Head-Fi.org › Forums › Equipment Forums › Computer Audio › Audiophilleo 1 and 2 USB to S/PDIF transport
New Posts  All Forums:Forum Nav:

Audiophilleo 1 and 2 USB to S/PDIF transport - Page 58

post #856 of 1173

Here's an interesting though - about how to test galvanic isolation (qualitative, not quantitative).

 

Turn the Audiophilleo, your DAC, and your computer on - select the DAC as your source and turn the volume up pretty loud.

DON'T PLAY ANYTHING.

 

Now, run some stuff on your computer, and listen to see if any funny noises are there to be heard

(if the isolation is flawed, then you would hear noises leaking through from the computer).

 

If your DAC mutes when nothing's playing, you could make up a WAV file that contained a very low level continuous tone;

now play that file and listen for other noises mixed in with it.

 

Assuming you hear something like funny noises that way,

connect up the Vaunix and see if they go away.......

 

 

 

Quote:
Originally Posted by Trogdor View Post


How? Right now the problem I see with all of this is that if the Vaunix is improving the AP2 wPP there is either an issue with Phillip's design or some other phenomenon occurring.
I'd like to understand more on WHY there is any audible differences.
So far no one has given me a technical reason that makes sense (at least to me!). I can only suspect that its placebo (no offense) without listening to it myself (anybody got a spare Vaunix laying around for me to borrow for a week?).
I understand that AP is now looking into it to see if there is indeed an issue, so I await Phillip's analysis.
Floating point arithmetic would be my guess (perhaps 32 vs 64-bit). Same comment goes to DaveBSC.
In the end Currawong, most of it is placebo. There were some USB audio bugs in 10.6 that caused issues with various devices (clicking noise mainly). AFAIK, Lion+, the players should sound the same.
Setting up a DBT with a friend or significant others is easy if you truly believe you can tell the difference between players. Try it! Its a lot of fun! :-)
That's a nice idea.
post #857 of 1173
Quote:
Originally Posted by kLevkoff View Post


The first issue is that many players either are not bit-prefect themselves, or cause the operating system to not be bit-perfect. Once you have the player changing bits then you would expect each one to change the bits in different ways, and so sound different. Likewise, the player may interact with the O/S to cause (or allow) changes to the bit-stream there. Some of the drivers themselves may not be bit-perfect. Luckily, if you have an AP1, you can verify once and for all that you're getting bit-perfect output.

That's a good point but for the most part, all the usual suspects on OSX are bit perfect including iTunes.
Quote:
The second issue is that every player (and every other program) interacts with the operating system itself, which is bound to affect the timing - which will end up causing or aggravating jitter. I don't think that, in this context, the jitter issue is as straightforward as we tend to believe. Assuming that each sample arrives approximately when it should (let's say "before the next one is due to arrive"), then removing jitter is simply a matter of re-clocking. But, what if the player (or the O/S) causes enough delay that we get several samples behind?

You will have drops not "audible differences."
Quote:
The third issue is power supply noise. The actual operation of the processor in the computer adds noise to the power and ground lines which is related to the power the computer is using - and what it is doing. If you connect a bad USB DAC to your computer, you can usually actually "hear" what program is running - the noise changes when you switch programs, or when a given program is "working harder".

A context switch should not be audible. Do you have any idea how many times a second a context switch happens? So this is a very bad theory! biggrin.gif
Quote:
Another (related) issue is that there may well be noise on the USB data lines themselves. (The data is ones and zeroes, but minor variations in those levels caused by noise, even though they're ignored by the digital circuitry, may "bleed through" and end up finding their way into the DAC and interfering with things.) With a device like the AP, which has galvanic isolation, both the ground and +5 supply are NOT passed through to the DAC, nor allowed to affect the output data, and the data is totally "rebuilt" by the output USB sender (also running off the isolated power), so this should also be eliminated. The real problem with "galvanic isolation" is that, while it's pretty simple to isolate the DC path, it's much more difficult to ensure that absolutely NO high-frequency noise can leak through. (At a dead minimum, with a single-box solution, you have the possibility that a minute bit of noise will be radiated by the wires coming in, and picked up by the wire going out.... therefore, it's POSSIBLE that reducing the noise going in will make a minute difference. Considering how well designed the AP is, though, I wouldn't be all that worried.
After all, at some level, having the computer within a few hundred feet of the DAC risks the possibility that noise being radiated through the air by the computer will be picked up by the DAC. Perfect isolation isn't even theoretically possible, so we'll have to settle for "inaudible" smily_headphones1.gif )

This was in essence what Phillip told me.

Btw the Vaunix is using a switcher which will be very hard to filter out 100%. If anything, using the Vaunix has a chance of making it worse than better.

Still confuzzled...
post #858 of 1173
Quote:
Originally Posted by kLevkoff View Post

Here's an interesting though - about how to test galvanic isolation (qualitative, not quantitative).

Turn the Audiophilleo, your DAC, and your computer on - select the DAC as your source and turn the volume up pretty loud.
DON'T PLAY ANYTHING.

Now, run some stuff on your computer, and listen to see if any funny noises are there to be heard
(if the isolation is flawed, then you would hear noises leaking through from the computer).

I don't have the Vaunix but this is indeed a good test. Silence here.
post #859 of 1173

If there's a design flaw with the modded AP+PP combo, I want it fixed so that I can get my $200 for the Vaunix back; the difference with the Vaunix while incremental, is not subtle. If you have the AP+PP and the Vaunix, it's easy to tell the difference; if you don't have the Vaunix, it's also easy to tell the difference--just buy the Vaunix. Then report back. 

post #860 of 1173
Quote:
Originally Posted by Trogdor View Post

That's a good point but for the most part, all the usual suspects on OSX are bit perfect including iTunes.
You will have drops not "audible differences."

 

Jitter doesn't cause a drop-out, not unless the signal is so bad that the DAC loses its lock. An incorrectly set computer buffer will cause skips and drops depending on the hardware. Here's a question. Why does JPlay's hibernation mode sound better than the normal, interactive mode? It's the same software, same configuration, same RAM playback in either case. The difference is in OS interaction. In hibernation mode everything is shut down except for the core of the OS. You can disconnect the hard drive and playback will continue, there is zero disk I/O. I have to assume that all of those little tasks and read/writes that Windows performs in the background take their toll on the correct timing of the digital output stream.

post #861 of 1173
Quote:
Originally Posted by DaveBSC View Post

Jitter doesn't cause a drop-out, not unless the signal is so bad that the DAC loses its lock. An incorrectly set computer buffer will cause skips and drops depending on the hardware.

He wasn't really talking about signal jitter due to common mode or other insidious causes. My response was to address the following question:

"But, what if the player (or the O/S) causes enough delay that we get several samples behind?"

Its called a drop and it will most certainly not be subtle.
Quote:
Here's a question. Why does JPlay's hibernation mode sound better than the normal, interactive mode? It's the same software, same configuration, same RAM playback in either case. The difference is in OS interaction. In hibernation mode everything is shut down except for the core of the OS. You can disconnect the hard drive and playback will continue, there is zero disk I/O. I have to assume that all of those little tasks and read/writes that Windows performs in the background take their toll on the correct timing of the digital output stream.

http://www.computeraudiophile.com/blogs/mitchco/jriver-vs-jplay-test-results-156

But I don't want to thread crap too much....back to the AP2/PP....need to figure out why there are any changes?!
post #862 of 1173

AP have a buffer ... and play music from the buffer. The timing in the PC is not important. The computer role is to ensure to fill the buffer to avoid it's become empty. When it's happen, you will hear a POP.

 

Make sure your jplay keep your bitperfect. cplay is not bitperfect ... in fact, the soundstage is larger ... you can enlarge more or less with a setting.

 

Play from ram, hd or whatever is not important with AP (buffer) ... but with soundcard it's matter (no buffer).

post #863 of 1173

Probably vaunix clean the signal from some of the noise of the usb ...

 

Not all USB is the same. Try different USB port, you may hear a difference ...

post #864 of 1173

My AP1 should come back with the PP upgrade tomorrow.  I intend to daisy chain (Vaunix -> PP -> AP1), rather than having both components connected to the Vaunix.  Anybody know a suitably short (6-8") A-to-B USB cable?  I'm having a hell of a hard time finding one for sale.

post #865 of 1173
Quote:
Originally Posted by kLevkoff View Post

 

There's a huge problem with your statement - actually two of them.

 

The first problem is that "expectation bias" will tend to incline us (myself included) to hear what we expect to hear - often leading us to hear things that aren't there. Worse, we may attribute things that are indeed present to the wrong causes. (Did DAC b actually have an ever-so-slightly cleaner high end, or was it just that the humidity was 5% lower when we listened to it?)

 

The second problem is that there are a huge number of products (and DIY options) out there - and very little opportunity to audition most of them in proper surroundings. Since we won't ever have an opportunity to hear even a minute fraction of them, we need some way to pre-screen the ones that are at least likely to be worthwhile. (And, yes, this often directly conflicts with the first problem.) As far as I'm concerned, the best screening method we have is to consider the scientific (engineering) justification of the product or technique.

 

In this particular case, the theory is that the AP is NOT insensitive to poor quality power (on the USB side).

If this is indeed true, then the Vaunix Hub, and several other alternatives, can be tried and would be expected

to produce similar results.

 

 

At least we can avoid wasting time on things that don't seem to have "legitimate justification" for our expecting them to work. (Is it possible that some dolt, whose scientific explanation is obviously pure drivel, somehow lucked onto something useful? Yes. Is it likely?)

 

In the case of the Vaunix Hub, I'm pretty sure that there's nothing magical about it in particular, so the engineering question becomes that of whether running the computer-powered-side of the AP off of cleaner power affects the output or not. If it really does, then there is some flaw in the design (since it theoretically should not). If so, then the Vaunix may help, but a better solution would probably be to identify and eliminate the weakness in the AP. If no such flaw exists, then we had best move on to other tweaks.......

 

The easiest way to find out if we're chasing our tails or not would be to cut up a USB wire and use it to supply clean power to the PP (on the computer side).

 

If USB bus power to the receiver is a problem it could be integrated in future purepower models meaning one less box is needed.  

 

I agree that it is useful to try to do tweaks that make sense from a scientific standpoint to avoid trips down the proverbial garden path, but sometimes something that seemed to look good in [someones] theory doesn't actually make an improvement because another factor has been overlooked, or vice-versa.

 

As for computer side influences on audio stream I don't think it is as simple as having dropouts or not - I can get dropout free audio from WASAPI or KS but both still sound slightly different, maybe less so than with my previous USB transport but still different.  I haven't blind tested it but this is what I have observed while playing around with software setting and the AP2/purepower.  Feel free to check if you observe something similar or not at your own leisure.tongue_smile.gif

 

I have found that keeping the AP2 buffer full is actually fairly tricky especially if you are using smaller software buffers.  Some streaming methods and players need larger buffer sizes than I was using with my previous transport, especially for 24/96 files.  

post #866 of 1173
Quote:
Originally Posted by Currawong View Post

The Vaunix hub resolved an issue with harshness in the treble when listening which I didn't like. I'm sold on power upgrades because all of my experiences with audio since I joined Head-Fi have proven the benefits to me. 

That being said, while I don't understand why it should make any difference,
I am very sensitive to the sound of instruments in jazz and classical so I notice, though now it is at the point of splitting hairs.
 

 

First sentence you spoke about highly resolved system, but next and most of the following was about "tonality"...Even the clocking statement that you do not understand why it makes a difference (I personally do not think it really does other than maybe what you heard=tonally).  Then the sensitivity-again, tonality, between different music players.

 

I personally do not have the same "sensitivity/tonality" issues as you do...well...I do, but when it gets as subtle as you are speaking of, which is the same way I hear my sound=everything is so subtle at this level, the tonality aspect really starts to go out the window and I have to know there really is NOTHING one can do about it besides equalizing the speaker/headphones a certain way to help give a little more tonal rightness OR just flat out buy a different speaker all-together that does sound "fundamentally" correct to you.

 

I do not have the kind of money to spend on Empirical stuff, but if there is anything computer based that sounds on a very high level, very refined, and that "tonal rightness" thing going on, it's the Off-Ramp.  You have as much money in your current setup as the most loaded Off-Ramp 4 is priced at right now.  And if you need more tonal rightness and production, I believe the OR5 adds just that bit more on these very discriminating levels of sound/tonality.


I do not like that I sound to be endorsing Empirical and this is not a thread to be mentioning it, but it's really the ONLY thing PC based that has had that truly "hi-end" level of tonal rightness, along with a very see-through signal pipeline that as a sum, sound correct.  Fact that I will not toss my money at it sucks because I'd love to have it, but for someone such as yourself that via unloading this setup OR just doing a practically free (only pay shipping) to demo the Off-Ramp 5 vs. this current setup, has that luxury to do so, I'd highly recommend trialing/demoing it.  I just had to mention this because as I say, because all that you said just made me think of why a lot of people prefer the Off-Ramp because they are ALWAYS mentioning the tonal correctness or "analog/organic" sound of it.  Now I would not use such words, but no product I can think of on the USB side has had so much "emphasis" regarding the tonal correctness of the device.

 

Hope this helps and nothing against the AP1/2 or trying to say one device is superior to another, but to toss this out as it really would be something to seriously consider in the least demoing  with respect to what is sounds like you are after sound wise.

 

Also wanted to say I always appreciate your comments and also Rdr's...others as well, but you two are very respectable people around here that I appreciate.  

post #867 of 1173
Thread Starter 
Trogdor, while I appreciate the limitations of being human, very often I notice differences in things when I expect there to be NO difference.

Though not the AP, I did take noise measurements of my ULN-2 while on bus power versus the included wall wart and the differences were very significant, even if it were just acting as a digital transport to another DAC.
post #868 of 1173
Quote:
Originally Posted by Currawong View Post

Trogdor, while I appreciate the limitations of being human, very often I notice differences in things when I expect there to be NO difference.

Currawong, that of course is being human... biggrin.gif

Its also one of the primary driving forces of the audiophile industry. Our innate urge to hear something different whether there is a difference or not (by no means am I calling you fallacious either, just stating a corollary to the conversation at hand).

So folks who have a Vaunix or AQVox and AP/PP, please post more impressions. I am curious if you hear any difference.
post #869 of 1173
I wonder if you guys can tell me your views on hOw big of a difference audiophiles USB cables
Make when connecting the pure power and audiophilio .
post #870 of 1173

very little as far as I can tell.  The AP2/ purepower I bought came with a Wireworld USB cable so if you get that package you could always compare yourself at that point.  Some cable companies offer trial periods so that may be another option.  I always hesitate to offer advice on cables as some people hear differences, some don't.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Computer Audio
Head-Fi.org › Forums › Equipment Forums › Computer Audio › Audiophilleo 1 and 2 USB to S/PDIF transport