs based on sigma delta chips (and opamps), it is likely that the Hiface will sound thin and dry simply because it will reveal the true sonic character of the DAC; other more jittery devices such as the Teralink seem to inject some kind of pleasing dither that makes the listening more tolerable and enjoyable (but less precise).
Here is again a few links to why sigma delta dacs are flawed and were only pursued to lower the cost of manufacturing in comparison with the R2R/multibit DAC chips.
Mother of Tone - Conversion Techniques
How DACs Work
J River Media Center vs. Foobar v1.0
A few days ago, J River Media gave away 5 free licences for people to try against their preferred media player (mainly Foobar).
The discussion starts here : http://www.head-fi.org/forums/f7/usb...ml#post6470058
After a few days of use, I found that there were subtle but noticeable differences.
To my ears, listening through J River (in comparison to Foobar v1.0) is like turning up the sharpness controll of a TV set. With J River all the apparent details are over emphasized but at the expense of the very fine and more subtle details. For example, the image outlines are sharper with J River, but the fine little ambient cues are less audible and you feel like everything has been recorded in a dead quiet studio.
So like for TVs, a little bit of added sharpness (with J River) could be either a good thing or too much and fatiguing on the long term.
Of course, one could also suggest that Foobar is dull and that J River is neutral, but that wouldn't explain how Foobar has more low level resolution.
Overall, it is a nice thing to have yet another mean to "tweak" the sound through the media players. And given that J River contains a lot more than a mere music player, I think it is a remarkable achievement when we compare it to Window Media Player/Media Center.
Note: All my comments about Foobar, concern the v1.0. For an unknown reason, I have always found the 0.9.x versions grainy sounding and kept using v0.8.3 until the release of the v1.0.
More on the Teralink-X drivers
I did some comments and measurements that are buried in this long thread, so I decided to include them in this follow-up.
Quick note/update: Since my comparison, Teralink has released the X2 version which uses a 24/96 Tenor usb chip. For future buyers, I think the extra cost of the X2 is worth it.
Quote:
Quote:
Follow-up 4 - 18/03/10 : Digital Filters, Minimum Phase and Upsampling:
While I have already mentioned digital filters and upsampling in separate posts, I thought it could be nice to compile some of that in a single post. So here are my findings on the subject: nothing new, just a compilation of thoughts.
Digital filters:
One of the discriminating factors between different DACs is the digital filters. Most modern chips include a built-in digital filter (for cost savings reasons) but they perform very poorly in general. But how much difference does it really make?
In my current DAC, I have the possibility to swap between 2 digital filters: the PMD100, which is a HDCD capable digital filter made by Pacific Microsonics, and the DF1704 digital filter made by TI. The two digital filters have 2 distinct sonic signatures (for those interested in the details, they can read my review of the audio-gd dac19mk3 where other users have reported similar differences between the 2 filters).
When playing 16/44 data, the DF1704 is fast and analytical, slightly on the bright side of neutral and the PMD100 is warmer and more “analogue” like.
While the PDM100 is supposed to be limited to 24/55, it seems to work at 24/88 (but not 24/96).
Well, by playing (native) 24/88 data on both, they sounded closer than they did at 16/44. The DF1704 smoothed out and the PMD100 had a little bit more sizzle on top.
That was similar to what I have experienced with some sigma-delta DACs that I have found to sound harsh at 16/44. Usually they sound much better with 24/96 data or upsampling to 24/96. The reason behind the perceived improvement is not due to the fact that we hear the extra data. It is rather due to the fact that we do not have to hear as much the nastiness of the digital filters operating close to our audibility range.
On many digital filters whether they are outboard filters (such as the DF1704) or built-in into the dac chip (such as the CS4398) the designer has sometimes the choice between a sharp/fast roll-off and a slow roll-off.
The sharp roll-off usually gives a flat measured frequency response on the 20-20,000Hz range, which looks good for published measurements. But that flat frequency response up to 20K comes at the expense of phase performance (which has an impact on soundstaging among other things).
The slow roll-off on the other hand is down from 1 to 3 db (in general) at 20 khz but has a better phase linearity, which is more audible to the human ear.
So when given the choice between what looks good on paper (a flat frequency response) and what sounds nice to the ear (phase response), most designers choose the first option.
I have looked at the measurement of many DACs on the stereophile website and some of the most expensive and better sounding DACs seem to have that roll-off on the highs. Weirdly enough most budget DACs and soundcards seem to have a flatter high frequency response. I am not saying that we shouldn’t pursue a flat high frequency response. But I am just saying that when you see a budget DAC that has a flat response for 16/44 data beyond 20 kHz, and if the DAC doesn’t use a fancy DSP, it is most probable that it has sacrificed the phase response for the good looking Frequency response measurements.
By the way, if you are still thinking that a flat frequency response is absolutely necessary to have a realistic sound, just think about the following.
Let’s suppose you listen at a violin in very reverberant room, then you listen to a violin in very damped room of a different size from a different distance. Then listen to a violin in a speaker system that has been equalized to have a flat frequency response. As far as I am concerned, while situation 1 and 2 will have different Frequency Response measurements, my brain will analyze and detect the sound as being real and live. When I will listen to situation 3, I most probably won’t be fooled thinking that the violin is real.
Fortunately, there is a way to go partially past those poor digital filters. By upsampling data to 24/96, you minimize the audible effects of those filters, as any aliasing/distortion they generate will be pushed further up in frequency range.
Impulse response & Minimum Phase Filter:
While I have mentioned the Frequency response and phase response of digital filters aspects, I haven’t mentioned yet the impulse response.
When compared to each other, a fast roll-off filter has a lot more pre and post ringing than a slow roll-off filter. That pre-ringing is one of the reasons some people have complained about digital playback and kept using analog sources (Vinyl, tapes…)
So by choosing a fast roll-off over slow roll-off, many DAC and sound card makers are willing to sacrifice the phase and impulse response in order to have a nice looking RMAA graph.
However, while the slow roll-off filter minimizes the pre-ringing, it doesn’t get rid of it entirely. There is a growing number of CD players and DACs that provide a new option which the Minimum phase filter. According to their research, the human ear would be less sensitive to post ringing than to pre-ringing.
For those who are interested, Ayre wrote an interesting paper on Minimal phase filters (here: http://www.ayre.com/pdf/Ayre_MP_White_Paper.pdf)
Again those limitations (pre and post response) are far worse at 16/44 than they are at 24/96.
That is once again (in my opinion) the reason why some people hear a lot of improvement when using up-sampling or native 24/96 data. I think that despite some measurements (SNR, THD), not many systems have a true resolution greater than 16 bits. If that were the case (let’s say 20 bit resolution), those systems would be indistinguishable from the reality which is not the case. So, playing back 24/96 data on the DACs can improve the playback experience in comparison to 16/44 simply because most digital filters are terrible at handling 16/44. That is perhaps one of the reasons that we see so many DACs today that use ASRC/upsampling. (But then again those ASRC chips do not seem to sonically sound good and can also benefit from a good upsampling from the source, but that is another story …)
SoX upsampling in Foobar:
Among the nice things that come along with using a computer as a source is the ability to try different playback methods and upsampling methods.
