Head-Fi.org › Forums › Equipment Forums › Computer Audio › Lilith Audio Player
New Posts  All Forums:Forum Nav:

Lilith Audio Player - Page 8

post #106 of 487
clearly, Linux should be way superior!

tbh, the first change will be that you'll have proper audio drivers made by skilled engineers....not the usual c-media/asus/creative crapola

and you can get direct access to hardware very easily...this is forbidden in the NT core AFAIK, and realtime isn't possible at all.

when they started BeOS, they were selling it as being a true realtime OS...I'm entirely willing to believe ppl who say that MacOSX/Linux sound better than windows, it makes perfect sense.

this said, uLilith+ASIO4ALL sounds amazing on XP....a far cry from everything else I've heard
post #107 of 487
The default linux scheduler is not "realtime". Nor does it need to be for audio. And Windows doesn't "throw data" anywhere, the soundcard accesses it via DMA. "Realtime" is entirely unnecessary for absolutely flawless audio playback on a PC. I suggest you spend some time with a book on psychosomatics.
post #108 of 487
hehehe, well linux can be made realtime if you compile the kernel to be so.

and yes, windows is always dependent on the system timer granularity and the tightness of the buffer sent to the soundcard.

I suggest you spend some time studying how audio works in computers: cMP² | CPlay / SoftwareInducedJitter
post #109 of 487
I'm not gonna join in the argument of what computer audio needs but I know one things for sure... The OS matters in the way the music sounds
post #110 of 487
That's why I wasted the necessary time to say "default" scheduler, and also why I pointed out that realtime is not required for perfect audio - the two points being related. But hey what would I know, I just write software for PCs and microcontrollers professionally.

By the way, if you go by what it says on that page, increasing the granularity of the scheduler would increase jitter due to more noise on the power rails thanks to the increased cpu switching noise. Some of the points made on that page are, if you have a really good imagination, plausible, but the "realtime"/small-buffer thing is misguided. It isn't the only thing either.
post #111 of 487
Quote:
Originally Posted by donunus View Post
I'm not gonna join in the argument of what computer audio needs but I know one things for sure... The OS matters in the way the music sounds
it's completely pointless anyway...see the 6 pages of threadcrap in this topic.

there's 2 sides:

-those that think that bit-perfect is exactly that..and that the OS/player/type of renderer does not matter, and that ppl who hear differences should stop smoking thai stick.

-those who hear differences between players/renderers/OS and think that non-believers have wooden ears and/or crappy gear

Amarra is said to be bit-perfect, yet it's also said to KILL the stock itunes....and there's zillions of testimonials like this one on the web: XXHighEnd Player Sounds!!!!! better than my foobar...
Quote:
The XX is the best player I've used in my system in terms of sound. Larger soundstage, everything fleshed out better/of a whole, low bass/bass/drums more impact better defined, everything highs on down more articulate.... [..]
My 11 year old daughter was brought in blind [..] she quickly responded there was a difference and she thought and chose XX as the better player.
you even told me in the past that you were also hearing differences between players...and sure as hell uLilith + ASIO4ALL sounds completely different from Reclock in KS...why is that so? I don't care tbh
post #112 of 487
I'm sure there is a lot more to it... The question is when will we ever know we are getting bit perfect audio. If OSa sounds different from OSb when both using this player with wasapi for example then how come we still get different sound quality from OS to OS? hmm or is it maybe like I said earlier where the cleaner the OS code, the variation between audio players for example becomes smaller and smaller when using the same audio output scheme.
post #113 of 487
Quote:
Originally Posted by b0dhi View Post
That's why I wasted the necessary time to say "default" scheduler, and also why I pointed out that realtime is not required for perfect audio - the two points being related. But hey what would I know, I just write software for PCs and microcontrollers professionally.

By the way, if you go by what it says on that page, increasing the granularity of the scheduler would increase jitter due to more noise on the power rails thanks to the increased cpu switching noise. Some of the points made on that page are, if you have a really good imagination, plausible, but the "realtime"/small-buffer thing is misguided. It isn't the only thing either.
ok my bad! what the XXHighEnd coder says sounds like major bs...but what about these results? Measuring XXHighEnd ...

anyway, I've spoken to the Reclock coder about this matter to great extend...he believes that it's all bs too.

but many ppl feel(including me) that Reclock sounds miles better than foobar: http://www.head-fi.org/forums/f46/tu...player-438010/

the Reclock coder has told me that he runs a small buffer in WASAPI exclusive in realtime priority.

I have to admit that it really annoys the hell outta me that all the BIT-PERFECT players sound different...because noone can come up w/ a proper explanation as to WHY???

when you ask the cPlay coder, he comes up w/ vastly smokey theories(his software is free btw), and if you ask the XXHighEnd coder...hahahaha

and if you ask about Amarra, they tell you: "well yeah"

I run my XP SP3 box on the PM Timer w/ the shortest granularity(9766 ms)..it does make a visible difference on the Reclock jitter...I can reach 0.17ppm
post #114 of 487
Quote:
Originally Posted by leeperry View Post
I have to admit that it really annoys the hell outta me that all the BIT-PERFECT players sound different...because noone can come up w/ a proper explanation as to WHY???
There... We both have the same question... When will there be a REAL bit-perfect hehehe

Were getting closer when we have a better OS like W7 since sound quality from player to player is harder to discern than when I was using Vista for example, but the phrase BITPERFECT? hmm I really dont know if we are really there yet. One speck of dust can already cause so many errors for example on a spinning cd to correct perfectly in real time that I highly doubt we will ever get there. How much more the combination of computer code, data reading from the transport, real time transfer of data to the dac, etc... too many places possible to lose data although getting almost to an amount that is insignificantly small but nonetheless Not BitPerfect
post #115 of 487
leeperry, I'm not saying that the software makes no difference. It may well, I don't know. However smaller buffers won't help there, and probably would make it a little bit worse.

The other, more important point IMO is that the way these software try to improve sound is backward. What I mean is they take a computer and reduce its functionality until its not really useful as a computer, only as a poorly designed music player. So why not just make a music player?

That's the DAC I'm designing now anyway. It doesn't go through any sound layer at all, audio data goes through ethernet to a microcontroller. Jitter is nonexistent due to the architecture. Only thing is it can only be used with a computer and with custom software, but at least the computer remains a fully functional computer
post #116 of 487
hehe, yes the cPlay coder is too extreme...tbh, if you got a XP box around try comparing foobar/Reclock and uLilith

I could share my thoughts on these 3 players, as I've A/B'ed them to death...but better not influence you.

there's also this minimalist WASAPI/ASIO player: StealthAudioPlayer

it also sounds very different from the 3 aforementioned players...all in bit-perfect audio renderers, using windows drivers that forbid resampling

I also tried XXHighEnd but the GUI is too annoying, and I've never managed to make it output any audio at all...plus it costs $100 so fuggedaboutit, and cPlay only plays FLAC and .CUE and has no GUI at all

to me the best music player so far is uLilith in ASIO on XP...try them if you can.
post #117 of 487
back to this player. I just tried it with asio4all and compared it to foobar using asio4all. Wow it sounds different now compared to foobar unlike when both using directsound. Now foobar sounds better. More defined, more slam, less of a distorted feeling to the sound. Maybe something with w7 again
post #118 of 487
I never said that uL was less "distorted" than foobar...or even worse than Reclock, but it sounds better to my ears. The sound is just less edgy and more laid back...why will have to remain a mystery I guess
post #119 of 487
now were back again to the age old question... which one is bitperfect hehehe Or from now on we should abolish that question and probably ask... Which one is closer to being BitPerfect?

This is nuts... I'm gonna stop this and be content with foobar hahaha
post #120 of 487
I'm exchanging emails w/ the uL coder, I will ask him! He will also send me the ressource file so I can fix all the typos
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Computer Audio
Head-Fi.org › Forums › Equipment Forums › Computer Audio › Lilith Audio Player