Quality of an Original CD compared to Copies...???

[VEBouto]

If i copied an original CD, then i made a copy of the copy.
I would have 3 CD's,
1. The Original.
2. The Copy of the original.
3. The Copy of the Copy.

My question is what would the sound quality of the copies be like.
If i kept copying the copy that i last did would the actual sound quality get worse and worse?
Because your just copying information from one disc to the next...i can't see how the actual sound quality would worsen.

Would CD 2 above be better than CD 3 and so on?
Lets assume i'm using the best quality blank CD's available.

[freestyler]

Bits are bits. Based on your assumptions,and generally speaking, the copies will be the same. Generational loss making 1:1 copies does not occur on [digital] optical media. Analog media (such as cassette or vhs tapes), on the other hand, is quite the opposite and prone to degradation over generations.

[ZippyP.]

Digital is an all or nothing kind of thing...if you've got all the bits copied then it's perfect. If you've got some bits missing you'd know it, your copying program will report errors

[VEBouto]

Yeah, thanks guys i was about 90% sure before that like you said with digital its the same if done without errors, but when using analog audio tapes and video tapes there will be a degredation of quality.
I have asked this question to people before but could never get a completely convincing answer.
So what better place to ask the question and get the right answer, than here.
Thanks again.

[Sifaga]

You should use quality CD's as well

when i use the cheap brands i sometimes get a clicking sound, i blame the cheap media

[vcdforme]

Why don't you do a test and let us know your results.
I would guess by the 20th disc you'd have some
serious audible problems.
You could ask someone like Glenn Meadows, or
Steven St. Croix of Mix Magazine.

I once did a test. You might say I had some time on
my hands...
I recorded a disc thru the analog ports of a cd player.
Then did it again, and again. By the 20th disc, things
were still sounding fair, a little scratchy. 50th disc
sounded like it was raining.
These were analog copies, not digital.

[guns1inger]

I once had a conversation with a guy who told me, in all seriousness, that the laser in the CD player was wearing the surface off his CDs and making them sound funny. He was also related to a woman who was sure the CIA were bugging her house and flying overhead in helicopters. Why the CIA would care about a whacko US ex-pat living in Frankston, Victoria, Australia was beyond me. Yanks seem to take strangeness and paranoia with them where ever they go. It must make them feel comfortable.

[DVD_Ripper]

If you are making direct CD copies then the 2nd and 2 millionth will be the same.

But if you are extracting the audio files to mp3 or wav files then creating a new disc of the songs then you will have some slight reduction in quality. Would be interesting to see how bad this degrates the sound after a few dozen copies (copying of course the latest copy)

[overloaded_ide]

As they say on the Butter Comercail, there is no difference!

Digital media like cd's and Dvds are stored as data, read by a computer as 0 and 1, no matter how many times your computer writes a 1 it is still a 1 it will never become a 3 or 4 nor will it ever reduce to just a fraction like 3/4

Truely the only way to reduce the quality would be to introduce physical damage to the disk (as far as I know of) or to write far above the disks rated speed. As long as you are making a copy of the disk, not converting to MP3 or other formats and back again, and you don't try writing a 16x disk at 48x burn speed then a copy is as good as the source for quality no matter how many times you write those 0's and 1's.

If your media is defective, thats a media problem. Scratches/ finger prints etc..may cause problems where they occure on the disk, but the rest of the cd where the problams do not ocure will still be original quality. Like if track 1 has a nasty finger print or scratch that causes pops/skips snaps etc.. track 7 will still be good as the original.

I use the cheapest junk cd's I can get for copies that may get trashed durring use, never had a problem unless I burn a 16x at 24x or higher then I have had a few that pop or crackle. Most the time even those turn out fine even burned to fast.

By trashed durring use I mean like in dirty dusty work areas or near welders that pop sparks that could burn and trash a disk easily. I know people that sit a stack of disks by their machines where they work so they can swap disk easily without getting out the case and such. They trash a few here and there.

I hear all the stories about bad cd r disks, some reason I never had one just go bad like I hear about, only burn problems are from burning above rated speeds like the 16x burned at 24x, well ya if your doing something your not supposed too!

I always used the cheapest I could get, free after rebate disks I love!! The ones not trashed by external forces or tossed because I just didn't want them any longer do still work fine. Some are several years old now, still work fine.

Anyway, copy to copy is same as the source for cd's or DVDs doesn't matter if it's data, audio, or video. Long as you don't make changes yourself and just make an exact copy then it is an exact copy. With the exception of any copy protection you may remove of course in order to make a copy.

[VEBouto]

Hi DVD_Ripper,
You say that converting to mp3 would degrade the quality which i totally agree with, but with a wave file aren't you still copyiing the exact song without any compression..inturn still keeping the same quality.
I'm not sure about this.
Thanks.

[Shocker Milwaukee]

Food for thought:

With DAE (Digital Audio Extraction) on older drives, I encountered pops and clicks within the extracted file. It was all digital to digital, but obviously errors were introduced....

Any thoughts?

[d_unbeliever]

no difference

[freestyler]

There is a difference. If a drive has poor DAE, you will get poor quality audio rips, even moreso if the drive caches data. Those familiar with Exact Audio Copy and the good way to rip audio can attest to this. Good drives are the ones that don't cache.

[vitualis]

There will probably be differences but depending on your technique and software, it may not be audible.

If you don't have perfect quality DAE, then you may accrue a few clicks here and there.

Even with perfect DAE, there is usually the loss or gain of a few blank sectors at the beginning. This is usually a characteristic of the drive. If you made multiple generational copies on the same drive you will probably eventually notice this.

Regards.

[proxyx99]

Since most reads/burns do introduce some errors (bad sectors, jitter, inconsistent media quality) you may have some trouble with say 20th copy (e.g. skipping, weird crackle noise). If the reading drive was 100% as well as burn was 100% on 100% media then the outcome of 20th copy would be exactly the same as the original.

[lordsmurf]

A 1x burn (DAE not needed, if memory serves correctly) should be perfection. On good media.

[proxyx99]

A 1x burn (DAE not needed, if memory serves correctly) should be perfection. On good media.

Tests run on a variety of CD recorders have proven that 1x option introduces most errors. 16x-burners tested (some time ago) have consistently delivered worst results at 1x. In these tests discs burned 4x and 8x were the best = least errors. Even some 16x burns were better then 1x. Conclusion: use mid-range speed. 40x CD burner is most consistant at 12x - 32x.

[lordsmurf]

A 1x burn (DAE not needed, if memory serves correctly) should be perfection. On good media.

Tests run on a variety of CD recorders have proven that 1x option introduces most errors. 16x-burners tested (some time ago) have consistently delivered worst results at 1x. In these tests discs burned 4x and 8x were the best = least errors. Even some 16x burns were better then 1x. Conclusion: use mid-range speed. 40x CD burner is most consistant at 12x - 32x.

I'm going off experience from copying audio CDs back in the days when 2x was a fast burned, and 4x was new.

I know a lot of the burner didn't have DAE back then. The first 4x I saw had DAE that made a mess, so we used 1x all the time.

My knowledge here I guess is a bit outdated.

I've been burning 16x on everything with a 16x LiteOn burner for about 4-5 years now. Was the first 16x burner with BurnProof. Never a problem.

The BTC DVD burners have 40x CD copying, but it never seem to finish much faster than the 16x burns. Not really a time saver, IMO. I slow them to 16x these days, more reliable.

[proxyx99]

Some (especially) early 40x were marketed as such even though 40x was a buirst speed that that drive couldn't maintain for more then few sec. Many, if not all, drives today are marketed as faster then they really are, claiming that burst speed is the sustained data transfer speed so I'm not surprised (none of the 52x CDRW has a sustained read/write speed of 52x to my knowledge). Never heard of such drive...
As to burning speeds 1x would be the one to avoid (jitter). On todays' drives I'd assume that speeds ranging from 12x to 24x will give the highest quality results

[overloaded_ide]

If i copied an original CD, then i made a copy of the copy.
I would have 3 CD's,

I beleave we are talking about different things here! If you are talking DAE your not talking about a direct copy of the cd as data.

If you copy the cd as data such as an ISO file you have nothing to do with Mp3, Wave, DAE, you are all 0's and 1's. If the PC, Drive, Burner, etc.. can read the 0 and 1 it can write the 0 and 1 and there will be no differnce at all, regaurdless of anything else. That is a copy of a cd!

If you have a problem the the 0 or 1 cannot be read or written, then you should get an error, and it's a media problem so use a different disk. Do not set to ignore errors, halt on errors instead.

If you are RIPPING the cd as a MP3 or Wave or any other format rather as a complete album or individual tracks then you would be getting into DAE, caching, drive quality etc.... that may introduce lower quality in various ways.

Making a ISO copy, then a ISO of the copy, then an ISO of that copy will result in perfect quality no matter how many times you copy the last copy.
All disks and quality will be the same, unless you induce external problems like defective disks, scratches, fingure prints etc...

Making a copy any other way than actuall data, yes you could induce quality variences just the same as you could by playing the original cd in various stereos. Play the master disk on a $500 stereo and you should have great quality, play the master on a $15 boombox and it is probably gonna sound like crap. It's still all the same data on the same disk, the difference is induced by the equipment.

Another thought on this. If I take about 30 mp3's and burn to a 16x CD at 32x using Music Match Jukebox the burn completes successfully but I get pops and snaps durring playback. That of course was caused by my burning too fast. However if I take the same songs and creat an ISO file and over speed the burn the same way, will the burn crash or complete?
Most likely it will crash and I have a coaster, unless I am using a program that will reduce the speed in which case I'm not burning 32x so that does not count.

The point is, one way I am burning audio files, the other way I am burning data, I will get 2 different results with the same source of the same 30 MP3's. So how you make the copy can make a difference. Data to data will result in the exact same data everytime, unless you induce external problems like burning the 16x disk at 32x. That is not a fault of the copy, it's user error and disk problems.

And I have burned Mp3's onto slow 16X disks at high 32X speeds and gotton pops and snaps. Then burned the exact same set of files to High 48X disks and got perfect disks. The difference in quality of the disks had nothing to do with the source of the files, it was induced externally by disk quality and burn speeds and user errors (though on purpose).

The same is true of Cd copies. If I copy Warcraft 20 times each copy made from the previous copy, it is still the same game and same quality! The graphics will not fade or get blurry right!! That is data copies and each one is exactly the same as the master, Audio cds the same thing, exact same data and quality as the master.

And one last point, I just realized I have actually tested this copy from a copy 20 times after all. Warcraft II does contain Audio! The music is the exact same quality now as it was on the master disk. I have made about 20 copies from copies of this game, I keep breaking them! I bought Warcraft 1, II, and battle net. Have not botherd with 3 yet. Anyway after breaking my master comercail disk of Warcraft II battlenet, I borrowed one and copied it because I did not have a backup of mine then (I use my key). Every since then I have 2 copies. When I break one, I copy the backup put it away then play the backup. So yes, I have gone through at least 20 generations of copies in the last few years and the one I play now has the exact same audio as when I play someone elses comercail disk at their house. Actaully mine is better because they don't have surround sound and I do, but that's equipment difference not source!!
But I guess that proves data to data copies do not degrade audio!

[vitualis]

I beleave we are talking about different things here! If you are talking DAE your not talking about a direct copy of the cd as data.

Sorry, but I think that you are mixing up some things.

When you copy an audio CD as an ISO, etc., the drive is still doing digital audio extraction. The reason why such a mode exists is that with audio tracks, essentially the entire sector is used for audio samples and that there is are fundamental limitations in the accuracy the drive can track the disc.

Making a ISO copy, then a ISO of the copy, then an ISO of that copy will result in perfect quality no matter how many times you copy the last copy. All disks and quality will be the same, unless you induce external problems like defective disks, scratches, fingure prints etc...

This statement is incorrect for several modes that a CD can be written in.

Audio tracks cannot 100% accurately be tracked. Thus, what I said before. Even with 100% correct digital audio extraction of the bits of an audio CD, there is usually some degree of error in a number of blank bits (CD drives cannot track an audio CD accurate at the level of smaller than a sector) at the beginning of the CD. That is, the audio data will be shifted by some milliseconds.

Multiple generational copies on the same drive will lead to this error being compounded. Obviously, in the pragmatic sense, this is essentially irrelevant, but a copy of an audio disc is almost never 100% identical.

Apart from audio CDs, there are obviously several other modes where "making an ISO" is not a 100% identical copy of the original. The various CD game copy protection mechanisms are good examples.

Regards.

[Cornucopia]

Yeah, what Vitualis said,

also this all goes back to the difference between track types.
While everything on any CD is digital 1's and 0's, the difference between what you get with Audio tracks vs. Mode1 tracks vs. Mode2Form2 tracks includes a number of things:

1. Sector Addressing

Vitualis said that Audio CD's can't address anything more specific than a sector, but actually that's not true. They really can't specifically address anything less than an AudioCD "Frame" (=1/75 of a second) as that is the basic unit for AudioCD's. Newer drives may consistently achieve better because they use the comparison logic that is incorporated in their "overlapping" DAE, but that's just a bonus and "consistent" doesn't equate with "100%".
If they want to go to a particular second, they have to read the discs table of contents, find the track it's on, find the time-in from the beginning of the track, and count until they get there. There are NO sector addresses, cuz the bytes that the sector address would use are taken up by audio samples. That complication is also why DAE speed is rarely as fast as the speed for sustained reads of data.
DataCD's are the ones that can address down to the sector level (~2048 bytes). There are 96 sectors in every Audio frame, so that's much better addressing.
Better Addressing=Better Read/Write comparison

2. Error Correction/Detection

Audio and Data Mode2Form2 tracks have only 2 levels of ECD/ECC. This allows statistically for an error in every billion bits. This gives an average of 5 1/2 expected errors for every 80 minute disc!
Data Mode1 and Mode2Form1 tracks have 3 levels of ECD/ECC. This yields statistically an error in every couple trillion bits, which makes most (but not all) discs completely error free.

What about all that "digital = no errors" stuff? The people who said that weren't telling you the whole truth. Both Analog and Digital have errors. Analog error curve is a long slope that starts at 99.99%. Digital is 100% ok until the errors are enough of a problem where they aren't correctable, then they're 0% until they get correctible again. Digital isn't a magic wand, it's just trading in for a different set of problems.

Well then, why is there a difference in the kind of problems between Audio vs. MPEG video vs. Data? The answer is fudging/interpolation. With Audio, if the error can't be corrected, it'll get concealed by interpolation--and this is where multiple generations will start to get noticeable. MPEG video is slightly worse at error correction, cuz so much of the signal redundancy has already be stripped out of the file--not so much "room for error", but a player can still cover over the actual errors that are there by using interpolation again.
As we all know, data doesn't forgive errors, so most data recording/playing/reading software is meant to come to a complete halt when an uncorrectable error is detected, unlike Audio/Video players and/or DAE/ripping.

BTW, as curious audio engineer, I have actually perform this 20-generation test on CD's before. Assuming high quality software/burners/readers/media and correct speed matching, you can get data CD's that are 100% copies, even down 20 generations. You can't, however, get Audio CD's--either via digital SPDIF playback, or via HQ DAE (like EAC).

More Food for thought,
Scott

p.s. DVD's are all just like big Mode1 DataCD's with even better error correction.

[vcdforme]

thanks Virtualis and Cornucopia for making sense out of all this
nonsense.

I'll say it another way: For audio,
A copy of a copy of a copy, is not an exact copy.
It's just close enough for us numbskulls to believe that it is.

[overloaded_ide]

Ok, you explained it so well there that I will take your word for it, though it goes against what I beleaved.

Since I have no actual way to read specific 1 and 0 data and compare for exact exactness disk to disk, plus the disks will probably vary abit anyway.

Hmm, not admitting I am wrong, just admitting a possibility I may not be correct!

I am not always perfect, I thought I made a mistake once. But I was wrong. :P

That's a joke,

[Moloko_Plus]

This is why cd and dvd media piracy is such a big problem, according to the bsa and riaa and mpaa. Unlike vhs or cassette tapes, there is no loss of quality with digital formats.
But then again, back in the day of vhs, they still complained and tried to outlaw vhs recorders. Go figure, right?

Wav files are untouched and lossless if youre ripping a music cd. I prefer eac to do that task.

Lordsmurf: Youve had the last burner for 5 years now? You must not make a lot of cds, lol. Ive ripped through 4 burners in the last...4 years id say. The first burner i had, was a a 4x, man that thing never worked at all, paid 150.00 for it too. Second was an Xtasy (Aka Lite-On) that was a 40x i believe. Lasted about a year. Then i had a um, Samsung i think..I dont know, that was on sale at wal-mart, that lasted about a year too. Now i have some standard 52x one, not sure what brand it is, i think "Mad Dog" rings a bell, costed like 20 bucks over the holidays when my samsung burner just decided to stop making cds for me.

I guess the best way to make a 1:1 copy is use clone cd, mount the image in daemon tools and burn that way. Who knows anymore, this is such an old deal, most people just do on the fly burns with easy cd creator nowadays.

[jtoolman2000]

One possible situation that could possibly cause degradation between copys is that yes, it is agreed that a ONE is ONE and a ZERO is a ZERO but,,,,,,
If you take a microscopic look at a pressed CD or DVD you will see almost perfect very distinct ( edges of the pit ( 1 ) very well defined ) where as on a burned CD or DVD that pit is more like a blurry rectangle and at wortst, a smudge. Create enough of these blurry rectangles ( 1-1-1 ) and you end up with a pretty indistinct area. Make a second copy of that ( 2nd generation ) and possibly, the resulting copy may or may not contain the exact code that was present on the original. I heard it said that the fact that a burned DVD can even be played back, is a small miracle in itself.

I think in practice, DVD and CD readers are smart enought to guess and interpolate almost missing or non optimally burned pits as 1s and so you get to hear or watch your backups as if they were perfect.

[Jester700]

One possible situation that could possibly cause degradation between copys is that yes, it is agreed that a ONE is ONE and a ZERO is a ZERO but,,,,,,
If you take a microscopic look at a pressed CD or DVD you will see almost perfect very distinct ( edges of the pit ( 1 ) very well defined ) where as on a burned CD or DVD that pit is more like a blurry rectangle and at wortst, a smudge. Create enough of these blurry rectangles ( 1-1-1 ) and you end up with a pretty indistinct area. Make a second copy of that ( 2nd generation ) and possibly, the resulting copy may or may not contain the exact code that was present on the original. I heard it said that the fact that a burned DVD can even be played back, is a small miracle in itself.

I think in practice, DVD and CD readers are smart enought to guess and interpolate almost missing or non optimally burned pits as 1s and so you get to hear or watch your backups as if they were perfect.

This is part of what Cornucopia was saying. PC file formats (including WAV) have better error correction than an audio CD, and can reconstruct data more perfectly (OK, maybe "less imperfectly").

Try it for yourself - rip a CD and burn the same material as an audio CD and as data WAV files - the WAVs take up more space. That extra overhead is another layer of error correction, and is why lots of audio people back up their productions in this format (well, that and the fact that you can easily save in 24 bit quality).

[FOO]

The original audio CD has about 30% overhead in Error Correction Bits.
It can replace the complete loss of about 2000 bits in a row.
If you read a CD with errors , they will be corrected. (up to some limit)
If you then write that to another CD there is some chance of error
again, but they aren't cumulative.
So there should be no generation loss in copying CDs.

As far as the fuzzy rectangles go... they are read and converted to digital
data before being written again. The fuzziness is not "copied" , it does
not increase.

[lordsmurf]

Information overload ... wow...

[VEBouto]

Wow, sure is great to get so much in put for my question from everyone...
But what about this theory...
If all of us had the same CD copying equipment that the record companys use, would the quality then be exactly 100%.
Cos if you look at the kind of blank CD's that we use, the information is burnt into the disc. Hence as suggested here, it's more prone to micro-errors after each copy.
But the Original ones are stamped or pressed, not burnt right? So instead of burning multiple copies on our cheap Cd/DVD Burners we had access to what the Record companies have, then the quality of copy to copy to copy would be the same...perfect would you not agree?
So doesn't it really come down to cheap PC Burners and dye based blank CD's that we use, as compared to state of the art CD manufacturing company's and there original type blank media that is pressed.
This is just a newbie theory of course.
Thanks.