Personally, I have settled for Foobar for quite a long time. And recently, I have been using SoX for 2 reasons: It is a relatively transparent upsampler (it doesn’t harm low level details) and it is very customizable.
In my case, I have found the most consistent and enjoyable results with upsampling to 24/96 (weirdly it sounding better than upsampling to 24/88) and with the minimum phase setting. It generally results in a more coherent soundstage that is pushed a little bit further back, and the sound is usually smoother. But those changes vary from one DAC to another, so it is impossible to make generalizations about the effects.
I have also tried different passband settings. The Stock setting is at 95%. When set at 99% the soundstage shrinks and everything gets tighter. When setting the passband at 90%, the soundstage becomes a little bit bigger and the sound and everything sounds a little bit softer and less defined.
Those different settings are a great tool for fine tuning the sound.
While I don’t use the SoX all the time with my reference DAC (which uses a slow roll off DF1704 digital filter), I found that it improved considerably the listening experience with the other DACs I have on hand.
Note:
I brought up the subject of digital filters so that people understand the effect of a low jitter sources on their converters.
When using a high jittery source, a lot of distortions and nastiness of the digital filters are masked/blured by the noise.
When using a low jitter source such as the hiface, there are improvements in many areas but it can also reveal/expose any flaws elswhere in the system.
(BTW, not high jittery sources have pleasing distortions, the emu 0404 usb is edgy sounding).
Doing tests with 24/96 source material will minimize the effect of the digital filtering and will probably allow for better comparisons of the intrinsic quality of different converters. If listenings tests are done at 16/44 on an average sounding digital filter, the tests won't reveal much and thy would be specific to that specific type of DAC implementation.
Follow-up - 23/03/10: USB Cables
Quote:
What most people fail to understand: is that jitter can be generated at many points:
1/ It can be generated at the recording (too bad, there is nothing we can do about it).
2/ It can be generated at the source
3/ It can be generated by the digital cable
4/ It can be generated inside the DAC itself
5/ It can be generated by any impedance mismatch from the source to the DAC
Unlike many people here who claim that zeros and ones are all that matter, I try to do as much research as possible. Last week alone, I bought an AES research paper (Benjamin and Gannon) and I also read this 90 pages+ report on the jitter in high resolution DACs (here: http://www.iet.ntnu.no/courses/fe811...t_audiodac.pdf). If you care to read it, you will see that the author mentions several times that digital cables can have a measured jitter of a few ns (even more than the source itself).
So is it the same for USB Cable?
In the case of an adaptive usb to spdif converter, the usb converter acts more or less like a digital coaxial cable: it streams real-time data (no error correction or feedback) and its performance (jitter/reflections) is affected by the quality of cable.
FYI, as I said it many times earlier a regular usb to spdif converter doesn't act like a usb hard drive. I streams real time data and if anything is lost there is no error recovery.
In the case of an async usb to spdif converter, it shouldn't matter in theory. However, there are at least 2 reasons that might impact their performance.
First, most of them are not galvanically isolated from the computer. So the USB cable doesn't only carry data but also the power supply. So in that case, a poorly shielded usb cable can introduce HF/RFI noise into the power supply of the converter and the clocks as well as other components) inside are very sensitive to such noise.
In fact, a very good usb cable can act a filter for high frequency noise (if the capacitance of the power lines is very high). However, the power lines and data lines should not be treated/shielded the same: The data line is transmitting high frequency signals while the power lines are transmitting low frequency/DC power.
Second, I have read somewhere (maybe Steve Nugent?) that while async converters act as a master clock for the data, those converters do not always implement error check and recovery. So in that case there could be a loss of data in the process (unlike a usb hard drive for example).
However, at the time I wrote the review, I didn't know the theory and I was as puzzled as many of you. But the test I did was pretty simple; I took the EMU 0404 usb as a source and used different usb cables. I tried to get the lowest possible ASIO latency setting with every cable before getting crackles and pops. And weirdly enough the cable that could achieve the lowest latency without problems was the Wireworld ultraviolet and the one with the worst performance was the stock usb cable of the emu (which is poorly constructed).
So of course, that was an indirect test, but I am pretty sure that if Stereophile did measurements with different usb cables, they will find differences in measurements like they did with digital cables and with different usb to spdif converters.
So, I understand that hearing differences between usb cables might come as shocking news. However, what I find more shocking is the number of people that state that it is foolish to hear differences between digital cables (whether it is coaxial or usb) without even knowing the science behind line induced jitter.
Of course, there is the problem of threshold of audibility of jitter. The AES study that Nick_Charles recommended me to read talks about thresholds levels of well above 1ns. However, that study was conducted more than 10 years ago and used a cheap sony headphone for the testing. (See the discussion with Nick Charles here: http://www.head-fi.org/forums/f133/p...ml#post6479343)
A more recent study that I have found recently about the temporal resolution of human hearing proves that with the right test equipment, the threshold of temporal resolution of human ears is far greater than we suspected (see here: http://www.physics.sc.edu/kunchur/pa...rge-Foster.pdf)
As you can see, I am not against a scientific debate on the subject. However, I am still on the process of collecting data. I will start a thread about usb cables, when I have accumulated enough data and info, and when I will have enough time to do it.
I took a little bit of time to write this post as it seems to be a recurrent subject. As I said, I will start a thread about usb cables in the sound science forum (in due time). However, the rules of this forum allow me to state freely my subjective listening experiences. So I am not going to do 30 DBT for each component I cite (usb cable, usb converter, digital cable, DAC, interconnect...) just to please a small portion of people.
To build a resolving and musical system, it takes a lot of time, research, and trial and error. I try to share the most significant components upgrade and tweaks. But I can’t spend endless hours just to do a proper DBT of a single usb cable. I would rather spend that time doing other things or just enjoying music.
Pictures of my system:
Purpose of this thread 24/03/10:
Since this thread has been assaulted many times by people wanting to discuss theory of jitter, what is possible to hear and what is not I decided to write in plain words, why I started it.
The reason I started this thread was to:
1. To share my experience with the 4 usb converters I have tested in my system
2. exchange any experience that helped you improve the source: whether it is a usb converter, a digital cable, a usb cable, a media player ...
However, if your main goal is to say that:
1. jitter is not audible in most cases
2. cables do not matter
3. all subjective hearing is placebo and only DBT can prove that there is a difference
Please read the following : http://www.head-fi.org/forums/f46/do...-forum-450574/
And in that case, I will ask you politely to not drag this thread off topic by discussing such subjects.
Such discussions are better suited for a separate thread. You can start a new one and post it in this thread. But PLEASE, do not discuss such subjects on this thread as it is not the purpose of this thread.
