02-tm-block.md

02 — The .TM block

Followed by pointer, not by offset, as the checklist requires: read the 'TM' tag in the PVD's application-use area, then the constant, then the length and the LBA it names.

tag offset in the PVD    888
tag                      'TM'
constant                 0x0014
length                   2048
LBA                      21
appears as a filesystem record   yes

Seven non-zero bytes in the whole 512-byte application-use area, and they are exactly this pointer.

The hashes

whole block, 2048 bytes    c5ffcef2a5e33d2df606185823cd95d1c174d65f
banner  0x000..0x44C       8d84115154d70360b3469acc99cdad3db0ed2c92
object  0x44C..0x7B8       690aae24a96b69659066e691d0b07db301260572

All three are byte-for-byte the values recorded for twelve of the thirteen CD32-era discs in the set.

The prediction, and why it is the interesting half of a pair

Written before the block was read (P10):

Twelve of thirteen CD32-era discs share three SHA-1s; the thirteenth is Speris, the other Team 17 disc, which carries 22,152 bytes of cdtv.device. Prediction: Superfrog matches the twelve — i.e. the .TM follows the TOOL, not the label. This deliberately points the opposite way from P3.

Correct. The .TM block follows the mastering tool.

The point of writing it that way was to have two same-disc predictions pointing in opposite directions, so that whichever way the disc fell, something got falsified. Both came out as written:

artefact	follows	evidence
cruncher	the label	Speris and Superfrog both use the Imploder; the other four studios use five different crunchers
.TM block	the tool	Superfrog matches the twelve ISOCD discs; Speris, same label, does not

That is a cleaner answer to open item 14 than either result on its own would be. "Does build practice belong to the studio, the label, or neither" is the wrong shape of question: different artefacts have different owners, and the owner can be read off per artefact. The cruncher is a choice made by whoever built the game; the .TM block is emitted by whoever cut the disc; and on this disc those are demonstrably different people, because the person who cut the disc signed the preparer field with his own name and it is not a Team 17 development credit but the author of the front-end menu.

Speris, for contrast

The other Team 17 disc has 22,152 bytes of cdtv.device where this one has the standard 2,048-byte block. So within one label, across two discs, the .TM differs — and across twelve discs and ten labels, cut with the same tool, it does not. The tool wins.