1) First thing would be to test if your files are really identical (that was part of the claim).
I use Linux so I have the tools available, but for Windows you might try:
MD5sums for Windows
or even better:
[Announce] sha1sum for MS Windows released
2) All else being equal (playback system) there exists no mechanism that would
allow identical files to sound different, but if you can prove this in a controlled test
this would certainly change the world of computer technology. Identical files that
are different at the same time, imagine the impact on on how we see the world... ;-)
I use Linux so I have the tools available, but for Windows you might try:
MD5sums for Windows
or even better:
[Announce] sha1sum for MS Windows released
2) All else being equal (playback system) there exists no mechanism that would
allow identical files to sound different, but if you can prove this in a controlled test
this would certainly change the world of computer technology. Identical files that
are different at the same time, imagine the impact on on how we see the world... ;-)
That a digital copy is a perfect copy .... or it isn't. If the CRC's are different then the copies aren't digitally perfect !!! Simple as that.
I understand the CRC will tell us nothing about the quality, but that is not the purpose yet. The first step is to determine if there IS a difference. I believe there will be, in fact i believe if we got 1000 CRC's, no two would be the same. I could be wrong, it wouldn't be the first time.
Once we determine if their is a variance between digital rips, then we can go about testing further, to see if perhaps some drives handle better, or maybe we need to upgrade power supplies in computers, Perhaps external CD drives with dedicated power is better, even turn off services in the computer which might make noise on the SATA bus. Optimising a computer to get the most out of the sound.
But first we must either prove or dispell the myth that a CD Rip is perfect.
OR, we can just keep bashing Erin saying he hasn't provided proof, when in reality he did ( he heard it ). Yet the people who bash haven't supported their case either. Prove that all CD rips are identical !!!!
There is a reason people spend good money on good transports and good CD players. To assume that a $10 computer CD drive in a $200 computer will perform the same as a $5000 dedicated CD player just because it's digital without testing is not valid.
I understand the CRC will tell us nothing about the quality, but that is not the purpose yet. The first step is to determine if there IS a difference. I believe there will be, in fact i believe if we got 1000 CRC's, no two would be the same. I could be wrong, it wouldn't be the first time.
Once we determine if their is a variance between digital rips, then we can go about testing further, to see if perhaps some drives handle better, or maybe we need to upgrade power supplies in computers, Perhaps external CD drives with dedicated power is better, even turn off services in the computer which might make noise on the SATA bus. Optimising a computer to get the most out of the sound.
But first we must either prove or dispell the myth that a CD Rip is perfect.
OR, we can just keep bashing Erin saying he hasn't provided proof, when in reality he did ( he heard it ). Yet the people who bash haven't supported their case either. Prove that all CD rips are identical !!!!
There is a reason people spend good money on good transports and good CD players. To assume that a $10 computer CD drive in a $200 computer will perform the same as a $5000 dedicated CD player just because it's digital without testing is not valid.
erin is trying one. Like I said before, no matter what the outcome, there will be a commotion, and that appeals to the sick puppy that is my soul.
DaveDaGr8,
In the first post, Erin said the the two files have the same checksum. As for not getting two matching checksums out of 1000 rips, have a read about AccurateRip. It relies on the majority of people being able to get the same checksum as each other.
Further, your $5000 dedicated CD player is still a CD player (red book), while a ripped .wav file is a data file. The computer CD drive is not needed for playback.
In the first post, Erin said the the two files have the same checksum. As for not getting two matching checksums out of 1000 rips, have a read about AccurateRip. It relies on the majority of people being able to get the same checksum as each other.
Further, your $5000 dedicated CD player is still a CD player (red book), while a ripped .wav file is a data file. The computer CD drive is not needed for playback.
Last edited:
That's almost a good idea except that there's probably a dozen different versions out there to start with (rereleases etc ).
A few years ago I tried to buy "Dark Side of the Moon" at a local store. They had several flavors of New improved / digitally remastered / now with added bonus tracks etc. Nothing that pretended to sound like the original though.
You'd think so, wouldn't you? However I remember a previous similar thread that claimed audible differences between bit-identical files remained even after the files were uploaded to a server, then downloaded by others.SY,
The test is irrelevant anyway, the bits will be screwed up during the email transfer.
Differences did remain, but low level detail suffered some degradation compared with the original files.
We are now sending them as Uncompressed Zip files,( a "zipcreator" option) which helps to reduce the Filemail Server degradation over these long haul routes.As Ripley would say, "Believe It Or Not " A USB memory stick is a far better option.
SandyK
Good thinking! Is it possible to dilute a data file with extra zeroes though? You'd need a sort of anti-compression algorithm.
Excellent! The principles of homeopathy have clearly inspired and informed subjectivist audiophiles. Following them, we inevitably reach the conclusion that music will affect us most deeply if we dilute it with zeroes to the point that none of the original bits remain. QED
Last edited:
Most of the manufacturers of DAC chips have been doing this for a decade or more - except they call it 'oversampling'.
OK,
So without being able to do any testing ( i'm away for another week) I can only rely on google.
AcurateRip, (what a gem, thanks Steve). It states that there are in fact differences in audio Rips, stands to reason otherwise the program would not exist. It also states that no rip is perfect and that the quality is dependant on the drive. Further it also compares a CRC with an internet database of CRC's. This sounds perfect except they don't tell us how they calculate CRC and if they're looking for exact or approximation or anything. The only thing they say they're looking for is silence and spikes, cliks and pops. SO i don't know what to make of this program other than it'll tell you if you've got dropouts or those horrible squelchy spikes.
The other program that shuts down services to make audio sound better
So back to Erin's original issue.
Given that we now know we won't ever get a perfect digital copy, how can we optimise the computer to get the best copy possible ???
So without being able to do any testing ( i'm away for another week) I can only rely on google.
AcurateRip, (what a gem, thanks Steve). It states that there are in fact differences in audio Rips, stands to reason otherwise the program would not exist. It also states that no rip is perfect and that the quality is dependant on the drive. Further it also compares a CRC with an internet database of CRC's. This sounds perfect except they don't tell us how they calculate CRC and if they're looking for exact or approximation or anything. The only thing they say they're looking for is silence and spikes, cliks and pops. SO i don't know what to make of this program other than it'll tell you if you've got dropouts or those horrible squelchy spikes.
The other program that shuts down services to make audio sound better
So back to Erin's original issue.
Given that we now know we won't ever get a perfect digital copy, how can we optimise the computer to get the best copy possible ???
Err no, that isn't what oversampling's about. Check this:
Oversampling - Wikipedia, the free encyclopedia
- Status
- Not open for further replies.