Data CD-R vs Music CD-R

[Cornucopia]

Your test is utterly inconclusive and unreliable.

Sounds like ROF is chickening out.....

Nope just saying the test is inconclusive since the equipment used would not be inherently available to the average or even the typical user. It would be irrelevant because the test method is flawed in many ways not the least of which is the type of music chosen which is far from representative of the music currently available in MP3 form and also doesn't take into account the pressed recordings from which most source material is converted into MP3 form.

It's a completely and utterly flawed testing method. It's unreliable in more ways than I've currently listed and can not represent a true model for MP3 conversion.

Up until your post, NO ONE has specified any "representative type" of music regarding their burns/rips/backups. This is irrelevant to a blind test.

This recording is quite similar to other "pressed" recordings, some that I've put out myself. I'd even put this on a pressing if the performance were just a little better.

The equipment used to create the original isn't available to many end users, but then neither are most multitrack facilities from which you seem to be referring that most end users will get their "pressings" from. That is also irrelevant to the test in that the lack of artifacts from that equipment won't mask differences you hear between the versions.
However, the encoding/ripping/media, etc. is WIDELY available to many end users of even modest means. This makes it more representative.

Oh yeah, you haven't even heard the music yet, how could you even dare to comment on it?

This type of music features a wide dynamic range and has good natural reverb, things that typically are more difficult for an mp3 encoding to adequately reproduce. That makes it more relevant to a criticism of mp3 encoding when doing comparisons.

If you have any "scientific" criticisms of this methodology, explain them to us, giving a better alternative. That is the only way people would respect you. Otherwise, just be quiet and don't take the test.

I could do a double-blind test, with mixed-in A/B switching, or use more variety of sources, but that would just extend the testing period and isn't particularly more revealing with the pertinent comparison objectives.

One thing that might bias people would be their listening environment. If you'd like, list your playback setup (PC card/connection to amp & speakers, room environment)...

Anybody else a taker?

Scott

[ROF]

I still say your testing method is flawed, incomplete, irrelevant, and ultimately inconclusive to the situation being discussed.

[smearbrick1]

Reading this thread is like listening to argument between two kindgarteners concerning which color is better; red or blue.

This argument could be justified if you were comparing competing mediums (vinyl and CD, cassette and vinyl, phonograph 78's and 8 track tapes, etc...) but to compare the sound of digital audio between data cd-rs and audio cd-rs is just insane. You rip a cd and copy the resulting wave file to a magnetic zip disc and still have the same quality! How can you argue this? Sure, their maybe some subtle variations in playback on stand alone devices caused by A/D D/A conversion, but to say this is caused by the media (audio vs data) is just ludicrous.

By the way...

Red is better than blue.

[garryheather]

And we all lived hapilly ever after... until someone mentioned SACD.

[gadgetguy]

Let me see if I have this right...

ROF is claiming that the test is flawed and inconclusive because the test environment is ideal and doesn't represent what the typical user has when doing conversion to MP3.

I agree. You cannot draw any general conclusions about MP3 conversion or on the differences between Data CDR and Music CDR from the test as described.

However, I read Cornucopia's test to be a challenge to anyone that either claims that, or wonders if, they can hear a difference between the original and a properly encoded MP3. To that end it seems a quite reasonable test.

[uxbridge]

I would not at all pretend to give you all a definitive answer, however what I know is this:

1. Music CD-R are the only one accepted by stand-alone recorders cause they read specific information, a CD burner will not. They were invented because somebody wanted to refrain people copying disks with stand-alone burners, assuming they would not know how to use a PC.

2. There is no difference in quality between a good data CDR and a good audio CDR

3. It is difficult to find bad quality music CDR, you can easily find crap data-CDR

4. the price difference between audio CDR and data is because of taxes, that is it.

5. an audio CD, regardless if it was born as "data" or "music" is recorded differently that CD-ROM, data is organised in sectors. CD players, even the finest, can read the data in less than perfect way and can be affected by jitter, in turn caused by the inner way the CD is burnt.
jiiter is one thing your ears can measure very well.

