Don't forget that tape has 4 tracks (used as 2x stereo in traditional audio tape). Other experiments have used 1 track as a pilot town to correct for flutter, but that still leaves 3x as much bandwidth as you were assuming.

[edit]

Found one of the ones I was thinking of. They were able to get about 24kbps on a single track, with a pilot tone on the other track to correct for flutter. This would imply 72kbps if using all 4 tracks simultaneously. At that rate, you probably get better quality audio with Opus than you would storing on a tape without some form of NR, though you only get half as much playback time.

https://redfrontdoor.org/blog/?p=795

▲

jedbrooke 3 hours ago | parent [-]

YES! I remember now. I have a notes file where all this is written down somewhere. A stock tape recorder can usually only read two channels at a time (stereo L and R), but then you’d still have the “flip it over” side A plus side B runtime.

Thanks for the link, definitely saving that one

	▲	aidenn0 3 hours ago \| parent [-]
		Note also that the linked article stops at QAM. This is a previously solved problem for phoneline communication; in the 1980s modems went from 9.6kbps to 33.6kbps[1] after adopting trellis coding. It's not impossible that a similar technique would work for audio cassettes. 1: https://www.itu.int/rec/T-REC-V.34-199802-I/en