System Used :
Main Chain :
Foobar 0.8.3 (KS and Otachan ASIO) --> Wireworld USB Ultraviolet --> USB TO SPDIF converter --> Hi-Fi Cables & Cie Sobek or 18 ft. Belden 1694 BNC Cable --> Audio-GD DAC-19MK3 --> Hi-Fi Cables & Cie Khnoum RCA interconnect --> Audio-GD C2C (w/ upgraded pot) --> Moon Audio Blue Dragon V3 cable --> Sennheiser HD-650
Usb to spdif converters :
EMU 0404 USB
Musiland Monitor 01 USD
Teralink-X
M2Tech hiFace
Power related accessories :
Hi-Fi Cables & Cie PowertransPlus Power Cords
Supra Mains Block
Essential Audio Tools Parallel Filter
Vibration Control :
E&T Spider Rack, Vibrapods, Vibracones, Sandboxes, Brass cones, Acrylic and Fiber carbon sheets,
Herbie's Audio Lab Tenderfoot, SuperSonic Component Stabilizer
Other gear :
Sources : Creative Audio 2ZS Notebook, EMU 0404 usb, Zero DAC, Audio-GD DAC-100
Headphone Amps : Little Dot MKIII, Audio-GD ST-3
Introduction:
Over the time I accumulated a few usb to spdif converters commonly discussed on head-fi, so I decided to write a comparative review of those units. Granted it is a bit tricky to describe the sound of a component that exclusively does its work on the digital level, however the differences do exist and I tried my best to describe how they compare to each other.
Jitter - theory :
There has been a lot of things about jitter. I found an interesting article here on TNT-Audio explaining jitter and the difference between transmitter jitter, line induced jitter and interfering-noise induced jitter.
There is a theoretical advantage of using a usb to spdif converters instead of a direct usb connection to the dac is an added layer of isolation and "jitter rejection". Below is a quote from an email exchange with Marco from m2tech : "S/PDIF receiver which, you certainly know, has a feature which is "jitter rejection", a figure that typically ranges 20 to 30dB: this means that the input jitter to an S/PDIF receiver is reduced by a factor of 10 to 30... 970ps become 100ps quite easily."
This might explain while I found that all 4 units performed better than the direct USB inputs of the USB dacs I have.
I even found out that when using the Musiland as a transport (using optical) for the EMU 0404 USB it gives better results than its USB input.
(Edit -- The intent of this thread is not to discuss jitter theory or treshold audibility of jitter. Please read the following post http://www.head-fi.org/forums/f21/do...-forum-227350/ before posting on those topics.)
Test Protocol :
Before writing my review, I did many A/B tests going back and forth between different units. And besides those A/B tests, I also did extended listening with all those units to get more familiar with their sound as sometimes
To make sure I was not adding any unforeseen parameters, for each A/B test I would use the same usb port, the same usb cable and the same digital cable (used in the same direction).
I did try various digital cables but did most my listening with the sobek digital cable as I was very familiar with its sound. For those who might think that digital cables do not matter, you can find Here a link of an old stereophile article that measures the jitter of various digital cables which show that even the direction of the cable can affect the sound. During my test I tried to beconsistent comparing units with the same digital cable (used at the same direction), and then compared again the units with a different digital cable.
I also tried different usb cables but I ended preferring the Wireworld Ultraviolet usb cable. I did most of my testing with that usb cable except with the m2tech who doesn't need a usb cable.
Ease of use & Drivers :
EMU 0404 USB :
The EMU 0404 usb is a very versatile unit. It can be used as a DAC, headphone amp, and also a USB to Spdif converter. However, it requires custom drivers and the digital output works only with ASIO.
It has optical and rca spdif output but it lacks a BNC output which might be a drawback for some.
The EMU can do 24/192 thanks to its custom drivers.
Musiland Monitor 01 USD :
The Musiland is a wonderful little unit. It needs custom drivers that work up to 24/192. It works with DS, KS and ASIO so that it can be used with any player.
It has BNC, Spdif and optical out. I am currently using it with the 1.0.3.2 driver and it works flawlessly.
Teralink-X :
The Teralink is the easiest to use. It doesn't need custom drivers and works with Direct Sound and ASIO (through ASIO4ALL).
It has I2S, BNC, Spdif and optical out. This one is a "plug and play" unit that performs flawlessly but is limited to 16/44 and 16/48.
It is possible to install custom drivers (thread here) but there are no sonic benefits in doing it (those drivers introduce many additional processing to the sound which I found to cause a degradation to the sound in my opinion).
M2Tech hiFace :
The M2Tech hiFace is a very minimalistic unit : it has only one output (I ordered mine with a BNC output) and it plugs directly into a usb receptacle without the need of a usb cable. For now, it has only drivers that work with Kernel Stream and it works with Foobar, Media Monkey and WinAmp.
DS and ASIO drivers are expected later down the road.
The unit does 24/192.
Sound :
First, all the usb to spdif converters listed in this review outperformed the usb inputs of the DACs I have on hand. So as long as they are used with a decent spdif cable (such as the affordable 18ft. Belden 1694 digital Cable), there should be sound improvement in many dacs (at least with the ones I tested).
Second, I have noticed big differences between usb cables (Edit -- For those skeptics on why there are differences between usb cables, please read the 23/03/10 follow-up). So I settled for the review on the best one I own (the Wireworld Ultraviolet USB) and used it for all the converters and the DACs except for the m2Tech Hiface. Therefore, it is important to note that the m2tech converter has a slight comparative advantage since it doesn't need the use of a usb cable.
Also, for those who might be interested, I wrote a full length review of the DAC-19mk3 (here) that I used for this review. It should give a good idea of a baseline of the sound in my system. I was using the Musiland back then with the Mode A digital filter for the DAC-19 mk3 and the ST-3 head-amp. Since then I upgraded to DF-1704 digital filter (mode B) and the C2C head-amp which made my system more revealing and transparent.
EMU 0404 USB :
I had the 0404 usb for a long time, so I was very familiar with its sound from its analog outputs. When used as usb DAC, the EMU 0404 USB is "bright" sounding, lacks deep bass and the overall sound can be described as "edgy". The sad thing is that those attributes also reveal themselves in the digital output which sound edgy and congested in the top and light in the bass in comparison with the other converters I have.
However, compared to the usb inputs of the Audio-gd DAC-19mk3 or the DAC-100, it is a noticeable step up in resolution and bass tightness.
Also, it is very sensitive to the latency settings : setting that latency from 2ms to 4ms can definitely improve the sound of the higher latencies. However, the EMU 0404 becomes more sensitive to anything else running in the computer and it shows trough crackles and pops. This is especially true forFLAC files as even 24/96 wav files can play at the lowest latencies without a glitch.
Note on USB cables :
I first discovered the effect of usb cables while using the EMU 0404 usb with the Audio-GD DAC-100.
I was doing a comparison between the spdif input of the dac-100 and its usb input, and for a few trials I preferred the spdif input using the EMU 0404 as a transport. When I did the same experiment a little bit later, I found that I preferred the sound of theusb input of the DAC-100 over that of EMU 0404 as a transport. The only thing that changed between experiments is that I used a Belkin cable with the DAC-100. I repeated the experiment many times and could detect the differences : the culprit was the stock usb cable that comes with the EMU 0404 that it is the worst sounding of all (it is also poorly constructed with thin conductors and ferrite shielding).
That led me to buying different usb cables from Monster, Real cable, Belkin Gold, and finally Wireworld Ultraviolet. All of them have different sounds but the Wireworld is clearly superior sounding and it also allowed me to reach the lowest latency settings with the EMU 0404usb without suffering from crackling and pops.
Musiland Monitor 01 USD :
The Musiland Monitor 01 USD is a definite step up in sound quality compared to the EMU 0404 USB (and incedently compared with the usb input of the dac-19mk3).
The highs are cleaner and the bass tighter. The soundstage is a little bigger and there is better separation of the instruments.
I tried different latency settings with the asio control panel of the musiland but it didn't seem to affect much its sound performance. The good thing about the Musiland is it always operates without glitches whatever the load is on the cpu.
Finally, while it is not harsh sounding like the EMU 0404 usb, it lacks some of the sparkle and life that the better converters (teralink, hiface) seem to inject to music. It is not bad by itself but the music seems to be constrained and tight once it is compared with better units.
Note on usb cables :
While the async protocol should make the kind of usb cable that is used irrelevant, the Musiland is however still affected by the choice of the usb cable. It is probably because it draws its power from the computer but is also possible that is still affected by incoming jitter from
Teralink-X :
When I bought the Teralink-X, I was not expecting much from it. At that time, I already had the Musiland unit and I was just curious to see how it performs because it uses high grade capacitors and a low jitter clock.
Straight out of the box it performed very well. After a few days of burn-in, the sound improved dramatically and it outperformed the Musiland unit easily.
The bass got deeper and more powerful, and the highs became more extended and sweeter at the same time. It is like lighting up a picture and discovering hidden details. But best off all, that increase in overall resolution did not increased the "edginess" of the sound. In fact, the sound became smoother.
The soundstage became bigger, almost limitless. In fact, on most recording I don't feel that I am listening to headphones, the only thing that keeps reminding that I am listening through headphones is the pressure of thesennheiser hd-650 on my head.
The imaging is precise, and hollographic in many recordings thanks to the precise separation of instruments and voices. Every instrument/singer has its own place on the soundstage.
Overall, I find the sound sometimes too good to be true. I got accustomed that only a few "audiophile" recordings sounded great in my system, however with theTeralink -X all recordings sound enjoyable in my system. I can still hear big differences between quality of recordings, the encoding used (mp3,flac, wav), but the Teralink-x, especially when paired with the Belden BNC cable, lets you focus on the positive sides of the recordings which is some might call "musicality".
Note on digital cables :
The Teralink-X works best in my system with the Belden BNC cable.
When paired with the Belden cable it gives a wide soundstage and a very smooth sound, "tubey like".
When paired with the Hifi Cables Sobek BNC cable, I get an increase in resolution, but a smaller soundstage and an increase in brightness "edginess".
Note on burn-in :
The Teralink-X needs a few days to sound its best. It doesn't sound bad at first but after a dew days of continuous play-back, the sound opened up and the bass got deeper. TheTeralink -X uses many capacitors inside that might explain the big change in sound that I did not notice to that extent (or at all) with the other converters
M2Tech Hiface :
The m2tech hiface is my newest usb to spdif converter, I have only owned it for a few days, but I have listened to it for enough hours now to get a sense of how it sounds. This minuscule unit is simply outstanding.
First, the music just makes more sense. The notes flow more easily and it requires less efforts from the listener to understand what the performers were trying to convey. There is also a general increase in resolution that makes the music more real, probably because of its better retrieval of low level information on the recordings which makes reverbs and ambiance hall more audible in Live recordings and Classical music.
Second, the soundstaging is wonderful. It has a better pinpoint and 3D imaging than the other units, and there is a even better separation of instruments and voices. In fact, whether it is because of better imaging or because of better dynamics, the result is that instruments and voices seem to have more acoustic power (while playing at the same volume levels as I did with the other converters), the presence factor of performers and instruments is much higher.
Also, while the soundstage with the Teralink-X is big, there is not as much differentiation between recordings. With the hiface, the soundstage changes a lot (in width, depth and height) between different recordings.
Finally, the frequency extremes seem to have been extended with a more shimmering and extended highs and more articulate and tuneful bass. Compared to theTeralink -X, there is also a shift in the tonal balance to a more neutral and faithful balance. There is a better differentiation between the quality of the recordings and it doesn't impose a sound signature on the recordings : Warm recordings come out as such and over-processed recordings are highlighted as such but are still enjoyable to listen to.
Note on digital cables :
Differences between digital cables are more easily spotted : the Belden is muffled, there is also haze and distance to its sound which gave me the impression of a big and wide soundstage.
The Sobek has a better resolution, is razor sharp, highs are more extended, and the overall perceived volume is louder probably because there is more information coming through to the dac.
Once I bought the 18 ft. Belden cable, I didn't like the sobek as much as I used to before when I used it with the other converters because it was more honest and showed the shortcomings of those units.
With the hiface I can enjoy the high resolution sound provided with the Sobek.
Note on media players :
Since I have had the hiface only for a few days, I did most of my testing with foobar 0.8.3 and foobar 0.9.6.4 with KS.
I preferred the sound using the old foobar 0.8.3 but the current version of the driver has a bug with 0.8.3, the time slider of track doesn't move when playing.
I tested briefly MediaMonkey and I made the Hiface work with ASIO4ALL. Then I tried ASIO4ALL in foobar with the hiface and it worked but the sound quality was not as good as with KS.
Final note on the Hiface :
Each time I was doing A/B comparisons with one of the other units, I would end up listening to the M2tech and forget about my review. Unplugging it to put another converter was a tough experience each time because I had to be pulled out of the music.
Beyond the criteria I described above (soundstage, resolution, ...), the M2TECH hiface was the most satisfying unit from a sonic point of view as it was the closest to real representation and offered the most believable representation of music.
Conclusion :
Getting a clean spdif signal from a usb port of a computer seems like a simple task but appears to be a rather complicated endeavour. There are many usb to spdif converters today in the market and it is hard to predict how they perform. Their performance is tied both by the quality of the drivers and the quality of the components (clocks,spdif transmitters, ...)
Out of the 4 converters I tested in my system, the m2tech hiface is the one that did the best job. I have indeed been very impressed by the performance this extremely small unit and it doesn't even require a usb cable. The designer of the hiface paid great attention not only to the proprietary drivers (like EMU and Musiland) but also to the quality of components and clocks that are used.
Granted it is the less flexible unit to use as it is limited for now to KS (Kernel Streaming) with limited media players but its superior sound quality makes you forget about its limitations.
The Teralink is the second best performing unit that is leaning a little bit on the "warm" side. However, it was only when I listened to theHiface converter that I became of that coloration.
If your library contains only CD files and mp3s, the Teralink-X is very good as it combines both a "high resolution" sound with a slight warmth that makes it more forgiving toward less than perfect recordings.
However, when you add the cost of a high quality usb cable, the overall cost exceeds that of the m2tech hiface. So it is hard to recommend it because it is limited to 16/48 and doesn't sound as good as the m2tech. Any "audiophile" who is willing to buy such a converter would likely either purchase theMusiland for its convenience or the m2tech for its performance.
The Musiland Monitor 01 USD is a very nice little unit that can do 24/192 and allows for high flexibility (ASIO, KS, DS). Sound wise, it is not a match (in my system) for the Hiface or the Teralink but I am pretty sure that if it used on other DACs that have better reclocking, the performance gap with the two converters might not be as noticeable.
The EMU 0404 usb comes last (old drivers, components of average quality, and requires a high quality usb cable). However, it is the only unit that has analog inputs and outputs. In my opinion, it should not be bought to serve only as a USB to Spdif converter. There are many units out there that are cheaper and that perform better.
Follow-up (26/10/09)
When I initially wrote my conclusion, I didn't intend to make it sound like the Musiland was a bad sounding unit. But I guess I didn't choose the appropriate words. What I meant to say is that in my system, and with my ears I preferred the Teralink-x and m2tech Hiface. The Musiland is still a very good unit.
I did some other testing since I first wrote the review with another good digital cable (the stereovox xv2) and ended up with this conclusion : I would rather use the Musiland with a good digital cable (Stereovox, Sobek) than using the m2tech hiface (my favorite of the group) with the Canare or the Belkin coaxial cable. This means that while there are clear and audible differences between those units, I was talking about the last few percents of performance in my system.
Also, there are people who found it curious that the order of purchase coincided with the order of prefrence. First, as I explained in the test protocol, I went back many times to each unit and did not do my review from memory, I gave a fair listen to all units. I also kept all 4 units on hand and won't be selling them in the near future.
Second, a little explaining about why I was "lucky" enought to have a better unit each time I upgraded. I owned the EMU 0404 usb for more than 2 years and used it mainly as a dac/headphone amp before using it as a transport for my dacs. So it was never my intention to use it as a usb to spdif converter at the time I bought it.
However, the spdif output of the EMU 0404 usb worked only with asio, which meant that I had to keep switching between the usb input of my DAC (for watching movies) and its spdif input (for music). It is at that time that I found a very promising unit the Musiland. When I had it I was very happy with it but I was a little bit disappointed about the marginal improvement it brought upon the EMU 0404 usb. I started reading about the Musiland unit and I learned that while the async protocol and use of a FPGA were good things, however it didn't use audiophile grade components (clocks, capacitors). It was when I was looking for low jitter clocks on ebay that I stumbled on the Teralink-X converter. On theory, it was supposed to be inferior to the Musiland (it uses an old cm-108 usb adaptive chip) but it used a low jitter clock, Oscon capacitors, ... and since it was pretty cheap (less than my Wireworld usb cable), I took the chance and ordered it. When I got it, the sound was superior to the Musiland unit which was rather surprising. This lead me to the quest of another usb converter but this time I was looking for an async usb converters with audiophile grade components. Then someone mentioned this great new usb to spdif converter on the Audio-gd 19 mk3 dac (I started). I went to their web site and I was positively surprised to read on their white paper about the use of 2 low jitter clocks and an async usb protocol. So this is how I ended up buying the m2tech. If I didn't learn about the m2tech I would probably have had my musiland modded with a low jitter clock. But for now I am happy with the m2tech hiface and hopefully won't be looking for other converters in the next few months
I hope this explanation can shed the light on my upgrade path and maybe also avoid skepticism about why I ended up prefering my newest unit each time
Jitter measurements:
While jitter measurements should be interpreted with great care (different results with different equipment and test protocol), it is sometimes a nice tool to make jitter comparisons in the same group of transports done with the same equipment.
Here you can find jitter measurements of the m2tech, a LynxTwo pci sound card, a Nagra DII and a Marantz CD player. In that test the m2tech was slightly better than the LynxTwo which is a professional grade PCI card, and better than the Nagra and Marantz cd player.
On a stereophile test, there were measurements of the EMU 0404 usb and other converters. The EMU had the worst jitter of the bench, and its jitter was 8 times that of the one measured with the m2tech (Granted the tests were done with different measuring equipment).
So, it seems that at least, I must have gotten the order of those 2 units right
Follow-up 2 (17/11/2009) : Digital Cables - Which length to choose ?
For those who are interested in any usb to spdif converter they are likely to use a spdif cable and wonder about the optimal length.
There are many people claiming that longer cables are better. There is even an article written by Steve Nugent on digital cables (here) which suggests that longer cables are better because they avoid reflections. While I have nothing against Steve Nugent, you have to be keep in mind that Empirical Audio also happens to be selling digital cables.
Dan Lavry, from Lavry Engineering, corrected some misconceptions about longer digital cables being better and explains why a shorter cable is better than a longer cable.
While I am no engineer, I believe that the facts stated by Dan Lavry are accurate for two reasons : Lavry has a high reputation in the pro audio world and since Lavry Enginnering doesn't sell digital cables, Dan Lavry has (in my opinion) no agenda or no interest in stating that a shorter cable is better than a longer one.
Below is an extract from the conversation and if you want to read the whole conversation, which goes on for a few pages, it starts here :
Quote:
| Originally Posted by Dan Lavry /img/forum/go_quote.gif Quote:
That is not so, and no self respecting radio frequency engineer or any other electrical engineer will come up with such false claim. In fact, the shorter the cable, the better you are. I am not suggesting to use 3 inches cables, but a 3 foot is better then 10 foot, and at over 30 feet you are certainly asking for trouble. You said the reason for keeping the length at least 2 feet had to do with reflections. Reflections have to do with MORE LENGTH, not with less length! Reflection becomes an issue when the cable becomes LONG, making the signal propagation delay longer (the signal travel time from the “driver” end of the cable to the destination “end”). What does longer time mean? Longer with respect to the digital signal rise (and fall) time. A typical cable delay is around 1.5 nano second (nsec) per foot. The velocity is slower then the speed of light, in the range of 1/3 to 2/3 of the speed of light, and it depends almost entirely on one factor - the cable inner material isolation (the dielectric). The rise time for the digital signal is between 5nsec and 30 nsec. 30nsec is slow but still within the specifications. 5-15 nsec is nice, and the reason that faster is not allowed has to do with setting a limit on the electromagnetic radiation (transmission of interference). At say 10 feet, the cable delay is around 15 nsec, and a 5nsec rise time is 3 times faster then the delay, so one DOES NEED to terminate the cable and do so properly. But at say 8 inches length, the delay is around 1nsec and even a fast 5nsec rise is 5 times slower then the cable delay, and the signal will have virtually no reflections at all. The shorter the cable, the better it is from reflections stand point as well as from many other standpoints. I am not suggesting 8 inch cables. I am not suggesting not terminating. In fact, as a rule the termination is built into the receiver side. The issue here is cable length, and the notion that there is a minimal cable length one should keep is just plain wrong. Regards Dan Lavry Lavry Engineering |
The intent of this follow-up is not to start up the debate about jitter and cable lengths, however, I felt that it was necessary for any reader to learn what Dan Lavry had to say on the subject.
Also, there is something that everybody seems to agree with. To get the best performing digital transmission, it is best to respect the 75 ohms characteristic impedance throughout the digital chain from the transport, the connectors, the digital cable to the digital receiver.
