[gevorg]

Folks at HDD Audio created several test files that contain various levels of jitter: 0ns, 3ns, 10ns, 30ns, and 100ns. As a point of reference, most PCM270x implementations have around 1ns of jitter, and EMU 0404 USB has about 0.5ns of jitter.

Is jitter audible? The jitter test 2. | Welcome to HDD audio

Can you hear the difference? Can you tell which one is which? What is your setup?

I will post my results soon. Lets keep this thread away from DBT talk and just share our findings.

 

[Reader]

This test is useless, your recording can only contain error, you can't have a file with jitter. Jitter is a signal problem, the file itself cannot have jitter. Jitter is only presented in final audio stream when it's higher than hardware tolerance, or, in case where anti jitter buffers or circuit is presented, exceed spec of these.

 

[ericj]

Quote:

Originally Posted by Reader /img/forum/go_quote.gif This test is useless, your recording can only contain error, you can't have a file with jitter. Jitter is a signal problem, the file itself cannot have jitter. Jitter is only presented in final audio stream when it's higher than hardware tolerance, or, in case where anti jitter buffers or circuit is presented, exceed spec of these.

if you click the link to the original test, they explain how they simulated the jitter.

 

[Reader]

Quote:

Originally Posted by ericj /img/forum/go_quote.gif if you click the link to the original test, they explain how they simulated the jitter.

I know, but the point is they are errors now and it will be in the final audio stream 100% of time.

 

[Pars]

I would agree with Reader. Your system always has some amount of jitter. Playing a file with supposedly calibrated jitter levels is useless as the lowest level of jitter you would hear will be the one with no jitter.

 

[gevorg]

Quote:

Originally Posted by Pars /img/forum/go_quote.gif I would agree with Reader. Your system always has some amount of jitter. Playing a file with supposedly calibrated jitter levels is useless as the lowest level of jitter you would hear will be the one with no jitter.

Of course you will always have jitter. The point is to compare the different levels of it and try to hear any audible difference.

When you play a 10ns jitter file, the actual jitter level you will hear will always be higher than 10ns, that is 10ns + current system jitter, but this will also be lower when you play a 1ns file (1ns + current system jitter). Think of this as a way to compare high jitter level to low jitter level. The actual jitter level will vary from system to system, but that not the point.

 

[gregorio]

Quote:

Originally Posted by Reader /img/forum/go_quote.gif This test is useless, your recording can only contain error, you can't have a file with jitter. Jitter is a signal problem, the file itself cannot have jitter. Jitter is only presented in final audio stream when it's higher than hardware tolerance, or, in case where anti jitter buffers or circuit is presented, exceed spec of these.

That's not exactly true, you can have jitter in the file, caused by the ADC during the recording process. However, I agree that the test is essentially useless for many consumers as most DACs have PLL (or other jitter rejection) circuitry which will effectively remove all transport induced jitter. Any jitter in the recording itself is pretty much there to stay as far as the consumer is concerned.

G

 

[rumatt]

If you click on the link to the original jitter test, the first reply is Steve N who pretty much trashes them. Hahaha. The rest of the replies seem to be spam?

This idea is pretty cool, but the details of how it's done determine whether it's a useful exercise.

 

[Reader]

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif That's not exactly true, you can have jitter in the file, caused by the ADC during the recording process. G

No you can't, jitter is about signal. What you have is ERROR caused by jitter.

 

[Slogra]

The samples don't work on Foobar2000: Unable to open item for playback (Unsupported format or corrupted file). Something must be wrong with them. Other m4a files i have work perfectly.

 

[audioengr]

Quote:

Originally Posted by Slogra /img/forum/go_quote.gif The samples don't work on Foobar2000: Unable to open item for playback (Unsupported format or corrupted file). Something must be wrong with them. Other m4a files i have work perfectly.

You must use dbpoweramp to convert them to .wav files first.

I would recommend using the first 4 test files, not the second. These are .wav files. The track selected for test 2 was not a very good choice it turns-out.

Here are the first 4 files:
http://hddaudio.net/wp-content/uploa...3/abxerapp.zip

The first tracks were noisy, but I was able to pick-out the jitter without too much trouble.

Steve N.
Empirical Audio

 

[nick_charles]

...

 

[gregorio]

Quote:

Originally Posted by Reader /img/forum/go_quote.gif No you can't, jitter is about signal. What you have is ERROR caused by jitter.

I don't really understand your point, jitter is positional (timing) errors in the sample points. So jitter is error and therefore any jitter caused by the ADC is effectively encoded into the digital data stream.

G

 

[nick_charles]

Quote:

Originally Posted by Slogra /img/forum/go_quote.gif The samples don't work on Foobar2000: Unable to open item for playback (Unsupported format or corrupted file). Something must be wrong with them. Other m4a files i have work perfectly.

Same here , I tried two different programs from the web to convert them to wav format and neither worked...

 

[leeperry]

they play just fine in the latest foobar(KS) w/ the ALAC component on XP SP3.

kinda makes you wonder when Asus says that their new ST sound card has 31% less audio jitter(than what? no idea)....there's 31% less audio jitter, that's all you need to know!!! don't ask no question, don't hear no lie

ASUSTeK Computer Inc.

and then you stumble upon this :

The second jitter test is online. (Page 4) - Equipment - HDD Audio Forum
Quote:

This one is difficult. Did you try the first one? It's easier. Hearing this with headphones will be difficult. Need good stereo imaging, not between your ears.

Quote:

The idea was to show that even when jitter is present at levels much higher than we expect in practice, it is extremely difficult to detect.

Quote:

There's no way nowadays that anyone is going to do better than buying DAC Chips from the likes of Wolfson, Analogue Devices, Burr Brown, Crystal et al. These are the world leaders and if manufacturers follow their instructions they will make a DAC as good as you can get.

all the DAC's from BB/AKM/AD have built-in jitter correction AFAIK, but prolly it takes $1800 silver cables, an acoustically controlled room and $5000 speakers...then you can "hear" audio jitter