RF & Audio

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
M

marce

Member
Joined 2007
The trade of between undesirable effects caused by fast rise times has to be balanced with the effects of slow rise times (such as increased ISI).
With the generaly uncontrolled impedance cables between components, RCA connectors etc, I still think 1ns Tr are going to cause more problems than say a 5ns Tr.

I am just going through the rather painful training to use the Cadstar Signal Integrity Verify tool (new toy:))! Though I have to admit it has become a necesity at work with the signal we are working with.
Once I become au fait with the tool, should be able to model the SPDIF interface and various cample combinations, including connectors, which should add fun to our discussions.
Goto admit the tool itself is easy to use, setting up the models and stimulous is the hard part.
Added a couple of links that reference Rise Time and Inter symbol interference that I have used for reference to get my head round all the concepts.


http://www.circuitsassembly.com/cms/magazine/209/9644/

http://www.serialata.org/documents/SATA26_ECN_019_v2_Cable_ISI_Test_000.pdf
 
J

jkeny

Banned
Joined 2007
marce said:
The trade of between undesirable effects caused by fast rise times has to be balanced with the effects of slow rise times (such as increased ISI).
With the generaly uncontrolled impedance cables between components, RCA connectors etc, I still think 1ns Tr are going to cause more problems than say a 5ns Tr.
Click to expand...

But if these reflection scan be ameliorated then we are left with the benefits of the low ISI. That's why I use Rf attenuators on SPDIF lines :) http://www.diyaudio.com/forums/digital-source/182760-rf-attenuators-again.html
 
wakibaki

wakibaki

Banned
Joined 2008
abraxalito said:
There's a thread about it over on Diyhifi too. Charles Hansen admits sound quality differences for his QB-9 asynch DAC with different USB cables. He says he doesn't know why.
Click to expand...

Charles Hansen said:
Everyone knows that things that SHOULDN'T affect a CD player will -- green paint, damping mats, disc treatment liquids, damping pads for the chips themselves, different feet for the player, and on and on and on. NOBODY knows why any of these things make a difference.
Click to expand...

You shouldn't take the stuff said on DiyHifi too seriously. Those guys are flakes. Bybee customers. Just because somebody can cobble together a few bits of digital technology into a DAC does not mean they are the full shilling.If this guy's DAC sounds different with different USB cables, then:

1. He should know the reason why. Isn't that the whole point of an asynch USB DAC? If it was my design I wouldn't rest till I'd figured it out. An unexplained phenomenon is a discovery waiting to be made.

2. It's a crap design in the first place

I found the thread you refer to, it's a complete furball of people hacking at each other over phenomena which are almost certainly imagined and are anyway of no concern to anybody other than the mentally ill. They've lost track of what is reality vs. superstition:- 'Everyone knows that things that SHOULDN'T affect a CD player will' - everyone damn well does not know any such thing, and when you start from that basis it's only a short step to being abducted and subjected to the anal probe. Hardly surprising tho' in the foetid atmosphere of audio appreciation, where an infinity of veils still remain to obscure the vision of the true acolyte.

If these people took some time out to learn to make music themselves with an instrument perhaps they'd have less frustrated energy spilling over into these masturbatory flights of fancy.

It's adolescent and pathetic.

w
 
wakibaki

wakibaki

Banned
Joined 2008
In post #137 (28 posts ago) 'buffering burst-download USB3 DACs', before anybody told me that such a thing had been built. You fail to realise JK, that as well as RF circuits I actually design high speed digital devices, and have been predicting jitter-independent DACs for at least 2 years. I wasn't aware that there were any on the market, and I cannot be au fait with the ins-and-outs of every digital device on the market, so I wan't actually aware that it could be done under USB2, or I might well have designed one myself, although it would never be a matter of high priority for me since it would only have been to prove a point, since I regard the majority of audiophile encumbrances as being as necessary as t1ts on a bull.

w
 
abraxalito

abraxalito

Member
Joined 2007
marce said:
The trade of between undesirable effects caused by fast rise times has to be balanced with the effects of slow rise times (such as increased ISI).
With the generaly uncontrolled impedance cables between components, RCA connectors etc, I still think 1ns Tr are going to cause more problems than say a 5ns Tr.
Click to expand...

Yep, agree there.

I am just going through the rather painful training to use the Cadstar Signal Integrity Verify tool (new toy:))!
Click to expand...

