Skip to content

6 June 2026

M. T. Kimmins edited this page Jun 6, 2026 · 5 revisions

Thinking about custom structure

My best guess is that since Segments 2.1 and 2.2 are essentially 'skipped,' then we would assume that the projector expects two junk audios before playing the 'third.' Perhaps one could get away with minimizing Segment 2 while completely removing Segment 4, but keep the addresses all just either the final byte after the audio, or maybe the beginning of any Region?

This would the put our time-capacity estimate at the following:

p25d80sh flash = 1048576 bytes
Segment 1 = 108 bytes
Region headers = 6 bytes each
Segment 2 must be minimized
Segment 4 can be deleted
Minimal packet of audio = 10 bytes (20 ms)
Each byte = 2 ms audio

$1048576-108-(2*(6+10))-(12*6)$

= 1048364 bytes

$1048364*2=2096728$

$2096728/1000=2096.728$

= 2096.728 seconds of audio

$2096.728/60=34.9454666667$

$\approx$ 35 minutes (regardless of how you account for minimized junk data)

Maximally, this would be about 3 minutes (2.912122) per frame. Essentially, a whole cartridge (as they currently are sold) per frame. We could stick the data of twelve cartridges in one.

For fun, let us estimate the financial value of this compression. Little Tikes sells three-cartridge packs on Amazon for $\approx$ 23 dollars CAD. If we can fit 12 cartridges, then that would be roughly 4 of these packages. So, we could fit almost $100 CAD ($92) worth of data on a single cartridge.

⟵ Older Table of Contents Newer ⟶

Clone this wiki locally