Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › Squeezebox 3 compares to which CDP/DAC?
New Posts  All Forums:Forum Nav:

Squeezebox 3 compares to which CDP/DAC? - Page 4

post #46 of 63
Quote:
Originally Posted by koob View Post
No peer-accepted electrical engineering authority has ever claimed that jitter below 200ps was audible and even this is conservative by several orders of magnitude. Through mathematical analysis we can prove that jitter below 100ps can't have any effect on 16bit/44.1KHz samples within the commonly accepted physical realities of our universe. No one that I know of has ever ABXed jitter in digital audio when the jitter was on the order of single-digit nanoseconds. Most commonly acceptable thresholds for jitter detection by human observers are on the orders of 10s to 100s of nanoseconds (not counting John Atkinson of course, who can detect single picosecond jitter without even connecting the transport to anything). Note that 10s to 100s of nanoseconds of jitter is 3 to 4 orders of magnitude worse than what the SB3 is producing.
Sorry to disagree with a fellow jitter-skeptic but your post has some factual inaccuracies. Firstly several peer reviewed journal papers have modelled jitter as audible in the 10 - 20 ps range, Dunn did this, of course his model relied on a 20K signal played at 120db above the threshold of hearing i.e ear-meltingly loud, nevertheless it was modelled as audible, even if not empirically tested, Dunn later cast doubt on his figures in the light of the Benjamin and Gannon paper.

Secondly according to Pohlmann jitter can go to 200ps before it degrades 16 bit audio to 15 bit resolution.

As for empirical audibility , this seems to depend on the type of jitter, the Ashihara paper (I have cited elsewhere) used random jitter and found this inaudible until 500ns. The Benjamin and Gannon paper used deterministic jitter which caused definite sidebands, here the worst case for audibility in music was 30ns , 10ns with a high frequency pure single tone.

Is jitter normally a problem, I remain skeptical, the only evidence for ps jitter audibility is poorly controlled and anecdotal tests.
post #47 of 63
Man! I have learned a lot from you guys. Got to love the references to peer reviewed Journals.
post #48 of 63
All this discussion of EE journals is miles over my 30 yr old poli sci degree. Let me ask if a different part of the SB3's job could account for the PS making a different sound. When I was researching buying a new DAC, (I ended up with the CI VDA-2), the threads here and the reviews at CI's website all recommended getting the optional outboard power supply with it. The consensus was the DAC sounded better with the upgraded PS. Could the upgrade to the SB's PS matter in the processing of the the data stream from FLAC/MP3/OGG to what I'm guessing is a wave format it would send to the DAC? Whether it goes to its own DAC or to an outboard one, if the PS is effecting that processing, an improvement, if any, should be equally noticeable. I use my SB with its included wall wart btw. No upgrade here.

Baz
post #49 of 63
Have to say that I find all the debate about whether a PSU makes a difference or not to be quite amusing.
I recently bought an SB+, which is a heavily modified SB3 with a seperate box linear PSU. I can happily state that even the burn in period for the PSU makes a difference to the sound quality of the digital output from the SB+.

That should nicely add to the debate. Before someone asks, nope wasn't using double blind testing. Does that mean my above comment should be automatically ignored, or can I still voice an opinion?
post #50 of 63
Quote:
Originally Posted by Mr_Sukebe View Post
Have to say that I find all the debate about whether a PSU makes a difference or not to be quite amusing.
I recently bought an SB+, which is a heavily modified SB3 with a seperate box linear PSU. I can happily state that even the burn in period for the PSU makes a difference to the sound quality of the digital output from the SB+.

That should nicely add to the debate. Before someone asks, nope wasn't using double blind testing. Does that mean my above comment should be automatically ignored, or can I still voice an opinion?
Silly question, but I presume you weren't burning in the SB3+ at the same time as the psu? If so, how do you know the changes are due to the psu..?
post #51 of 63
Quote:
Originally Posted by pedalhead View Post
Silly question, but I presume you weren't burning in the SB3+ at the same time as the psu? If so, how do you know the changes are due to the psu..?
Good question, with an easy answer. The designer popped around last Sunday with an older copy of the SB+ and swapped over the two boxes. Couldn't hear any difference between the control boxes, but changing the PSUs had quite a profound effect, despite them apparently consisting of the same components. The only difference being that the older one has had 3 months of running and burn in.
post #52 of 63
As to the stock SB3 power supply, there is a theory that the RF giving off by the cheap switching supply could be causing problems for any electrical device in the system (DACS, EQs, Pre-amps, etc.). It's a well known fact that the stock supply gives off a ton of RF noise, just turn on an AM tuner any where near it and you'll hear all the crap coming out of it. One way to test this theory is to try a new PS and if you can hear a difference in sound quality try listening with an without the stock PS plugged into the wall at the same time (but with the new supply hooked into the SB).
post #53 of 63
Quote:
Originally Posted by hciman77 View Post
Sorry to disagree with a fellow jitter-skeptic but your post has some factual inaccuracies. Firstly several peer reviewed journal papers have modelled jitter as audible in the 10 - 20 ps range, Dunn did this, of course his model relied on a 20K signal played at 120db above the threshold of hearing i.e ear-meltingly loud, nevertheless it was modelled as audible, even if not empirically tested, Dunn later cast doubt on his figures in the light of the Benjamin and Gannon paper. .
Agreed, though Dunn's modelling doesn't constrain to redbook CD standard audio. Granted I didn't state that constraint either, but given that 100% of my collection is in that format, that's where my headspace is.

Quote:
Originally Posted by hciman77 View Post
Secondly according to Pohlmann jitter can go to 200ps before it degrades 16 bit audio to 15 bit resolution.
If Pohlmann states that jitter above 200ps _can_ affect 16 bit audio and I made the assertion that jitter below 100ps _can't_ affect 16 bit audio, those 2 statements are tautologically compatible, so this doesn't represent an inaccuracy. If we make X the threshold expressed in picoseconds at which jitter has an effect, Pholmann states X < 200 and I state that X > 100, so we have 100 < X < 200.

Quote:
Originally Posted by hciman77 View Post
As for empirical audibility , this seems to depend on the type of jitter, the Ashihara paper (I have cited elsewhere) used random jitter and found this inaudible until 500ns. The Benjamin and Gannon paper used deterministic jitter which caused definite sidebands, here the worst case for audibility in music was 30ns , 10ns with a high frequency pure single tone.
The Benjamin and Gannon study, in addition to using the worst case scenario, also allowed the participants to decide when the jitter was no longer having an effect. Ashihara's study forced each participant to ABX the effect of jitter at a given level before lowering the jitter to the next level. As you say, some particpants were able to ABX 500ns, no one was able to ABX 250ns. As you also point out, the way jitter was added differed in the 2 studies as well.

But again, even if one takes the Benjamin & Gannon study as the low water mark of demonstrated ability to discern jitter, we are still talking several orders of magnitude beyond the SB3's worst case performance.
post #54 of 63
Quote:
Originally Posted by koob View Post
Agreed, though Dunn's modelling doesn't constrain to redbook CD standard audio. Granted I didn't state that constraint either, but given that 100% of my collection is in that format, that's where my headspace is.



If Pohlmann states that jitter above 200ps _can_ affect 16 bit audio and I made the assertion that jitter below 100ps _can't_ affect 16 bit audio, those 2 statements are tautologically compatible, so this doesn't represent an inaccuracy. If we make X the threshold expressed in picoseconds at which jitter has an effect, Pholmann states X < 200 and I state that X > 100, so we have 100 < X < 200.



The Benjamin and Gannon study, in addition to using the worst case scenario, also allowed the participants to decide when the jitter was no longer having an effect. Ashihara's study forced each participant to ABX the effect of jitter at a given level before lowering the jitter to the next level. As you say, some particpants were able to ABX 500ns, no one was able to ABX 250ns. As you also point out, the way jitter was added differed in the 2 studies as well.

But again, even if one takes the Benjamin & Gannon study as the low water mark of demonstrated ability to discern jitter, we are still talking several orders of magnitude beyond the SB3's worst case performance.

Was this random jitter or was it centered around the human speach frequencies? You can run pinknoise over your favorite CD at 5 db's and you won't even hear it but run a 5db sine wave at 500 hz and you sure will.
post #55 of 63
Quote:
Originally Posted by koob View Post
The discussion got threadjacked when the suggestion was made that a PSU upgrade would help the digital coax out: the OP asked for comments on the SB3 as a DAC.

My response was simply a tongue-in-cheek attempt at pointing out the absurdity of the PSU suggestion. The PSU has no impact on the _timing_ of the stream of bits being fed out of the SB3. If the PSU had enough impact on the _quality of signal_ from the digital coax to the point that it impacted the downstream DAC, this would manifest itself as a garbled stream, not as a subtle difference requiring listening sessions to discern.

Sorry for the thread jack. I'm sorry you don't understand the PSU effect. I can certainly hear it, easily on my system. I have yet to read about anyone trying the linear PSU and not hear a different from the standard switching walwort.

Sorry again about the thread jack.
post #56 of 63
Quote:
Originally Posted by utep10 View Post
Sorry for the thread jack. I'm sorry you don't understand the PSU effect. I can certainly hear it, easily on my system. I have yet to read about anyone trying the linear PSU and not hear a different from the standard switching walwort.

Sorry again about the thread jack.
There are a few folks on Slim Devices webforum that have NOT heard a difference when changing from the stock to the linear PSU.
post #57 of 63
Quote:
Originally Posted by regal View Post
Was this random jitter or was it centered around the human speach frequencies? You can run pinknoise over your favorite CD at 5 db's and you won't even hear it but run a 5db sine wave at 500 hz and you sure will.
Benjamin and Gannon used Jitter frequencies between 1530 Hz and 1850 Hz. These frequencies were chosen as they were thought to cause to most easily detectable distortion i.e worst case - the Ashihara et al study used random jitter which seems to mostly make the signal generally noisier.

As you imply general noise is pretty hard to detect. Given the magnitude of the jitter Ashihara et al were able to add without it being detectable there is an implication that these subjects (this is me being circumspect) showed an inability to detect a lot of extra noise. We know that jitter above 121ps or 200ps (depending on authority cited) degrades resolution of 16 bit systems to 15 bits. The implication being that these subjects in their tests at least could not tell the difference between 16 bit resolution and 15 bits or possibly even less.
post #58 of 63
so back to the original post....SB3 and it's DAC...
post #59 of 63
I'm going to add my 2c worth on this very intriguing subject...

I bought my first Squeezebox this week for $230 and received it yesterday, hooked it up to my main stereo. I got the Squeezebox for a second bedroom system that I am working on and so far only have the SB, amp and cables for it. I quickly connected it to my main system via its coaxial digital output to my Hawk Audio MP-DAC MkIII, configured the wireless connection, and queued up a wonderful Shoutcast station streaming very high quality Baroque music.

Main system is compromised of Audio Note CDT-2 transport, Hawk Audio DAC, Bent Audio TVC preamp, Nosvalves modded Dynaco ST-70 with KT-77 tubes, Bob Brines' FT-1600 Fostex-based MLTL speakers.

The stream sounded awesome! About the same level of quality as a decent classical CD playing via the AN CDT. So already the SB has exceeded my expectations.

However, this is due to the fact that it's output is connected to my DAC. And yes, I am still using the SB's stock switching supply. I have the $6 linear supply shipping to me from Hosfelt. So now since this SB is moving to my bedroom system, which I had originally planned on not including an external DAC, I've come to re-evaluate that decision.

The DAC is an extremely intriguing device and I have previously owned the modded Art DI/O. Before, my system went like this: NAD 521i CDP, Art DI/O, passive resistive volume control, Bryston 3B-ST, and the same FT-1600 speakers. This system sounds like JUNK compared to my current system, where I've changed everything BUT the speakers! I'm now also using Audio Art IC's which have made a huge improvement to my system.

What I'm getting at here is that if I had plugged this new SB into my previous incarnation system, I seriously doubt that I would hear a huge difference between the SB's DAC and the Art DI/O DAC, or the types of power supplies used. The components downstream make a HUGE difference and based on these quoted "tests," I fail to see how they are able to quantify so many other unaccounted variables. You'd think that my older main system wasn't that bad, with the highly regarded Art DI/O and the old stand-by Bryston amp. Well I can assure you that the Hawk Audio DAC is worlds better than the DI/O. Where in its circuitry are the real differences made? Who knows?

Yes our mathematics and physics can disprove that there are any discernible differences to be made between cables, DAC's, or power supplies, but our ears are always telling us otherwise!
post #60 of 63
Quote:
Originally Posted by n_maher View Post
You have to consider that everything starts with the power. Think about it like drinking water - sure you can drink just about anything that comes out of your tap but run it through a filter and while it might not look a whole lot different it clearly tastes better. At least that's how it's been explained to me.
yeah, to me too - but when we consider it's not tap water but beer, say, or wine we're talking here, then some audio tweaks can certainly seem the equivalent of voodoo corkscrews and magnetic bottleholders: a matter of superstition.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › Squeezebox 3 compares to which CDP/DAC?