Blue or red pill?

[pkane2001]

Just now, manisandher said:

 

There were absolutely no drop-outs during the A/B/X. They are irrelevant.

 

Mani.

 

How did you determine this, Mani? If drop outs are a tiny fraction of a second, these may not be easily recognized as drop-outs, although they may affect sound quality.

 

[pkane2001]

Just now, manisandher said:

 

All 30 digital captures taken in real-time during the A/B/X are bit-identical.

 

Mani.

 

That's a good starting point, but it's only proving bit-perfect data prior to the DAC. If analog captures show drop-outs, this could indicate that the DAC or the receiver circuit had some difficulties with timing or other aspects of the digital signal.

 

[pkane2001]

4 minutes ago, mansr said:

Correct.

 

If the recording is long enough, sure. In this case, it was around 140 seconds. I trust the equipment to remain stable for that duration, especially seeing as it is always on, so there should be no warmup effects.

 

What I think audible is neither here nor there. That was the whole point of this exercise.

 

Agreed, though I doubt there were any during the test. That should have been readily audible. I didn't hear the playback though, so I can't be sure.

 

If a fraction of a Herz jitter sideband with only a few dB difference in level is audible, then this is a fundamental finding, and either Mani has super-human hearing, or most of what we know about jitter audibility needs to be thrown out 

[pkane2001]

4 hours ago, mansr said:

Mani did some longer captures of a 10 kHz sine tone using both SFS settings. Here's an FFT of those:

 

 

Looks like there's a little more jitter with SFS=0.1. I'm not saying this is necessarily what Mani heard, only that it's an identifiable difference. Maybe. It could also be caused by some unrelated change in the environment that occurred between the two recordings.

 

Both of the long captures also had a dropout of a few milliseconds about 140 seconds in. The FFT is from the portion preceding this glitch.

 

Mans, can you also post a wider plot of the same data? At least a KHz on each side, and may be the full 20-20KHz range?

[pkane2001]

30 minutes ago, STC said:

 

Is there any reference to explain the graph where one can conclude that the difference seen in the graph must be due to jitter?

 

Jitter produces symmetric sidebands around the fundamental frequency. Close-in jitter tends to appear as widening of the fundamental frequency at the base.

[pkane2001]

57 minutes ago, mansr said:

Sure.

 

 

Here's another plot showing a minor difference:

Thank you, Mans! 

 

Interesting. There is quite a bit of noise with some noise spikes rising nearly 40dB above what appears to be the noise floor. In the zoomed in portion all the spikes appear to be at 100Hz interval, is that right? Seems a bit strange, even for a 50Hz mains frequency.

 

Do you have a calibration scan of a 10KHz sine wave using your ADC? Something we could compare these scans with without Mani's PC and DAC in the circuit?

 

[pkane2001]

16 minutes ago, mansr said:

Yes, there's a spike every 100 Hz. No, that's not normal.

 

We used Mani's ADC. What would you use as a reference?

 

Ah, I didn't realize it was Mani's ADC. A sine wave generator, if you have one. If not, any well-known, quality DAC (preferably with measurements) with a digital sine wave generator to feed the ADC. Worst case, any other DAC, again preferably with measurements.

 

[pkane2001]

38 minutes ago, manisandher said:

 

Mans has some analogue captures of a 10k tone being played through the Phasure NOS1, taken with the same ADC.

 

Mani.

 

That might be useful to compare to, to eliminate the ADC as the possible source of noise, jitter and 100Hz harmonic content.

[pkane2001]

43 minutes ago, Superdad said:

 

I wish the CA search engine worked right.  Because I'm quite sure I made a post at the beginning of this thread predicting that no matter how well you passed the test the results would be rejected. 

 

Were they rejected? I thought they are still being reviewed and discussed. Maybe I missed the rejection note.

[pkane2001]

1 hour ago, manisandher said:

I'll upload* the digital captures taken in real-time during the A/B/X. You'll then be able to verify things for yourself and drop the "we are told".

 

I'll upload* the analogue captures too. Do with them what you like.

