Taperssection.com
Gear / Technical Help => Post-Processing, Computer / Streaming / Internet Devices & Related Activity => Topic started by: blu666z on January 14, 2004, 02:23:48 PM
-
Just want to make sure I'm verifying FLACs correctly. Just hitting the test button on the Frontend doesn't prove that you have the originals. You actually have to compare the fingerprints correct? Any software to do that?
-Kevin
-
dont have FLAC installed on this system, but "verify" should be used to check that the FLAC fiiles are error free
-
But that doesn't mean that they are the original FLACs the taper seeded though.
-Kevin
-
guess I'm not understanding...FLAC is like SHN. if the fingerprint/md5 verifies then it is effectively the same as the master original the taper/seeder made. thats the whole point of the fingerprint/md5, to verify the files are error free and that are the same as the original seed
-
Lemme take a stab at what I think is the original question, Kevin:
Loading the FLAC files into FLAC FrontEnd and pressing the Test button will test that each individual FLAC file is not corrupt. But it does not validate the FLAC files against the signatures within the fingerprint file. How does one validate the FLAC files against the fingerprint signatures?
-
How does one validate the FLAC files against the fingerprint signatures?
Nice stab. This is the question I was trying to ask. The only way I know is to drop the FLACs in the frontend, create a fingerprint file and compare that manually to the fingerprint file provided(hopefully) by the taper. If there isn't a program to do that, maybe I could create one.
-
guess I'm not understanding...FLAC is like SHN. if the fingerprint/md5 verifies then it is effectively the same as the master original the taper/seeder made. thats the whole point of the fingerprint/md5, to verify the files are error free and that are the same as the original seed
That doesn't prove they are the originals. ie. I get a show in from you in FLAC. I decode track 1, fuck around with it in SoundForge and then encode it back to FLAC. Now if I run the test on it, it will test fine but it is not the original file you sent me. Does that make sense? To relate that to SHNs, just because the files match the MD5 they were sent with doesn't mean that MD5 is the original seeded by the taper.
-Kevin
-
guess I'm not understanding...FLAC is like SHN. if the fingerprint/md5 verifies then it is effectively the same as the master original the taper/seeder made. thats the whole point of the fingerprint/md5, to verify the files are error free and that are the same as the original seed
That doesn't prove they are the originals. ie. I get a show in from you in FLAC. I decode track 1, fuck around with it in SoundForge and then encode it back to FLAC. Now if I run the test on it, it will test fine but it is not the original file you sent me. Does that make sense? To relate that to SHNs, just because the files match the MD5 they were sent with doesn't mean that MD5 is the original seeded by the taper.
-Kevin
if you alter a file in any way, then the fingerprint/md5 will NOT match the file you downloaded.
as for: "To relate that to SHNs, just because the files match the MD5 they were sent with doesn't mean that MD5 is the original seeded by the taper."
well, I guess thats where the fingerprint/md5 that is hosted on etree/archive, etc comes into play...you can match the the fingerprint/md5 with what is listed on those sites to see if it matches the "original"
-
The question remains:
HOW does one verify FLAC files against the fingerprint file signatures (NOT the internal FLAC file checksums)?
-
Looks like I may have to write a little utility to do that.
-Kevin
-
you have to do a side by side comparison of the fingerprint's numerical output.
Just go to the db (if it is in the db) and find the original fingerprint......and compare it to yours.
-
you have to do a side by side comparison of the fingerprint's numerical output.
Just go to the db (if it is in the db) and find the original fingerprint......and compare it to yours.
Right, but this still doesn't address the central question:
How does one validate the FLAC files against the fingerprint signatures (not against their internal checksum, not validating one fingerprint file against another)?
-
How does one validate the FLAC files against the fingerprint signatures (not against their internal checksum, not validating one fingerprint file against another)?
Wow - it actually sounds like it is a manual comparison unless you have a mac and that x-act software (it has a feature allowing a fingerprint check).
I must say, this is surprising. Sounds like in some ways, the md5 check is better because it validates both file integrity and a check that the files are the originals...
-
"HOW does one verify FLAC files against the fingerprint file signatures (NOT the internal FLAC file checksums)?"
I guess I'm missing something obvious...arent those(signatures/checksums) the same thing, just under a different name? when you verify the file it compares it to the fingerprint signatures/checksum sent with the files.
to verify that this is the same fileset that the taper/seeder originally seeded you would just compare the same as with shn/md5 files, either visually compare the checksums or run the original md5 against the fileset you have. either way produces the same result
-
How does one validate the FLAC files against the fingerprint signatures (not against their internal checksum, not validating one fingerprint file against another)?
you could run the test on the checksums. If verified, you could output a fingerprint of your own and compare it to the one in the .txt file.....if those match, compare them to the ones in a database, or the ones referenced as the original files. am i still missing something?
-
How does one validate the FLAC files against the fingerprint signatures (not against their internal checksum, not validating one fingerprint file against another)?
you could run the test on the checksums. If verified, you could output a fingerprint of your own and compare it to the one in the .txt file.....if those match, compare them to the ones in a database, or the ones referenced as the original files. am i still missing something?
That would do it. A lot more work than double clicking on the md5 :P
-
That would do it. A lot more work than double clicking on the md5
agreed.
-
That would do it. A lot more work than double clicking on the md5
agreed.
Okay, that's kinda what I figured. Thanks for helping clear this up in my mind, everyone.
-
yall are confusing 8)
-
One thing that I have run across with both flac and shn is that they will not convert to wav if there is something wrong with the flac fingerprint or md5...I don't know if that info is what you're looking for, but I thought I'd toss it out there.
-
Another tid bit.. If you got the file from bit torrent, and it said "Downloaded Succeeded", then it's been verified to be an exact duplicate of the original. Bit Torrent has error checking and verifing built into it. beauty eh??
-
I thought the same thing, that it was more work but it actually isnt, at least the way I do things.
Lets start old school, you d/led a show in SHN format via FTP. Its all there so you run the md5s to check it all made it out ok, and it did. Now, to be sure this is the original seed most usually check the etree db to match up md5s. You can either do this by d/ling the md5s and running them against the files to verify, or you can do it by eye... which is what I always did. Now youre all good, can decode to wav and drop on a CD to enjoy whever you please...
New school, you get a show in FLAC format via FTP. You can then check the ffp by eye to the ffp on the db to be sure you have the right source with the correct lineage that hasnt been played around with. Then you can decode to wav AND verify to be sure its all there in one step. Seems to me it has taken out a step and is therefore less work than it actually seems. Throw in the BT auto check, and its a nicer touch ;D
-Stiles :coolguy:
-
So sounds like the visual check is all you have...I may work on a some little utility to do it.
-Kevin
-
yeah, but dont you think a visual check is actually faster?? The only time i can see its not faster is when you have about 10 shows on your HD you can just queue up, walk away and come back.
-
you can sreate a md5 for the origianl Flac files. md5 is not for shn only but any data. right click on anything (data file) and you get create md5 option,
-
>>Wow - it actually sounds like it is a manual comparison unless you have a mac and that x-act software (it has a feature allowing a fingerprint check).<<
can i get a +T? hehe
actually, i believe caleb's util will do it on windows:
http://www.bklyn.org/~cae/md5check
a flac file has an internal md5sum. creating a fingerprint file just extracts the internal md5s stored with the flac files and writes them all to a disc
so all I did in xACT, and I'm assuming all caleb did in his scripts, is store all of the signatures in a structure and then call "metaflac --show-md5sum file" on each file and compare to the one in the fingerprint file. it just automates the process you guys are doing manually now