One more thing, Dan Lavry said the discussed issue is similar with usb : "what I stated would be good information for ALL cables including USB".
As a result the usb cable should be kept as short as short as possible (no surprise here).
This is good news for our wallets
Follow-up 3 - 16/03/10:
I have been using the Hiface for months now and I am still amazed about how good it sounds. Since I wrote my usb to spdif shoot out, it seems that the Hiface has gained popularity.
Below are the link to 2 reviews by EnjoytheMusic.com and 6 moons.
http://www.enjoythemusic.com/magazin...hiface_usb.htm
http://www.sixmoons.com/audioreviews/m2tech/hiface.html
More on the USB to SPDIF converters ...
Comparison between usb to spdif converters:
So far, I have compared the Hiface to 4 other usb to spdif converters. I have testes each of those converters with different drivers and usb cables (Wireworld Ultraviolet, Monster, Belkin Gold, Real Cable...). And so far, the Hiface seems the most neutral in comparison.
Here is a quick recap of how they sound:
-Emu 0404 USB: thin and edgy sounding, small soundstage, limited extension at the frequency extremes
-Musiland 01 USD: relatively neutral balance, better extension at the frequency extremes than the EMU 0404
-Teralink-X: relatively warm sounding, better extension at the frequency extremes than the Musiland
-Purepiper usb to spdif converter: relatively warm sounding, limited extension at the frequency extremes.
-Hiface: The most transparent of the group
Using the Hiface with different digital cables:
I tried the Hiface with many digital cables. Here are some quick impressions using the dac19mk3 as a DAC.
-Hiface + Canare cable: The sound is constricted, edgy, the soundstage is small
-Hiface + Belden cable: The sound is warm, the soundstage bigger than life (but not very precise), the transients are slowed and muffled in comparison with higher end cables
-Hiface + Sobek (modded): The sound is slightly on the warm side of neutral. The soundstage is not as big as with the Belden but is more defined and varies from one recording to another. There is a big increase in details.
-Hiface + Stereovox XV2: The sound is slight on the thin side of neutral, a little bit bright/metallic timbre. However, it is more detailed and spacious than the Sobek.
-Hiface + Oyaide DB-510: The sound is neutral and transparent. It the most detailed but also the most natural combination.
Using the Hiface and Oyaide with different DACs:
Since I have settled for the Hiface + Oyaide as my reference front end, I have used it to test different DACs and here is how they sound
-Hiface + Oyaide + Purepiper DAC A-1: It has a relatively neutral tonal balance with a slight emphasis on the upper midrange. It is a little bit dry sounding. With the wrong associated equipment it can sound harsh.
-Hiface + Oyaide + FUN version A (with the AD1852 dac chip): The sound is darker than the Purepiper. It is smoother; it has a lot less apparent details (different voicing) but has more low level details. It is also richer sounding.
-Hiface + Oyaide + FUN version A (with the WM8740 dac chip): The sound is warmer than with the AD1852 but there is a definite loss of details and a little bit of mid bass bloat.
-Hiface + Oyaide + FUN version B (with AD1852): It has a more limited bandwidth than version A and has dirtier highs. The sound is constricted and edgy in comparison with version A.
-Hiface + Oyaide + DAC19mk3 (with the DF1704 digital filter): The soundstage is huge and well defined, with holographic imaging. The transients are faster. There is a great sense of transparency. The bass is deep and accurate. The highs are cleaner than any other combination above.
-Hiface + Oyaide + DAC19mk3 (with the PMD100 digital filter): While the PMD100 is not as fast as the DF1704 filter, it has a purer tone with greater tonal density. It is “analog heaven” not because it adds pleasing distortion but because it has purer highs.
Conclusion
Overall, I am not saying that the Hiface + Oyaide should be used in any system as component matching is important to get a musically satisfying result.
But both the Hiface as well as the Oyaide digital cable are very transparent components. It is possible to achieve a similar tonal balance with other combination, but by using lesser components, you loose at the same time low level details, frequency extension, soundstage and imaging clues, and timbre subtleties that cannot be recovered elsewhere in the chain.
So while I understand that a specific system can sound thin and edgy with Hiface, I believe it is not because of the Hiface but because of other components in the chain (digital cable, DAC) that are more likely to be the culprit.
Side Note: I suspect that in highly resolving systems when using DACs based on sigma delta chips (and opamps), it is likely that the Hiface will sound thin and dry simply because it will reveal the true sonic character of the DAC; other more jittery devices such as the Teralink seem to inject some kind of pleasing dither that makes the listening more tolerable and enjoyable (but less precise).
Here is again a few links to why sigma delta dacs are flawed and were only pursued to lower the cost of manufacturing in comparison with the R2R/multibit DAC chips.
Mother of Tone - Conversion Techniques
How DACs Work
J River Media Center vs. Foobar v1.0
A few days ago, J River Media gave away 5 free licences for people to try against their preferred media player (mainly Foobar).
The discussion starts here : http://www.head-fi.org/forums/f7/usb...ml#post6470058
After a few days of use, I found that there were subtle but noticeable differences.
To my ears, listening through J River (in comparison to Foobar v1.0) is like turning up the sharpness controll of a TV set. With J River all the apparent details are over emphasized but at the expense of the very fine and more subtle details. For example, the image outlines are sharper with J River, but the fine little ambient cues are less audible and you feel like everything has been recorded in a dead quiet studio.
So like for TVs, a little bit of added sharpness (with J River) could be either a good thing or too much and fatiguing on the long term.
Of course, one could also suggest that Foobar is dull and that J River is neutral, but that wouldn't explain how Foobar has more low level resolution.
Overall, it is a nice thing to have yet another mean to "tweak" the sound through the media players. And given that J River contains a lot more than a mere music player, I think it is a remarkable achievement when we compare it to Window Media Player/Media Center.
Note: All my comments about Foobar, concern the v1.0. For an unknown reason, I have always found the 0.9.x versions grainy sounding and kept using v0.8.3 until the release of the v1.0.
More on the Teralink-X drivers
I did some comments and measurements that are buried in this long thread, so I decided to include them in this follow-up.
Quick note/update: Since my comparison, Teralink has released the X2 version which uses a 24/96 Tenor usb chip. For future buyers, I think the extra cost of the X2 is worth it.
Quote:
| Originally Posted by slim.a /img/forum/go_quote.gif I have tried 3 drivers with the Teralink-x: the stock driver that automatically installs, then the CM-108 Driver v.5.12.8.2119, and finally the Ploytec usb asio. The CM-108 Driver v.5.12.8.2119: I tried after a few weeks of using the Teralink and I found the sound horrible. I tried to disable all the dsp effects but there was always a loss in resolution and sound quality. I did some research and found out it had only 14 bits of resolution vs. 16 bits of the stock windows drivers. I uninstalled it and went back to using the stock windows drivers. The stock windows drivers that installs automatically: This is a good driver. I did most of my listening with this one. However, the sound is the on warm side of neutral compared to all the other converters. After trying many digital cables to get a better perception of its sound, I noticed that there is a "dulling" of the sound and a smoothing/slowing of the transients. It has a pleasing effect but it is not accurate. The soundstage is very big but not very well defined. Anyway, I am sure that people who like the "tube" sound will prefer it to most other settings or even converters for that matter. As for me, once I detected the added warmth, I could not continue to appreciate it. The Ploytec usb asio driver : This one improves the sound to a more neutral balance. There is less "dulling" of the sound. The soundstage size remained the same but the imaging improved. This driver is closer in tonal balance to the sound of the Musiland and the Hiface which leads me to believe it is closer to the "truth". Overall, I think that not all people are looking for "cleaner" sounding sources. In my experience, lowering the jitter (improving the quality of the transport) result in the following results : Bigger soundstage, less mid-bass warmth/bloat, more defined and deeper bass. Usually, you get less mid bass and more deep bass (if your equipment let you hear that) which might not be what people expect or want to hear. |
Quote:
| Originally Posted by slim.a /img/forum/go_quote.gif I read about the cmedia drivers here : Homebrew CMI 8738 drivers - Hydrogenaudio Forums By the way, when I chose to uninstall the CMedia drivers the first time I tried them it was based only on my subjective listening. You can read my comment about them here : http://www.head-fi.org/forums/f7/rev...ml#post6042974 Also, since I was in a curious mood today, I did some RMAA measurements this morning for both the generic/stock driver of the Teralink and the CM-108 Driver v.5.12.8.2119. I put the results in a Zip file if anyone is intersted in them. First, how did I test it ? I used to the EMU 0404 usb to record the output of my audio-gd dac-100 using the Teralink as a transport. Since I have tested the EMU with a SNR of 113 db (the SNR drops to 96 db when tested in 16bits which is to be expected) and since the audio-gd dac-100 has a SNR over 100db, if anything is done wrong in the digital domain it would show in the analog stage. If data is lost somewhere in the path it cannot be retrieved. For what is it worth, my findings are as follow : The generic windows drivers do not seem to mess with the data. The CM-108 Driver v.5.12.8.2119 seem to have trouble outputting correctly 16/44 without messing with the data. The SNR is worse by a 4 (and up to 6db) compared to the best results I had with the generic windows drivers. To my surprise, the 24/96 test with the CM-108 Driver v.5.12.8.2119 drivers improved the results in SNR over 16/44. However the drivers are limited to 48khz. There is a cut-off at 48 khz. After trying 24/48 and 24/44 I concluded that the benefits comes from going to 24 bits rather than from "upsampling" to 96. I read in the Valab thread people linking the 24/96 upsampling with the CM-108 drivers. Since I don't own a NOS DAC, I cannot comment on that. All I know is that there is a real loss in transparency using the CM-108 drivers in my system. I understand however that some people might like the "sonic signature" of the CM-108 drivers, but it is not the most accurate driver in my opinion. |
Follow-up 4 - 18/03/10 : Digital Filters, Minimum Phase and Upsampling:
While I have already mentioned digital filters and upsampling in separate posts, I thought it could be nice to compile some of that in a single post. So here are my findings on the subject: nothing new, just a compilation of thoughts.
Digital filters:
One of the discriminating factors between different DACs is the digital filters. Most modern chips include a built-in digital filter (for cost savings reasons) but they perform very poorly in general. But how much difference does it really make?
In my current DAC, I have the possibility to swap between 2 digital filters: the PMD100, which is a HDCD capable digital filter made by Pacific Microsonics, and the DF1704 digital filter made by TI. The two digital filters have 2 distinct sonic signatures (for those interested in the details, they can read my review of the audio-gd dac19mk3 where other users have reported similar differences between the 2 filters).
When playing 16/44 data, the DF1704 is fast and analytical, slightly on the bright side of neutral and the PMD100 is warmer and more “analogue” like.
While the PDM100 is supposed to be limited to 24/55, it seems to work at 24/88 (but not 24/96).
Well, by playing (native) 24/88 data on both, they sounded closer than they did at 16/44. The DF1704 smoothed out and the PMD100 had a little bit more sizzle on top.
That was similar to what I have experienced with some sigma-delta DACs that I have found to sound harsh at 16/44. Usually they sound much better with 24/96 data or upsampling to 24/96. The reason behind the perceived improvement is not due to the fact that we hear the extra data. It is rather due to the fact that we do not have to hear as much the nastiness of the digital filters operating close to our audibility range.
On many digital filters whether they are outboard filters (such as the DF1704) or built-in into the dac chip (such as the CS4398) the designer has sometimes the choice between a sharp/fast roll-off and a slow roll-off.
The sharp roll-off usually gives a flat measured frequency response on the 20-20,000Hz range, which looks good for published measurements. But that flat frequency response up to 20K comes at the expense of phase performance (which has an impact on soundstaging among other things).
The slow roll-off on the other hand is down from 1 to 3 db (in general) at 20 khz but has a better phase linearity, which is more audible to the human ear.
So when given the choice between what looks good on paper (a flat frequency response) and what sounds nice to the ear (phase response), most designers choose the first option.
I have looked at the measurement of many DACs on the stereophile website and some of the most expensive and better sounding DACs seem to have that roll-off on the highs. Weirdly enough most budget DACs and soundcards seem to have a flatter high frequency response. I am not saying that we shouldn’t pursue a flat high frequency response. But I am just saying that when you see a budget DAC that has a flat response for 16/44 data beyond 20 kHz, and if the DAC doesn’t use a fancy DSP, it is most probable that it has sacrificed the phase response for the good looking Frequency response measurements.
By the way, if you are still thinking that a flat frequency response is absolutely necessary to have a realistic sound, just think about the following.
Let’s suppose you listen at a violin in very reverberant room, then you listen to a violin in very damped room of a different size from a different distance. Then listen to a violin in a speaker system that has been equalized to have a flat frequency response. As far as I am concerned, while situation 1 and 2 will have different Frequency Response measurements, my brain will analyze and detect the sound as being real and live. When I will listen to situation 3, I most probably won’t be fooled thinking that the violin is real.
Fortunately, there is a way to go partially past those poor digital filters. By upsampling data to 24/96, you minimize the audible effects of those filters, as any aliasing/distortion they generate will be pushed further up in frequency range.
Impulse response & Minimum Phase Filter:
While I have mentioned the Frequency response and phase response of digital filters aspects, I haven’t mentioned yet the impulse response.
When compared to each other, a fast roll-off filter has a lot more pre and post ringing than a slow roll-off filter. That pre-ringing is one of the reasons some people have complained about digital playback and kept using analog sources (Vinyl, tapes…)
So by choosing a fast roll-off over slow roll-off, many DAC and sound card makers are willing to sacrifice the phase and impulse response in order to have a nice looking RMAA graph.
However, while the slow roll-off filter minimizes the pre-ringing, it doesn’t get rid of it entirely. There is a growing number of CD players and DACs that provide a new option which the Minimum phase filter. According to their research, the human ear would be less sensitive to post ringing than to pre-ringing.
For those who are interested, Ayre wrote an interesting paper on Minimal phase filters (here: http://www.ayre.com/pdf/Ayre_MP_White_Paper.pdf)
Again those limitations (pre and post response) are far worse at 16/44 than they are at 24/96.
That is once again (in my opinion) the reason why some people hear a lot of improvement when using up-sampling or native 24/96 data. I think that despite some measurements (SNR, THD), not many systems have a true resolution greater than 16 bits. If that were the case (let’s say 20 bit resolution), those systems would be indistinguishable from the reality which is not the case. So, playing back 24/96 data on the DACs can improve the playback experience in comparison to 16/44 simply because most digital filters are terrible at handling 16/44. That is perhaps one of the reasons that we see so many DACs today that use ASRC/upsampling. (But then again those ASRC chips do not seem to sonically sound good and can also benefit from a good upsampling from the source, but that is another story …)
SoX upsampling in Foobar:
Among the nice things that come along with using a computer as a source is the ability to try different playback methods and upsampling methods.
Personally, I have settled for Foobar for quite a long time. And recently, I have been using SoX for 2 reasons: It is a relatively transparent upsampler (it doesn’t harm low level details) and it is very customizable.
In my case, I have found the most consistent and enjoyable results with upsampling to 24/96 (weirdly it sounding better than upsampling to 24/88) and with the minimum phase setting. It generally results in a more coherent soundstage that is pushed a little bit further back, and the sound is usually smoother. But those changes vary from one DAC to another, so it is impossible to make generalizations about the effects.
I have also tried different passband settings. The Stock setting is at 95%. When set at 99% the soundstage shrinks and everything gets tighter. When setting the passband at 90%, the soundstage becomes a little bit bigger and the sound and everything sounds a little bit softer and less defined.
Those different settings are a great tool for fine tuning the sound.
While I don’t use the SoX all the time with my reference DAC (which uses a slow roll off DF1704 digital filter), I found that it improved considerably the listening experience with the other DACs I have on hand.
Note:
I brought up the subject of digital filters so that people understand the effect of a low jitter sources on their converters.
When using a high jittery source, a lot of distortions and nastiness of the digital filters are masked/blured by the noise.
When using a low jitter source such as the hiface, there are improvements in many areas but it can also reveal/expose any flaws elswhere in the system.
(BTW, not high jittery sources have pleasing distortions, the emu 0404 usb is edgy sounding).
Doing tests with 24/96 source material will minimize the effect of the digital filtering and will probably allow for better comparisons of the intrinsic quality of different converters. If listenings tests are done at 16/44 on an average sounding digital filter, the tests won't reveal much and thy would be specific to that specific type of DAC implementation.
Follow-up - 23/03/10: USB Cables
Quote:
| Originally Posted by upstateguy /img/forum/go_quote.gif I just found your review. I also have found that my DACs with built in USB performed better with transports than they did by themselves. I scanned through what seemed to be a very long and comprehensive review until I came to the bit about USB cables. It was like reading a review where all of a sudden the author stated that elephants could fly. It was very hard to take anything seriously after that. I was wondering if you would bring your discussion of USB and digital cables over to the Sound Science forum where it might be discussed in greater detail and without restrictions as to what can be addressed? USG |
What most people fail to understand: is that jitter can be generated at many points:
1/ It can be generated at the recording (too bad, there is nothing we can do about it).
2/ It can be generated at the source
3/ It can be generated by the digital cable
4/ It can be generated inside the DAC itself
5/ It can be generated by any impedance mismatch from the source to the DAC
Unlike many people here who claim that zeros and ones are all that matter, I try to do as much research as possible. Last week alone, I bought an AES research paper (Benjamin and Gannon) and I also read this 90 pages+ report on the jitter in high resolution DACs (here: http://www.iet.ntnu.no/courses/fe811...t_audiodac.pdf). If you care to read it, you will see that the author mentions several times that digital cables can have a measured jitter of a few ns (even more than the source itself).
So is it the same for USB Cable?
In the case of an adaptive usb to spdif converter, the usb converter acts more or less like a digital coaxial cable: it streams real-time data (no error correction or feedback) and its performance (jitter/reflections) is affected by the quality of cable.
FYI, as I said it many times earlier a regular usb to spdif converter doesn't act like a usb hard drive. I streams real time data and if anything is lost there is no error recovery.
In the case of an async usb to spdif converter, it shouldn't matter in theory. However, there are at least 2 reasons that might impact their performance.
First, most of them are not galvanically isolated from the computer. So the USB cable doesn't only carry data but also the power supply. So in that case, a poorly shielded usb cable can introduce HF/RFI noise into the power supply of the converter and the clocks as well as other components) inside are very sensitive to such noise.
In fact, a very good usb cable can act a filter for high frequency noise (if the capacitance of the power lines is very high). However, the power lines and data lines should not be treated/shielded the same: The data line is transmitting high frequency signals while the power lines are transmitting low frequency/DC power.
Second, I have read somewhere (maybe Steve Nugent?) that while async converters act as a master clock for the data, those converters do not always implement error check and recovery. So in that case there could be a loss of data in the process (unlike a usb hard drive for example).
However, at the time I wrote the review, I didn't know the theory and I was as puzzled as many of you. But the test I did was pretty simple; I took the EMU 0404 usb as a source and used different usb cables. I tried to get the lowest possible ASIO latency setting with every cable before getting crackles and pops. And weirdly enough the cable that could achieve the lowest latency without problems was the Wireworld ultraviolet and the one with the worst performance was the stock usb cable of the emu (which is poorly constructed).
So of course, that was an indirect test, but I am pretty sure that if Stereophile did measurements with different usb cables, they will find differences in measurements like they did with digital cables and with different usb to spdif converters.
So, I understand that hearing differences between usb cables might come as shocking news. However, what I find more shocking is the number of people that state that it is foolish to hear differences between digital cables (whether it is coaxial or usb) without even knowing the science behind line induced jitter.
Of course, there is the problem of threshold of audibility of jitter. The AES study that Nick_Charles recommended me to read talks about thresholds levels of well above 1ns. However, that study was conducted more than 10 years ago and used a cheap sony headphone for the testing. (See the discussion with Nick Charles here: http://www.head-fi.org/forums/f133/p...ml#post6479343)
A more recent study that I have found recently about the temporal resolution of human hearing proves that with the right test equipment, the threshold of temporal resolution of human ears is far greater than we suspected (see here: http://www.physics.sc.edu/kunchur/pa...rge-Foster.pdf)
As you can see, I am not against a scientific debate on the subject. However, I am still on the process of collecting data. I will start a thread about usb cables, when I have accumulated enough data and info, and when I will have enough time to do it.
I took a little bit of time to write this post as it seems to be a recurrent subject. As I said, I will start a thread about usb cables in the sound science forum (in due time). However, the rules of this forum allow me to state freely my subjective listening experiences. So I am not going to do 30 DBT for each component I cite (usb cable, usb converter, digital cable, DAC, interconnect...) just to please a small portion of people.
To build a resolving and musical system, it takes a lot of time, research, and trial and error. I try to share the most significant components upgrade and tweaks. But I can’t spend endless hours just to do a proper DBT of a single usb cable. I would rather spend that time doing other things or just enjoying music.
Pictures of my system:
