|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
99.67.90.88
what does everyone think about their conclusion that WAV files sound better than FLAC? They also concluded that FLAC files converted back to WAV still sound inferior to the original WAV file. In addition, they said that the more you compress a supposedly lossless FLAC, the worse it sounds. I have never done a direct comparison between WAV and FLAC, but I find that fully compressed FLAC files sound very good in my system.
What say you?
Steve
Follow Ups:
Over the years (34 in high-end audio, 20 in IT, as classically-trained musician, sound engineer, etc) I often chuckle at how many self-anointed digital audio "experts" (aka "absolutists") ruminate in nano-circles. Many claim irrefutable empirical/mathematical "evidence" e.g., "USB must be better than SPDIF", "FLAC is lossless - therefore there can be no sonic difference from WAV discernible by anyone" etc.
It helps to occasionally grapple with the big picture. If one accepts that digital storage media are progressively becoming more affordable, in relative and absolute terms, then why on earth bother wasting time, energy and expense in compressing/processing original .wav files? Just rip/store those fully-fledged audio files and enjoy!
For the record I can discern the sonic inferiority of FLAC-encoded files (compared to equivalent original .wav files, on my equipment) within 10 seconds of such tracks starting up. Blind test? No need. It's painfully obvious.
-spindrifter-
"We have now sunk to a depth at which the restatement of the (bleeding) obvious is the first duty of intelligent men." — George Orwell
"For the record I can discern the sonic inferiority of FLAC-encoded files (compared to equivalent original .wav files, on my equipment) within 10 seconds of such tracks starting up. Blind test? No need. It's painfully obvious."
This says more about your digital playback setup than about the FLAC format.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Yes, Tony. Meaning my "FLAC playback" is flawed? "I'm all ears".
-spindrifter-
"We have now sunk to a depth at which the restatement of the (bleeding) obvious is the first duty of intelligent men." — George Orwell
Edits: 01/21/12
It's either your player software, your operating system, your sound card driver, your sound card, the rest of your PC, your digital cable, your power cable(s) or your DAC. :-)
The easiest place to start is with the player software. However, this isn't necessarily the best solution to this "problem" since the root cause of the problem is that the DAC is not treating the signals coming from the computer as "bits". All things being equal a DAC with better isolation will sound better on WAV files as well and will make the choice of computer, digital cables, etc. less critical. Of course such a DAC may have other sonic problems...
I use cPlay. 99.5% of the time it is not decoding the FLAC, therefore there is no sonic degradation due to FLAC 99.5% of the time. Most of the remaining 0.5% is the start of a track where there is no music playing. I am unable to hear any degradation of the silence. So I don't worry about sonic degradation from FLAC. However, if I did I would use dBpoweramp and convert an entire album worth of FLAC files to WAV and then play those. This would involve only one or two mouse clicks and about 20 seconds wait for the conversion to complete. (Most of the new music that I get is already in the form of FLAC files, as it is downloaded from various web sites.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
My current PC playback environment: Linux OS 64-bit (various versions lately including Mint and Archbang, presently ALSA sound module, also Pulse Audio, rip apps include Asunder and K3B, playback apps include DeadBeef and VLC, ASUS mainboard, AMD 3.2GHz quad core, 8GB RAM, VIA VT1708S DAC/ADC codec (http://www.via.com.tw/en/resources/pressroom/2006_archive/pr060222VT1708-HD-codec.jsp), several external USB 7200rpm Samsung 1TB HDDs, internal Kingston 96GB SSD, customised Fostex T50RP headphones, filtered mains board, better quality mains cable.In a digital playback sense what am I doing wrong? Are the likely causes of my FLAC playback flaws due to DAC mis-processing and electrically-induced distortions?
Is there a blueprint for ideal digital red-book/FLAC or high-res/FLAC playback? Is there an agreed industry minimum standard or set-up? Is your set-up ideal? Do you ever store and listen to .wav files? Or is FLAC your standard "go-to"?
Do you listen to acoustically-recorded and other music? Have you recorded, mixed, edited and/or (re)mastered? Are you an advocate for "high-res" formats?
Then, may I also ask, do you presume an individual's sonic perception is quantifiable, empirically-measurable or objectively comparable?
-spindrifter-
"We have now sunk to a depth at which the restatement of the (bleeding) obvious is the first duty of intelligent men." — George Orwell
Edits: 01/21/12
Last year I put together what's going on while ripping and/or converting to flac/wav and compared the results bitwise.http://soundcheck-audio.blogspot.com/2011/05/cd-extraction-study.html
Conclusion:
Decrypted .flacs are 100 identical to .wav files on the PCM data
From a data value perspective there can't be any audible difference.Since I also experience a sound difference, doesn't matter what DAC or OS or system I'm using, something else must be causing the difference.
There are certain factors which do or might have an influence on sound.1. The flac realtime decoding process on the renderer causes
additional non linear load. That non linear additional load
translates into a pyhsical impact on the data stream.
Note: The bits received by the dac will have the same value!!
What can be different though is the shape of a bit, the timing, the
noise surrounding it.
That's what causes the audible differences on the DAC side.
2. If two .wav files sound different even though they are 100%
identical, the storage location (fragmentation or device type) of that
file might be causing a difference and some indirect effects while
reading and processing them.
Maybe even the RAM storage situation might cause a slight difference.
If the testcase is done right I wouldn't expect any differences
on 100% identical .wav files.
To eliminate most of those audible realtime decoding effects offline flac decoding is recommended.
You'd need a sw-player which decodes the entire flac to wav (PCM) prior to playback and plays those decoded PCM data from RAM (disc/cache).All this is a workaround to cover the actual problem.
We wouldn't experience any differences, if audiointerfaces would be able to decouple from PC induced distortions, such as timing variations and noise.
The whole mess is related to the limited capability of todays audiointerfaces to cope with those physical transport induced distortions.I found a workaround for myself. On my Squeezebox setup I decode flacs on the server.
The renderer doesn't cause any audible differences that way. I send a decoded flac or a wav stream. The renderer will get exactly the same
data and does the same processing on both base formats.Enjoy.
Cheers
--------------------------------------------------------------------------------------------
::: Squeezebox Touch Toolbox 3.0 and more ::: by soundcheck
Edits: 01/12/12 01/12/12
If that is what is going on, then those using async and/or reclocking would not notice it, right? A simple fix then?
It's a common misconception -- pretty much driven by respective marketing messages, the press and customers who bought into those messages --
that asynchronous USB dacs would solve ALL the issues associated to PC based audio.Asynchronous USB is an approach to solve one out of many problems associated to PC induced distortions.
If you'd followed what's been reported over here and elsewhere, you'd know that slightest changes/optimizations (HW and SW!!!) on the PC/transport/renderer side will have an impact on the vast majority of DACs out there.
I'm not aware of any DAC which wouldn't respond to transport based optimizations. (And I havn't found anybody who would have pointed me
to such a device.)
Cheers
--------------------------------------------------------------------------------------------
::: Squeezebox Touch Toolbox 3.0 and more ::: by soundcheck
Edits: 01/14/12
"It's a common misconception -- pretty much driven by respective marketing messages, the press and customers who bought into those messages --
that asynchronous USB dacs would solve ALL the issues associated to PC based audio."
Agreed.
Why would anyone expect Async USB to remove all the issues? Async USB remotes the PC issues across the USB cable into the DAC box. The USB circuitry continues to operate as part of the PC.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I would rather have FW connections but few manufacturers are willing to pay for FW.
I also have FW dacs, which until USB3 are faster, not that this has been very important other than in downloading pictures and videos. I guess that I have never used USB for anything come to think about it apart from some early dacs.
Data, music, what's the difference when one is talking digits? It's how the interconnect is used that matters and how this use is actually implemented. In other words, it's about products not standards.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It is also about speed of transfer in music. USB is three different standards, with only USB3 being faster than FW. The other two may be fine for simple data.
Your are under two misapprehensions.
Here are the facts:
1. The timing of data transfer across the USB is irrelevant. The only timing that is relevant is the timing of the clock at the DAC chip that controls the time at which the analog signal changes. The time at which the data is transferred across the USB is no more relevant than the time at which the CDR copy you are playing happened to be burned.
2. Faster components have nothing to do with the accuracy of timing. That is controlled by the quality of the clock architecture and implementation. Lack of speed in data transfer is of relevance only if the transfer is delayed so long that it does not arrive in time. It is perfectly possible to play music transferred over a connection that operates slower than the data transfer rate if sufficient buffering is accorded and the play list has finite duration.
It is interesting to see where these myths originate and what their actual relationship to reality might be.
1. Poor interconnect designs transmit the clock across the interconnect. A prime example of this is SPDIF or AES/EBU where the clock is provided by the transport. With this poor clock architecture any noise on the clock arriving at the DAC will affect the clock that operates the DAC chip, unless heroic measures are employed to prevent these. These measures are seldom completely effective and they come at a price, namely that the local oscillator has to be variable which means it can not be the highest possible quality without further costly and heroic measures. With SPDIF/AES there is a simple solution to this, namely to run an extra wire and send a clock from the DAC back to the transport. This eliminates the entire problem if done correctly, but of course at some extra cost. With USB the situation is far worse, because the streaming equivalent (adaptive USB) has tremendous jitter because the clock is completely unrelated to the audio speed, a characteristic of the packet system. However, it is possible to run USB in such a way that the clocking in the DAC is independent. This requires sending control information from the DAC to the Transport telling it when to send packets. There are two ways of doing this and there are products that use both approaches (block mode or "asynchronous").
2. Faster components may allow for sloppier design practices, so if a mediocre engineer (or a competent engineer whom circumstances force to feed his family by shipping products on command of bean counters before their design has been completed) may get better results by working with faster components. However, if a competent engineer does a good design, it will not matter how fast the components are, assuming that they are fast enough. I have designed real time hardware/software that ran reliably where the components were less than one percent faster than the theoretical minimum and where the presence of a single extra instruction or machine cycle in the inner loop would have caused the system to fail. USB 2.0 is fast enough for all possible two channel sample rates with a huge margin to spare. The extra speed of FW 800 or USB 3.0 is irrelevant for audio applications. (These high speed interconnects were developed for bulk data movement applications such as file backup.)
Another reason why faster components may sound better is related to why they are faster. It's not the speed itself, it's the characteristics of the underlying semiconductor technology. The speed of newer computers comes in two ways: faster circuitry and more circuitry (more parallel operations). This becomes possible because the transistors get smaller and as a result can operate faster (less capacitance). This means there is less power consumption for a given task. The resulting reduction in power consumption means less electrical noise to disrupt analog signals. The benefit comes from the smaller transistors, not their speed. In many cases even more benefit can be realized by slowing down the clock speeds which permits lower voltages and even lower power consumption and hence lower noise.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I heard the BMC DAC1 with their new USB port. For the first time I would have to concede that USB can sound very good. With their first USB port, the FW Weiss DAC202 blew it away.
"On my Squeezebox setup I decode flacs on the server."
Can you tell me how you set this up? Right now I am using my computer to feed FLAC files to my SB Touch.
Thanks,
Steve
Remember, it's all about the music.
checkout my sbt tollbox blog. cheers
--------------------------------------------------------------------------------------------
::: Squeezebox Touch Toolbox 3.0 and more ::: by soundcheck
Installed your modifications on my Touch, although I left the screen, remote, and Wifi on so my wife can use it. Also did all of the server optimizations you recommended. The improvement in the sound is amazing! I really didn't expect it to sound so much better. Good work! Thanks so much for taking the time to do all of the work necessary to produce this mod!
Steve
Remember, it's all about the music.
See Samoore. You went from saying that fully compressed Flac files sound very good in your system to being stunned at how much better things can sound.
You arent alone. I have walked several pc users through this and the settings matter, and flac decoding can affect the sound. This is why I just use .wav and agree with half of what TAS says.
I have a hard time buying there corruption idea that once a file is turned into flac, it is ruined.
Afterwards we discovered faith; it's all you need
sensible. I personally do not use any processing within player.
Hey, did't we have another thread about this in the last week?I'm one who has been using FLAC for > 7 years and have yet to convince myself there is *any* audible difference through headphones or speakers. (Okay, maybe there's a difference on my Celeron 600MHz computer from ~1997 but that was more popping and obvious decoding problems when multitasking!).
I have played FLAC vs. WAV to friends and my wife with NOBODY being able to tell a difference if I asked which "copy" of the song they preferred (I usually will tell a little white lie like "I just bought a new copy from Japan and wanted to see if there was a difference".)
Therefore, IMO, there's no problem to be worried about at all. There are many other more worthy factors to worry about in audio reproduction.
Edits: 01/11/12 01/11/12
I always use cPlay to play my FLAC files. Here the extra processing manifests as a slight delay when loading a playlist. The FLAC processing for the first track is (usually) completed before any music plays so there is no degradation. On most tracks there is no further FLAC processing until the next track begins, at which point there will be FLAC processing for a second or so, but here there is usually no music, so again there is no potential degradation of sound quality.
For long tracks (or average length tracks at hi-res) cPlay sometimes breaks a track into multiple loads, so midway through a track it may decide to load another chunk. There will be a brief period while the processor is running at a high load, and there could be audible degradation. However, I've not heard it. There are several possible explanations why I don't hear this, one being that the effect is smaller than my ability to hear it, another being that the effect comes in steady bursts, so there is no cyclic load causing periodic jitter at an audible frequency, and finally that the usual effect of jitter, degradation of sound stage, is not perceived for brief periods as the mind has already "locked in" the mental illusion during the 99.5% of the time there is no FLAC processing.
When I play music with my editing programs I always convert FLACs to WAV. Soundforge 9 used to have problems editing FLAC files, the GUI got missynced from the audio samples, so I got in the habit of always editing WAV files. I believe this was fixed in Soundforge 10, but I don't like the idea of editing my archival files so making a copy that includes a conversion is an extra step that's a good idea. iZotope RX 2 Advanced does not support FLAC files so the question does not arise.
If I am doing audiophile tests for sound quality, I always convert files to WAV, not because I believe this results in better sound, but reducing the number of formats being compared is a basic matter of good experimentation: reduce the number of possible confounding variables. Doing experimentation is not the main reason why I play recordings, however. :-)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
"Doing experimentation is not the main reason why I play recordings, however. :-)"Really... What else is here there??? You can't possibly enjoy listening to music now , can you??? ;)
Edits: 01/12/12
samoore posted this on digital. Since there is no verbiage to tell inmates to post this kind of stuff in the PC audio section and not the general digital section, he didnt know, and the post was moved here.
What system are you using?
Afterwards we discovered faith; it's all you need
Last time I tested this with a friend:
Logitech Transporter (stream native format from server FLAC or WAV file) --> Vitus Audio SIA-025 (25wpc Class A) --> Gemme Audio Katana speakersMy usual computer setup:
Win 7 PC --> Benchmark DAC1 [recently sold but used most over the last 2 years] --> AudioEngine A2 or headphones (Sony MDR-V6 / Audio-Technica M50 / Sennheiser HD650)Good enough? Like I said, no difference IMO.
Edits: 01/12/12
Hey Arch,
On the transporter I think it matters where the decoding is done. IIRC you could hear a difference if you have the transporter do the decoding.
On the W7 machine I am not surprised. You dont mention any tweaking and IMHO that needs to be done and is harder to do on w7. For instance my dedicated xp box with the same card and dac has so much more resolution than the untweaked W7 machine using the same mobo and similar processor.
Anyhow I suppose hearing it or not is system dependent and definitely can be affected by the computer settings. Just having antivirus installed messes up the sound for instance.
Afterwards we discovered faith; it's all you need
"For instance my dedicated xp box with the same card and dac has so much more resolution than the untweaked W7 machine using the same mobo and similar processor."
But does the music sound better?
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Hey Tony,
YES! You just hear more and what you hear is more musical. The main difference is that the dedicated box is more involving.
Afterwards we discovered faith; it's all you need
.
"In this land right now, some are insane and they're in charge. To hell with poverty, we'll get drunk on cheap wine."
See response above. Is this supposed to be funny?
.
"In this land right now, some are insane and they're in charge. To hell with poverty, we'll get drunk on cheap wine."
Seriously, what kind of setup are you using and what kind of test have you tried to say with such apparent authority that you hear a difference? Otherwise there's no serious discussion possible.
The answer is YES! What you hear is what you hear. What is 'there' in an external, scientific proof sense is a different matter...and only in some respects a more important one. Put it another way: when he says he hears it, you can ask why or how he does and wonder if suggestion or inner state or motivation plays a part in it. But to question that he hears it, it's THAT that makes discussion impossible!
I agree. The subjective evaluation is all that matters when it comes to musical enjoyment. I can enjoy a 192kbps MP3 of my favourite tune just fine though could potentially enjoy it even more uncompressed with good equipment...However, I fully acknowledge that a major element to my enjoyment of music has much to do with the psychological state at the time and am completely fine with accepting that I can err when listening to different cables and thinking that one is better than another but can't prove it when a buddy switches them out.
But when an article talks about attributing differences in sound quality based on what seems to be identical digital files (virgin WAV vs. WAV reconstituted from FLAC!), one must be forced to reckon with external objective reality.
Edits: 01/17/12
talking about the files but the player playing back those files....
"In this land right now, some are insane and they're in charge. To hell with poverty, we'll get drunk on cheap wine."
.
"In this land right now, some are insane and they're in charge. To hell with poverty, we'll get drunk on cheap wine."
what one imagines. However, in my profession it's common to run into the misperceptions of the mind all the time.
In reference to the topic in this thread, as I said, I have not run into anyone able to back up the claims to which they believe (ie. there's a difference between WAV & FLAC) when presented with a test.
You haven't answered my question above however.
anything,
I have answered your question very, very, clearly.
Once again, - with feeling this time.....
They are not talking about the files, - but the player playing back those files.
"In this land right now, some are insane and they're in charge. To hell with poverty, we'll get drunk on cheap wine."
Sorry about that typo... LOL.
I read the article today. I don't think I'm misreading it - they're not talking about just the player when it comes to FLAC vs WAV.
Kind of sad, really. I guess maybe it is like asking people who have spent a lifetime reviewing horse carriages to write reviews about (and discuss the technology found in) those newfangled horseless carriages.
.
"In this land right now, some are insane and they're in charge. To hell with poverty, we'll get drunk on cheap wine."
Just assuming that WAV sounds ever so slightly better than FLAC, I wonder if the supposed difference gap can be minimized with a higher performance CPU?
The only thing I can think of that might make FLAC worse would be the extra processing power needed to uncompress the FLAC file during playback. Just a thought... which would fly in the face of those who build flea power low clock rate music servers AND play FLAC files.
.
Makes as much sense as anything else here.
It would be interesting to compare the clocking and data out of a spdif connection to verify wav and flac are the same.
Considering both should be bit for bit the same, there would have to be something else at play unless it was psychological. We know that won't ever occur though!
You are right...the data should be the same.
Which means the problem is jitter. The difference in the jitter between FLAC and WAV playback should be measurable. And if there is a measurable difference, the problem is with the computer hardware (it can't deal with FLAC decompression, input from the storage device, and outgoing data transfer to the audio device at the same time).
And, even if there is a measurable difference in jitter, a modern DAC shouldn't be affected by jitter in the incoming signal. So the DAC is at fault as well.
All of which has nothing to do with the FLAC file format itself (hence the continuing chorus of opinions that these people who pronounce one lossless file format to somehow be "better" than another, data-wise at least, are clueless about the basic concept of digital data).
Here is a post you might appreciate that has a theory as to what might be happening
Note that the theory concerns a potential defect (albeit an obscure one) in the power supply of the DAC. If the power wiring is the channel for the unwanted information ("Is it FLAC or is it WAV?") then one might be able to demonstrate this by driving the DAC with a different source while using the computer to silently decode and send out either a WAV file or a FLAC file to a second DAC that is not connected to an amp. (Details left as an exercise...)
There are people who take information leakage really seriously. These are the spook people who worry about crypto keys leaking out of encryption devices used for top secret government communication. It behooves engineers who are interested in building DACs that actually work correctly to investigate this field. Spies have died because "bits weren't just bits". This goes back many decades. Part of the problem with audio comes from a narrow "digital" or "analog" or "hardware" or "software" focus, rather than a broad system perspective which is characteristic of the best engineering. Of course, many talented engineers go where the most money is to be found, and it's in killing people, not making better sounding recorded music.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
"Killing people?"
How about protecting it's own citizens? That's how I view it. ;-)
" ... there would have to be something else at play unless it was psychological. We know that won't ever occur though!"
Could be that the jitter of the FLAC may be higher than WAV due to the processing - out the SPDIF port.
Could be psychological - but it would be better to understand the issue than to dismiss it -
There's a much longer discussion on FLAC files in the PC / Computer Audio section of the Asylum.
Since the original WAV file converted to FLAC, and the FLAC file converted back to WAV remain bit for bit identical, it stands to reason that any audible differences must be caused during playback, and not due to any corruption of the FLAC file itself.
"Since the original WAV file converted to FLAC, and the FLAC file converted back to WAV remain bit for bit identical, it stands to reason that any audible differences must be caused during playback, and not due to any corruption of the FLAC file itself."
Reason is inapplicable.
I figured if files can be converted back and forth bit for bit, any audible difference must be due to something other than the files themselves. Wouldn't it be reasonable then to consider the playback process on the files?
Tony stated "reason is inapplicable". Maybe not your reason.
Either way, if the file is converted back and it is equivalent to the .wav file than it should be the same when played - as long as the processing of the flac to wav is not occurring at the same time as the playback. If the conversion occurs at the same time then we have timing, power consumption, and noise considerations that will possibly affect the sound.
This is one reason to just have the full file in uncompressed form in memory to avoid the - read from disk, decompression, and playback at the same time. These actions all have impacts on interrupts and timings in the computer (discounting power supply noise and draw). The ear may be able to perceive them. Bottom line is if the data is identical then it is something outside of the data causing a problem and does not prove one format worse than the other.
Imagine if your compressed zip files couldn't be uncompressed without modifying your name and address :)
What I meant to say is that for some people reason is not part of their process of reaching conclusions. Let's leave it at that, I'd rather not get too specific. :-)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I have tried ripping a few familiar tracks from a cd into both flac and wav files and them played them back to back to see if I could discern any differences. In most cases the wav files sounded a little better. I couldn't tell a big difference but the most noticeable was a slight reduction in fine details, especially in the higher frequencies on the flac files. On two or three tracks, I couldn't tell much of any difference. I did this mainly for my own curiosity as I am just getting started ripping my discs to my hard drive and was trying to figure out which decoder to use. I think I will use wav most of the time, especially for those discs I want to listen to critically. At this point, I am really pleased with the sound quality of the wav files played back from my hard drive (using JRiver MC 17 with wasapi; also using JRiver to rip).
I have been saying this for years. They are right on with these results, with the exception of saying that FLAC converted back to .wav is somehow compromised. I have not found this to be the case, but I'm using XLD for that conversion. The software used is critically important IMO.
I am convinced that even though static testing with FLAC files shows that the conversions are bit-perfect, that when these CODEC tools are run real-time, some kinds of errors occur. Some manufacturers now convert from FLAC to .wav and then play the file in their servers. They know about this effect.
Steve N.
This guy repeatedly demonstrates that he doesn't understand even basic concepts when it comes to digital audio. And this has been going on for years!I find it personally humiliating that I was ever duped into buying an Empirical Audio product (and am quite happy that I found someone willing to buy it off of me soon afterwards).
Edits: 01/12/12
And you basically demonstrate over and over again that you are a TROLL!!If you weren't so damn cheap, you would have gotten the upgrades to your product and it would have beaten whatever you ended-up with. You didn't even take advantage of the free upgrades.
I know what I hear, and that is what I'm talking about here.
Edits: 01/12/12
Since the only difference between choosing FLAC or WAV is going to be the amount of incoming jitter to the DAC (because we know the data will be the same), it then logically follows that if you are hearing a significant difference between these two file formats then the Empirical Audio DAC you are using is highly sensitive to incoming jitter, and should be avoided by anyone seeking a quality DAC.
Moderators, can we please finally ban this salesman from using misinformation to steer people to make poor decisions about their audio purchases?
You cannot honestly believe that. Just record the digital signal and do a bit compare of the recording to the WAV original.
Either a decoder works correctly or not. There are of course software bugs but these code bases have been beaten to death by now.
Cheers
Thomas
Its not the bit-compare that I am debating, its the real-time behavior of the CODEC. Something must be going wrong.
But the real time behavior of the codec is exactly what you see when you record that digital output. All timing and RFI issues aside you can verify for yourself with a wide range of material that the decoder is working just fine whether you use it standalone and asynchronous or synchronous during playback.
After all you are in the business of selling USB DACs. The drivers and firmware to deal with that are at least as complex as a lossless decoder counting the number of lines of code and the timing issues involved. One way of looking at it is that there is an encoder that takes a PCM signal and encodes it for transport over USB and firmware in the DAC that decodes it back to PCM. If you do not believe this can be made to work then I question many more things.
Cheers
Thomas
Have you done listening tests yourself on a resolving system?
I only know what I hear and it's broken. It's too bad you were not in my room at RMAF. I could have played the demo for you there.
Yes, I do listening tests on a resolving system every day although my ears are no longer what they used to be...There can be many reasons why playback can be impacted using FLAC. Intermittent real time errors from the decoder are not a likely candidate. As I said there is a higher probability for you to have a software issue in your driver or firmware.
Cheers
Thomas
Edits: 01/13/12
"If you do not believe this can be made to work then I question many more things."
It does work. It just doesn't work perfectly. In as much as at least one third of any DAC constitutes analog circuitry the Uncertainty Principle guarantees that it can not work perfectly. I am not nit-picking. Voltaire wrote, "Perfect is the enemy of good."
This is nothing but a (hard) engineering problem. One needs to define what it means to be good enough and then ascertain what level of performance is necessary. Then one needs to devise a way of reliably and quickly measuring whether a component is performing at the requisite level. At this point one can use any number of relatively straightforward hardware design techniques to achieve the result. It is possible that the resulting product will be "high end", i.e. affordable by only a minority of audiophiles, but I suspect that if this is true the situation won't last for long.
Note: it is unlikely that anyone whose mantra is "everything matters" will ever succeed in such a quest. They are defeated from the get-go by their attitude. One must believe in one's goal. This is the first prerequisite for success in any endeavor.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
''This is nothing but a (hard) engineering problem. One needs to define what it means to be good enough and then ascertain what level of performance is necessary. Then one needs to devise a way of reliably and quickly measuring whether a component is performing at the requisite level.''
It is a hard software/hardware inetraction issue. Player software makes a difference when the writer of software fails to define a good enough convergence criterion when driving a defined piece of hardware.
I don't believe you've got it quite right. The problem is that the software runs on a given piece of hardware and this hardware does not come with a sufficiently accurate specification, i.e. the hardware is not sufficiently defined.
There is no "best" software possible for all hardware, because one player may work better with one set of hardware and a different player with different hardware. (Here the difference in "hardware" might just be a different cable connecting a given computer and DAC.)
This is why I have said before and will say again that the problem is not in the computer or in the software, it's in the DAC. This is the only place where the problem can be solved in general. One can cobble together various combinations of (off the shelf or modified) computer systems, cables and DACs and run various "player of the month" software and get different sounds, but one will never achieve what is possible except by solving the problem at the DAC end. This is a hardware problem, and it concerns the correct operation of the mixed signal and analog circuitry in the DAC, together with protecting the operation of this critical circuitry by providing sufficient electrical isolation from the purely digital portions of the DAC as well as the overall electrical environment in which the DAC is operating.
It is the stubborn rejection of this simple truth by audiophiles that prevents them from forcing the manufacturers of DACs to do their homework and build ones that work properly.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
You need to take onboard the integration of electronics, mechanics and computing as a discipline to understand what I said. There are various names given to such an appraoch, eg at Philips Einhoven, they recognise it as Mechatronics.
We were instrumental in introducing such approaches in UK universities and most engineering courses now take on an interdicipinary approach towards invention and design.
I understand perfectly what you said. If one designs a complete system one can of course adjust the individual components to give a higher level of performance within given design constraints. But this is completely impossible in the context of Computer Audio, which is predicated in building systems that leverage off of mass market hardware and commercial off the shelf software. However, this does not contradict what I have proposed. For example, if one tweaks the software one can reduce the amount of electrical noise that has to be rejected in clock and signal circuitry of a DAC. This may be a cost-effective way in building a one-box system, but in the context of computer audio it is completely impractical, among other reasons being the proprietary nature of the inner workings of the component parts of the computer system. In any event, this bespoke approach is fraught with many perils and comes at great expense. Indeed, this complex approach has failed. If not, why aren't you just using your DCS gear instead of dabbling in computer audio?
One can build a complete DAC that provides a high degree of isolation from the vagaries of amplitude and phase noise on the input digital stream while outputing an accurate representation of the encoded information provided by the format. In the case of a DSD DAC this can be done using no more than a few hundred transistors. One has some hope of getting good results with a careful execution of such a design. Contrast this simplicity with a typical PC. My audio PC uses about 100 billion transistors.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
''But this is completely impossible in the context of Computer Audio, which is predicated in building systems that leverage off of mass market hardware and commercial off the shelf software.''
Highend and pro audio seek to do this, although at a price. What you said explains why PC audio is pot luck because there is inadequate understanding of what is going on.
Voltaire wrote, "Perfect is the enemy of good."
But what the heck did he know about audio??? :)
Am I wrong or is it my understanding that the only way you can get high resolution downloads is through FLAC. If thats the case who cares about WAV.
"Am I wrong or is it my understanding that the only way you can get high resolution downloads is through FLAC. If thats the case who cares about WAV."
Forget about high resolution downloads. They wouldn't be any good even if they were WAV instead of FLAC. The download servers are connected to their router using non-audiophile approved Ethernet cable (wrong color insulation). The sound quality has been irreparably degraded.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Hi,
> They are right on with these results, with the exception of saying that
> FLAC converted back to .wav is somehow compromised.
So, you mean right on with their results, except their results disagree with yours. Funny, that.
Ciao T
Steve, what's your opinion of FLAC uncompressed (via dBpoweramp) vs. WAV? My understanding is that FLAC uncompressed gives you the sonic benefit of WAV PCM, but with FLAC's better tagging capabilities.
I have quite a few tracks that I have done this with, but I never compared them. I'll try to find the time, but I'm really busy. Designed 3 new products in the last 2 months.
My 2 cents is that drive space is super cheap and flac doesnt provide any tagging benefits that you cant get on the right player supporting wave directory structure or .cue sheets.
If there is even the possibility of flac degrading sound on playback why mess with it?
Maybe a better way of saying this is that you have to set up your library in a specific way, and are limited to specific players that support cue sheets well, in order to try to match the tagging benefits that FLAC provides.
Alan
Hey Alan,
It is true. But there are some good players out there that handle .wav. and yeah I think most of us hopefully ARE limited in that the best possible sound is the goal. So the number of players are limited anyhow in that only a few sound great and given the hoops we all jump through for sq using a certain directory structure or players that support cue sheets is a trivial sacrifice IMHO.
Anyhow .wav gets IMHO a bad wrap in that if you do it right you can get the benefits of tagging without the sq issues of flac. An example of this is the squeezebox. Go to their forums and they say it doesnt support .wav and you have to use flac for tagging issues. Their software does support a directory structure and consequently you end up with album and artist and track titles just like flac.
Anyhow if I have to give up some convenience for sq, I dont mind doing that especially when I cant see what I am giving up!!!
Afterwards we discovered faith; it's all you need
Funny how the once "Professionals" are now Novices.The paradigm shift from traditional audio to computer audio may prove to be too big for yesterdays reviewers to be of any use.
What the computer audio world needs are "True" computer geeks who are also Audiophiles. Not hackers who are only adept at hacking into existing codes and tweaking them here or there, but true code writers who know how to write from scratch.
Not a lot of people want to learn the current, complex techniques for tweaking a general-purpose PC to get decent sound through a complex, expensive chain of boxes and cabling coming between it and their preamplifier. I just want to listen to music…
Your reply points to a design - an integrated computer/audio architecture with eptified software. Picture a special-purpose computer that's purely a hi-fi component. Numerous products claim to be this, but most seem to boil down to the patched-up hardware and software you refer to.
I would buy this *box* and take it home, but I want to hear the traditional old-world audiophile press say it sounded good. Don’t be too quick to jettison those dudes… Like it always has, it comes down to trusting your human ear and weighing what you and the reviewers heard.
Hi,
Well, let's see...
USB is worse than SPDIF according to them (in a correctly set up and implemented system the reverse is in fact true).
FLAC degrades Wave Files permanently (there is no evidence of this, as long as care is take to remove other variables such as disk fragmentation for the un-flacéd file).
Their batting average is not going up.
If one was willing to see things negative one may be tempted to conclude that they have agenda to make computer based audio appear as "not high end"...
I think it is just down to the people who write these articles being completely clueless when it comes to Computer Audio (or computers) though.
Time for TAS to hire some writers who know their way around computers.
Their UK sister publication HiFi+ seems to be doing rather better and seems to have by far less beruehrungs angst when it comes to Computers and Audio.
Ciao T
My limited experience with commercially available products shows that authors of the article are right on re: USB vs. SPDIF.
With devices where both inputs are represented, USB (async or not) is inferior, sometimes vastly, to SPDIF, fed from the same optimized for audio (a la cMP) computer via USB and SPDIF out of soundcard of reasonable quality, respectively.
I am willing to accept a notion that what I heard weren't best implementations of USB - hence the question in the subject.
Hi,
I cannot comment on what particular designs you have experimented with nor do I know and I do not wish to be accused to use this forum to promote my own designs.
The comment above stands as made and needs no addition.
Ciao T
Sometimes I'd like to be the water
sometimes shallow, sometimes wild.
Born high in the mountains,
even the seas would be mine.
(Translated from the song "Aus der ferne" by City)
.
"In this land right now, some are insane and they're in charge. To hell with poverty, we'll get drunk on cheap wine."
To me this is a trade promotion magazine. Maybe this fits your bill.
And what is the Absolute Sound?
ever read this? Probably not. Usual case of shouting instead of hearing.
Actually did...
"And what is the Absolute Sound?"
POS; a magazine with right close to zero credibility; a farce.
I agree in part and use .wav exclusively. I can see where the extra processing could affect the sound via the changed electrical environment.
Though I have a very hard time with their assertion that changing from .wav to .flac to .wav to .flac degrades the file.
Personally I think all this kind of "testing" is far to complex. There are tons of variables. For instance did they take into account the placement of these files? Were the .flacs in one directory or drive and the .wavs in another?? Was there data degradation or where the files identical?
Anyhow you should do a comparison and see what you hear or what you dont. Who cares what Tas concludes if you hear differently?
You got it!
They are likely noticing effects that are secondary effects of the computer - not the files themselves.
A file encoded with FLAC will take more processing power to reconstruct than WAV. If you were to take several computers with different amounts of processing power, I'd bed you'd notice more jitter induced sound degradation in the less powerful ones.
Having said that, USB is a very poor spec for high quality sound reproduction - and it has been all the work to try to reduce jitter with the USB "asynchronous" modes that has made it into the best performer to date. I cannot help thinking that if all computers weren't fitted with USB's, how important that standard would be to the audiophile world. (My personal favortie is Firewire, even though I don't have any firewire based audio - because the specification allows you to allocate bandwidth to a task, and it cannot be interrupted by anything else - which is why it is used by pro audio pretty often.)
Him
The U in USB stands for Universal.
It is de facto the choice for computer peripherals, so we better deal with it.
Firewire is basically dead. Even it's own inventor and biggest supporter Apple is dropping firewire support from their hardware.
Further, non of the currently extant Firewire Audio Solutions are asyncronous. So they all use the same principle of clock recovery as classic adaptive USB Audio. Duet to the usually much higher processing requirements for Firewire transmissions you have more issues with jitter entering through PSU modulation.
Asyn USB is the way forward. In the last DAC I helped design we also added Äsync SPDIF" to be "way backwards compatible".
We actually started the design with a solution that supported both FW and USB, but FW support was the first thing to be dropped, as no-one wanted it (no-one asked had FW support on the "must have list"), but USB (and Async USB) was top list item.
Ciao T
USB is cheap enough. If it is truly superior to spdif et.al when will industry start using USB-out on CDP etc.
We should be seeing digital Receivers, EQ's, CDP, and frankly every component that can interface with a Dac should have USB out...that is if it us superior.
I have no opinion as to which is better.
Hi,
> USB is cheap enough. If it is truly superior to spdif et.al when
> will industry start using USB-out on CDP etc.
Old style CD-Players are mostly history. Pure CD-Mech's are next to extinct (most CD-Drives nowadays are actually DVD Drives).
Todays CD-Player is called iPod or ipad. And they are fitted with USB Digital Outs (on a non-standard connector and with the need for key chip, so apple can make money, but it's USB - early iPod's had Firewire BTW, later ones dropped it).
> We should be seeing digital Receivers, EQ's, CDP, and frankly every
> component that can interface with a Dac should have USB out...that
> is if it us superior.
USB PC Audio well implemented using generic USB Chips has less jitter than SPDIF well implemented using generic SPDIF Chips.
I do not see USB displacing SPDIF for consumer digital connections or AES/EBU, simply because of the installed base and because generic SPDIF is "good enough" for most users (they do not care about quality).
If you go further you can make the "transport layer" (that is USB, SPDIF, Firewire, I2S, Whatever) irrelevant by removing the jitter from all of them.
So, SPDIF and USB will coexist for a while. Eventually both will be phased out for a new protocol (whatever it may be). Until then Asynchronous USB will be the standard by which SPDIF implementations must be judged and most will be found wanting.
As said, I was involved with designing a sub-system that takes basically I2S in on a hardware level (which may be created by a USB, Firewire, SPDIF, AES/EBU, Toslink or even CD?DVD Mechanism) and regardless of source reliably kills the source jitter. It basically does for ANY source what Async USB does for USB only.
Sooner or later I expect we will see such a system as a SOC (system on a chip) for a few bucks (it is certainly easy enough - we already most of it using logic chips) if there is enough demand for bitperfect low jitter operation (which I doubt)...
Ciao T
> > I do not see USB displacing SPDIF for consumer digital connections or AES/EBU, simply because of the installed base and because generic SPDIF is "good enough" for most users (they do not care about quality). < < <
Just curious, are you inferring that those who do care about quality should or do you USB?
Can I assume then that many/most/or all of the top recording studios that do care about quality in their Mastering studios DO use USB as the primary digital interface?
Hi,
> Just curious, are you inferring that those who do care about quality
> should or do you USB?
I am inferring that tey SHOULD use systems that minimise problems, whatever these systems may be.
> Can I assume then that many/most/or all of the top recording studios
> that do care about quality in their Mastering studios DO use USB as
> the primary digital interface?
I am not that current, but most probably use pro-tools which is propriatry and neither USB or Firewire. And they probably use analog mixing desks for summing and loop back into pro-tools for the mixdown.
Due to wiring already installed (viz installed base and USB's distance limitations as well) other than that ADAT and AES/EBU are still the most used. Unlike HiFi studio's usually have a central clock everything relevant is slaved to, so things are not as bad as they may be.
It is a different envoironment with different requirements.
Again, much depends on the context.
If, right now, you wanted to be able to plug a 2-Channel (or more) DAC into a Window$, O$X or 'Nux source and get lowest possible jitter as well as bitperfect playback of music files on the hard drive (or network) you would VERY hard pressed to do better than asynchronous UAC2 (UAC2 = USB Audio Class 2) devices implemented correctly, if not the least because each of these devices will have a USB jack and the neccesarry software support for USB (Window$ still needs 3rd party drivers though).
Any of the current ready firewire solutions as well as SPDIF solutions need much additional circuitry to clean up problems to a similar level.
As to what the existing installed base does?
That is another story.
I recently moved to a new place and was able to start from almost an empty structure. So I ground-up designed and installed the lighting using LED Lights (warm white) and using many low power, low voltage spots on tracks.
The results are extremely impressive (the place at night has such an airy and open feel, it is absurd) and my whole place with all lights on (all rooms and lights) uses less power for light than one classic 60W lightbulb (and much less than my tube amp which idles at around 150W).
So, why do we not see LED's everywhere?
Before my old place already had ton's of halogen downlighters (which I changed to Leds), but the whole infrastructure just did not suit the LED approach (with it's usually much narrower angle of light coverage) and to redo each and every light fixture to get the best out of LED lights would have meant major building work.
Well, something to think about anyway, the most sold car engine in the world is also the worth for efficiency or performance, all else being equal (but cheapest to make).
Ciao T
.
makes assertions, not provide info.
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: