Computer Audio Asylum

Music servers and other computer based digital audio technologies.

Return to Computer Audio Asylum

Ayre QB-9 jitter?

66.192.47.2
Posted on October 26, 2009 at 13:47:01
Vik
Audiophile

Posts: 1031
Location: Rochester, NY
Joined: January 1, 2002
 I was reading the measurements section in the recent Stereophile review and noticed that they didn't publish jitter measurements for the Ayre QB-9.

I understand that jitter is just one element in how a DAC sounds, but I would like to know what the jitter (in picoseconds)is on the DAC. I'm not smart enough to read the graphs as posted so I apologize if the answer is staring me right in the face.
Vik
Hung up on the wrong measurements, posted on November 2, 2009 at 19:18:51
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 Jitter measurements are a really poor indicator to make a buying decision IMO. These measurements tell you very little about how the thing sounds.

Measurements published dont tell the whole story, namely the important part of what the spectra of the jitter is.

I have two oscillators that have identical specs, 2psec of RMS jitter. The two oscillators sound radically different when used in the same system. It means very little. Only a macro indicator.

Read the post again, posted on November 2, 2009 at 19:36:18
Vik
Audiophile

Posts: 1031
Location: Rochester, NY
Joined: January 1, 2002
 "I understand that jitter is just one element in how a DAC sounds, but I would like to know what the jitter (in picoseconds)is on the DAC."

Apparently you missed this sentence in my post. Clearly I understand jitter as just one measure of a DAC and not the sole determining factor in a particular unit's sonic merit.

120 psec, posted on October 26, 2009 at 23:14:10
Charles Hansen
Manufacturer

Posts: 4321
Joined: August 1, 2001
 That is the limit of the test used with 16 bit data. Go to the online article linked below and look at figure 2. That is a computer-generated, theoretically perfect picture of what the playback waveform should look like.

In the case of the QB-9, the noise floor was right below the tips of the harmonic series of the 229 Hz square wave. There were no peaks above what was supposed to be there in the theoretically perfect signal. Compare that with some of the players on the second page with horrible jitter spectra.

The final graph in the article is of the Ayre C-5xe. That is about as close to the theoretically perfect spectrum as you can get. But the QB-9 sounds about equal to the C-5xe. So my conclusion is that uncorrelated noise that is 120 dB below the signal is not all that audible.

RE: 120 psec, posted on October 31, 2009 at 19:19:08
jkeny
Audiophile

Posts: 182
Joined: May 4, 2001
 Charles,
On Diyhifi you castigated the AP2722 as "AP only specs that model to have a jitter measurement floor of 1000 psec" and as a result rubbished the Musiland DAC whose jitter noise was measured in comparison to the AP2722.

Now you seem quiet happy to reference the graphs produced by this sub-standard equipment as showing your DAC is near perfect!

Which is it? Some consistency would be good!

RE: Ayre QB-9 jitter?, posted on October 26, 2009 at 15:55:14
Old Listener
Audiophile

Posts: 704
Location: SF Bay area
Joined: February 6, 2005
 > they didn't publish jitter measurements for the Ayre QB-9.

Figure 14 and the paragraph above the figure.

I much prefer to see the graph rather than be given a single number without a graph.

Bill
Graph vs number, posted on October 27, 2009 at 06:22:45
Gordon Rankin
Manufacturer

Posts: 2247
Joined: June 9, 2000
 Gang,

The problem is that some of the Graphs have a lot of spuria on them and people think they have terrible jitter, when in fact it is actually much better.

Unless you understand this a number is better.

Both Charlie and I own a Wavecrest. These are jitter measurement tools and we can test the actual jitter on the Word Clock going into the dac. Again we use Word Clock because it is the slowest of the Audio Clocks in the interface and is correlated to the Master Clock so it gives the best representation of the overall jitter.

The problem with giving this number is two fold....

1) Most current dacs have jitter rejection built in. This is why we do analog tests as well with the graphs.

2) Most current dac chips use MCLK more than Word Clock as a reference, especially I2S which only uses Word Clock as an enabler and then this is clocked by the Bit Clock.

I am sure there is more but I need more coffee.

Thanks
Gordon
J. Gordon Rankin
No number was published?, posted on October 26, 2009 at 16:12:24
Vik
Audiophile

Posts: 1031
Location: Rochester, NY
Joined: January 1, 2002
 Again, I don't see a measure of jitter in picoseconds. All the paragraph you cited indicates is that measurements were done, how they were done and that the results were great. I just want to know, in picoseconds, what the jitter was.

Any help?

RE: No number was published?, posted on October 26, 2009 at 19:41:17
Sunya
Audiophile

Posts: 386
Joined: July 11, 2007
 If you check the last digital sources JA measured you'll see that he doesn't give a psec number anymore... Don't know why he decided to skip it.

Why JA doesn't give jitter numbers anymore, posted on October 27, 2009 at 21:41:54
Charles Hansen
Manufacturer

Posts: 4321
Joined: August 1, 2001
 The jitter number in picoseconds is generated by a piece of software from Miller Audio Research. The problem is that software only works with the National Instruments data acquisition cards. These cards have far too much noise to get an accurate reading.

JA has switched to using an AP 2722 to do the spectral analysis for his jitter plots. The noise floor of the hardware is probably around 20 dB (10x) lower than the National Instruments card. But the software won't spit out a number.

So he has a choice of giving an accurate graph of the jitter spectrum or else using software that only works with cards that are so noisy that they corrupt the measurement. He has apparently chosen to forgo the number in favor of the accurate graph.

RE: Why JA doesn't give jitter numbers anymore, posted on October 28, 2009 at 00:47:04
Sunya
Audiophile

Posts: 386
Joined: July 11, 2007
 Yes, I know this, but I wanted him to keep giving a peak-peak number because even if it's not 100% accurate I still find it relevant in the context of all the measurements he's made for Stereophile in giving a clue of how good a digital source was engineered.

Take for example the Bel Canto DAC3 and the Weiss Medea:

http://stereophile.com/digitalprocessors/1107bc/index4.html
http://stereophile.com/digitalprocessors/781/index5.html

and the Zanden 5000:
http://stereophile.com/cdplayers/1106zanden/index4.html
http://stereophile.com/cdplayers/1106zanden/index8.html

Maybe a Prism dScope would be useful as well along to the Audio Precision:

http://www.prismsound.com/test_measure/products_subs/dscope/dscope_home.php

Yes, wouldn't it be great..., posted on October 28, 2009 at 11:56:04
Charles Hansen
Manufacturer

Posts: 4321
Joined: August 1, 2001
 ...if we all had a single number that would tell us how good a piece of equipment was.

It's really a drag that we have to look at a graph and use our heads and actually think about what might be better. Or worse still, actually *listen* to something to see how good it is. Life sucks.

Maybe some like the numbers to help justify the price., posted on October 28, 2009 at 15:00:28
subdud
Audiophile

Posts: 497
Location: Indy
Joined: June 1, 2004
 If we weren't such a picky bunch then maybe we would be satisfied with something costing much less yet sounds pert near the same.

RE: Yes, wouldn't it be great..., posted on October 28, 2009 at 12:51:40
Old Listener
Audiophile

Posts: 704
Location: SF Bay area
Joined: February 6, 2005
 > ...if we all had a single number that would tell us how good a piece of
> equipment was.

"I'd give it a 77. It has a beat and you can dance to it."

Heard on American Bandstand.

> It's really a drag that we have to look at a graph and use our heads
> and actually think about what might be better.

I like pretty pictures of a tall peak with a nearly flat valley floor 120 dB below.

> Or worse still, actually *listen* to something to see how good it is.
> Life sucks.

It is a dilemma for me. If I listen to some hi-res material, I get bored with the music and just want it to end. If I'm listening to a fine performance of music I like, I forget all about listening for differences.

Bill
http://carsmusicandnature.blogspot.com/

Page processed in 0.042 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2009, Audio Asylum, all rights reserved ] [ Go Green ] [ Take5 ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]