Jump to content
IGNORED

red or blue pill - Part II?


Recommended Posts

3 hours ago, manisandher said:

I reckon a 10-run ABX would take around 10-12 minutes. I'd go for 3 of these in total, with a small break between each. This would give a total sample size of 30. Would you agree that a score of >25/30 could be taken as a positive?

 

Yes, 25/30 is an excellent result.

 

3 hours ago, manisandher said:

There was no issue with the digital feed - it proved that the DAC received bit-identical signals during the ABX, where I scored 9/10.

 

I used a Tascam DA-3000 ADC for the initial analysis of the analogue output of the DAC. Here are its specs:

1932218338_TascamDA-3000recordingspecs.JPG.6e1bac63e22af9a65336a48612cdfb93.JPG

 

As I recall, the digital feed recordings had issues with not synchronizing right away. Some number of samples were different at the beginning. Since these were the main record of what had actually transpired, I would have to assume that the two captures were not bit perfect, at least at the start of each track. Again, an explanation of what went wrong isn't really a substitute for the real capture being bit-perfect :)

 

Tascam THD+N of 0.003% can be a lot better. I have a simple pro Apogee interface (paid about $350 for it used) that does about 10x better (0.0003%). I think Benchmark or RME equipment was mentioned before in this thread, and these do 8-10dB better than my little Apogee. The point is to have as little distortion added to the recording as possible by the test equipment. Things like jitter and phase distortions need to be looked at, as well, since these can create audible differences when large enough. A few simple loop-back recordings of DAC/ADC playing music for 60 seconds would give us enough to analyze to see how the equipment behaves before the test.

 

With DeltaWave software, I've compared now a large number of DAC/ADC loop-back recordings from various sources. The better equipment produces a null of -90dB or better. The recordings your sessions produced generated, at best, a -50dB null. That's a really poor conversion quality compared to some of the better equipment out there.

 

Link to comment
3 hours ago, manisandher said:

 

I find it useful to consider this in two parts:

 

1. the ABX listening test itself

 

The specific DAC, amps and speakers used for the ABX, and where they're situated, etc. are all totally irrelevant. It's a red herring on Mans's part. The only thing that matters is showing that the DAC received bit-identical signals when replaying A and B. The digital feed to the DAC was captured throughout the ABX, and it was proven that the DAC had indeed received bit-identical signals throughout. And yet I heard consistent differences between A and B, as shown by my 9/10 score.

 

2. the analysis of the analogue output of the DAC

 

This proved difficult. I've posted the specs of the ADC used in the original analysis, which show nothing untoward. But on analysing the analogue captures, the ADC proved unsatisfactory.

 

Mani.

 

Mani, there are some things that need to be considered in (1) as well. Bit-perfect transmission is not sufficient if the interface (SPDIF) also carries the clock signal used to drive the DAC output. Large amount of jitter on the interface can cause timing errors. For this reason, using a USB or another asynchronous connection to the DAC would be a much better choice, IMO.

Link to comment
2 hours ago, manisandher said:

 

So, use a USB DAC with an spdif output, and capture spdif output in real-time to test that A and B are bit-identical?

 

Any way of checking A and B are bit-identical using a USB DAC without an spdif output?

 

Mani.

 

There are very inexpensive USB to SPDIF converters out there that could be used, but that wouldn't work simultaneously with the DAC unless there's some PC sound driver software used that can feed the same input to multiple outputs (easy to do on the Mac, not sure about Windows).

Link to comment
2 hours ago, manisandher said:

 

I've got the following ADCs to hand:

- RME Fireface 800 (need to repair the SMPS first)

- MOTU 896HD (mkI)

- Prism AD124 (44.1/48 only)

- Tascam DA-3000

 

(Shame I don't still have my Pacific Microsonics Model Two :(.)

 

The Tascam should be fine for the digital captures, as there's no actual conversion going on.

 

I'll do some DA/AD loops with my USB DAC and the other converters to figure out how they fare. Would a simple sine wave suffice, or would real music content better?