 

*Won't be able to do this until the weekend.

 

Mani.

 

That would be great, Mani!

[pkane2001]

30 minutes ago, PeterSt said:

Lots of hits on 400Hz (isolation) transformers ...

 

So what explains the regular interval 100Hz spikes in the older DAC captures? Same cause as 400Hz with NOS1? Strange, especially that the magnitude of the spikes is about the same in both cases.

 

EDIT: interesting that 176.4KHz is exactly 4 times 44.1Khz, isn’t it??? 

[pkane2001]

1 hour ago, STC said:

 

The bits are the same but how the bits reached the recorder or DAC can be different. For an example, A CD drive could take one minute to write a track while a newer one could take about 50 seconds. The timing of the data accumulated at the recorder varies but they should be bit-identical to the original file. Correct me if I am wrong, isn't a DAC receives data in real time stream and the timing of how the data is converted to analogue need to be precise compared to how the data is written to a drive digitally. I am asking since I do not know.

 

 

 

I have been a audiophile for long enough to believe that there should be audible difference with SPDIF vs Toslink Vs BNC vs XLR connection. Are you telling there cannot be any difference here.

 

 

This chart was taken from https://audiosciencereview.com/forum/index.php?threads/understanding-hdmi-and-potential-for-hdmi-cable-differences.104/ 

 

Amir said "This is showing the *analog* output of the AVR DAC. Purple is S/PDIF. Yellow is HDMI. Identical digital data was sent to each input. Yet what came out of the DAC was much more distorted in the case of HDMI."

 

Assuming that the data reached the DAC were identical to what arrived at the Tascam the noise or jitter cannot be conclusively said to be similar before going through the BNC to RCA adapter. I am sure many here would argue that they could hear a difference with BNC and SPDIF itself but that is not the point of contention here. It would be easier to concentrate on the difference alone that Mani heard by eliminating the different cable used for Tascam and DAC. Ideally, both should have used BNC or RCA.

 

 

Jitter can be audible and can distort playback, there's no doubt in my mind that this can happen as I spent years in the 90's measuring, testing, and trying to eliminate jitter from a CD-based system. 

 

In the HDMI vs SPDIF jitter is responsible for the large difference in the width of the 12KHz spike. This is because HDMI does not carry a separate audio clock. The receiver derives the timing for audio from the video signal, which often adds a ton of jitter to the process, so this plot is not surprising at all.

 

 

[pkane2001]

1 minute ago, jabbr said:

Ok then. May I summarize my impression after 77 pages?

 

There is a software setting that the developer says affects “jitter”.

 

A listener claims to be able to hear changes in this setting. The has been verified by a single A/B/X test. 

 

The fact thst this setting affects jitter (phase error) has not yet been independently verified.

 

At the moment we have evidence that digital phase error is audible. 

 

Perfect summary.

 

[pkane2001]

 

6 minutes ago, PeterSt said:

 

It is noise already - not a test signal ...

 

Right. This is noise or some sort of interference, but at a very high level. What's more, these peaks are throughout the audible range of the recorded signal, resulting in significant distortion. There's something not right with either the playback or the recording part of the setup.

[pkane2001]

1 minute ago, PeterSt said:

Maybe it is useless to look at these spikes because it is just a noisy DAC. We can try to explain those things, but we better don't ...

 

But it looks like the noise patterns are different between two SFS settings, which may account for what Mani was able to hear. 

[pkane2001]

3 hours ago, STC said:

 

How about the Altmann DAC? Can the data after the SFS changes reaching the buffer before the DAC subject to timing difference?

 

Sure.  Timing errors at the SPDIF input can result in all kinds of difficulties at the DAC, from PLL errors to skipped samples to excessive jitter. A better/more modern SPDIF receiver in the TASCAM recorder may have none of the same difficulties. The fairly regular noise spikes we are seeing in the analog captures can easily be caused by the confusion within Altmann DAC.

 

[pkane2001]

 

12 minutes ago, PeterSt said:

 

That can be. But what I tried to say is that when noise is so high, something else is going on and it just varies. For example, transformer noise/hum also varies all the time. I am not even sure why but it could just be the mains (voltage level which varies) itself. And this *is* in the lower regions, and close to 50Hz (or 100Hz the way Mans presented it) harmonics.

I wanted to ask Mans whether these spikes may be "walking" more or less (and this regarding my "close to" 50Hz/100Hz). But then I thought, well, what you quoted ...

 

What's strange is that even with NOS1a 10KHz capture, there's  a similar set of spikes, except at 400Hz intervals. Either coming from Mani's environment, or present in the original 10KHz waveform, or possibly caused by the recorder itself?

[pkane2001]

2 minutes ago, PeterSt said:

 

Nah ... and when I said "with some effort you can see the 400Hz in there too", I didn't even calculate/multiply/divide correctly.

 

 

It is 100Hz, if anything ... (and that is allowed for 50Hz mains which I have here (Mani and Mans too btw))

 

Wasn't looking at your chart, but the one Mans posted earlier:

 

5 hours ago, mansr said:

Same FFT parameters as before:

 

 

The noise floor is considerably lower. There are spikes every 400 Hz, which is still not normal.

 

[pkane2001]

7 minutes ago, jabbr said:

That plot  does not support the assertion that close-in phase error is the difference

 

Where was this assertion made?

 

[pkane2001]

16 minutes ago, manisandher said:

 

I think we can all agree that the Altmann is not a particularly good-measuring DAC, based as it is on the late '80s1543 chip. The key question is: did this compromise the A/B/X in any way? I'd say absolutely not.

 

Mani.

 

Mani, if you are still talking about proving that you could hear the difference between  two different SPDIF streams with various amounts of jitter or noise containing the same bits, then I’d say you have proven that. But, that lack of jitter rejection or other issues within your DAC  could have caused you to hear a difference, where a better DAC or a USB reclocking interface would eliminate them, I’m afraid you didn’t test for at all.

[pkane2001]

6 minutes ago, acg said:

 

Except the dac that Mani usually listens to is sota in the ways that you mention and SFS changes are audible with it as well... just not proven in this test of course. 

 

Yes, may be true, but was not tested for in this test.

[pkane2001]

28 minutes ago, psjug said:

Are any difference in the analog captures (the music ones) apparent when comparing small sections of time series plots?

 

There are in a simple 10KHz sine wave signal capture, so you can assume there will be at least as much, but probably much more, with a more complex signal. Analog captures will never match exactly even under the best conditions.

[pkane2001]

40 minutes ago, manisandher said:

 

If a DAC were perfectly isolated, I'd agree.

 

 

We first need to answer the question, "Isolation from what?" And we can't answer this until we've fully articulated the exact mechanism that causes different jitter signatures with bit-identical replay.

 

Mani.

 

Mani, there are further tests that can help pinpoint the source of the audible difference.

 

[pkane2001]

11 minutes ago, STC said:

 

Can someone give a proper definition for bit-identical data, please? My understanding is when we say something is bit identical that means it does not include any other code of the sender along with the original bits. In Tascam, the SFS was probably ignored when the data were captured. In the case of DAC, although it received bit-identical data it also had received variable in the form of SFS which could have handled and processed differently. 

 

The use of bit-identical term to be included in the explanation of the audible difference is IMO may confuse the issue further.

 

SPDIF protocol does not carry SFS data (bits) of any sort.

 

SFS is processing done by XXHE software on the PC, before data is converted to SPDIF stream and output to the DAC. The bit-identical in this test refers to exactly the same samples being transmitted from the PC to the DAC with two different SFS settings.

 

What this doesn't account for is any variation in sample timing (jitter) or noise being carried with the digital signal or along other pathways. Both can potentially have an audible effect on the output of the DAC, despite exactly the same bits (samples) reaching it in the two test scenarios.

 

[pkane2001]

48 minutes ago, semente said:

You seem to be pointing at jitter, other to phase...

 

Whats the difference between phase noise and jitter?