<_whitenotifier-8>
[scopehal-apps] azonenberg 5cf011d - Updated to latest scopehal
someone-else has quit [Quit: Connection closed]
<azonenberg>
ok, three different test boards just came in
<azonenberg>
Assembling shortly...
nelgau has joined #scopehal
nelgau has quit [Remote host closed the connection]
Degi_ has joined #scopehal
Degi has quit [Ping timeout: 256 seconds]
Degi_ is now known as Degi
<d1b2>
<mubes> @dannas the timeout waiting for a scope response can look like a hang sometimes. It's worth checking if you're actually waiting for something that's not arriving.
<_whitenotifier-8>
[scopehal] mubes e63ca9d - Runt trigger upper and lower limit reversal fix
<_whitenotifier-8>
[scopehal] azonenberg 5a32781 - Merge pull request #533 from mubes/fix_runt_limits Runt trigger upper and lower limit reversal fix
<azonenberg>
ok soooo i'm trying to speed up the downconversion filter
<azonenberg>
I had to convert some math from single to double precision because it was losing precision and causing phase shifts in long captures
<azonenberg>
but i want to see if there is any way i can cleverly get back to single precision somehow since ultimately all i need is sin(timestamp*freq) and cos(timestamp*freq)
<azonenberg>
(and do it in a way that is vectorizable, so avoiding persistent accumulators if i can)
<_whitenotifier-8>
[scopehal] azonenberg cce0595 - DownconvertFilter: Use SetupOutputWaveform() to avoid unnecessary timestamp copies etc
<_whitenotifier-8>
[scopehal] azonenberg 8df31a8 - DownconvertFilter: switched to integrating single precision phase with subtraction to keep results within fp32 range
<_whitenotifier-8>
[scopehal] azonenberg e067be8 - DownsampleFilter: added option to turn off antialiasing filter and just do nearest neighbor sampling if input already has had high freq content removed