 

Mani.

 

For loop back, both, a simple 1kHz sine wave and a music files would both be useful. Sine wave is easier to interpret (get actual THD+N, etc.) but a music file will exercise a lot more of the converter chain. The Gearslutz interface loopback thread uses this WAV file (24/44.1kHz).

 

 

Link to comment
51 minutes ago, PeterSt said:

All right Paul, fair enough. Then : Could you please take distance from your own remark that some random ADC requires its time to settle and (level) trigger the recording, so that it is perfectly allowed to chop off the irregular beginnings in order to next have comparable files.

I think you can.

 

DeltaWave can chop off any mismatches at the start and even the end of the track, so no worries. It can do this automatically, or manually. The problem with mismatched data was not on ADC recordings in the first test, it was with the digital recorder. That leaves the question whether the issue was with the recorder, or the source. In the absence of additional testing, I don't know the answer. And if the error was with the source, then that could easily cause some audible differences at the start of the track and so needs to be eliminated as a possibility.

 

51 minutes ago, PeterSt said:

Next up are the recordings already present. So what can your software do with them ?

The whole point is: if that brings nothing, it most certainly does not testify of Mani cheating or anything of the sort. It would tell, however, that the test means are still too weak.

 

The problem is that my software shows very large differences between the ADC-recorded track and the original digital one. Larger than any good DAC/ADC loop would produce. I would be very surprised if this was caused by this extremely low-level noise that you are talking about: -50dB or so difference is certainly in the audible range. Which is why I have to question the quality of the equipment and how it was set up.

 

The other issue I find often in trying to take ADC captures is ground loops and other electrical interactions between components that are not normally part of the playback system. The result can easily be a very distorted ADC capture, while the DAC is producing a reasonably clean and accurate output. That's why I suggested a preliminary end-to-end/loop-back test using simple sine-wave signal as that will help to reveal any such issues, as well as a baseline recording using a standard music track.

 

51 minutes ago, PeterSt said:

FYI, and just in case you didn't know it yet: I can show you and everyone the most easily discernible differences between whatever dozens of means, all leading to bit identical looped back data, those means including USB cables. It goes from, way too wide sound stages, to way to flat ones, compared to narrow and deep. Don't ask me why you and so many others have difficulties with getting that. All I know myself by guarantee, is that somehow we all (!) are not able to find test means showing it. So ... let's have it puuulease ?

 

All not meant as teasing or anything, but I feel we are in a deadly loop.

 

The reason this is questioned is that unless you do the test under properly controlled blind conditions, you are using your brain's interpretation of what you see in combination with what you hear.  I think you'll agree that your ability to see is not what we want to test when doing proper listening tests :)

 

If you can demonstrate (just like Mani is trying to do) that you can distinguish your USB cable under double-blind conditions then that will break the "deadly loop"! Even more so, if the measurements/ADC captures still don't reveal any differences.

 

Link to comment
14 minutes ago, manisandher said:

 

Paul, as you know, I've been exploring getting hold of an ADC that would be up to the job of measuring the bit-identical differences I'm hearing. Here are the loopback measurements of the ADCs I currently have to hand:

image.thumb.png.681fb93d158f574ac46da0afe8e8a63f.png

(* Potential future purchase.)

 

I suspect that in the majority of cases it's the ADC that's the limiting factor in these measurements, and not the DAC. So, it seems to me that none of these are really be up to the job at hand... and the RME is considered one of the best currently available!

 

So where to now?

 

Mani.

 

The Gearslutz measurements are a little out of date :) DeltaWave corrects for the non-linear phase effects (which is usually caused by the filters in the DAC and the ADC). Since these are usually not audible differences, the Gearslutz results are way too low. I've had a conversation with Didier who is running the list about updating the list, at least to add the new, corrected measurements. But he's reluctant to do so, claiming that the results are already too confusing and complex and he doesn't want to add to the confusion :) Maybe at some point I'll just set up DeltaWave to run through all the uploaded files on that thread and produce and publish my own list of results...

