Apk Perfect Player

[Shantelle Wenske]

Isaw a interesting discussion elsewhere and thought I would pose the question here. I know a lot of the software manufactures visit this sight and there are a lot of knowledgeable people here. The question is,

Bitperfect means that all 1 and 0 are exactly the same arriving at the dac.It doesnt matter what player is used, if it is bitperfect, the dac should be receiving exactly the same data from all of them. They should sound the same.

Some say it is from jitter. If a reclocking devise or a dac that stores in memory and then reclocks it, would it not eliminate most of the jitter. At least it should take away any timing difference. While jitter can cause some harshness or a lack of clarity, can it change the tone making the highs of one player more pronounced and other players to have a richer deeper tone? All of these tone and clarity differences while having bit perfect output doesnt make sense.

I hear the differences and feel that it is real that these players do have different sound signatures. Is it because the players are altering the data and they arent bitperfect or is there other variables that can affect a bitperfect output?

This is THE big unanswered question for the computer audiophile. Theory tells us it has ONLY to do with the bits and the timing of them. Given the different player are all bit-perfect, is it possible a difference in jitter results in the different sonic signatures? Perhaps the algorithms in some players tend more towards precisely timed sampling, some a bit more sloppy in the timing (warm sounding?), etc. If this is not the case then there have to be additional factors that we dont understand or are not aware of.

What is needed here is an in-depth academic study on the subject. A detailed analysis of the output of different players, in what ways the bit streams differ, measuring the obvious variables of bit values and timing and looking at any other potential differences. Not sure what measuring equipment is available that can realize this.

Bottom line is that this is currently not understood, I am not sure the programmers of the various software really know why their software is sounding better (just listen to Jon Reichbach's vague answer to this exact question in Chris's podcast-see thread linked above). How good would it be though if ALL players outputted the EXACT same bit stream and we can rely purely on real audio gear to 'tune' the sound; the DAC, amplification and speakers. I think this is the biggest limitation in computer based audio, however I guess this has never really been the case though as different CD-transports have different signatures.

Ok, it will not be new all over, and maybe it's a somewhat childish approach, but I'll try it like this, this time. I hope you nor anyone is offended, but if so I guess it's up to them to decide on *why* :

I am not sure the programmers of the various software really know why their software is sounding better (just listen to Jon Reichbach's vague answer to this exact question in Chris's podcast-see thread linked above).

I don't know of any single piece of playback software that explicitly works on - and/or is able to make better sound just by means of, well, that software. I'm not a JR and have a forum full of it. It's all continuous explicit development and it exists. Now look for such a development elsewhere. Therefore it is coincidence, and therefore people complain at a next version and have no clues (never answers either). If anyone thinks this is not true, please point it out.

I am not sure this would be good as such. I mean, now at least you have the opportunity to make your DAC sound better than ... well, it originally is ? this is pretentious, but I think it can be approached like this (DACs never bing correct anyway).

But for you information and honesty, I ever began developing my DAC to just attack that : the difference stupid software can make, therewith eliminating my own (commercial) software. I know how it works and what it does. STILL my little project (of ages) failed. Conclusion : I don't know what I am doing or what is happening afterall.

By now (and again being honest), I can't understand myself anymore HOW the enormous differences I am able to create in one single player are always be "bit perfect". I know you are not going to listen, and you don't need to (nor do I want it). But *if* you'd do it, you'd drop dead because of not understanding. Right now you're only half dead. It may be better.

Lastly, before you think all the credit should go to me, It really should not. Oh, a little maybe for starting all this (some 5 years back). But the remainder is really from all those users giving feedback. Too few bass this time ? ok, I'll add some. That this may take 6 months is something else, because it is all a most indirect way of working. It's not DSP ...

Starts a few posts earlier. Maybe don't read it, because it is not important at all. What is important maybe, is that it's just "us" creating hypes around software to next blame the poor developer he can't decently explain why his software sounds good. It's all in that thread, and I guess nothing changed since.

For fun : if in any place software can't change sound, it is in Hydrogen. Try tell there that it does whithout scientifical studies and proof and everything, and you're banned right away (forum rules). Too bad Foobar also sounds different. Must be the one and only best I suppose ?

My personal conclusion (and take-away from your post Peter) is that the different "coincidences" that are happening in the different software players result in different jitter spectrums and this drives the different sonic signatures.

I mean it's staring us in the face, there are 2 variables, the bits are identical so this is eliminated. The jitter is not fully understood or ever really measured and an EXTREMELY sensitive quality of the audio signal (affected by things like wired vs wireless internet, hard drive access vs solid state, the enabling/disabling of bluetooth, type of digital cable used, and finally implementation of the playback software). People claim to hear improvements when these aspects are changed, I believe them and believe its due to their effect on jitter in the signal.

As I mentioned, the proof would be a detailed analysis of the bit-streams from different players. The bit values would be the same, but the JITTER would differ. Conclusive data on this would validate this theory, however by process of elimination this is what I personally have concluded.

1 is way too complex for me to answer because I'm sure there are a million potential reasons both internal to the "box" and external to it. But the answer to 2 is easier I think. If you hold the equivalent of a electrical magnitude in a sample and hold circuit for a longer or shorter duration than it is supposed to be held you change the waveform. Since these changes will occur in the 10s of microsecond range for a 44.1 kHz sample rate it will affect the naturalness of the higher harmonics of instruments. In otherwords they will sound like crap.

Edit: it just occurred to me that subroutines called by a processor that correlate in time to the sample rate that is occurring would probably have the most effect on jitter. That is just an intuitive guess, nothing more.

(1) The system playing back the music must be sensitive enough to resolve the difference. Digitally stored music as a general rule, sounds colder, harsher, or slightly more mechanical to some folks, as compared to the same music played from a high quality analog source. The better the sysem doing the playback, the more apparent this difference will be.

That isn't a hard and fast fact by the way, other folks say digital music sounds more precise, more detailed, and clearer than music from an analog source. There ae advantages and tradeoffs all over the place.

(2) Bit perfect, as used by some audio folks, is a somewhat tricky term. You must remember all our digital data is currently transmitted via analog transmissions! This is true even when you read data from a hrddisk, cd-rom, or even directly from memory in your computer. There is quite a bit of lattitude in what you can do to that analog signal and still call the result bit perfect. Quite a bit indeed, and most of the things you can do result in a perceived change in the final sound that is output.

People like Peter get very clever in manipulating the digital data, and thus claim to have improved the sound without changing the data. This is true, as they did not change the mening of the data (1 or zero) but they sure transformed it a few times.

Jitter is the great boogeyman of the audiophile world, but you can bet jitter is well handled by folks making top quality DACs these days. Programs like Amarra, Pure Music, XXHigh end and so forth are making digital music sound more appealing by using algorithms that modify how the analog transports perform, for the most part.

Buying gear that is most sensitive to jitter in order to hear differences that shouldn't be there would be kind of dumb thing to do... Arranging cabling and grounding so that there are huge ground currents going through (digital) cables will surely also help maximizing jitter in most cases. And then trying to tame this by tweaking the software... Ehh...

Would be sort of a dream that one day the income would cover the hardware costs, or even better the time spent... Catching up with the past costs, would I dare to dream? Well, I wasn't on the list so maybe I don't count.

MacMini 8Gb OSX > Pure Music / Bitperfect / Amarra / iTunes > Synology DS215J NAS > Schiit Wyrd > Stello U3 > Naim Uniti Atom, Harbeth P3ESR. Meier Corda Arietta Headphone Amp > Sennhieser HD650 Phones (Cardas rewire). Isol-8 Powerline Axis. Isotek GII Orion Power Conditioner. Cardas Clear USB Cable. Tellurium Q Black Speaker Cable. All other cables by Mark Grant.

"the most stringent tolerances for jitter suggested in this report are more than an order greater than those recommended elsewhere for vision circuits; it seems safe to assume therefore that, where tolerance limits are appropriate for the transmission of vision signals, they should also be adequate for transmission of sound signals."

3a8082e126