spectrum analysis and comparison of the "boards of canada - geogaddi" releases.
what you see above is a spectral comparison of the first track. the ksi cd release is in white, and the unknown release is in black. the graphs are almost identical. notice the big drop right around 11.5khz in both releases.
what you see above is a spectral comparison of the fourth track. the ksi CD release is in white, and the unknown release is in black. again, the graphs are almost identical. Notice the big drop right around 11.5khz is still present in both releases.
what you see above is a spectral comparison of the fourteenth track. the ksi CD release is in white, and the unknown release is in black. once more, the graphs are almost identical. The big drop right around 11.5khz in both releases is still there.
now for the money shot..
what you see above is a spectral comparison of the fourteenth track. this time we're comparing the ksi vinyl release and the ksi CD release. the ksi CD release is in white, and ksi vinyl release is in black. Notice the big drop right around 11.5khz is NOT PRESENT in the vinyl release.
the ksi CD at 192kbit release and the unknown at 96kbit are virtually identical.they both exhibit the same drop out at the 11.5khz mark.
the ksi CD at 192kbit and the ksi vinyl at 192kbit are spectrally very different. the data above 11.5khz is not missing in the vinyl release.
we can conclude one of two things at this point..
conclusion one: the ksi 192kbit CD release and the unknown 96kbit release come from the same CD source, but are two different encodings.
according to the name ksi gave their release, these mp3s came from the real CD. the release name is: Boards_of_Canada-Geogaddi_(WARPCD101)-2002-KSi). they even give the catalog number of the CD.
so, if this release actually came from the real CD, like ksi is indicating in the release name, that would mean the CD is missing virtually all sound data above 11.5khz. the spectral analysis of the vinyl release proves that these songs have frequencies above 11.5khz. i find it very unlikely that a real CD rip would be missing this data. remember, cds support 44.1khz data, by nature..
this brings us to conclusion two: the ksi release is a re-encode of the unknown release.
which conclusion is correct? my money is on the second one.. but it's up to you to decide for yourself.
one small note: the images above are filtered screenshots, they were combined in paint shop pro for comparison. you can see the unaltered originals here.
sorry for any errors, i'm only human.. and so is ksi ;)
UPDATE: ok, someone mentioned to me that some people may not know anything about the 96kbit release. basically, some unknown person released a 96kbit mp3 rip of the cd on the web, and the next day, ksi put out a 192kbit release of the cd.