 

Let me run some of these through DW (assuming the loopback files are available) to get proper results.

Link to comment
1 hour ago, manisandher said:

 

Paul, as you know, I've been exploring getting hold of an ADC that would be up to the job of measuring the bit-identical differences I'm hearing. Here are the loopback measurements (most taken from the Gearslutz DA_AD thread) of the ADCs I currently have to hand:

image.thumb.png.681fb93d158f574ac46da0afe8e8a63f.png

(* Potential future purchase.)

 

I suspect that in the majority of cases it's the ADC that's the limiting factor in these measurements, and not the DAC. So, it seems to me that none of these are really be up to the job at hand... and the RME is considered one of the best currently available!

 

So where to now?

 

Mani.

 

Here are the DeltaWave results for the few loop back files I could find. The others were no longer available, so I couldn't test them.

 

Unfortunately, can't confirm that, for example, RME FF800 was not recorded through a digital loopback, similar to your first attempt with M2. In any case, Motu M2 doesn't look that bad at all. Remember that this is the combination of the DAC and ADC, so individually, ADC and/or DAC may perform even better.

 

image.png.b40bc8541ae57583cfac0655f7605dc0.png

Link to comment
4 minutes ago, manisandher said:

 

 

Thanks Paul, that's really, really helpful. I might hang on to my new M2 after all :).

 

I've got a Fireface 800, so will do some loopbacks (once I've fixed its SMPS - second one to fail on me on an FF800!).

 

The ADI-2 Pro FS is supposed to be the king of modern 'affordable pro' ADCs. Any idea why it performs so badly in the loopback?

 

Mani.

 

I'll try to dig into it, didn't have the time to look in detail. Unfortunately, it's not clear under what conditions the loopback recording was made, with what settings, etc.  Let's see if DW can help to find what was causing the large error.

Link to comment
44 minutes ago, manisandher said:

 

 

Thanks Paul, that's really, really helpful. I might hang on to my new M2 after all :).

 

I've got a Fireface 800, so will do some loopbacks (once I've fixed its SMPS - second one to fail on me on an FF800!).

 

The ADI-2 Pro FS is supposed to be the king of modern 'affordable pro' ADCs. Any idea why it performs so badly in the loopback?

 

Mani.

 

Hold the presses! Found another loopback recording. This one is described as using ADI-2 Pro FS with sharp filter setting for both, DAC and ADC. A much better result (see last item):

 

image.png.9117d2c358f9effd7c4dfc7f4a274a3b.png

Link to comment
58 minutes ago, manisandher said:

 

Great! I'll go ahead and order an ADI-2 Pro FS.

 

(The M2 seems pretty good for the price though.)

 

Mani.

 

To close the loop. Looking into the other ADI-2 Pro FS capture with lower scores, it appears to be using a different filter, and has an excessive amount of jitter. The first capture shows around 13μs RMS jitter over the entire recording, while the second one shows 186ns. The shape of the phase difference plot is also obviously different between the two. The first one is shaped as an inverted letter U, the second one is nearly a straight line.

Link to comment
2 minutes ago, fas42 said:

Gee whiz ... the technical, measured accuracy of a piece of hardware dramatically changes, just by altering a setting or two of how it works ... guess that proves that all one needs is dig out of a set of numbers from somewhere, to be fully assured "how good something is" ... 🙃.

 

Good try, but in this case, there's no indication as to how the 'bad' capture was produced. There's so much jitter in it that I'd be surprised it had anything to do with settings. At least I can't imagine what setting would cause this.

 

Link to comment
2 hours ago, fas42 said:

Which underlies the problem with taking measurments in audio, in general - unless one has absolutely full understanding of every detail that may be pertinent to the numbers derived, then it is Russian roulette, often times, whether they have any bearing on "what it sounds like" ...


Yes, if you don’t know what you’re doing, you’re likely to mess up. Surprising, right? ;)

 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...