|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
67.177.242.32
While visiting RMAF, I asked one of the Wadia guys if they were running Apple Lossless via their iPod and Wadia 170i iPod dock. He said they ran uncompressed .WAV files because they sound slightly better.
Has anyone experimented with this and what is your conclusion?
Follow Ups:
I have once or twice compare the same selection direct from CD, from WAV, from ALAC, and from FLAC. I haven't heard any significant differences. OK, maybe I have indifferent equipment and bad ears.Dropouts have always been my biggest issue with computer playback regardless of file format. However recently, (on good advise from Dawnrazor, I think), I set Foobar2000's 'full file input buffering' to 200000kb. So far this seems to have almost eliminated dropouts.
Something else that has made an objective difference is selecting my M-Audio Revolution 7.1's ASIO driver in Foobar. Using this driver with either ALAC or FLAC, the HDCD indicator on my DAC lights up; this isn't the case using Direct Sound, nor with iTunes playback of ALAC. However truthfully I can's say I hear a difference on account of it.
___
Feanor's Classical Survey: 250 Exemplary Compositions
Edits: 10/16/08 10/16/08
Fean,
Glad things are improving.
YOu might want to try adjusting the ASIO buffer in either Foobar or with the Maudio card.
Unless you have them at the highest now, I bet that helps a lot.
nt
___
Feanor's Classical Survey: 250 Exemplary Compositions
In general, you are likely to get better sound quality (between glitches) when you use fewer buffers and fewer glitches when you use additional buffers. You should use as few buffers as possible. It is best to experiment with your system and find out what software is causing glitches. You can kill off all unnecessary processes and system services. (It may not be so easy to find out which are "necessary". If you kill a necessary one you will have to reboot your system.) I say "in general" because the effects on sound quality are system dependent. For example, if your DAC has very good jitter rejection then software timing changes in the PC may have little or no effect on sound quality. Similarly, if your DAC has a huge amount of its internal jitter, then any extra from the PC won't matter. So if you don't hear differences, it may be a good thing or a bad thing, technically. (Musically, it's good if you don't have to obsess over subtle differences in sound quality!)
I used the task manager to look at running processes and observed their cpu usage, disk accesses, and page faults as a way of finding ones that I should kill off. This isn't a perfect way, since the task manager uses sufficient resources so that it disrupts the system, e.g. causes audio glitches itself.
I would start by disabling your network adapter or physically disconnecting the network cable. Then uninstall your anti-virus software. (With Norton, simply disabling the various features did not prevent the computer from going on holiday for extended periods.) There may be anti-virus programs that can be temporarily disabled without leaving a significant effect on audio performance, but I am not familiar with them. I don't use any anti-virus software on my audio workstation, but then it is behind a hardware firewall from the Internet and only connected to my LAN intermittently. I don't web surf or do email on this machine.
Running an ultra lightweight audio application such as cPlay helps as well. This does most of the work before the music playback starts, including FLAC decoding. If you have the RAM, then this is the way to go. Even upsampling is very efficient with this software. Of course, without the overhead comes the lack of fancy features.
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
I have kept my dedicated music PC pretty lean. I like to stream internet radio hence I'm a bit reluctant to disable the anti-virus, albeit the risk is probably small if I restrict Internet use in general.In Task Manager I set Foobar to 'Above Normal' priority. This made no difference in sound quality I could detect though it might have reduced dropouts ("glitches") a bit. What really got ride of the dropouts, as I mentioned, was specifying as generous input buffer in Foobar, on the other hand I keep the output buffer small when using ASIO.
I can't really judge the jitter tolerance of my DAC. My older Assemblage DAC has no special jitter provisions that I'm aware of. The only comparison I've had is to my CD player which is entry-level Sony; the Assembage is slightly better, being a tad airier and perhaps very slightly better resolved.
As I mentioned also, using ASIO apparently sends the DAC a better bit stream than does Direct Sound since with the former the DAC recognizes an HDCD signal. However I guess this doesn't say anything about jitter.
Fortunately I'm not obsessive about tiny differences. If differences such as there might be aren't pretty apparent to me, I don't go looking for them.
___
Feanor's Classical Survey: 250 Exemplary Compositions
Edits: 10/21/08
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
nt
___
Feanor's Classical Survey: 250 Exemplary Compositions
Hi.
My 2 cent.
One more time discussing this issue.
Yesterday I did some testing on my Linux machine.
I experienced approx. 10% higher CPU consumption for .flac decoding,
while pure PCM streaming directly to the output (no - we don't need
ASIO or KS in Linux) didn't really show any differences on CPU
performance.
Now, if you follow -let's call it the "CICS"-logic, meaning the more you got going on on the PC the more distortion you'll experience, which is pretty much valid for rather poorly implemented DACs (actually the majority of DACs), then you might guess that there must be some truth behind the audible difference wav vs. flac claimed by quite a number of people including me.
Looking forward to the next discussion about this issue.
Cheers.
As to Linux, I downloaded the ubuntu "Hardy Heron. Audacity is the player I use. It sounds great! I have medium fan noise. Whether I use Windows or Linux, doesn't matter. I am running line-in to Insignia 500w Amp, internal Dac. I built this AMD platform 4200+ cpu and probably need an aftermarket piped heatsink. Just my 2 cents, I use WMA Lossless or EAC.
Now I Understand
Hi.
Try MPD - Music Player Daemon - and Minion (A FireFox-Add-On MPD FrontEnd").
If you like Audacious, you'll love MPD. The sound is even better.
The big advantage is that the daemon doesn't need a front-end by itself, thus using very limited resources.
And you can easily control the daemon from any FireFox browser in your
home.
Highly recommended.
Cheers
If used for storage only, lossless is perfectly fine, as long as it's converted back to WAV prior to playback.
The decoding of lossless files played back in real time to me seems to introduce a signal-correlated jitter component into the audio signal, and it seems consistent in character from device to device. I get a real headache listening to it. The music sounds broken. With most lossless, the artifact seems to "smooth out" vibrato, inflections, and even tape flutter. (If there was an ABX software suite that could utilize lossless formats, I'd run a listening evaluation to either validate or refute what I think I'm hearing.)
The only lossless that I find tolerable in real time playback is OptimFROG, followed by Monkey's Audio. I cannot stand FLAC or Apple Lossless.
My favorite form of WAV playback is broadcast wave format (BWF)..... Awave Audio suite provides a conversion to this format, and I now use it almost exclusively.
To each his own, but my experience is that real-time FLAC playback beats spinning the identical music on RBCD when played back through the same DAC. And The Absolute Sound's Robert Harley agrees. See "The Brave New World of Music Servers," The Absolute Sound, Issue 177, where Mr. Harley preferred FLAC playback via a music server to the corresponding RBCD on a multi-kilobuck transport.
Mud,
THose are 2 different things. The thread was about .wav vs. lossless. WHat you are talking about is lossless vs. optical transport.
I don't think most here would disagree that lossless can sound better than most optical transports.
But if you take the same media server and compare flac vs. .wav files, some say there is a difference....but no actual disks are involved after they are ripped.
I understand the difference between .Wav versus lossless, on the one hand, and lossless versus optical transport, on the other. The point of my 10/19 post's response to Todd's dislike for FLAC was simply that I and at least some others think streaming FLAC sounds excellent REGARDLESS of what it is shot out against, including state-of-the-art RBCD playback.
Edits: 10/21/08
d
Gang,
You may recall we visited topic this last year at CES. Slow computers cannot decode the Lossless fast enough and therefore do not sound as good. Though we could not hear the difference between the AIFF or WAV files with these computers as speed is not really a problem.
For the lonely iPod running at much lower speed, it would have even more problems with the lossless files.
Thanks
Gordon
J. Gordon Rankin
If you want to believe that two losslessly-encoded files *are* different when decoded, then you are free to believe that, although you are wrong.
If you want to believe that they *sound* different, then you can create justifications for your belief if you want to.
Doing so has no purpose that I can tell.
In a sample of 10 people; if 5 say they prefer .wav to .flac (after listening to both) do you then believe they are “wrong” because you are in the group of the other 5 and happen to think/believe differently? I have A-B’d .wav against .flac and prefer (the sound of) .wav. Others have the same experience and have said so. This is “evidence” to suggest that the matter is, at the least, inconclusive. You need to think a little more about the distinction between certainty and belief and how both things are founded. There was nothing in your post to convince me that my belief should defer to yours.
Bling,
If you encode a WAV file with lossless compression, then decode it, the result is the exact same as the file you started with. If not, then by definition it's not lossless compression.
There are two playback stages:
- Get the bits to the DAC
- Convert to analog, get analog to your ears.
By definition the bits-to-the-DAC are identical.
So if anything affects the analog result or the experience of it, then something's wrong. There might be something in the coupling of the decode process with the DAC's clock, for example, that would induce signal-related jitter. But if that's the case then you should separate the decode process from the DAC clock by the greatest possible amount. But that coupling, in any halfway-sensibly engineered system, should be so small as to be completely undetectable - and certainly minuscule in comparison with other parts of the system where you're in much more control, such as for example the position of loudspeakers in your room. If it's detectable then there's something wrong with your system and you should probably try find a system which doesn't suffer from that problem. After all, if you think A sounds different from B when A and B are mathematically identical, then what chance do you have of telling whether A or B is really *better*?
So I feel that agonizing over one lossless compression format versus another is a fool's errand.
I'm not saying you shouldn't believe that WAV sounds better than FLAC, or FLAC than AIFF, or whatever, on your system. Feel free. But is it an interesting question? Relative to anything of any importance at all, is it worth an ounce of effort? Especially if the result is partially to confuse those who don't understand the starting point -- that lossless compression systems, by definition, don't make any change in the digital audio signal itself -- then, I think, no.
Something is indeed wrong. People who hear differences have an imperfect DAC. But who has a perfect DAC? As you have stated, the way to improve the sound of compressed files is to increase the isolation of the DAC. But how?
One expensive and potentially imperfect way is to spend big bucks on a super high quality DAC, reclocker, power isolator, etc. An inexpensive way is to expand lossless compressed files back to their original form before the music starts playing. One can do this by storing them that way on the hard drive as some people are doing (which costs a tiny amount for extra hard drive space) or by decoding them into RAM (which costs a small time delay when starting playback with a memory player). There are pros and cons of all these approaches. IMO, it's best to take a practical approach and not a theoretical/religious approach.
There are real arguments between the various lossless formats. Mostly they have to do with platform support (O/S and applications), features (e.g. 24 bit samples), bugs, and trust (in supplier). Perhaps there are differences in CPU resource consumption and indirectly in sound quality on many systems, but reports of this type are rare. Mostly, people compare sound quality of lossless compression vs. uncompressed PCM (WAV or AIFF).
There is also the possibility that some of the people who hear the differences are deluded. These people can choose fancy hardware or simple software changes, according to their budget/ego.
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
If two groups report two different results then perhaps you need to look at your argument. It might be that your premise is wrong or that you are overlooking some factor that is relevant to the experience in question.
Believe me, I do not “agonise” over this issue. Rather I am amused when someone claims to be in possession of certainty and is so dismissive of others that do not support their POV. A group of people (within this thread) record that they have an experience that goes counter to the claim that A is identical to B. Most, I think, are not arguing about “mathematical identity” but are saying that A sounds different to B. My thought is that you may challenge their justifications but not their experience. Experience has just as much value and relevance to the question as your proposition about “mathematical identity”.
Bling,
"If two groups report two different results then perhaps you need to look at your argument. It might be that your premise is wrong or that you are overlooking some factor that is relevant to the experience in question."
And yet none of the posters provide any scientific data to prove the existence of audible differences.. Is that to much to ask of anyone here claiming to have golden ears?? When I posted about DBT all I get is foolish responses that do not address the specific questions asked. Would any of the posters believing they can hear the differences in cables, IC's, or power cords please prove or disprove their position with factual scientific data!
The majority of responses become defensive and immature when asking these important questions.
MAK
Racer,I did not know this thread was about factual scientific methodologies. I thought it was, amongst other things, a place where audiophiles came to share their opinions and experience.. With respect, I am amused that you want to make this a discussion about the “scientific method” or similar. OK, you be the first; set out your proposition(s) using the following (scientific) outline and we can proceed:
i. Question.
ii. Observation.
iii. Induction.
iv. Theory.
v. Verification.My point to inguz is that theory alone is not enough if you want a proposition to be accepted as “knowledge”. In relation to his particular proposition re: “mathematical identity”, my point is that there is anecdotal evidence that runs counter to the outcome his “theory” predicts and this (perhaps) is significant. Inguz seems to want to account for this by saying that I am “delusional” and/or my system is deficient. Well, pardon me if I do not accept this on his (or your) say so! It is also possible that his theory overlooks other factors. In fact, I would go further; my observation, given that it is supported by a number of other people, suggests strongly that there are other factors.
Bling,
Edits: 10/16/08 10/16/08
To me, question is meaningless. If, after assembling a decent system, cleaning out the ears (if necessary), and getting all preconcieved notions out of your head, you sit, take a listen and don't hear any differences ... that means there are none - FOR YOU. No scientific evidence is going to change that.
It doesn't affect in any way the simple fact that they are there - FOR ME.
Yea really, like if it was proven with scientific data, would he then be able to hear it. I don't think so!! If you can't hear, you can't hear it!!
It's clear that the fundamental knowledgebase of electronics, built on for many years and which provides the foundation of the hobby, is of no interest to some people, and is discredited by other people. Those who offer their expertise are dismissed as over-analytical objectivists.
So the dominant modality becomes: "my opinion is as valuable and credible as fact-based knowledge"; "measurements are useless - - no one hears measurements"...and so on.
Intelligent people, who in their everyday lives accept the role of science and engineering, make intense arguments against the relevance of knowing electronics in their pursuit of the audio hobby. They cannot accept the idea that this knowledge is empowering and will assist in making informed choices among the critical components (like speakers). Rather, the ear is to be the trusted informant.
These are the same people who rely on science and engineering every day. The people who don't think twice about riding an elevator, driving across a bridge, working in a hazardous occupation, seeking medical help, etc. These are the same people who take a medication whose availability is completely tied to its safety and efficacy (through double-blind tests, trials, peer-reviewed studies, etc.), but believe that DBT has no relevance to audio.