Brings back memories of the time when I still had two braincells to rub together. I started using Cadstar when it was still called 'Racal-Rediac something-or-other' in the mid 80s. Incidentally speaking of signal integrity modelling, there used to be a package advertised in E&WW which would take a PCB track layout and turn it into a schematic of inductors, caps and suchlike. Do you know if anything like that is still available (hobbyist pricing, not Cadstar-level I hasten to add!)?
 
abraxalito

abraxalito

Member
Joined 2007
wakibaki said:
You shouldn't take the stuff said on DiyHifi too seriously.
Click to expand...

I don't take anything I read anywhere too seriously waki, including what you write here :D

Those guys are flakes. Bybee customers. Just because somebody can cobble together a few bits of digital technology into a DAC does not mean they are the full shilling.If this guy's DAC sounds different with different USB cables, then:

1. He should know the reason why. Isn't that the whole point of an asynch USB DAC? If it was my design I wouldn't rest till I'd figured it out. An unexplained phenomenon is a discovery waiting to be made.
Click to expand...

Can't disagree here. I was wondering why he didn't know too. But his lack of understanding is my market opportunity. Remember Ricardo's dictum? Profits are made from differential stupidity.

2. It's a crap design in the first place
Click to expand...

If by 'crap' you mean it has room for improvement then yes, I agree.

I found the thread you refer to, it's a complete furball of people hacking at each other over phenomena which are almost certainly imagined and are anyway of no concern to anybody other than the mentally ill. They've lost track of what is reality vs. superstition:- 'Everyone knows that things that SHOULDN'T affect a CD player will' - everyone damn well does not know any such thing, and when you start from that basis it's only a short step to being abducted and subjected to the anal probe. Hardly surprising tho' in the foetid atmosphere of audio appreciation, where an infinity of veils still remain to obscure the vision of the true acolyte.
Click to expand...

Well there is certainly that stuff there, just as there is here from time to time. But I've learned to filter all the ego-stoking and stroking entertainment out and cut through to what really matters. If you find no useful nuggets down amongst the dross I'd say that's your loss :D
 
M

marce

Member
Joined 2007
The days of pen plotters, 24 pin dot matrix colour printers, photoplots sent in the post, and a max clock of about 20MHz. Never imagined doing 100+ MHz DDR ram interfaces or 0201 chip capacitors. Remember Redcad, we started using it a few years after Redboard.
I think being a brought up a yorkshireman the phrase is: "ee when I was a lad:)
On a more on topic note, when I get the hang of the new software it'll be interesting to start a new thread and investigate more signal integrity issues.
Have Fun
Marc
 
J

jkeny

Banned
Joined 2007
Guys,
There is a brilliant piece of seminal work that JoesphK is doing over on DiyHiFi.org which is repeating the Dunn & Hawksford experiments & confirming their findings about jitter & it's effects. Why I refer to it here is that it shows that RF effects distortion folds back to the audio band causing audible distortion, particularly in the sensitive 2-4KHz frequency range. (BTW, I was looking for a critique of the audibility threshold of 2nS of jitter & I forgot that the critique is contained in that Dunn & Hawskford paper - read that 1992 AES paper & you will see that pS of jitter is considered the audible threshold :))

Anyway, back to JosephK's up-to-date confirmation of Dunn & Hawksford's premise - he is using real world audio devices (EMU 1616M card & a reclocked CDRom as well as a Hiface device bot stock & modified by him) & taking measurements using a modern, sophisticated piece of equipment - a Lecroy WaveRunner 204Mxi-A (to be joined by a WavePro 735Zi)

It is well worth reading the thread for the text alone (as a non-member you won't be able to see the plots). The text is very enlightening & shows also his journey of discovery from getting to know the Lecroy & what it can do, to measuring real deterministic jitter spectra on real world devices but at RF frequencies & finally to being able to see the folding-back of these RF freqs into the audio band (visible when he used an even accurate scale of 5psec / div). Just to anticipate the nay-sayers :) here's his final quote (& I hope he doesn't mind me posting it here:
For the end: I'm already hearing somebody saying: boh, what the heck do you want with "only" 6-8psec jitter peaks?

Well, I do think that I can constantly, safely distinguish the two versions of hiface at my disposition.
The unmodded one is showing ~10-12psec jitter, in the highest peak. And it is NOT correlated jitter, much more benign.. And that is all the difference between them.

Ciao, George
Click to expand...

Again, I do think I had seen some simple things confirmed: I can repeat the Hawksford test, and see the same thing again. For me it also shows the "validity" of my test setup.
Also, consider that those simulations/ measurements were done at 1nsec jitter level, in his article!
I'm repeating it at a much lower level, and the phenomenon still valid and functional, with a real device, much better performing than the original setup was.