So in the end there is no difference in quality between a good audio CDR and a good music CDR. The first are cheaper.
If you burn music onto a crap CDR, chances are you won't notice the difference, but some people with some equipment (usually very good and expensive players) could indeed.

About reflectivity, cyanine disks (blue) are more reflective than phtalocyanine (gold) so in general the blue ones are a better choice, but you can get very good phtalocyanine disks for little money.
I use RITEK.

About MP3, panel of listeners, quoting books (Pohlman, Fundamentals of Digital audio) find MP3 to be transparent at 192kbps joint stereo.
This is an actual fact.
On the basis of this, the nickname for 320kbps Mp3 encoding is "insane".
Some "golden years" will notice differences but it is less than a person in 1000.

If you need to reencode however MP3 material, 192kbps will not produce good results, that's why some people use 320 kbps for "archival" purposes.


Ciao

[cpotoso]

If you are making backups of your music CDs I'd recommend using specialized media. You will notice your music has a crisper sound on the music CD-R. Some can't tell the difference, but I know I can. If there is more than $1 difference between individual discs (Data Vs. Music) I'd go with data CD-Rs. Usually the price difference is 25 or 50 cents which can be justified by a better quality sound reproduction. I've also noticed that my music CD-Rs have a better compatibility with all CD Players. If you've ever burnt a music CD that plays fine in your equipment, but not your friends, you know what I mean.

I assume you're making a joke! :P Otherwise you're completely nuts!!! Repeat with me: a bit is a bit is a bit is a bit. Whether it is on music CDR or data CDR it does not matter (as long as it can be read).

[Cornucopia]

Here it is!

It's slightly larger than I though it would be 'cuz I used a longer segment (also more recent song). You can access it as 1 combined zip file (24MB). They have had their names changed to protect the innocent and date/times are all the same. There is a very minor variation in filesize, but don't go trying to find any clues that way.

http://www.cornucopiadm.com/goodies/CD_WAV_MP3_TEST.zip for the Zip file

I haven't listened to them (except the original of course) myself yet, so this could get interesting...

Scott

[lordsmurf]

People that think media brand makes an audio quality difference (burn errors excluded) ........ or that they can hear the difference between pre-recorded digital formats of adequate bitrate ....... only hear quality difference because they want to.

It's all in their imagination, whether they want to acknowledge it or not. Usually you find this from people with mild megalomaniacal tendencies (power or control freaks, for the layman).

[vitualis]

Sorry ROF, but your arguments stink.

"Test environment not ideal"...

As I stated before, MP3 is transparent to source at 320 kbit/s all other things being equal. Most "source" CDs sound pretty crappy when listened through $10 speakers through the PC as well. These limitations are independent of the MP3 format and say nothing about its quality.

If you can only fully "appreciate" the quality of a CD on a high end hi-fi system with good speakers, then a fair test would be to listen to the MP3 files on exactly the same system. If you do this, MP3s at that bitrate are transparent to source as have been extensively tested by proper laboratories before.

The only way that Cornucopia's test isn't ideal is that there probably won't be a large enough sample, the clip is too short, it hasn't been tested for statistical power (I assume) and it isn't double blinded. Otherwise, it is a reasonable test. I fail to see how you can criticise it so extensively when you haven't even performed it yet. Seems like "denial" to me.

Regards.

[vitualis]

Here's a test ROF.

Since you CLAIM that every MP3 you encode at 320 kbit/s sound palpably different from source, it should be easy for you to upload just one example where the 320 kbit/s clip sounds obviously inferior. Provide an upload of the relevant clip with source (keep it short enough so it stays within "fair use").

I guarantee that you cannot provide one, or at least not one where I cannot prove "user error" (because I can encode exactly that same stretch of audio with MP3 and yet sound exactly like the original).

Regards.

[ROF]

I have no need or desire to upload anything. If you can't hear the difference between 320K bitrate encoding and over 1000 bitrate encoding so be it. I know I can. I know others that can. But if it will help to alleviate the conversation I concede that your 320K MP3 is the best sound reproduction ever. There should never be a need to buy an audio CD again. The audio CD is just wasted space since at 320K you should be able to fit 4-7 albums worth of info on a single disc.

Of course, this arguement should carry over to HD DVDs since it's been announced that they are going to be using lower quality encodings. It shouldn't make a difference to anyone arguing your point here since as has been said, a bit is a bit is a bit. It's all digital and won't effect output. For those who disagree I guess we'll be wanting more from our digital recordings but you can stick with 320K.

[vitualis]

"I know I can"...



Just like there are those who reckon you can "hear" the difference between a short $10 optical cable vs a $500 one.

The size of "bitrate" is irrelevant between different encoding algorithms.

Lossless audio encoding codecs are significantly smaller than raw PCM too but they computationally are definitely identical bit for bit. Depending on the absolute bitrate, an MPEG-4 based video encoding with a lower average bitrate may well have superior video quality to a higher bitrate MPEG-1.

ROF, the burden of proof is on YOU. YOU have claimed that every MP3 you encode sound different / worse than the source even at 320 kbit/s. Neither I or Cornucopia can prove a negative. However, you can PROVE that MP3 at 320 kbit/s is inferior to source by providing just ONE example, of which you should have dozens.

In fact, I'll be VERY interested to discover a piece of audio or music where MP3 does fail even at the top bitrate.

However, unless you can put your money where your mouth is, then I reckon that you are posting BS. I demand proof and you claim to have evidence. Well, let's see it then and educate us all.

Regards.

[ROF]

No need for me to prove anything. I've already said that if you believe an MP3 sounds as good as a CD encoding than that's good enough for me. I know that I'm not the only one who can hear the difference but to avoid continuing arguing such nonsense I'll just admit you are right and that MP3s are the best sound reproducing capsules available to date.

[smearbrick1]

No need for me to prove anything. I've already said that if you believe an MP3 sounds as good as a CD encoding than that's good enough for me. I know that I'm not the only one who can hear the difference but to avoid continuing arguing such nonsense I'll just admit you are right and that MP3s are the best sound reproducing capsules available to date.

Now that is just silly. If you believe something, stand up for it and try to prove yourself right. Otherwise you will just become a laughing stock of the entire forum. Perfect example:

I read your results from your RootKit testing. It turns out that your own tests changed your entire opinion about the rootkit. I also thought your test results were very informative and helpful.

If you believe you are right, make a comparison and post your results. Just make sure you define all the variables involved so that others can reproduce your results.

Just agreeing with someone for the sake of agreement is just silly, and not what debate is all about.

[lumis]

ROF,

if you've heard a lot of 320kbps mp3 recordings that sound worse than the original audio cd, just tell them which one it was.. hell give them 4 or 5..
evidence & proof are on the side of everyone elses side.. and you could easily prove them wrong if what you say is true.. but instead you keep saying stuff like "no need for me to prove anything" "i have no need or desire to upload anything"..

you're not right "just because"..

honestly though, i think there is something seriously wrong with you, either that or you're playing a pretty good joke, one that has gone too far.

[Paul_G]

Misleading people is wrong. I personally find it disgusting that ROF is allowed to continue to post misleading infomation on a site dedicated to helping people.

Music CD-Rs offer nothing more than Data CD-Rs, apart from costing the unsuspecting mugs more $$.

[ROF]

Now that is just silly. If you believe something, stand up for it and try to prove yourself right. Otherwise you will just become a laughing stock of the entire forum. Perfect example:

I read your results from your RootKit testing. It turns out that your own tests changed your entire opinion about the rootkit. I also thought your test results were very informative and helpful.

If you believe you are right, make a comparison and post your results. Just make sure you define all the variables involved so that others can reproduce your results.

Just agreeing with someone for the sake of agreement is just silly, and not what debate is all about.

That's exactly why I have said that the other side is correct. If I were to post results the topic would get reported, locked, a select few would continue to post nonsense about MP3s sounding better than CD Audio, and the point of me doing so would be useless. Similiar to how my results from my rootkit testing were bashed as useless. It's much easier to just agree on this or at least just say that the other side is right, and move on rather than argue a moot point that's a proven fact.

[smearbrick1]

Now that is just silly. If you believe something, stand up for it and try to prove yourself right. Otherwise you will just become a laughing stock of the entire forum. Perfect example:

I read your results from your RootKit testing. It turns out that your own tests changed your entire opinion about the rootkit. I also thought your test results were very informative and helpful.

If you believe you are right, make a comparison and post your results. Just make sure you define all the variables involved so that others can reproduce your results.

Just agreeing with someone for the sake of agreement is just silly, and not what debate is all about.

That's exactly why I have said that the other side is correct. If I were to post results the topic would get reported, locked, a select few would continue to post nonsense about MP3s sounding better than CD Audio, and the point of me doing so would be useless. Similiar to how my results from my rootkit testing were bashed as useless. It's much easier to just agree on this or at least just say that the other side is right, and move on rather than argue a moot point that's a proven fact.

But that is where you are not quite right. You can't state it is a proven fact and bail. You must offer up supporting, "factual" evidence of such. You can't just say "396kb/s mp3 are not transparent with the original because they aren't". That's not supporting your arguement. If you want to win, or at least survive the arguement you have to back it up. That's all.

It would equate to someone saying "bigfoot is real because I saw him!"

You can't say "factual" or "proven facts" without providing said facts. Besides, I don't think this topic will get locked as it has remained fairly civil. If it would be locked due to your fact posting, I would be the first one to complain on the grounds that the lock is baseless and biased. Vitualis seems like a fair guy. He also seems like he likes a good debate. I don't think he's that small minded the he (or any other MOD) would lock a topic for disagreement. Unless, of course, someone became completely beligerent and started threatening violence.

[ROF]

Just do a search for MP3 Vs. CD Audio and you can see the results.

two such tests include:

http://www.lincomatic.com/mp3/mp3quality.html

Nice graphical Comparisons

or

http://www.pcworld.com/reviews/article/0,aid,64123,pg,1,00.asp

A magazine I subscribe to.

Both state that while the resulting MP3 is close to the CD Audio there are always those individuals who can distinguish between the two. Does anyone have a link to any site which states otherwise? If not, I'm going to call it a fact that MP3 quality music is less quality than CD audio.

[Rich86]

Lossless audio encoding codecs are significantly smaller than raw PCM too but they computationally are definitely identical bit for bit. Depending on the absolute bitrate, an MPEG-4 based video encoding with a lower average bitrate may well have superior video quality to a higher bitrate MPEG-1.

ROF, the burden of proof is on YOU. YOU have claimed that every MP3 you encode sound different / worse than the source even at 320 kbit/s. Neither I or Cornucopia can prove a negative. However, you can PROVE that MP3 at 320 kbit/s is inferior to source by providing just ONE example, of which you should have dozens.

In fact, I'll be VERY interested to discover a piece of audio or music where MP3 does fail even at the top bitrate.

However, unless you can put your money where your mouth is, then I reckon that you are posting BS. I demand proof and you claim to have evidence. Well, let's see it then and educate us all.

Regards.

Although there are some "lossless" compression algorithms, mp3 is not one of those. The bits are definitely changed when converting a standard 44.1khz digital audio track to a compressed mp3. The higher the bit rate used during that process, the less the data is changed. But it is definitely changed. I don't understand how anyone can argue that point.
Whether the compressed mp3 file sounds better, worse, different to any one person has more to do with that person and what equipment and environment is used to listen. It is all a matter of personal experience. I've heard mp3's that sound terrific. I've heard others that sound like junk. High bitrate mp3's certainly sound quite good - but they are still, by definition, different than the full 44.1khz source audio track, if that is where the mp3 was made from.
So - if a person is happy with their mp3 listening experiences, who are we to argue? Maybe many folks satisfaction with today's available compressed audio formats is why download music sales are skyrocketing and cd sales are falling?
But, I confess to being old fashioned - and still take my digital music straight up - at the full 44.1khz rate.

[adam]

Rich86 and ROF, no one is saying that the mp3 will be identical to the source or that there will not be measurable quality loss. What most people are saying is that these differences cannot be distinguished by the human ear and the playback quality will therefore be transparent. As to whether this is true or not I simply don't know or care.

ROF both of those sources truly just prove nothing at all. The first one is just some guy. He runs a blind test on 2 folks and they can't tell any difference. Then he runs a non-blinded test on himself and says oh yeah I can tell a difference. That's not science that's bias and he even admits it in his writing. The graphical analysis he notes is immaterial as, again we are talking about audible transparency not bitwise compares or signal/noise comparisons.

As for the second link, "We compressed...using...MusicMatch Jukebox Plus for the MP3 format." Nuff said. And even still, the vast majority of people could not notice a difference at high bitrates.

[ROF]

Rich86 and ROF, no one is saying that the mp3 will be identical to the source or that there will not be measurable quality loss. What most people are saying is that these differences cannot be measured by the human ear and are therefore transparent. As to whether this is true or not I simply don't know or care.

which is why I didn't care to continue the discussion. I can hear the difference, others can hear the difference, there is factually a difference, yet there are a few here that think there is no difference simply because they can't hear it. So be it I say. if you think an MP3 sounds just as good as CD Audio that's fine but some here wanted to claim that as a fact when in fact(no pun) the opposite is the truth. There was even someone earlier who felt like all they could add to the discussion was slander against me.

The truth is while some may not be able to to tell the difference, no matter what the bit rate encoded, an MP3 is different sounding than a Retail Audio CD. Arguing otherwise is moot which is why i tried to leave this subject to those who feel they must be right.

[Cornucopia]

Just do a search for MP3 Vs. CD Audio and you can see the results.

two such tests include:

http://www.lincomatic.com/mp3/mp3quality.html

Nice graphical Comparisons

or

http://www.pcworld.com/reviews/article/0,aid,64123,pg,1,00.asp

A magazine I subscribe to.

Both state that while the resulting MP3 is close to the CD Audio there are always those individuals who can distinguish between the two. Does anyone have a link to any site which states otherwise? If not, I'm going to call it a fact that MP3 quality music is less quality than CD audio.

You mean the 1st link, where the guy summarizes like this?:

Surprisingly, 256Kbps and 320Kbps MP3 were virtually indistinguishable by my ears from the CD, except for a light boost in the upper midrange and treble.

And that's using BladeEnc from 2001, which has been far surpassed in quality by Lame (or the Fraunhoefer). He also could use a much better system to listen on!

Or the other link?
They only tested up to a bitrate of 256kbps (with a no-name codec), and at that level 73% said they couldn't tell the difference between it and uncompressed, and on a scale of 1(bad) to 5(perfect), 256kbps mp3 was a 4.9. That test was mainly focused on efficiency/quality with regards to LOW bitrates of 64, 96, 128, etc.

You need to read those again.

Scott

[ROF]

i posted them because the results verify that there is indeed an audible difference between an mp3 and CD Audio. I also posted them because the results show that not everyone can hear this difference but there is in fact a difference. That's my point. Some here claim otherwise when the fact is CD Audio is better quality then any encoded MP3 at any bitrate.

anyone got a link to test results that prove an MP3 is identical to CD Audio? If not I guess the discussion is over.

[smearbrick1]

With these results it can be stated that to 70% to 80% of people polled could not hear an audible difference in the two recordings. The other 20% were either lucky at guessing, or could genuinely hear a difference. I wouldn't argue either way. I have friends that hear pitches such as 7th tones, diminished (which I thought were minor), or augmented chords (which I mistook for major).

I don't doubt that their is someone out there who can tell the difference. As far as the majority of people... the everyday Joe Moron... (not saying everyone who can't tell the difference is a Moron. Just my term for common folk.) Those people don't know the difference between 128kb/s and CD audio. Hence, one of the most popular bitrates for MP3 is 128.

Personally, I prefer WMA or AAC.

[ROF]

If there is a difference and it's proven that there is what the heck was arguement about? What test was I suppose to post? Why was I supposed to waste my time proving something that's already a proven fact? People can use the word transparent or whatever other glossed over terminology but in the end an MP3 no matter how it's encoded or at any bit rate it is still inferior to CD Audio. That's a fact!

[adam]

in the end an MP3 no matter how it's encoded or at any bit rate it is still inferior to CD Audio. That's a fact!

And no one in this thread ever said otherwise. The only statements you are disagreeing with are the ones you have put in other people's mouths. Yes we all agree that Mp3 is a lossy compression scheme. Yes there is a difference! Yes the mp3 will always be lower in quality than its source, whatever format that may be. The question is whether anyone can hear the difference or not and your own sources prove that a substantial majority cannot given that particular environment and mp3 encoder...and things have come a LONG way since then so the numbers can only go up. What others in this thread ARE saying is that given a controlled enviorment and testing method and using the latest compressors, virtually no one on this planet can distinguish an audible difference. I have to admit that the sources you posted actually make me more inclinded to believe that this is true.

[ROF]

@Adam

have you ever seen anyone post in this thread that a bit is a bit is a bit, Digital is Digital, or when Vitualis wrote:

I guarantee that you cannot provide one, or at least not one where I cannot prove "user error" (because I can encode exactly that same stretch of audio with MP3 and yet sound exactly like the original).

Sound Exactly like the original? Are you sure no one has said that MP3 is not a lossy compression scheme? I normally agree with alot of what you post but in this case I believe you are in error. Either that or "sounds exactly like" has a different meaning than what I'm used to.

The links I posted show that there are some people who can tell the difference while there are those who can't. They also help back up the case that MP3 is lossy compression scheme no matter who does the compression or at what bit rate it's at. Some people will still be able to tell the difference because it doesn't sound exactly like the original. That's a fact.

[Cornucopia]

i posted them because the results verify that there is indeed an audible difference between an mp3 and CD Audio. I also posted them because the results show that not everyone can hear this difference but there is in fact a difference. That's my point. Some here claim otherwise when the fact is CD Audio is better quality then any encoded MP3 at any bitrate.

anyone got a link to test results that prove an MP3 is identical to CD Audio? If not I guess the discussion is over.

You really must not be paying attention. No one has said that MP3 is identical to Uncompressed audio, except maybe you when you are "paraphrasing" someone else's post.

Here's how it works:
1. LPCM ("Uncompressed") audio is the "standard".
2. Losslessly compressed audio (monkey's, zip, etc) is BIT FOR BIT IDENTICAL to uncompressed, upon decompression. But it doesn't compress that much (~2.2:1 is good)
3. Lossy compressed audio is NEVER BIT FOR BIT IDENTICAL to uncompressed (upon decompression), but because of the nifty psychoacoustic methods used, it doesn't matter--it doesn't have to be. At or near the highest bitrates available to the compressed format (be it MP3, AC3, DTS, AAC, etc), it is INDISTINGUISHABLE fromUuncompressed to the VAST majority of people. I'm talking ~99%. Not 100%? Nope, statistical testing doesn't really work that way.

Can you distinguish it? You say you can, but won't submit to a blind or double-blind test, so your argument is weak. And from your past posts, I'm not sure I'd put you in the "Golden Ears" category. Most of them are long-time professional musicians and audio engineers, who have been TRAINED to be able to recognize differences in tonality.

You can take the test and talk about it, or not. But if you don't and don't have a better/more scientific alternative, you shouldn't bash it like you have been doing.

Note: That test of mine DOES include comparisons between the MusicCD media and DataCD media--the original thread of this post. IIRC, you ROF were the one who started harping on MP3's anyway. My test was just made to cover both questions.

Scott