Tootfinder

Opt-in global Mastodon full text search. Join the index!

@azonenberg@ioc.exchange
2026-02-11 20:43:07

And another nice shader speedup. Not quite the massive boost of the vector frequency/phase filters, but it's next in line after them.
I'm gonna use this random-access LFSR trick on so many filters that work with scrambled data lol.
v0.1.1 (20M point memory depth)
* PRBS31 generation: 65 ms
* PRBS31 verification: 360 ms
Latest (20M point depth)
* PRBS31 generation: 3 ms (21.6x)
* PRBS31 verification: 1.7 ms (211x)

@azonenberg@ioc.exchange
2026-01-27 00:56:51

Current summary of all of the GPU acceleration work I've been doing in ngscopeclient since v0.1.1.
Other than the CDR PLL and eye pattern which were high-priority targets that were worth the effort, and the Ethernet which I got nerdsniped on, these are all trivial inner loops of just a few lines that I threw at the GPU without any significant tuning, so there's probably more performance upside.
But I'm not complaining with what I've pulled off so far. (As you ca…

Logarithmic bar graph of filters and the speedups obtained since shipping v0.1.1

Filter  Speedup
AC Couple       10
Average 5.6
Base    17
CDR PLL 7.5
Clip    4
DDJ     16
Downconvert     5.8
Downsample (AA filter)  16.3
Downsample (no AA)      22
Duty Cycle      8
Emphasis        13.2
Envelope        14.5
Ethernet - 100baseTX    10
Eye pattern     25
Histogram       12
PAM Edge Detector       2
TIE     5.3
Vector Frequency        1040
Vector Phase    243