logoalt Hacker News

Lammytoday at 2:11 AM2 repliesview on HN

> Just don’t forget to skip track 1.

There is very slightly more to it than just the mixed-mode sessions: the audio tracks on the Quake CD were mastered with pre-emphasis, so your ripper or player needs to apply corresponding de-emphasis EQ in order for them to sound correct. 1996 was actually a strangely late year to find any CD mastered with pre-emph; it was mostly an '80s thing: https://www.studio-nibble.com/cd/index.php?title=Pre-emphasi...

On top of being a technique that had fallen out of common use by that time, Quake has the pre-emphasis metadata flagged in the subcodes but not flagged in the TOC, so some software players/rippers like (annoyingly) Exact Audio Copy won't heed it: https://wiki.hydrogenaudio.org/index.php?title=Comparison_of...

Here's a side-by-side comparison — my EAC rip of my MacSoft version of Quake (left) alongside the output of `sox_ng deemph` for the same file (right): https://i.ibb.co/9m18CSsk/Emi-2026-08-17-18-52-48.png

And if you're curious, here's the pre-emphasis subcode channel Q flag and pre-emphasis frequency response as described in the Red Book standard, and sox_ng's code for undoing it:

https://audio-union.com/wp-content/uploads/2023/02/Red-Book-...

https://audio-union.com/wp-content/uploads/2023/02/Red-Book-...

https://codeberg.org/sox_ng/sox_ng/src/branch/main/src/deemp...

Although if someone has been unknowingly listening to the pre-emphasized tracks for thirty years, maybe the ‘incorrect’ one is what sounds right to them? :)


Replies

protimewastertoday at 5:19 AM

IIRC, it's actually debatable if the pre-emphasis flag is supposed to be set. I don't remember the exact details anymore, but I think it was discovered that there multiple releases with bit-identical audio tracks, but some of them have the flag set while others don't. They can't both be correct, because the audio data should've changed if the emphasis was removed.

show 1 reply