What is the best format to save your music from a CD to a PC? MP3 (if so, what bitrate?) WMA, WMA lossless? I know the lossless are suppose to be the best, but does the amount of HDD space it takes up justify it for the average person? I should mention that the PC and is hooked up to a Yamaha YSP-1 Digital Sound projector (one component that produces 5.1 surround sound). My company sells these, so customers would be listening to the music from the PC. The connection is with a SPDIF optical cable. I just want a format that would take advantage of the sound system without unnecessarily wasting HDD space. thanks for any advice.

After doing a little research, I decided to go with FLAC. I use Exact Audio Copy when I rip my CD's. A FLAC file is usually 30%-50% of the original CD file, but more importantly, if you rip your collection to lossless, you won't have to rip it again in the future. If you rip it as MP3, you'll NEVER gain back the original quality.

IMO, 750GB and 1TB hard drives are very cheap right now. You should be able to store a lot of CD's with those sizes.

I'm currently ripping all of my cds using Exact Audio Copy(works great with my scratched cd's but takes forever) to MP3 320 Lame v3.97. I'm ripping to MP3 to use on my PS3 and MP3 320 lame sounds cd like to me on my less than audiophile system and widely is reported to be "transparent"by others much more knowledgeable than me. The typical file size is 90 to 130 mb. Which is large for MP3 but about 1/4 to 1/3 of Flac. 256 vbr is supposed just as good as 320 cbr and a little smaller but I stuck with the 320.

With that said I would use Flac if you have the Disk space and as tokerblue, said you'll never have to rerip again like I will if I want lossless. I just didn't want to eat up the 80 gig PS3 with cd's, my computer is ancient and haven't picked up a external hd yet.

I have not been able to tell the difference between 320 kbps mp3 and lossless. I used to rip 320, now I rip lossless - just in case once day I can tell the difference. I don't want to rip again. If you are trying to show off equipment rip lossless then you do not have to worry if the rip was the cause of any problems you might hear.

Any opinions on WMA lossless? I notice that the file size of WMA lossless is much less than wav lossless.

You can get a 1 TB external hard drive for under $200. I would bet that would hold over 15,000 WAV files, possibly 18,000. You can multiply that by 2 or 3 if you use FLAC. Why do people worry about saving space?

If you have a big collection, I recommend you use lossless. No need to comprimise SQ at all. I bet we will have 300-500 GB IPODS in the near future.This way you will never have to re-rip your collection.

If you make WAV files, you can compress them to put on portable equipment if you choose. Most software makes this conversion without problem.

What program(s) is everyone using to rip FLAC files?

I use 320 MP3. The reason is that I can find no audible difference between it and FLAC or WAV, for that matter in bias controlled listening tests. MP3 is an industry standard supported by virtually every one and, as an added benefit, the files are 1/2 the size of FLAC files. I would use FLAC also if they sounded better but they don't.

Any opinions on WMA lossless? I notice that the file size of WMA lossless is much less than wav lossless.I ripped my whole collection to WMA Lossless and have been very happy with it. SQ is (of course) great. And WMA Lossless only takes about 40% of the space of the WAV files (and sounds identical).

The one thing I am still looking for is a car deck that will play WMA Lossless from a USB stick (or SD card). They all play WMA, but I can't find one that plays WMA Lossless. Anyone?

Mike

Soundwise it does not matter what lossless format you use. The compressed lossless files take about 50% of the space of uncompressed - but the content is the same - basically think of it as zip for audio, the details are different, but the result is the same - a bit perfect copy that takes less space. Sounds like you are using a PC as a source. One issue on PC's is to get bit perfect output from the soundcard - lots of them apparently interpolate everything to 48khz before outputting the signal. I don't know PC's, but you might want to look into this if you want the best quality output.
You might want to pick a song or two and try the different options - see if you can hear the difference. Maybe you could even demonstrate to your customers the difference between the various options.

If you can't tell the difference between a 320 MP3 and a lossless wav, the system you're playing this stuff on is not high fidelity enough. If you're dumbing it down to an MP3 to listen to it on a MP3 player with ear buds, MP3s sound fine. If you're listening to this stuff on good equipment and still can't tell the difference, you need to get your hearing checked. I personally will not compromise quality. I will buy a few hard drives and store them as wavs. I no longer rip music anyway. I record each song real time at 96kHz/32-bit with Audacity to improve the original content. The file comes out twice as big a lot of times, but sounds better. If you're serious about sound quality, you won't think high quality MP3s are good enough and definitely not worry about space on a HDD. HDDs are cheap.

If you can't tell the difference between a 320 MP3 and a lossless wav, the system you're playing this stuff on is not high fidelity enough.

Unfortunately, you are incorrect Mr. Audio. I suggest you make the bias controlled listening tests for yourself rather than criticize mine.

I no longer rip music anyway. I record each song real time at 96kHz/32-bit with Audacity to improve the original content. The file comes out twice as big a lot of times, but sounds better.

This has to be a put-on, but I enjoyed it.

So the best format to rip to using Windows Media Player and not caring about HDD space is .................... ?


MP3? no

WMA? depends?

WMA Pro? depends?

WMA VBR? depends?

WMA Lossless? yes?

WAV? yes?

If you don't care about disk space, use any lossless or uncompressed format which supports the original bit rate, depth, etc. They will all sound identical. I'd probably use whatever WMP supports natively, in order to make my life easier. You can always convert it to something else later if you've got any reason to do so. I use iTunes, so I use Apple Lossless for convenience.

Why would you want to rip them? It is hardly any effort to simply play the CDs when you want to listen to music.

Why would you want to rip them? It is hardly any effort to simply play the CDs when you want to listen to music.

Future media center control (my thought process anyway).

So the best format to rip to using Windows Media Player and not caring about HDD space is .................... ?


MP3? no

WMA? depends?

WMA Pro? depends?

WMA VBR? depends?

WMA Lossless? yes?

WAV? yes?

If you have any scratched/ damaged cds Use Exact Audio Copy instead of media player, I'm using it as we speak. I gotten clean rips on some older cds which were so scratched up I almost threw them away, takes a while though. You can read about it over at Hydrogen Audio.

As far as lossless use what ever floats your boat. I'm not sure which one keeps the metadata the best though. I played around with Wav and had to reenter all the data again when I converted them to MP3 with Foobar2000(good program, better than Media Player as well), could be me though I just started getting serious about ripping all my stuff.

Why would you want to rip them? Two words. Squeezebox Duet. :) Works with FLAC, WMA lossless, Apple lossless and many others.

http://www.slimdevices.com/pi_duet.html

http://www.slimdevices.com/images/duetgallery/1.jpg

Two words. Squeezebox Duet. :) Works with FLAC, WMA lossless, Apple lossless and many others.


So? Sorry, I fail to be impressed by gadgets like that.

So? Sorry, I fail to be impressed by gadgets like that.
Well, different strokes. Having instant access to any song on 1000 CDs without having to get up and root through all the CD cases impresses the hell out of me. :)

The playback quality is better too, especially with scratched CDs. Ripping programs like EAC or dB Poweramp do a pretty good job of reading damaged disks that your CD player would have a problem with. And the jitter of the Duet is 50 ps, better than all but the most expensive CD players. Ripping will even remove jitter that was on the original disk -- happened a lot in the past with certain pressing plants.

Why would you want to rip them? It is hardly any effort to simply play the CDs when you want to listen to music.Sure it is. Why even have to get up to touch the CDs or a CD player in the first place, when you can have your entire CD collection on a hard drive at exactly the same quality (if you wish) as the CDs?


So? Sorry, I fail to be impressed by gadgets like that.As evidenced by your previous question, you are clueless. These "gadgets" can give you iimmediate access to every single song on every single CD you own, at the same CD quality, at the touch of a button, with the artwork as well as improved track information. So you can put your comparatively cumbersome CDs in the attic.

What is the best format to save your music from a CD to a PC? MP3 (if so, what bitrate?) WMA, WMA lossless? I know the lossless are suppose to be the best, but does the amount of HDD space it takes up justify it for the average person? I should mention that the PC and is hooked up to a Yamaha YSP-1 Digital Sound projector (one component that produces 5.1 surround sound). My company sells these, so customers would be listening to the music from the PC. The connection is with a SPDIF optical cable. I just want a format that would take advantage of the sound system without unnecessarily wasting HDD space. thanks for any advice.

It seems like threads like these get off on tangents not really related to the OP's question. Within a handful of posts we are recommending lossless rips and spending $200 on 1TB external HD's.

The question was, for the "average person" listening through a PC and a Yamaha YSP-1, is the amount of HDD space it takes up to go lossless worth it for the improvements. And the answer to that is a resounding no!

Even if it's not about the money -- and $200 for a 1TB external drive just for that last drop of incremental improvement is not small potatoes for most folks -- for the vast majority of people listening on a non-audiophile setup (i.e. 99% of folks) you can rip it to MP3's at 192VBR (or even less!) at less than 10MB per song and they will get 100% enjoyment.

Again: "I just want a format that would take advantage of the sound system without unnecessarily wasting HDD space."

Do you really think he should be telling his customers to go buy an external HDD because their shiny new laptop with a 160GB HDD isn't going to cut it?

(BTW - before anyone lashes out, I'm not saying any of you are *wrong* about FLAC, lossless, lame, whatever -- well, except Mr. Audio :D -- just that it's not the best advice for the OP's specific situation)

EDIT: on a related note to the OP's question, MP3 is nice because it is such a widely-accepted industry standard and will play on virtually anything. that will give you maximum portability and compatibility for future hardware

It seems like threads like these get off on tangents not really related to the OP's question. Within a handful of posts we are recommending lossless rips and spending $200 on 1TB external HD's.

The question was, for the "average person" listening through a PC and a Yamaha YSP-1, is the amount of HDD space it takes up to go lossless worth it for the improvements. And the answer to that is a resounding no!

Even if it's not about the money -- and $200 for a 1TB external drive just for that last drop of incremental improvement is not small potatoes for most folks -- for the vast majority of people listening on a non-audiophile setup (i.e. 99% of folks) you can rip it to MP3's at 192VBR (or even less!) at less than 10MB per song and they will get 100% enjoyment.

Again: "I just want a format that would take advantage of the sound system without unnecessarily wasting HDD space."

Do you really think he should be telling his customers to go buy an external HDD because their shiny new laptop with a 160GB HDD isn't going to cut it?

(BTW - before anyone lashes out, I'm not saying any of you are *wrong* about FLAC, lossless, lame, whatever -- well, except Mr. Audio :D -- just that it's not the best advice for the OP's specific situation)

EDIT: on a related note to the OP's question, MP3 is nice because it is such a widely-accepted industry standard and will play on virtually anything. that will give you maximum portability and compatibility for future hardware

While I sort of agree with you on principle HOWEVER I have to wonder if it is thoughts/actions like this that have started the so called "loudness wars" which caused the overall quality of newly recorded material to suck.
Rip to flac and you never have to rip again most devices will transcode to mp3 on the fly . Once the information in music is "downgraded" to mp3 its gone forever, if you get better equipment and you notice a loss of sound quality in your mp3 than oops gotta rip again

Unfortunately, you are incorrect Mr. Audio. I suggest you make the bias controlled listening tests for yourself rather than criticize mine.

Biased? It is in fact a FACT that any MP3 no matter how good the quality will not sound as good as a wav. If you think it sounds close enough, congratulations, you save a whole bunch of HDD space. Yayyyy. Just because you can't tell doesn't make me incorrect. It just makes you less picky.

This has to be a put-on, but I enjoyed it.

Did you enjoy that? Now go try it and prove me wrong.

Do you really think he should be telling his customers to go buy an external HDD because their shiny new laptop with a 160GB HDD isn't going to cut it? 160GB will hold something like 640 CDs in FLAC format. Anyone with that many CDs would probably appreciate the advantages of lossless compression. Even if he couldn't hear the difference right now, he'd probably want to go lossless because he'd never ever want to rip that many CDs again if he changed his mind in the future. :)

FLAC to MP3 can be done as a single batch conversion -- tell the computer to do it and let it grind for a few hours. MP3 to FLAC, no point, the fine detail has already been lost -- rip them all again.

Go with FLAC.

Why would you want to rip them? It is hardly any effort to simply play the CDs when you want to listen to music.

Are you the forum donkey ??

Have you read about full house audio systems. I can click a album, song, artist, playist and play it in a room through cool little touch screens I have in every room. Overall, I can play up to 4 songs simultaneously throughout my house or share in any song. The ONLY way to do this is with ripped music.

I can understand that you do not think like this because you live in a one bedroom "pad" all by yourself but may I suggest you for once think less about your needs and try to understand the possibilities that exists


btw, to the OP Will is right if space is UNLIMITED go with FLAC!!

Did you enjoy that? Now go try it and prove me wrong.

Hey, if it sounds better to you, I sure can't disprove that. The resulting files can't be any more accurate than the originals though.

Hey, if it sounds better to you, I sure can't disprove that. The resulting files can't be any more accurate than the originals though.

Not true. Anything can be improved on. Don't think it's possible? There are DACs that up-sample original 44.1kHz to 192kHz. There are TVs that turn 24 frame movies into 48 frames. There are DVD players that turn 480i resolution into 1080p. How is this done? Interpolation. Yes, it's not real, it's artificial but still it's effects are desirable because it brings the original content closer to where we want it to be. I don't find 120Hz mode on TVs to be a good thing because I think it makes things look like a cheap soap opera on daytime TV. Nevertheless it is quite impressive that they took 24 frames and turned it into double that with no additional frame information. Ofcourse when you artificially up-sample anything it is foolish to think it's just as good as HD. There is no replacement for DVD-Audio or SACD, but they are gone now and interpolation is all that is left. No your CDs won't sound like DVD-A, but they will sound closer, and I'll take that over being a purist who adds expensive tube amps and buys overly "warm" speakers to aid the problem. Audacity is a free program. I challenge you to try it and give it an honest run. It will take not even a minute to download and install. Record a song at the highest bit rate and sample frequency and compare them. Just try it, it's free.

Do you really think he should be telling his customers to go buy an external HDD because their shiny new laptop with a 160GB HDD isn't going to cut it?


here is the thing, if you are going to do it once, then do it right. I have tons of MP3s actually and its my choice format because I use Ipods, etc and space is a permium, 40Gig isnt much!

Again, if space is a permium then MP3 is the format to rip the stuff into but down the road once you listen to lossless you seem to want everything else to be lossless and you find yourself re-ripping more then 10 thousand songs. ;)

The better the system the more you can hear difference too....Ipod and headphones, no issue...in house system?? You can hear the flaws in those MP3s. Im okay with it in the end but sometimes to truely test my system I like the best recording possible.

But we're not talking about YOU, or ME, or even most AVS'ers. We're talking about the OP's original post, which is for his customers listening from a PC connected to a Yamaha soundbar. He didn't specify, but it didn't strike me that he's selling to a bunch of obsessive audiophile types like us AVS geeks. That's all I'm saying. How many people are actually "critical listeners"? How many people have ears/systems that can really tell the difference between an 8mb, 256kb VBR mp3 and a 50mb lossless copy? How many people have 10,000 songs?

We live in a world of folks happily enjoying their mp3's on ipods at 128kb and plugging into their stereos through the headphone jack, things that would be heresy to most of us but these people are 100% satisfied.

batpig, I posted both options but if space is unlimited there is no reason not to rip in lossless. The user does not care what format it is period and there is no cost difference, why not? If space is limited then just rip into MP3, That is what I posted. No reason not to rip in the better format if a person can do it! I do not know how this can be even argued either.....ripping software is easy, songs honestly do not take up that much space period and you do not know the size of his collection.

Trust me, I have all the mp3s, even 128bit ones that I still listen too I definitely not a critical listening type person, I actually run fast away from anything setup like that too.


In the end we can only post ALL OPTIONS available and its the OP that can decide which is right not you and not me. The difference is that Im posting the options, you are arguing.

Honestly, if its a company and its for demos. HE BETTER rip to lossless, why risk something sound like crap to a potential customer?

Geesh, This is simply a no-brainer. I run my own company and what ever we can do to sell our company we have to do it 150%.

He is selling AV stuff too so some of his customer WILL know that he has crappy MP3 128bit music playing and if it was me I would think his business is kind of a joke if they can not even rip lossless audio. Sorry but those things do matter in a demo!

Biased? It is in fact a FACT that any MP3 no matter how good the quality will not sound as good as a wav..... etc.

I suggest you re-read the post you are responding to, and start over again from scratch. hint: biased vs. bias-controlled test. different context. and have your corrections on my desk by 5, thanks.

I suggest you re-read the post you are responding to, and start over again from scratch. hint: biased vs. bias-controlled test. different context. and have your corrections on my desk by 5, thanks.

I suggest you let the original poster of the post that I was responding to speak for himself. hint: no need for bias-controlled testing since a 320 MP3 is factually inferior to a WAV file. same context. and have your desk cleaned out by 5, that's my seat you're sittin' in.

Not true. Anything can be improved on. Don't think it's possible? There are DACs that up-sample original 44.1kHz to 192kHz. There are TVs that turn 24 frame movies into 48 frames. There are DVD players that turn 480i resolution into 1080p. How is this done? Interpolation.

One thing to be aware of is that while video upconversion employs a number of tricks which modify the content (e.g. edge enhancement, etc), audio upsampling aims to preserve the original frequency content as exactly as possible. Upsampling DACs, for example, use a method called zero-stuffing which produces a signal with spectrum identical to the original (plus images).

I've actually done a number of experiments myself, when I was trying to decide how to rip my own music; my conclusions differ from yours though.

OP, if you're still around, my opinion is that high-bitrate mp3 should be audibly fine (I rip in lossless mostly for archival purposes), but I also agree with the poster who said lossless might be a good thing for marketing purposes.

my opinion is that high-bitrate mp3 should be audibly fine (I rip in lossless mostly for archival purposes)

since you've done the experiments, out of curiosity at what bitrates do you consider the results "audibly fine"?

since you've done the experiments, out of curiosity at what bitrates do you consider the results "audibly fine"?

I was able to do some true ABX testing using my Squeezebox and a Perl script which generated playlists of uncompressed and compressed mp3s. I tried to use recording which brought out the flaws of mp3 as I knew them, e.g. recordings with a lot of cymbals, transients, etc.

I felt that the 'knee' in quality truly is around 128kb; differences tend to be noticeable there, but if you don't have a lot of high-frequency content in the recording you may not lose much. As I moved up, I found that there are definitely recordings where I can't reliably tell the difference between 192kb mp3 and lossless. Some recordings lose more than others. I'd be impressed to meet anyone who could tell the difference between 320kb and lossless, reliably, in an ABX setting. Or if someone can point me to a clip to use which brings it out, that'd be pretty cool too.

I suggest you let the original poster of the post that I was responding to speak for himself. hint: no need for bias-controlled testing since a 320 MP3 is factually inferior to a WAV file. same context. and have your desk cleaned out by 5, that's my seat you're sittin' in.

320 MP3 is technically inferior to WAV, agreed, but the more important question is whether or not the difference in quality is truly discernible to human ears. The answer is far from clear on that front. That's where the previous suggestion for bias-controlled testing comes in. enjoyed your snappy retort though. well played.

One thing to be aware of is that while video upconversion employs a number of tricks which modify the content (e.g. edge enhancement, etc), audio upsampling aims to preserve the original frequency content as exactly as possible. Upsampling DACs, for example, use a method called zero-stuffing which produces a signal with spectrum identical to the original (plus images).

I've actually done a number of experiments myself, when I was trying to decide how to rip my own music; my conclusions differ from yours though.


I have read a few articles on this subject that aim to disprove up-sampling. Honestly I tried this experiment on a whim. Before I read anything on it I was just wanting a program that could cut and paste wavs for a project I'm working on. I ran into Audacity and played around with it's many features. I decided to just toy with the sampling frequency and bit rate just to see what would happen. I remember reading reviews on a dual cd recorder deck from Sony that claimed to record at 96kHz/24bit and most of the reviewers were astonished that their recordings came out better than the original. They didn't understand why and I certainly thought they were hearing things. However I discovered after listening to what I just recorded on Audacity for grins that there was an audible difference and it was a good difference. I then started looking up articles to find an explanation why. Most of the articles were on external real time DACs because there isn't much information about Audacity in the home music world. However I did find that Audacity is a highly regarded sound editor due to it's list of features, interface friendliness, and it's ability to up and down sample. The program allows you to see the waveform and soom in to see the sample markers and you can put the original 44.1kHz up to the up-sampled recording and see how the jagged edges have been eliminated and the waveform actually appears to be a lot more of an analog waveform than it's 44.1/16 counterpart. Not only does it look more natural but sounds more natural. I can send you a sample if you'd like.

320 MP3 is technically inferior to WAV, agreed, but the more important question is whether or not the difference in quality is truly discernible to human ears. The answer is far from clear on that front. That's where the previous suggestion for bias-controlled testing comes in. enjoyed your snappy retort though. well played.

While I see your point that most would not even notice or care because they are pretty close, the best MP3 just doesn't have the little extra bite and soundstage that a WAV does. I find it extremely noticeable. I would be able to pick them out in a controlled test with a good pair of speakers for sure, side by side. However the point I think you're trying to make is that most likely nobody would be able to tell if it was a high grade MP3 or a WAV without having them side by side. That is probably true especially if you're playing music the customer has never heard before and most people don't take hours to listen to many types of music on speakers or a system they're thinking about buying like I do. You got me there.

I would be able to pick them out in a controlled test with a good pair of speakers for sure, side by side.

Have you actually done it, or are you just assuming that it's the case? Because it's pretty amazing what happens when you set up a bias-controlled blind listening test. You may be surprised.

Have you actually done it, or are you just assuming that it's the case? Because it's pretty amazing what happens when you set up a bias-controlled blind listening test. You may be surprised.

No I have not done bias controlled testing on this matter. However I think of it this way. I want the one that takes up less space to win. So I would already be sort of biased already in the MP3's favor. Sometimes that kind of bias would possibly cloud your judgment as well. So you have the bias of the fact that the WAV should sound better, even just a little vs. the bias that the MP3 is significantly smaller therefore not needing to buy an extra HDD which saves money. I believe you can achieve the sound judgment needed without blindfolds and a jury of people if you are completely honest with yourself. Plus, if you can't hear the difference and there is actually one you just cannot hear it or can hear it but think the difference is negligible, that's fine too. You save money and space. In the end, you're the only one you have to prove it to anyway. However in the case of a demo of speakers and amps, I do seriously doubt that you'll lose a customer because they can detect that you're playing a high quality MP3 vs. a WAV. I don't think anyone would be able to unless they heard the WAV side by side. a 192 or 128, most people would be able to pick that out especially if they know the material you're playing.

So there are biases on both sides. Most of the time they cancel each other out. It really matters how honest you are with yourself. If you can't do that you'll need to call your wife or your best friend into your geek cave and put a blindfold on.

So you have the bias of the fact that the WAV should sound better, even just a little vs. the bias that the MP3 is significantly smaller therefore not needing to buy an extra HDD which saves money. I believe you can achieve the sound judgment needed without blindfolds and a jury of people if you are completely honest with yourself.

That would all be fine and dandy if the human brain was designed to "be honest with yourself", but unfortunately (well, fortunately for speaker cable manufacturers! :rolleyes:) the human brain has many functional "self-deceptions" that are essentially built in. We have an amazing ability to engage in all sorts of self-rationalization that can cloud our judgment.

Your logic about how biases should "cancel out", and your belief that you can defeat your own perceptual biases by "being honest with yourself", sounds reasonable, but is flat-out contradicted by actual empirical testing and our knowledge of human cognition.

Again, just try the tests, you may be surprised.

EAC + lame and VBR (variable bit rate) set on the high end.

That is where the point of optimum quality versus file size exists. Even if harddrive space is cheap, why bog it up with wav files that are unnecessary and nearly indiscernible from the mp3s?

My input. rip to flac. You then can make mp3's with those if you choose. My only suggestion is if you go mp3, 320 variable bit rate. that is all..
PS: have a squeezebox and some nice gear, sounds awesome ( I stream flac...i can tell the difference )
PSS: Hey Batpig!

Even if harddrive space is cheap, why bog it up with wav files that are unnecessary and nearly indiscernible from the mp3s?

Because even at 384, MP3 still isn't quite the best, but adequate.
192 is the minimum bitrate for those that "don't care" and want to crunch 50lbs of crap into a 5lb. bag.:D

It boils down to convenience vs. quality. Your choice.;)

The program allows you to see the waveform and soom in to see the sample markers and you can put the original 44.1kHz up to the up-sampled recording and see how the jagged edges have been eliminated and the waveform actually appears to be a lot more of an analog waveform than it's 44.1/16 counterpart. Not only does it look more natural but sounds more natural. I can send you a sample if you'd like.

No, I understand how upsampling works; oversampling DACs work well. But what's going on in your signal chain appears to be this:

1. Start with a 44.1kHz audio file
2. Upsample to whatever
3. Send upsampled data to DAC
4. DAC again upsamples/oversamples to whatever it uses internally

As I see it, the only thing you get by upsampling in two steps is some small amount of additional round-off error. If there were an audible difference, I'd guess there were an interaction with something else in the signal chain, or some problem that this works around in Windows or something like that.

Because even at 384, MP3 still isn't quite the best, but adequate.
192 is the minimum bitrate for those that "don't care" and want to crunch 50lbs of crap into a 5lb. bag.:D

It boils down to convenience vs. quality. Your choice.;)

but its not crap. The analogy is like buying a BMW with or without the sport package. The base model is still sufficient.

It boils down to convenience vs. quality. Your choice.;)

I don't think that's an accurate assessment of the tradeoff, at least at higher bitrates. If it's not audible discernible, are you actually sacrificing quality for that convenience?

192 is the minimum bitrate for those that "don't care" and want to crunch 50lbs of crap into a 5lb. bag.:D192 VBR is usually what is cited. Jus' sayin', Ratman. No offense. Still inferior to lossless, AFAIC. ;)

EAC + lame and VBR (variable bit rate) set on the high end.

That is where the point of optimum quality versus file size exists. Even if harddrive space is cheap, why bog it up with wav files that are unnecessary and nearly indiscernible from the mp3s?

I agree, I find it comical someone could consistently(if ever at all) tell the difference between lame 320 or v0 256 vbr and lossless. Unless you really want to archive in lossless and have the space or you may want change formats down the road, I personally don't see the need for lossless over mp3. But thats just me.

VBR or not... MP3 or any other codec using compression algorithms is a loss of quality.

I personally don't care what any individual chooses to use but, the facts are the facts. It's not a "like for like" copy. ;)

Nobody is claiming MP3's are a "like for like" copy as you put it, just that there is no discernible(they are transparent) difference between 320 cbr, v0 256 vbr and any lossless codec :)

No, I understand how upsampling works; oversampling DACs work well. But what's going on in your signal chain appears to be this:

1. Start with a 44.1kHz audio file
2. Upsample to whatever
3. Send upsampled data to DAC
4. DAC again upsamples/oversamples to whatever it uses internally

As I see it, the only thing you get by upsampling in two steps is some small amount of additional round-off error. If there were an audible difference, I'd guess there were an interaction with something else in the signal chain, or some problem that this works around in Windows or something like that.

Hearing is believing. I just did a bias controlled test with my brother tonight who is not into sound like I am. I told him to listen to both versions of the song. I didn't tell him which one should sound better, I just asked him if there was a difference between the two. He not only detected the difference right away, but said that one of them was a lot clearer. He could hardly explain the improvement because it wasn't one thing like crisper highs, it was just more open and clearer sounding. Now my brother has told me on many occasions when I have asked him to listen to something that he can't hear what I heard. He just really doesn't care most of the time. His system is a poorly calibrated cheap Sony receiver with mix and matched speakers. He clearly doesn't care about quality, just that it's kinda loud and kinda booms. Not tonight though.

Try to reason it all out all you want why it can't sound better, it still won't change the fact that it does. However if you don't wanna try it out for yourself it's no skin off my back. I really don't care if everyone on this forum thinks I'm hearing things, I'll just have to enjoy this all by myself. It's stupid if you don't try it, it's totally free and takes less than 5 minutes to convert a song. I mean wouldn't you love for your music to sound better?

Mr. Audio, I've heard lots of 96kHz audio, and I've tried recording at different sample rates, and bit depths and I know what I heard. What's the rest of your signal chain after you upsample?

What's the rest of your signal chain after you upsample?

I'm not sure I know what you mean.

I've bias-controlled tested digital sound formats until I'm blue in the face. The fact is people can't hear any difference between a 320 MP3 and a WAV file. It doesn't matter what music it is or how much high frequency content it has or who is doing the listening. 320kbps compression doesn't remove anything meaningfully audible. You can sum out the files and listen to only what was removed. It sounds like a barely audible breeze right at about the internal noise level of a microphone.

There simply isn't any point to using anything else because MP3 is THE industry standard. Almost every CD player plays MP3 files. Almost no CD player plays FLAC file. If you rip at a high bit rate, you lose data but no sound. You can have terrabytes of file space but the MP3 is still the industry standard. Audiophiles need to get past the fear mongering that hits them from every direction.

I'm not sure I know what you mean.

What do you use to play the files back: software, DAC/soundcard, amps/receiver, etc.

What do you use to play the files back: software, DAC/soundcard, amps/receiver, etc.

Everything from Audacity itself, Windows Media Player 11 after it's been exported to a wave with the integrated sound card of which I cannot remember the name, my audio system in my house and cd player in my car after I've burned it to a CD. My home system is a Sony DVP-NS75H DVD player hooked optically into my Yamaha RXV-659.

In which setup do you hear a difference? When listening on the PC, or burned to CD on the home or car setup?

Hearing is believing.

Umm, no.

In which setup do you hear a difference? When listening on the PC, or burned to CD on the home or car setup?

All of them.

Umm, no.

Back off scientest. If numbers explained everything, there would be nobody claiming in this thread that MP3s sound just as good as WAVs. I know you don't like using your ears because you don't trust them, so perhaps you should hook your amplifier's outputs up to to multimeters and oscilloscopes and take in your music and movies that way. You could totally geek out writing down every measurement and adding it all up and having a nice mental chart of what everything should sound like. Or you could use your ears and say, "I like the way that sounds." or "I don't like the way that sounds." Option B would be more fun, but you don't seem to be interested in having any other fun other than telling people they are wrong because they use their ears to determine what they like. Hey, whatever floats your boat. If being a pest is what makes you smile I guess you should just stick with what you're good at.

Back off scientest.

Um, no...

Do you think only your opinions should be considered?

Have you ever had your hearing evaluated?

Have you ever done any psychoacoustic tests?

If you think you can "believe" what you hear you don't understand how your hearing works.

Sorry that upsets you. If it does, I'd suggest you not post your opinions on a public forum.

Edit: and yes, I like using my ears.

Edit again: you are making some rather stupid assumptions about what I believe, you obviously know nothing about me.

the best MP3 just doesn't have the little extra bite and soundstage that a WAV does.

Since when do 0's and 1's bite people and have soundstage? :rolleyes:

When you use digital "manipulation" of the original format for the sake of storage space. ;)

There's always good, better, best. Like for like is best IMO.

Since when do 0's and 1's bite people and have soundstage? :rolleyes:

I dunno. Perhaps I could give you an answer if I was actually listening to 1's and 0's, but I happen to listen to sound, not numbers. :rolleyes:

All of them.

Seriously? :confused: You've got to realize that the burned CD's are once again 44.1kHz audio, so sample rate has nothing to do with any difference in sound.

At best, you're introducing some kind of distortion in the process of upsampling your rips to to 96kHz and then downsampling to 44.1kHz when you burn. I seriously doubt that there's any audible difference. If you want to save some space, rip your burned CD's again and store that data with a lossless codec...

Seriously? :confused: You've got to realize that the burned CD's are once again 44.1kHz audio, so sample rate has nothing to do with any difference in sound.

At best, you're introducing some kind of distortion in the process of upsampling your rips to to 96kHz and then downsampling to 44.1kHz when you burn. I seriously doubt that there's any audible difference. If you want to save some space, rip your burned CD's again and store that data with a lossless codec...

Hey man look. I'm not trying to prove anything to you. You asked. I don't need to prove it to anyone else because I already proved it to myself. Why should I spend tons of energy typing and getting into a useless argument when you could hear it out for yourself with no money spent and hardly any time spent. I don't have all the answers why yet. I'm getting closer to the reasons why but I don't have them all yet. When I have them, then I'll know and be happy. In the mean time and thereafter you can continue to think my brain is tricking my ears to hear something different that makes the music sound better to me. You certainly have a lot of people studying in your corner to prove that the human body and brain as a whole is a worthless imperfect instrument that can never be trusted with any type of judgment. The bottom line is what you think sounds amazing may sound no different than a Sony HTIB to someone else simply because they just don't care to hear a difference if there is one. I challenge you one last time to try it and hear it for yourself instead of trying to convince me that I'm all wet, that's all I'm saying. I don't wanna argue with you, try it if you want or don't try it, it's up to you.

Do you think only your opinions should be considered?
Never said such a thing. Who's making assumptions again? I'm confused.
Have you ever had your hearing evaluated?
No I have not. However this is not even relevant since all my tests are relative. Even a person with 20/100 vision can see the difference between standard and high definition when put side to side. A person who has average hearing loss can still hear which sound is clearer than the other.
Have you ever done any psychoacoustic tests?
No, but how does this disqualify me from having an opinion?
If you think you can "believe" what you hear you don't understand how your hearing works. Ok either you're really reaching for an argument here or you really don't understand the phrase I used. Either way you're incorrect. Let's take your translation which would be that you can't hear something that you wanna hear and don't hear things you don't wanna hear. Is that not the root of psycho acoustics? Now my translation. Hearing is believing meaning that hear the difference between the two recordings and you will believe that improving CD audio is possible or even not possible. You will believe something either way. I believe right now either way that you're wrong.:D

Hey man look. I'm not trying to prove anything to you. ... I challenge you one last time to try it and hear it for yourself instead of trying to convince me that I'm all wet, that's all I'm saying. I don't wanna argue with you, try it if you want or don't try it, it's up to you.

The thing is, this came up when the OP asked what format to rip in. I wouldn't have made much fuss, were it not for the suggestion that he upsample to 96kHz and roughly double his disk usage. Since you still hear a difference with the downsampled 44.1 audio, it seems clear that the sample rate doesn't explain the difference you're hearing. This seemed important in a discussion about bit rates.

As for whether there's an audible difference, if you can make two clips available, I'll take a listen. Keep in mind I've already said I sometimes can't tell the difference between 192kbs mp3 and lossless. Can you put a couple of short clips on a webserver or make them downloadable somewhere?

What program(s) is everyone using to rip FLAC files?

I'm not usng a program, per se. Instead, I am using my Escient MX-111 to rip my CDs to FLAC. However, I think the Escient is first converting my copy-protected CDs to analog, and digitizing the analog to FLAC (it seems to do this to all of my CDs). I think the audio quality of the rips do not sound nearly as good as my original CDs. I'm thinking of trying a CD ripping software program. I've looked into Exact Audio Copy, but it appears to provide the user with a great many options. I don't think I want to deal with that many options. Plus, it requires external programs to do the file conversion, like LAME, etc.

Therefore, I am thinking of using something like Easy CD-DA Extractor as it has received good reviews, has fewer options, frips to FLAC, and is self-contained (See http://www.download.com/Easy-CD-DA-Extractor/3000-2140_4-10018859.html?tag=mncol;pop&cdlPid=10853169 for reviews).

Does anyone have any experience with this program? Will this create a file/files which I can then directly import into my MX-111?

Never said such a thing. Who's making assumptions again? I'm confused.

I made a two word statement, you're immediate response was "Back off". seems to me you don't want to hear any opinions contrary to your own beliefs. That would also be confirmed by your response:

"I don't need to prove it to anyone else because I already proved it to myself."

Basically, as far as I can tell you're not very open minded. You've made a statement about something that is rather strange and now you're leaping at anyone who even questions it in the least.

I actually believe, that depending on how you've arranged your signal chain, you may have heard a real difference. I won't dispute that you believe one version is "better" than the other. I will however question whether any of your observations are applicable to the rest of the human populace.

Basically, if you are going to post something here, then you should be prepared to "prove it to anyone else" who wishes to understand what you are going on about.

No I have not. However this is not even relevant since all my tests are relative. Even a person with 20/100 vision can see the difference between standard and high definition when put side to side. A person who has average hearing loss can still hear which sound is clearer than the other.

Again, no. To expand on your analogy; if you have red green blindness then your statements as to the color accuracy of any given reproduction are useless for 93% of the population or so (and probably don't mean much to the remaining 7% or so). Similarly, if you have deficiencies in your hearing than any statements you might make audio reproduction process are questionable. You cannot make relative judgments if you do not have an accurate base from which to start. Without a basic hearing test you don't even know if you can actually hear the full dynamic range of a CD or the entire frequency range of a CD least of all any higher bit rate encoding. Eg. if you're a male over 30 years old I'd be willing to bet you dollars to donuts that you can't hear anything over 16kHz. As such a statement like "this version had more detail" tells us absolutely nothing without two common references: 1) what does "detail" mean; and 2) how well can your hearing discriminate between deviations from this first reference.

No, but how does this disqualify me from having an opinion?

You're welcome to your opinion. However, if you haven't participated in any psychoacoustic testing you should be open to explanations as to what you are hearing that don't require suspension of belief in basic physics and biology on the behalf of the rest of the world. In particular, you may want to pay careful attention to the people that suggest you repeat your tests in as a double blind ABX test. At this point you have a "belief" that you know what's going on, but you have done nothing that confirms your belief correlates to the experience of the rest of the world. As a result your posting of your belief here leaves you open to ridicule by those who believe that such tests are necessary.

Ok either you're really reaching for an argument here or you really don't understand the phrase I used. Either way you're incorrect. Let's take your translation which would be that you can't hear something that you wanna hear and don't hear things you don't wanna hear. Is that not the root of psycho acoustics? Now my translation. Hearing is believing meaning that hear the difference between the two recordings and you will believe that improving CD audio is possible or even not possible. You will believe something either way. I believe right now either way that you're wrong.:D

This is a hard paragraph to parse. However, I think you're basically assuming that the human auditory system is not subject to illusions?

Most moderately educated people have a basic understanding of optical illusions. If you're more widely read you may be familiar with tests into witness reliability. For example, there's a test video of a crime scene in which an actor in a gorilla suit runs across the background. The majority of people that view this video do not notice the gorilla. What people seem to be unaware of is that the same kinds of things go on with the auditory system but to a far greater degree. Partially this is because we are simply not trained to listen for such things. Partly this is because the parts of the brain associated with auditory processing are more primitive than those associated with vision.

I've pointed at the research on psychoacoustics testing elsewhere on this board so I'd rather not clutter up this thread with it again but the bottom line is that our auditory system lies to us all the time. At the level sorta equivalent to a simple optical illusion we fill in missing harmonics and fundamentals, compensate for pitch and phase skew and miss micro timing differences. At the level of detail recall (ie; the gorilla suit) most people have good memories for the human voice and familiar melodies to the point that their expectations cause them to miss deviations in reproductions from these standards. There's much more to it, but the bottom line is that this expectation bias leads people into believing they have heard things when, in fact, the parts of the brain that respond to auditory stimuli are in fact not even activated at all.

Now at this point, you are free to say "so what, I'm happy with what I'm hearing" and you are perfectly welcome to do so. What you shouldn't do is assume that anyone else in the world will share your experience or your beliefs. Until you can demonstrate a difference that is reliably detectable by a large portion of the human populace there is simply reason to expect the rest of us to participate in your "tweak" or even understand why you might be happy with it.

At best, you're introducing some kind of distortion in the process of upsampling your rips to to 96kHz and then downsampling to 44.1kHz when you burn.

Depending on his hardware he should be able to decode directly from 96kHz to analog without requiring that he down sample. If he's upconverting in the digital domain then there certainly is some requirement for interpolation between the sample rates that would leave room for errors. If instead he's converting to analog and then sampling to 96kHz who knows what he's doing to the signal. Either way, the differences could be audible (almost certainly so in the second case) and may in fact sound "better" to him, even if they are not accurate representations of the original source.

Edit: reading back it appears he's doing it with Audacity in the digital domain. It's a good program but who knows how it would do interpolation.

Now at this point, you are free to say "so what, I'm happy with what I'm hearing" and you are perfectly welcome to do so. What you shouldn't do is assume that anyone else in the world will share your experience or your beliefs. Until you can demonstrate a difference that is reliably detectable by a large portion of the human populace there is simply reason to expect the rest of us to participate in your "tweak" or even understand why you might be happy with it.

Your points in your last post were well stated in unlike your first post. If you didn't mean it to be sarcastic I apologize but remember, we're just typing here not talking face to face.

Anyway I fully understand that I left myself open for ridicule even mentioning that such a thing like improving an audio CD is possible, and for free at that. I really don't care about the ridiculing because I was basically throwing it out there and see if anyone would bite. I have found no post of anyone who had tried this probably because they thought it wasn't going to do anything but waste their time. I tried it because I just like to play around with things and see what effect it would have and it would rob of only 30 seconds of my life so it wasn't too costly. I thought like everyone else that it would do nothing and possibly F up the file. I took an honest listen and found there was a difference that was good. I then tried it with several songs and then burned it to CD. When I was convinced that I had found something better I looked around to see if anyone else had found it first. Nobody.

Now Audacity even claims that true record high resolution audio recordings such as DVD-A and SACD that the differences between those and the CD many people honest still cannot hear. As for SpectraID, he possibly will not be able to hear the difference between the original and the modified that I may send him because he claims to not be able to hear a difference between a 192 MP3 and a WAV. Does that make him wrong. No, unless he says "There is factually no difference between a 192 MP3 and a WAV."

Quite possibly the reason why people can't tell between a more lossless format is because the majority of us seem to judge sound quality by how low the lows are, how high the highs are, and how warm the mids are. Most people don't notice the depth of sound that is lost as quality drops. Like poor quality video just seems dull and flat like a coarsely painted picture as where good quality is not only sharper, but actually appears to be more 3D. You can see more depth as sharpness, color fidelity, and gray scales improve. Audio works similarly. MP3s work well for most people because they still capture the meat of the information but leave small subtle information behind. This is not just high frequencies, but also overtones, ambient tones, identifying characteristics of voices and instruments and not to mention volume of each instrument and singer mashes closer and closer together helping them become more like one unified sound other than separate sounds. This is why a lot of people can't tell between even true HD audio. There is a thread on this forum that says "Why does Dolby TrueHD suck?" That guy can't hear a difference because he is looking for something in the sound to jump out and tackle him. That's not how it works. The difference between 8 track tapes and cassette was huge. The difference between cassette was a quantum leap. The difference between CD and SACD is not nearly as huge of a step up in relation to the other two. That's what turns most people off. Most wouldn't buy a new game console that had graphics just a little better than the last one. No. The majority of us wanna see a big change. The jump to HDTV was big. The jump to 1080p from 720p not big. Popular though because of extremely aggressive sales and marketing kinda like Bose.:D All this is really tied in with money. Do you wanna pay big money for a small change? Depends on how bad you want it because it usually costs. However this thing I did with Audacity does not cost a penny and my overall point was try it you have nothing to lose. I'm not tell you to buy a 3000 dollar CD player. At that price the difference better be big. :eek:

I'm sorry if it appeared that I was trying to shove a "belief" down someones throat. That is clearly not the case. I am an audiophile but I have serious restrictions on what I will do to achieve better sound. Paying big money to hear a nuances no. Paying no money to hear them, you bet your a$$.:D

The answers to why there is a positive difference audibly when you're takin' 44.1kHz/16bit and converting them to 96kHz/32bit and then back to 44.1kHz/16bit, I don't know for sure yet. I am taking to a guy right now in the Audacity forum about it to see if I can come up with a logical answer. For right now I know it works, but not sure how or why.

Your points in your last post were well stated in unlike your first post.

Ah, but my first post was well stated. It was just very succinct....;)

It was more of a test to see if you meant what you wrote or if you were simply being hasty in your writing. I certainly wasn't expecting the response I got, but no matter...

If you didn't mean it to be sarcastic I apologize but remember, we're just typing here not talking face to face.

Uh, it wasn't sarcastic, it was rude. However, it's "teh Interents", I could care less....

Anyway I fully understand that I left myself open for ridicule even mentioning that such a thing like improving an audio CD is possible, and for free at that. I really don't care about the ridiculing because I was basically throwing it out there and see if anyone would bite. I have found no post of anyone who had tried this probably because they thought it wasn't going to do anything but waste their time. I tried it because I just like to play around with things and see what effect it would have and it would rob of only 30 seconds of my life so it wasn't too costly. I thought like everyone else that it would do nothing and possibly F up the file. I took an honest listen and found there was a difference that was good. I then tried it with several songs and then burned it to CD. When I was convinced that I had found something better I looked around to see if anyone else had found it first. Nobody.

Thing is, at this point you really don't know if you've found something or not. You think you've found something audible, but that could just be expectation bias caused by the fact that you see nicer looking wave forms in Audacity. Or it could be differences in playback levels due to going though different decoders and processing chains. Until you perform a rigorous, level matched, double blind ABX test you really don't know to what extent you're onto something or not. Yeah, that may seem like a lot of song and dance but it's the only way to really know.

If you _really_ think there is a difference you owe it to yourself to figure this out. Nice thing is, since you're dealing with digital signals you don't even need to use an ABX switch box, this can done with only a PC using Foobar 2000 (I think there's also some standalone ABX players out there that you can probably find via Google). And yes, there may actually be some audible difference, that doesn't mean the 96/32 version is more accurate...

[QUOTE]Thing is, at this point you really don't know if you've found something or not. You think you've found something audible, but that could just be expectation bias caused by the fact that you see nicer looking wave forms in Audacity. Or it could be differences in playback levels due to going though different decoders and processing chains. Until you perform a rigorous, level matched, double blind ABX test you really don't know to what extent you're onto something or not. Yeah, that may seem like a lot of song and dance but it's the only way to really know.

Ahh, but the waveform when it's exported goes back to 44.1kHz/16bit. I know this because I took a look at it when I imported the WAV file I had just exported back to Audacity and even compared them against each other and the sample markers were just as spread out as they were on the original and the waveform had mostly subtle differences that were almost not noticeable enough to catch my eye. I was zoomed in however looking at the waves in milliseconds, so any change even small looking at a wave that closely will have to yield a different result. As far as DAC and processing, I notice a difference like I stated across the board. That was one of my tests. My computer is an integrated mediocre piece of crap that I wanna throw out the window most of the time because it does nothing exceptionally well except make me angry. I'm sure especially since Audacity is work strictly in the digital domain that my crappy integrated sound card can't be of any sort of credit to the results I'm getting. I mentioned this discovery on the forum for the reason you speak of. I need to do rigorous testing. I can call all my friends and family into my HT and ask them what they think, but most of them don't care about sound reproduction so that test would hold no weight. I thought since this is a relatively simple process and the software is free that others who do have the ability to do such testing would try it out and test it further than I can. I can't sell this idea because it would involve selling copy written material which is definitely illegal.
If you _really_ think there is a difference you owe it to yourself to figure this out. Nice thing is, since you're dealing with digital signals you don't even need to use an ABX switch box, this can done with only a PC using Foobar 2000 (I think there's also some standalone ABX players out there that you can probably find via Google).accurate... I am working hard on finding an answer to this because I really do wanna know why.
And yes, there may actually be some audible difference, that doesn't mean the 96/32 version is more accurate...
You may be right in one respect. No it is definitely distorted from the original waveform, I have seen the waveforms with my eyes. However audibly it comes out sounding more correct than the original. I have to end this post now because I gotta go to an appointment. Bye for now.

Like I was saying it comes out sounding more correct but not in the fashion that everyone thinks it should. It's not brighter and it's not warmer. The part where most people would have a tough time discerning the difference between the original and the modified would be that you will not hear something obvious right off the bat like more bass, more highs, sweeter more tamed mid-range. The reason why MP3s suffer from less high information is because the has been raped so much that it starts taking out the next smallest details like high frequencies in general. 320 doesn't take much from CD but 192 and 128 certainly do. Most think the smallest details are the highs only, but overtones and ambient sound are much smaller in amplitude and get skipped over first. This "improvement" that I think I have found is not sensitive to certain types of speakers or amplifiers. Because it's not a frequency "boost" of any kind, it can be played on any system and even my crap car system and sound better than the original material. So if you guys want me to somehow send you sample files I will. You can make it easier on your ears if you tell me what kind of music you prefer. I have all kinds. It doesn't matter the style or genre of music it is, you will still hear improvement. I'm not sure of the best way to get these files out there to you guys if you just wanna give me alittle help with that we can get this thing goin'.

Ahh, but the waveform when it's exported goes back to 44.1kHz/16bit.

So perhaps all you've done is run some expansion or compression on the original file? I'll do that on occasion with Audacity; if I run across something with horribly compressed dynamics I'll run some expansion to make them a tad more sane. You can't get it back to normal (since you don't have any reference point) but you can make it less fatiguing. I'm not perfectly clear on what you're actually doing, but it doesn't sound like it has much relevance to this thread...

I'm not perfectly clear on actually doing,

This is exactly what I'm doing. First make sure you are running Audacity V 1.2.6. Set the default bit rate to 32 bit, and make sure the High-Quality Sample Rate converter is set to High-Quality Sinc Interpolation and High-Quality Dither is set to Triangle in preferences first. Close and then reopen the program to initiate the new settings. Import any song ro Audacity that is WAV format. Change the sample rate to 96kHz. The waveform will become half it's previous size so you must slow it down. Then highlight the whole waveform and choose change speed under effects. Change the speed -53.990%. After that then go to choose to export as WAV and there you have it. Listen and compare it to the original on whatever program you use to play music on your computer. Disclaimer is if your speakers on your PC suck really bad you may not be able to hear a difference. My speakers are Logitech x240s so they're pretty good. If your speakers really suck on your PC just burn the files to CD and play it on your good stuff. Good luck.

but it doesn't sound like it has much relevance to this thread...

I didn't know my mention of this thing I've done would expand into what it has. Shall I start a new official "Improving CD audio" thread?

This is exactly what I'm doing. [...] Shall I start a new official "Improving CD audio" thread?

Looks like you ran an interpolation algorithm and redithered the original. Guess it could certainly sound different. No reason to think it is "better" and again you should test both in a double blind ABX; can you detect the difference and if so which sounds "better".

If you start you're own thread I wouldn't call it "Improving" since that's not necessarily what you are doing...

[QUOTE]Looks like you ran an interpolation algorithm and redithered the original.
I know that. It is definitely the interpolation and dithering that is making the difference. The thing that I'm still not sure about is why it still sounds just as good when I export to a WAV. Possibly since my sound card is cheap and integrated and certainly cannot reproduce sounds over a 44.1kHz/16bit ceiling, there is probably no difference audibly between the 16 bit and the 32 bit waveform because my sound card is most likely dumbing it back down to 16 bit when I play the raw version through Audacity. It would probably sound different if my sound card were more capable. If that is right then that would explain why it does sound I'll say "modified" to the original and sounds the same no matter where I play it.
Guess it could certainly sound different. No reason to think it is "better" and again you should test both in a double blind ABX; can you detect the difference and if so which sounds "better".
I'm working with spectra on this. I'm going to send him some clips and is gonna listen and analyze them through software to visually see the differences.
If you start you're own thread I wouldn't call it "Improving" since that's not necessarily what you are doing...

I still disagree with you on this one. If it does indeed sound better to the ear, you have made an improvement. Just like Audessey EQs and room correction doesn't actually correct the signal to make the system sound better, rather it does distort the signal quite a bit to even out being pushed through the room. CD audio blows. I think most of us know that. Whether you rerecord a song off a CD and EQ it to help even out it's short comings like when studios "remaster" recordings or artificially inject interpolated information that is pretty accurate to the signal's original intent before it was butchered down to 44.1kHz then you have still improved it if it does indeed sound better. Good interpolation results can be striking.

If it does indeed sound better to the ear, you have made an improvement.

Umm, no....

You want the lecture on expectation biases again? :D

Umm, no....

You want the lecture on expectation biases again? :D

No because we're never gonna see eye to eye on this so might as well call it a draw. I see it as CD audio is not accurate in the first place because it has already been tampered with when it is compressed. Artificially un-compressing by interpolation is no more wrong than the record companies compressing it. I'm not disagreeing with expectation bias but I don't think that it really applies here. However you think it does. So lets just understand each other until this "modification" further more investigated. If others don't agree after investigation, that's fine too. I'll still modify my music and everyone else can wait for a new lossless and "correct" format to be released if that ever happens again and wait in line at Best Buy to buy their music collection all over again at a much higher price that they bought their CD collection if they can even find all the music they already own depending highly on how good the selection is. It failed once and it probably will again so I'm not holding my breath.

I see it as CD audio is not accurate in the first place because it has already been tampered with when it is compressed. Artificially un-compressing by interpolation is no more wrong than the record companies compressing it.

Interpolation will _not_ change the compression one iota. In this particular case it could introduce some errors but it can't fix anything. Similarly, I don't think the redithering can do anything but change the noise floor (probably add to it).

I'm not disagreeing with expectation bias but I don't think that it really applies here. However you think it does. So lets just understand each other until this "modification" further more investigated.

That's just the thing; you can test for expectation bias. It's more-or-less trivial using your current setup, just download Foobar and load the tracks up. If you think this project is worth doing then you owe it to yourself to do the blinded ABX. Make that part of your "investigation"...

FWIW, I did try your tests on a track. The resampled track came out playing much louder, I didn't think I had any volume normalization enabled but I'll have to go back and check again but don't have time to play with it today. I mention it because it is well known that given no other differences but volume people will very consistently pick the louder of two samples as being "better"...

Mr. scientest. I am a very passionate person. I am also a very analytical person and have good attention to detail. My job requires that of me. I always play the devil's advocate. I do believe that I give everything and everyone a fair shake. I do give everything an honest run, I really do. Do I open my mouth a little too soon before that honest run is over? Sometimes.....I guess you can tell where I'm headed with this. I downloaded foobar and really didn't hear any differences between the tracks. I then went back to windows media player and found I could do blind testing with it. I loaded the two tracks to compare and minimized the playlist so that I couldn't see which one was playing. I pressed the "next track" button about 10 to 15 times so I really would not know which one was playing. I gave both a listen and found I could hardly tell them apart as before I could swear it was almost night and day. Key word is hardly. Even more interesting is that there was definitely a difference even though now it was really small and hardly noticeable, but it was in favor of the ORIGINAL!!! The original did sound slightly BETTER! DAMN IT! Oh well. I would rather know the truth. Amazing. The difference was so much bigger when I thought the modified sounded better. Even when I wasn't comparing the two because I guess subconsciously I knew that it was the modified and it biased my hearing even when I wasn't thinking about listening while I was driving. The funny thing is the day that I thought of playing around with Audacity I knew logically it was not even possible. I figured that you can't take a turd and stick in a flower vase and expect it to actually smell like roses. I suppose you could if you sense of smell was biased enough to make it seam to smell like a rose, but that is a stretch. Well my first thoughts and impressions were correct, but I have no idea what happened after that. At least I did learn something first hand and that's always good. I learned that you indeed CAN'T trust your ears. I'm also an honest man and admit that I'm wrong when I am. I was wrong. Is hearing believing?

Umm, no -sceintest

I just had to add that this whole thing did me a huge disservice. I was in happy land and everything sounded so good even though in reality it sounded not as good. Man! I do agree with the phrase: "Ignorance is Bliss" Oh well. On the flip side all my originals do sound better now than they did before. :D

I just had to add that this whole thing did me a huge disservice. I was in happy land and everything sounded so good even though in reality it sounded not as good. Man! I do agree with the phrase: "Ignorance is Bliss" Oh well. On the flip side all my originals do sound better now than they did before. :D

Man, don't know whether to be happy or sad for you. Congratulations on doing the test and kudos to you for telling us what you found. I hope maybe this is a good experience for you; you didn't end up spending some $40,000 on high end equipment before you figured out what expectation bias was all about and you didn't have to spend 10 years learning how your hearing plays tricks on you.

The first time I realized I didn't really know what my hearing was really doing was many years ago. I must have spent several days pissed at myself for failing what I thought was an easy test. Hopefully you won't do that. Instead relax, enjoy the music and have a good weekend while doing so.

Cheers....

Back off scientest. If numbers explained everything, there would be nobody claiming in this thread that MP3s sound just as good as WAVs.

They can. And you know how they got that way? It involves sciency stuff like psychoacoustics (psymodels). And blind listening tests. Science never said that two files with different 'numbers' must sound different. In fact, the people who routinely seem to expect two things to sound different, because something was done to one but not the other, are nonscientists.

You see, 'science' isn't just measuring stuff with oscilliscopes. It's a way of rationally separating signal from noise. As a result of careful sciency investigation leading to development of current MP3 codecs, we now can create two sound files that measure quite differently in some respects, have notably different waveforms...yet really are likely to SOUND the same. Wild, wacky stuff.

So, perhaps you should abandon your cartoon view of science, and try a blind listening test of a good MP3 versus source. It might rock your world.

So, perhaps you should abandon your cartoon view of science, and try a blind listening test of a good MP3 versus source. It might rock your world.

Krab, it's too bad that the forums lost so much content. After a lot of back and forth Mr. Audio finally did break down and do a double blind comparison using Foobar (and then some other mechanism of his own devising). The end result was that he did in fact discover he was experiencing expectation bias. Everything was cool here until the AVS forums lost the last couple of pages...

I know that. It is definitely the interpolation and dithering that is making the difference. The thing that I'm still not sure about is why it still sounds just as good when I export to a WAV.

um...you are aware that WAV != 16/44.1, yes? They aren't the same thing. A wav file can be any number of bit depths/sample rates. You can have a 24bit/96Khz WAV file, for example.


Possibly since my sound card is cheap and integrated and certainly cannot reproduce sounds over a 44.1kHz/16bit ceiling,


You sure about that? Depending on date of manufacture, it might well do 'higher number' formats.


There is probably no difference audibly between the 16 bit and the 32 bit waveform because my sound card is most likely dumbing it back down to 16 bit

well, if that's what's happening, it might make an audible difference, if the reformatting is done poorly.


when I play the raw version through Audacity. It would probably sound different if my sound card were more capable.

Actually, you should not expect 16 vs 32 to sound different at all, except when you take the very quietest part of you track, and play it very loudly.


I'm working with spectra on this. I'm going to send him some clips and is gonna listen and analyze them through software to visually see the differences.

Spectra won't always tell you if a difference is audible.


I still disagree with you on this one. If it does indeed sound better to the ear, you have made an improvement.

But you haven't determined if that's due to the difference, or to your imagination.


Just like Audessey EQs and room correction doesn't actually correct the signal to make the system sound better, rather it does distort the signal quite a bit to even out being pushed through the room.

Well, Audyssey's doing that to counteract audible, measured distortions introduced by the room and loudspeakers and layout. So the result at the listening position is intended to SOUND more like the recording is supposed to.
It's not solving a fake problem, like your 'interpolation' is, nor is it relying on novel interpretations of science to explain.


CD audio blows. I think most of us know that.

I think you're wrongly assuming that 'most of us' have bought into audiophile nonsense the way you seem to have. In fact, CD audio is likely to be indistinguishable from 'hi rez' audio under nonpathological conditions.


Whether you rerecord a song off a CD and EQ it to help even out it's short comings like when studios "remaster" recordings or artificially inject interpolated information that is pretty accurate to the signal's original intent before it was butchered down to 44.1kHz then you have still improved it if it does indeed sound better. Good interpolation results can be striking.

Your 'interpolation' cannot possibly 'inject information', nor is the audible band in any meaningful sense 'butchered down' merely by rendering it in 16/44.1

Really, it's obvious you don't really know what you're talking about, and are just juggling technical terms without understanding them, possibly after having read some audiophile pseudoscience sites.

Krab, it's too bad that the forums lost so much content. After a lot of back and forth Mr. Audio finally did break down and do a double blind comparison using Foobar (and then some other mechanism of his own devising). The end result was that he did in fact discover he was experiencing expectation bias. Everything was cool here until the AVS forums lost the last couple of pages...

Ah. So, since 8/01 , he's actually compared MP3 to source with Foobar? The post I was replying to was his specific claim about MP3.

I saw that he tried an ABX...I saw that he ABX'd his 'upconverted' file to the original, and got a null, but then when he did a somethingorother with WMP that he claims is a blind test, he still asserts a *definite* difference (now favoring the original). That's post #89 from 8/01.

Did he progress from there?

Ah. So, since 8/01 , he's actually compared MP3 to source with Foobar? The post I was replying to was his specific claim about MP3.

I saw that he tried an ABX...I saw that he ABX'd his 'upconverted' file to the original, and got a null, but then when he did a somethingorother with WMP that he claims is a blind test, he still asserts a *definite* difference (now favoring the original). That's post #89 from 8/01.

Did he progress from there?

Your post two previous attributes stuff to me that was actually from him but, no that's more or less where things ended: with Foobar he could not hear any difference; with WMP and his somewhat suspect testing methodology he ended up preferring the original. There was a bit more discussion but nothing significant that I recall. Given your reaction I thought the discussion must have got lost when the database died, but I see now that's not the case. I'm guessing now you just hadn't read through the whole thread?

I was pretty impressed with Mr. Audio for actually doing the test and posting that he couldn't really hear a difference when he listened blind, or at best it was a tiny difference; a lot of folks wouldn't have done that. There were a few posts to that effect, including my last posts, which got trashed in the database failure. I think all his posts are still here in the thread.

There was also a short discussion about whether lossless compression could make some difference, to which the answer was, concisely, "no".

Your post two previous attributes stuff to me that was actually from him

Sorry! That was unintentional. Fixed.

but, no that's more or less where things ended: with Foobar he could not hear any difference; with WMP and his somewhat suspect testing methodology he ended up preferring the original. There was a bit more discussion but nothing significant that I recall. Given your reaction I thought the discussion must have got lost when the database died, but I see now that's not the case. I'm guessing now you just hadn't read through the whole thread?

I had (excluding the lost stuff), but my point is, is it true or not, that he never actually compared MP3 to source? It seemed to me, again, that all he has done viz blind testing is compared some track created by his curious Audacity-based resampling/reformatting method -- not a well-made MP3 -- with source. My entry point here was him making unfounded claims about MP3s. And lossy vs lossless was what the OP was most concerned with.

well, it's been a while since reading this thread (and I'm too lazy to wade back through it now) but the gist of his POV here was that he could "doctor up" music files from his redbook CD sources and they would be superior to those source discs. He wasn't sure why they sounded better to him (and others he played them for), but he was utterly convinced that his doctored files did in fact sound better. He later acquiesced on this point, as described by scientest.

If he ever made the additional claim that mp3 was significantly more subpar than CD, I don't recall it. I don't think he presented an opinion one way or another on that question. If he did, he didn't spend much time or interest trying to defend it. He was a whole lot more passionate arguing for the superiority of his home-brewed aural concoctions juxtaposed against the lowly 16 bit/44.1 redbook CD. His energies were focused on demonstrating the insufficiency of this format, not mp3.

Bottom line... he called "UNCLE!" a long time ago. I could point you in the direction of at least a half dozen other threads which are in much more dire need of your attention. :D

I had (excluding the lost stuff), but my point is, is it true or not, that he never actually compared MP3 to source? It seemed to me, again, that all he has done viz blind testing is compared some track created by his curious Audacity-based resampling/reformatting method -- not a well-made MP3 -- with source. My entry point here was him making unfounded claims about MP3s. And lossy vs lossless was what the OP was most concerned with.

As far as we know he hasn't compared mp3's to uncompressed; I suggested he might want to do so, based on the outcome of the 'upsampling' experiment. Most of the controversy was due the claim about upconversion, so once he tested that, the thread was pretty much done. I think any claims about mp3's were dissolved by the result of the upsampling test...

well, it's been a while since reading this thread (and I'm too lazy to wade back through it now) but the gist of his POV here was that he could "doctor up" music files from his redbook CD sources and they would be superior to those source discs. He wasn't sure why they sounded better to him (and others he played them for), but he was utterly convinced that his doctored files did in fact sound better. He later acquiesced on this point, as described by scientest.

If he ever made the additional claim that mp3 was significantly more subpar than CD, I don't recall it.

I don't think he presented an opinion one way or another on that question.


He did. Wade back to my first post on this thread...it's a reply to his quote about MP3.
Or more precisely, his incredulity that people could actually find an MP3 to be audibly indistinguishable from lossless.

If he did, he didn't spend much time or interest trying to defend it.

He didn't, at least in what's still visible. Perhaps that's because it was just part an parcel of the standard package of 'audiophile' myths he embraced (e.g., 'CD audio is crap').

He was a whole lot more passionate arguing for the superiority of his home-brewed aural concoctions juxtaposed against the lowly 16 bit/44.1 redbook CD. His energies were focused on demonstrating the insufficiency of this format, not mp3.

Yes, but the 'MP3 is crap' meme is both more central the subject of the thread, and far more widespread, than the issue posed by kooky home-brewed upsampling. Ditto the 'CD audio blows' meme, which he also offered up. Along with the implicit belief that 'higher numbers are better" Those are what I mainly addressed, because those are the myths that would be more likely to be held by the average drive-by reader.

Bottom line... he called "UNCLE!" a long time ago. I could point you in the direction of at least a half dozen other threads which are in much more dire need of your attention. :D

Feel free. But the OP's question in this thread is a very common one.

You raise some valid points, of course. But since the focus of the dwindling thread participants turned solely to Mr. Audio's upsampling claims, your attempts at addressing the mp3 quality question seems a bit unnecessary now, at this late stage, especially since the OP and others who might have been helped by it had abandoned the thread well before then.

Ratman, for one, bailed as soon as he got a glimpse of his thread ally conceding defeat at the hands of what I suspect Ratman views as his own personal enemy, the blind testing weapon of belief destruction, and hasn't been seen or heard from since. :p

Huh? :rolleyes:
Just stated my opinion and chose to stop beating the dead horse. :p

You raise some valid points, of course. But since the focus of the dwindling thread participants turned solely to Mr. Audio's upsampling claims, your attempts at addressing the mp3 quality question seems a bit unnecessary now, at this late stage, especially since the OP and others who might have been helped by it had abandoned the thread well before then.

Ratman, for one, bailed as soon as he got a glimpse of his thread ally conceding defeat at the hands of what I suspect Ratman views as his own personal enemy, the blind testing weapon of belief destruction, and hasn't been seen or heard from since. :p


Well, I guess my work here is done, then.

(/holsters guns, walks off into sunset)
(cue Morricone theme)

Huh? :rolleyes:
Just stated my opinion and chose to stop beating the dead horse. :p

Now if we could only get you to stop sniffin' the glue. :p

Yeah... funny. But, why be condescending?
IMO, that's no way to make your opinion more valid than mine.

I know I'm out of my league here. You guys know your stuff.

Anyway, this is how I handled my migration to digital audio.

1) use EAC to rip all of your music to .wav's.

Considerations
pay extra attention to which CD-ROM drive you choose to use and the setting you use in EAC! This step is critical and will affect sound quality on every step thereafter.


If you are most concerned with fidelity and have scads of time, use EAC's more rigorous ripping algorithms. They'll give you good results in my experience, but took to long for me to deal with. (I sort of regret not putting in the effort here, but doing this may take 5 times longer)


take extra time to define how EAC will save the file. I set it up to spit out the files like itunes. - artist/album/track # - song title. Since .wav's don't have id3 tags, this step is important to allow mediamonkey to automatically read the metadata later so it can tag your compressed files. This will save you a lot of time later, and if you don't do it, you'll have to doctor things up manually (I did this the 1st time - fool me once!)


Figure about 700 MB of space per cd that you encode to .wav.


____________________________________________________________ ___
Now that you've built an organized, uncompressed, high quality library you can change the formats at any time with having to drag your cd's out. That's the strength of this method. The weakness is that you need a heap of HD space to handle more than 1 library - .wav and your compressed files.
____________________________________________________________ ___


2) Use mediamonkey to organize and compress the files.

-First, I got mediamonkey to build a library from the metadata contained in the folder structure I defined in EAC. This gives you a nice clean library of .wav's with a minimum of hassle.

-Next, use mediamonkey to compress the .wavs. You can choose any format you want. Lossless, mp3, aac, ogg vorbis, whatever. For maximum compatibility with portable media players (ie you don't want to be tied to the ipod by using AAC), use mp3 at a bitrate that best fits you. I chose 192 kbps CBR. It sounds pretty good for the ole' 30 GB ipod 5.5g.

considerations


mediamonkey is just personal preference. I use it for its library management power. Note that after the 30 day'ish trial you can't batch rip without paying for the software.
AAC probably sounds better at this bitrate, but locks you into Apple portable media players. I'm not comfortable with that. Maybe you are.


Conclusion

This was my first whack at building a digital media library. What I think about compression? I think there are formats superior mp3, but they lack compatibility with devices.

Lossless? I can't cram a lossless library onto my measly 30 GB ipod. Also, lossless compression has no format like mp3 to unify the platforms, so I am still on the fence. A fence made of .wav's.

I'm currently ripping all of my cds using Exact Audio Copy(works great with my scratched cd's but takes forever) to MP3 320 Lame v3.97. I'm ripping to MP3 to use on my PS3 and MP3 320 lame sounds cd like to me on my less than audiophile system and widely is reported to be "transparent"by others much more knowledgeable than me. The typical file size is 90 to 130 mb. Which is large for MP3 but about 1/4 to 1/3 of Flac. 256 vbr is supposed just as good as 320 cbr and a little smaller but I stuck with the 320.
.



I rip to flac, and I seem to end up around 150-200 MB per album. I can't see why you'd rip lossy to save that little space, especially when 320gb drives can be had for $40.


Nevermind, 350mbish. Still, hard drives are SOOOO cheap.

pay extra attention to which CD-ROM drive you choose to use and the setting you use in EAC! This step is critical and will affect sound quality on every step thereafter.


Less of an issue if you use paranoid mode....


If you are most concerned with fidelity and have scads of time, use EAC's more rigorous ripping algorithms. They'll give you good results in my experience, but took to long for me to deal with. (I sort of regret not putting in the effort here, but doing this may take 5 times longer)


I can't see why anyone would use EAC and NOT use paranoid mode? Take the time, and make sure you get a good rip!


Figure about 700 MB of space per cd that you encode to .wav.


Encode to FLAC. We lost a previous post about this in the crash, but a FLAC will create a WAV file that is bit for bit identical to the original WAV file if you want. More importantly, the FLAC file will take up 1/2 to 1/3 less space than the WAV file.

I haven't tried it yet but I've been hearing good things about dBpoweramp for ripping. It's from the developers of AccurateRip (rip database used by EAC) and supposedly it improves on EAC's accuracy and runs faster as well. They give full credit to the developer of EAC for pioneering many of the concepts used. One nice plus is it includes access to 4 different tagging databases and compares them to make automatic tagging more accurate.

How it works: http://www.dbpoweramp.com/secure-ripper.htm

More importantly, the FLAC file will take up 1/2 to 1/3 less space than the WAV file. And you can tag the files when you rip instead of having to do it as a separate step later. You'll also get all the tags, not just the ones pulled from your directory structure.

What is the best format to save your music from a CD to a PC? MP3 (if so, what bitrate?) WMA, WMA lossless?

Imho, the only way to go is the mp3 guide at uberstandard.org. Those files sound indistinguishable from the original even on pro equipment, they're very low in file size, and mp3 is supported by everything.

What is the best format to save your music from a CD to a PC? MP3 (if so, what bitrate?) WMA, WMA lossless? I know the lossless are suppose to be the best, but does the amount of HDD space it takes up justify it for the average person? I should mention that the PC and is hooked up to a Yamaha YSP-1 Digital Sound projector (one component that produces 5.1 surround sound). My company sells these, so customers would be listening to the music from the PC. The connection is with a SPDIF optical cable. I just want a format that would take advantage of the sound system without unnecessarily wasting HDD space. thanks for any advice.
Right now WMA Pro provides the best sound quality for any given bit rate, and is supported by a lot of things. Also i have use WMA lossless which provides one of the best compression ratio but takes more processing power to decode. I would use and do use (i have a zune) WMA Pro and Lossless. I have downloaded some stuff that is flac but it isn't supported by much. The only thing i don't like about WMA is that the ps3 won't play lossless that i try to stream. Also wma lossless is about 7-8 mega bytes per minute so like a typical cd would be like 300 megabytes

Right now WMA Pro provides the best sound quality for any given bit rate

I doubt that. In the most recent 128kbps listening test at Hydrogenaudio.org, which included WMA Pro, there was no clear winner. All of the tested codecs did about equally well overall.

http://www.rjamorim.com/test/128extension/results.html


WMA Pro was not the clear winner in any of the previous tests either, though it always did well.

This suggests one could do just as well using a codec with wider hardware support than WMA Pro.

Less of an issue if you use paranoid mode....



I can't see why anyone would use EAC and NOT use paranoid mode? Take the time, and make sure you get a good rip!

Or use the Accuraterip plugin (after configuring it with a two CDs that are in the AR database). dbPoweramp uses it too, both methods are faster than EAC maximum security mode. Alternately, EAC test & copy, where you generate the CRC then match it on the second round, can be faster than paranoid (though not strictly immune from error...it's remotely possible to get the same 'wrong' CRC twice). Any track that doesn't pass these, can be re-ripped slowly in secure mode.



Encode to FLAC. We lost a previous post about this in the crash, but a FLAC will create a WAV file that is bit for bit identical to the original WAV file if you want. More importantly, the FLAC file will take up 1/2 to 1/3 less space than the WAV file.


Yup. Everything I rip goes immediately to FLAC, at level 8 (maximum compression).

Or use the Accuraterip plugin (after configuring it with a two CDs that are in the AR database). dbPoweramp uses it too, both methods are faster than EAC maximum security mode. Alternately, EAC test & copy, where you generate the CRC then match it on the second round, can be faster than paranoid (though not strictly immune from error...it's remotely possible to get the same 'wrong' CRC twice). Any track that doesn't pass these, can be re-ripped slowly in secure mode.

I've read some pretty serious questioning of Accuraterip. Don't recall exactly what the issues were anymore but you may want to do some digging if you're relying on it. Think the discussion may have been on Hydrogen...

I think the longer Accuraterip is in existence, the better it gets. The only serious objection I've heard is your disk may not be in the database or not enough people may have ripped that disk to have a high confidence level.

I like the way dB Poweramp works in that it tries a fast rip first and calls it good if it gets a good match in Accuraterip. If not, it tries again with the slower methods. Basically nothing EAC can't do with some user intervention but the whole thing is automated so it ends up faster and easier overall.