Another evening, another few filters getting OOM speedups.
Tonight it was invert (27.3x, trivial memory bound shader that just outputs negative x[i] for each output sample) and the 8B/10B decode (didn't even bother to GPU it, just removing the redundant sampling operation by using the new CDR recovered-data output was enough for a 12.1x speedup and 6ms on my 50M point benchmark is fast enough I'm in no hurry to GPU... but at faster data rates it may still be worth it)
Andrew Zonenberg
npub1cd…z9vpd
2026-02-01 05:45:40 UTC
Author Public Key
npub1cddglts94qutscms0qpmk87lel9m8xku7q0wr20u2th5fxvvunqqxz9vpdSeen on
wss://relay.ditto.pubPublished at
2026-02-01 05:45:40 UTCEvent JSON
{
"id": "0f7da92a020e66738707c8c8405e32ca15e00445c953c49c5e9c29c8efabc525",
"pubkey": "c35a8fae05a838b863707803bb1fdfcfcbb39adcf01ee1a9fc52ef44998ce4c0",
"created_at": 1769924740,
"kind": 1,
"tags": [
[
"proxy",
"https://ioc.exchange/users/azonenberg/statuses/115993787794704825",
"activitypub"
],
[
"client",
"Mostr",
"31990:6be38f8c63df7dbf84db7ec4a6e6fbbd8d19dca3b980efad18585c46f04b26f9:mostr",
"wss://relay.ditto.pub"
]
],
"content": "Another evening, another few filters getting OOM speedups.\n\nTonight it was invert (27.3x, trivial memory bound shader that just outputs negative x[i] for each output sample) and the 8B/10B decode (didn't even bother to GPU it, just removing the redundant sampling operation by using the new CDR recovered-data output was enough for a 12.1x speedup and 6ms on my 50M point benchmark is fast enough I'm in no hurry to GPU... but at faster data rates it may still be worth it)",
"sig": "ef34169762e2a6b716415ebeb6934f0c07e09352f77821e54ccbafef74919cc76ee2d149cc18026cfcca2ca5c48d33564c973a6f15167a0e9579aa91328d8053"
}