So, it's possible to step out of superstitions and folk wisdom of the audio medieval times that many hobbyists want to live in... and step into the 21st. century. For starters, the next time your dealer tells you to buy the $250 interconnects because they're "better" than the $50 pair, ask him to prove it.
racerxnet2000, glad you liked my post on AVS Forum enough to quote it almost verbatim here on AA! How about some attribution the next time.
post on another lousy, yet different, audio blog, I'll assume he knows absolutely nothing, has no fresh or new ideas of his own worth posting, and is unable to string words together in a manner befitting a complete sentence. Yep, I think I'll pretty much ignore any more of his postings.
Others are free to follow my lead....
"Dammit..."
"scientific evidence" to their particular listening situation. Would they magically start to hear the differences, or (holding my breath) will just stop posting about DBT nonsence?
My guess is - neither...
when decoded real-time? In other words, not decoded into WAV in-memory before playing?
If the answer is positive, you're definitely in the wrong forum - HydrogenSulfideAudio is waiting for you.
Lossless was all i could tolerate in my mind. First i burned my cd collection at 320 mp3... listened for a while and decided to do it all over again as lossless .wav files. Mp3 sounded worse compared to the original cd... that's what made me spend the time doing it over. I like knowing my collection is as close to the originals as possible. BUT, i am always watching HD prices. 500 GB around $100 these days. I didn't bother trying any of the claimed lossless formats because i was concerned i'd have to burn again!
Staples is having a sale this week. WD my book 1T HD 149.99
Whether you hear a difference between a lossless version of a file an a WAV (or AIFF) version of the file will depend on your playback hardware and software. If your playback software expands the lossless data while music is playing it is possible (some would say likely) that this processing will indirectly affect the sound. If your playback software does the expansion before the music starts, as with a memory player like cPlay, then it is highly unlikely that there will be any difference in the sound output. Similarly, if you have a good sounding WAV file, you compress it losslessly and then expand it back to a second WAV file, it is highly unlikely that the two wave files will sound different. If they do, most likely you can cure the problem by defragmenting your drive.
I store audio files lossless except if I have been editing them, and then only because my editor has poor support for lossless files. No matter how big the disks are that I purchase, it's been my experience that they quickly fill up and I am always purchasing more. Perhaps I have too many recordings and too much high res material. :-)
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
> Similarly, if you have a good sounding WAV file, you compress it losslessly and then expand
> it back to a second WAV file, it is highly unlikely that the two wave files will sound different.
What about a bad sounding WAV file? Not so highly unlikely?
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
Most people probably play their own CD's, I wonder how much they have been Doctored or EQ'd....
It would be interesting to hear some regular RedBook CD's played instead of their own music.
Vista 32bit Laptop---> Lavry DA10---> Acoustic Zen Silver RefII---> Aragon 8008BB---> Dynaudio Audience 82
Karma Means Never Having To Say You're Sorry
I recently had to replace my music encoded in Apple lossless, so went with FLAC this time, for my SBs. Although there was no side by side comparison, I felt there were some improvements.
Edits: 10/13/08
A number of my customers have reported that AL sounds a bit "threadbare" compared to .wav. Mostly those with SET and high-efficiency speakers. It's a really subtle difference in my "muscle-amp" system.
I've compared Flac and WAV and have heard no difference. Never used Apple lossless.
But who's to say what someone else hears or thinks that he hears? Some people have claimed Mp3 sounds better to them. Some think the brand of hard drive makes an audible difference.
More than anything else, people seem to want to be told what sounds better, rather than listening for themselves. But that's probably no different from questions like "What's the best tube?" or "Which speakers do I buy?".
...three years ago, I thought I could hear a difference between AIFF (WAV) and ALL on some recordings, particlarly older ones.
This was through Shure E5c earbuds, since I don't use my iPod on my main system. But I do use it on my smaller family room AV system.
I went back and re-recorded those same cuts after demagnetizing the discs and adjusting the computer processes to prioritize iTunes. That seemed to improve the quality enough so that I only use ALL now.
The issue for me isn't the cheap availability of hard discs, but the iPod memory max of 160Gb.
Hey Abe,
It is probably more system dependent than a general thing.
I use and advocate the use of .wavs. Hard drives are soooo cheap these days it just makes sense IMHO.
ALso, if you do a search on the general asylum under flac I believe there is an interesting post from an inmate where he a-bs flac vs. wav...something about the transients.
Anyhow, that like a lot of things audio is a big can of worms. You'll get posts about it being bit identical and the "need" for tags etc.
Oh, and IIRC you have a sqeeuze box. DO a search here for post about flac sounding different to John Swenson on his squeezboxes....
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: