757e312602
Link to kfest 2022 presentation
2023-07-06 22:19:06 +01:00
ca6973f2a2
Merge pull request #1 from KrisKennaway/1mhz-opt-osc
...
Complete overhaul
- model as damped harmonic oscillator instead of RC circuit
- two-stage end-of-frame processing to control clicks
- use simplest approach for core audio playback to allow access at any cycle intervals >= 10 - trying to amortize multiple speaker accesses per operation just led to pathological edge cases, e.g. when trying to reproduce 0
2022-07-22 15:01:01 +01:00
2dff0c242c
Merge branch 'master' into 1mhz-opt-osc
2022-07-22 14:58:47 +01:00
a7c8a2ff2b
Add links and update to reference slot 3
2022-07-22 14:53:00 +01:00
cf24da1228
Add example audio file
2022-07-22 14:52:10 +01:00
5abad38314
Improve comments
2022-07-22 14:36:09 +01:00
9129cdccfe
Rename toggles to voltages; get working again
2022-07-22 14:32:01 +01:00
2907ba583d
Tidy
2022-07-22 14:30:54 +01:00
893e872da4
Tidy
2022-07-22 14:29:31 +01:00
b243e7afb4
Overhaul README
2022-07-22 14:28:48 +01:00
30c79accd7
Add a requirements.txt
2022-07-22 12:22:56 +01:00
5d628aec0f
Tidy
2022-07-22 12:22:45 +01:00
e4f20ea9c7
Tidy
2022-07-22 12:12:46 +01:00
183781da60
Tidy
2022-07-22 12:02:35 +01:00
cf7716d068
Tidy a bit
2022-07-22 11:39:05 +01:00
f64b1a6e2c
- parametrize frame size. 4KB has too much buffering though
...
- Correct speaker model to apply coefficients for a square wave impulse
- Parametrize speaker scaling factor
- Flush wav file output after 1MB
2022-07-09 11:20:03 +01:00
e79ed985bc
Support 11 cycle cadence using STA $BFFF,Y(=31). This slightly improves quality
2022-07-08 23:28:35 +01:00
a356f6a3a7
It works!
2022-07-04 11:03:19 +01:00
cea66fe916
Generate stage 3 jump tables
2022-07-04 09:46:56 +01:00
21c2257e5c
Tidy
2022-07-03 23:18:08 +01:00
8eeb26379d
Oops, add missing file
2022-07-03 23:11:20 +01:00
6ff35df1ae
Oops, add build file for lookahead.pyx
2022-07-03 23:10:43 +01:00
233e94d94f
Tidy
2022-07-03 23:09:04 +01:00
f65aaba765
Write out wav files incrementally during processing
2022-07-03 22:42:45 +01:00
2e3e51bb99
Chunk the output resampling to limit memory use
2022-07-03 22:33:01 +01:00
18a16999b0
Tweak to handle full audio files without crashing
2022-07-03 22:32:36 +01:00
a9e7a1bbff
Implement (10,10) eof duty cycle as a special case
2022-07-02 21:59:08 +01:00
6772b579ca
Dramatically reduce memory use
2022-07-02 21:20:16 +01:00
0d217e9326
Improved quality and speed
2022-07-02 20:53:12 +01:00
d153119231
Allow writing out a noise wav file containing the difference between input and output audio
2022-07-02 16:50:42 +01:00
535656aceb
Working!
2022-07-02 14:15:41 +01:00
2a20bebb37
Tidy
2022-07-02 11:43:35 +01:00
3458d3a1de
Assert combined EOF operations have the expected (a, b) cadence
2022-07-02 11:42:36 +01:00
2016e4a5f1
Tidy
2022-07-02 10:58:38 +01:00
622583696d
Simplify player op generation
2022-07-02 10:52:23 +01:00
efb66e03d8
Begin to rework player operation generation
2022-07-02 10:05:21 +01:00
695ddb2c10
Checkpoint
2022-07-02 00:27:27 +01:00
445ec18f34
WIP - tidy and compute the minimum duty cycle for variants that have a STA $C030 in stage 2 trampoline
2022-07-01 23:17:25 +01:00
47c7f441cf
WIP - generate 3 stage EOF trampoline parametrized by (a, b) duty cycles
2022-07-01 22:22:38 +01:00
c33e835f7a
Simulate at clock speed and downsample output
2022-06-28 22:09:42 +01:00
9a9434e9aa
Tweak
2022-06-28 22:09:21 +01:00
d2cdceb04e
Regenerate
2022-06-28 22:09:09 +01:00
11afaa152b
- Only use the minimal player, quality is better and we don't often need more than 6 cycle gaps
...
- Generate 1-cycle EOF opcodes for now, but we'll need to tweak further.
- Generate both duty cycle orderings (a,b) and (b,a) since we anchor the EOF opcodes with a STA $C030 at the beginning
- Sort EOF by total length so it's clearer when looking at the usage profile which ones are needed
2022-06-28 22:06:38 +01:00
346c647c40
Tweaks
2022-06-28 22:04:57 +01:00
73ef91b5ad
Simulate PWM audio filtered through speaker
2022-06-26 21:50:29 +01:00
804ce60434
Clean up some more and restore fake EOF opcodes
2022-06-25 15:26:14 +01:00
efa63a5ab2
Tidy
2022-06-25 15:15:40 +01:00
e0908a9ced
WIP - refactor player generation to be a) deterministic b) use opcodes_6502 c) statically enumerate a minimal set of opcode sequences. Since we're giving up on using arbitrary opcode combinations this is much simpler, and maybe more correct.
2022-06-25 15:14:17 +01:00
bc032ae3ca
Autogenerate eof with speaker opcode interleaving
2022-06-21 22:32:12 +01:00
cb5de62cc8
Look ahead beyond the EOF opcodes for better prediction quality
...
Normalize based on 100th percentile
2022-06-21 21:33:19 +01:00
KrisKennaway