For me the "jitter foldback" effect was unknown up to know.
But, it would reinforce a good bit of my real life experience: that jitter is easily mucking up also the low frequency range, not only at the upper end.
This small stupid "violins" test suggests that a hard, "brrigth" effect can be produced at midband, while playing a range of only high frequencies. That is quite like the "digititis" what I know oh so well..

Then, about the missing modulation with music: I think it is a moving target. How is the dynamics are evolving in the music? Arches are always relatively low dense spray of high frequency excitation - and the response can be ~ relatively high jitter products in the low-midband!
Also, what if we were looking only at those "slices" of the audio signal, which are in the vicinity of zero crossing. I could imagine, that in that restricted sample range the temporary jitter spectra could show higher jitter levels even with real music..

Ciao, George
Click to expand...
 
Last edited:
J

jkeny

Banned
Joined 2007
Maybe this is worth a whole thread to itself as the information contained in his posts & measurements are startling if they are verified & repeatable! He states that the only difference between his stock & modified Hiface (his mods) are this 10-12psec of RANDOM jitter & not correlated jitter. He states & I concur (as do hundreds of others) that this is an easily audible difference between stock & modified. Startling stuff
 
T

Telstar

Member
Joined 2007
Similar findings are those of Gordon Rankin, according to whom around 5ps are audible.
I think that the limits of studies such as Hawksword are to limit The bandwith to 22khz and not to use higher resolution material. The 8ps estimated figure for 20bit data should be a hint :) For 24 bit it is 2ps IIRC.
 
J

jkeny

Banned
Joined 2007
Telstar said:
Similar findings are those of Gordon Rankin, according to whom around 5ps are audible.
I think that the limits of studies such as Hawksword are to limit The bandwith to 22khz and not to use higher resolution material. The 8ps estimated figure for 20bit data should be a hint :) For 24 bit it is 2ps IIRC.
Click to expand...

Thanks Telstar
Yes this quote from the Dunn & Hawksford paper echoes what you say
A better estimate of the audible jitter threshold can be obtained by examining the jitter error sequence and assuming that it will be inaudible if below the level of quantisation noise present in the system for any possible excitation frequency. Lidbetter [20] thus arrives at a value of 120 ps for a 16 bit 100% sample DAC and an incredibly low 8 ps for 20 bit system. Shelton [21], Fourre [22], Harris [8], van Willenswaard [7] and the recommendations embodied in AESll-1991 [23] ail quote similar values.
Click to expand...

Anybody who quotes 20ns or the 35ns jitter audibility threshold should really update their knowledge rather than quoting from the 1974 paper that this 35ns comes from, especially if they say they design high-speed devices & are working in the area :)

What I find startling is that I always assumed that random gaussian jitter was simply audible as white noise & therefore possibly not really of much consequence. The measurements of George (JosephK) & his stated audible differences (which I & hundreds of others) agree with, are more than simple background hiss - they fundamentally effect the presentation of the music!
 
Last edited:
J

jkeny

Banned
Joined 2007
I don't know if the above posts should be split off into a separate thread as I believe it is seminal work & pushing the envelope which deserves more prominence than it being buried in a thread but anyway, I wanted to follow it up with a quote which might tie into the suspicion that 10-12ps of jitter could not be of audible significance. Now I know what I'm about to quote is going to be dismissed by some who are going to claim that it comes from a manufacturer who is selling a product but first have a look at the credentials of this designer/manufacturer & secondly look at the products design - on both counts I don't think you will find a lack of engineering! BTW, look at the common mode chokes on the digital inputs - RF eliminators!
Review: Anedio D1 DAC - my new reference DAC - Head-Fi.org Community
From the owner of Anedio?
“One thing that I've learned from my feeble attempts to correlate listening with measurements is that the presence of low-level non-harmonic tones seems to make audible differences. Some of the commercial DACs perform well in the traditional THD+N measurement, but when you examine their high-resolution FFT plots, you would notice a lot of non-harmonic tones. These tones may be very low in amplitude, below -130 dB or even -135 dB, so they would appear indistinguishable from random noise or quantization noise in low-resolution FFT plots shown in audio magazines. But these tones would be revealed in high-resolution FFT plots (1024K point FFT) if the ADC of the test equipment is sufficiently clean. Some people might wonder about the jitter spectrum of the Anedio D1 DAC shown on the web site, and ask what the point is of getting these non-harmonic tones down to -150 dB. Well, I thought it would make no difference in sonic quality, but sometimes it does. This still amazes me because I don't have anything like the golden ear. But in other cases, it's very difficult to hear any difference. There is still much more to learn about the psychoacoustics of hearing.”
Click to expand...

Even if you are suspicious of the origin of this quote it is worth pondering!
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom