For what it's worth, it looks like the TOC and subcode match. Here's the result of cdda2wav -J:
Type: ROM, Vendor 'HL-DT-ST' Model 'DVD-RAM GH10L ' Revision 'FC0B' MMC+CDDA
983040 bytes buffer memory requested, transfer size 64512 bytes, 4 buffers, 27 sectors
#Cdda2wav version 3.00_cygwin32_nt_1.7.5(0.225-5-3)_i686_i686, real time sched., soundcard, libparanoia support
54420 sectors of data before track #2, ignoring.
DATAtrack recorded copy-permitted tracktype
1- 1 uninterrupted no data
AUDIOtrack pre-emphasis copy-permitted tracktype channels
2-11 yes no audio 2
Table of Contents: total tracks:11, (total time 71:01.36)
1.[12:05.45], 2.( 5:08.45), 3.( 2:26.01), 4.( 8:20.19), 5.( 6:05.29),
6.( 7:24.44), 7.( 8:38.17), 8.( 5:35.69), 9.( 6:28.69), 10.( 3:32.42),
11.( 5:15.31)
Table of Contents: starting sectors
1.( 0), 2.( 54420), 3.( 77565), 4.( 88516), 5.( 126035),
6.( 153439), 7.( 186783), 8.( 225650), 9.( 250844), 10.( 280013),
11.( 295955), lead-out( 319611)
CDINDEX discid: n_9sMj0uUigoQi.U9QxEqhHrhnY-
CDDB discid: 0x9510a50b
CD-Text: not detected
CD-Extra: not detected
No media catalog number present.
scanning for ISRCs: 11 ...
Skipping data track 1...
index scan: 11...
I tried applying a de-emphasis filter to track 1 of my Quake soundtrack rip (using "sox track1.flac track1_deemph.flac deemph") and compared the two.
The first and most immediate impression is that the de-emphasized track sounds muffled compared to the straight rip. Actually, I would say that it sounds more like a typical 22.050hz (or lower) samplerate early-to-mid-90s game soundtrack, lacking in treble.
I definitely prefer the straight rip, it sounds sharper and more industrial to me, and that's the sound I remember from playing the game. I'd never heard it in a de-emphasized version before today.
I think only Trent Reznor himself can tell us what the soundtrack is really supposed to sound like, but I honestly think the pre-emphasis flag in the TOC is an error.
Thanks, I appreciate it. I think I'll just stick with the straight rip as well.