azonenberg changed the topic of #scopehal to: libscopehal, libscopeprotocols, and glscopeclient development and testing | https://github.com/glscopeclient/scopehal-apps | Logs: https://libera.irclog.whitequark.org/scopehal
<d1b2> <johnsel> In that case, I found a bug
<d1b2> <johnsel> If the peak is position such that the label will be just on th edge/partially outside the window it gets blurry presumably because it keeps changing position a (few) pixel(s)
<azonenberg> So, when a peak disappears completely
<azonenberg> the marker is supposed to fade out over several frames, but if the peak returns it'll come back
<azonenberg> this is probably an edge case where every frame or two it is/isnt on screen
<azonenberg> maybe it's showing up as two separate peaks
<azonenberg> By all means poke at it, but it's low priority in comparison to some of the other stuff i'm working on right now
<d1b2> <johnsel> That could be it, I can't replicate it now
<d1b2> <johnsel> I did have a second crash while zooming through the FFT window horizontally and vertically
<d1b2> <johnsel> I can scroll and middle click let's say 10-15 times but at some point it crashes. Will try to pull some debug info next time
Degi has quit [Ping timeout: 255 seconds]
Degi has joined #scopehal
josuah has quit [Remote host closed the connection]
<_whitenotifier-f> [scopehal] azonenberg pushed 6 commits to master [+0/-0/±17] https://github.com/glscopeclient/scopehal/compare/f8c94209777b...b63c57e4ebfc
<_whitenotifier-f> [scopehal] azonenberg c0845a1 - Initial vkFFT conversion of SpectrogramFilter, super inefficient. See #660
<_whitenotifier-f> [scopehal] azonenberg 9165fcd - Removed FFTS referenecs from vestigial OFDM decode block. See #660
<_whitenotifier-f> [scopehal] azonenberg 5dd56ad - Removed FFTS calls in vestigial TDR de-embed filter. See #660
<_whitenotifier-f> [scopehal] ... and 3 more commits.
<_whitenotifier-f> [scopehal] niklaut starred scopehal - https://github.com/niklaut
Abhishek__ has joined #scopehal
esden_ has joined #scopehal
Abhishek_ has quit [*.net *.split]
esden has quit [*.net *.split]
esden_ is now known as esden
Abhishek__ is now known as Abhishek_
sgstair has quit [Ping timeout: 260 seconds]
sgstair has joined #scopehal
<_whitenotifier-f> [scopehal] azonenberg pushed 3 commits to master [+0/-0/±10] https://github.com/glscopeclient/scopehal/compare/b63c57e4ebfc...dd39d0f58def
<_whitenotifier-f> [scopehal] azonenberg 7c276dd - ComputePipeline: Added DispatchNoRebind
<_whitenotifier-f> [scopehal] azonenberg ba386c8 - Window functions now support offsets allowing multiple invocations within a single buffer. Spectrogram now uses GPU based windowing
<_whitenotifier-f> [scopehal] azonenberg dd39d0f - Removed false graphwidget dependency from apple silicon scopehal build
<_whitenotifier-f> [scopehal-apps] azonenberg pushed 2 commits to master [+0/-0/±3] https://github.com/glscopeclient/scopehal-apps/compare/bcfe3096b6dd...fa479f1da225
<_whitenotifier-f> [scopehal-apps] azonenberg ab0f8b6 - Updated tests to use ngscopeclient shaders
<_whitenotifier-f> [scopehal-apps] azonenberg fa479f1 - Updated submodules
<_whitenotifier-f> [scopehal] azonenberg pushed 4 commits to master [+0/-0/±14] https://github.com/glscopeclient/scopehal/compare/dd39d0f58def...654d56bc604e
<_whitenotifier-f> [scopehal] azonenberg 4785295 - Marked a bunch of overriden virtuals as "override"
<_whitenotifier-f> [scopehal] azonenberg 3e6eb24 - SpectrogramFilter: build on apple silicon now that FFTS dependency has been removed
<_whitenotifier-f> [scopehal] azonenberg 051f763 - Marked more functions as override
<_whitenotifier-f> [scopehal] azonenberg 654d56b - Fixed a bunch more override warnings
<_whitenotifier-f> [scopehal-apps] azonenberg pushed 1 commit to master [+0/-0/±1] https://github.com/glscopeclient/scopehal-apps/compare/fa479f1da225...30029d93abdd
<_whitenotifier-f> [scopehal-apps] azonenberg 30029d9 - Updated submodules
bvernoux has joined #scopehal
<_whitenotifier-f> [scopehal] azonenberg commented on issue #803: Rigol: Floating point exception - https://github.com/glscopeclient/scopehal/issues/803#issuecomment-1734216948
bvernoux has quit [Quit: Leaving]
<d1b2> <johnsel> wtf all of a sudden ngscopeclient won't run
<d1b2> <johnsel> ... name=FilterGraphExecutor[6].queue QueueManager creating family=0 index=15 name=FilterGraphExecutor[7].queue PS C:\Program Files\ngscopeclient>
<azonenberg> @johnsel huh, try a clean build?
<d1b2> <johnsel> this is a binary install
<azonenberg> oh interesting
<d1b2> <johnsel> I reinstalled the binary install too
<d1b2> <johnsel> Cursed PC syndrome again I think
<d1b2> <johnsel> I bought a new motherboard that should be good, I think the CPU + RAM + PSU are next
<d1b2> <johnsel> I really can't deal with these weird random issues anymore
<d1b2> <johnsel> I may have a borked Windows install again. It keeps BSODing too every once in 2 weeks or so while idling
<azonenberg> did you try an older binary build in case it was a regression? ... oh
<azonenberg> lol
<azonenberg> in that case i cant help you :p
<d1b2> <johnsel> it won't even write a dump to the disk that's how bad it BSODs
<d1b2> <johnsel> I really think it must be the CPU itself at this point
<d1b2> <johnsel> Yeah I'd try that, if it wasn't for the same version worked last night
<d1b2> <johnsel> I literally ran the same install and now it crashes with an access violation error
<azonenberg> yeah your hardware / OS sounds well and truly screwed lol
<azonenberg> random memory corruption or something
<d1b2> <johnsel> yeah you'd think memory, it feels like memory. But I have ran hours and hours of memtest, and it clears just fine
<azonenberg> reminds me of the time i was asked to help a classmate as an undergrad with a compsci 1 homework assignment that was failing to compile under visual studio
<d1b2> <johnsel> I've removed DDR sticks
<azonenberg> turns out one of the libc header files had a bitflip in it
<d1b2> <johnsel> run the system on 1 stick
<d1b2> <johnsel> no differnce
<azonenberg> and was giving syntax errors lol
<juh> i wish the fact that i do server maintenance could help here, but at work we would just casually "swap the cpu" and then "swap the motherboard again" if it didn't work
<azonenberg> lol
<azonenberg> yeah if you have a whole datacenter worth of spare parts and just need the thing running again thats the way to do it
<d1b2> <johnsel> well I started my professional career doing sysadmin, I know how diagnose these issues properly I just don't have/didn't have the budget to buy another set of everything. An
<juh> the amount of money just floating around on shelves in there is mind boggling
<d1b2> <johnsel> well yeah, downtime is even more expensive
<d1b2> <johnsel> I think by law I can order online and have a period of 30 days where I can return without giving the seller a reason for it
<d1b2> <johnsel> So I will probably just do that with a big vendor that can take the costs of restocking it
<d1b2> <johnsel> Just buy everything (except the mobo which I alread bought) and swap it out piece by piece until I figure out what is borked
josuah has joined #scopehal