* GyrosGeier
heard back from Debian FTP team about his ffts upload
<GyrosGeier>
it got rejected because the copyright file was incomplete /o\
<GyrosGeier>
but at least there is progress
<GyrosGeier>
I can fix that, and reupload
<azonenberg>
Ok good to hear
<azonenberg>
We're still a ways from being ready to think about getting glscopeclient proper into anything resembling a release
<azonenberg>
but getting closer every day
<GyrosGeier>
yes
<GyrosGeier>
also ffts will need a few rounds, because the porters will probably want to take a look at it
<GyrosGeier>
a friend of mine is checking MacOS
<GyrosGeier>
I'm checking amd64 and ppc64el
<GyrosGeier>
m68k is rather low on my priority list :>
<azonenberg>
ffts has a lot of dynamic code generation
<azonenberg>
i suspect it only works on a fairly short list of ISAs
<azonenberg>
Replacing it with something else longer term is very much on the wishlist but unfortunately FFTW, the 500 pound gorilla of FFT libs, is GPL. What's the go-to FFT library on BSD, anybody know?
massi_ has joined #scopehal
massi has quit [Ping timeout: 256 seconds]
<GyrosGeier>
it does have a fallback
<azonenberg>
oh ok
<azonenberg>
i thought it was codegen only
<azonenberg>
m68k is also big endian which will present problems as pretty much none of the file and network I/O does endian conversion
<azonenberg>
because all of the scopes and initial target platforms are little endian
<azonenberg>
so for now i wouldn't even attempt to get it on m68k
<azonenberg>
ffts might be able to take it
<azonenberg>
but scopehal/glscopeclient won't
massi_ has quit [Remote host closed the connection]
<azonenberg>
i can't imagine anyone will actually try to do scope waveform analysis on a 68k in 2022 :p
<d1b2>
<DriftingPancake> Aren't you going to run into opengl issues way before that, anyway?
mewt has joined #scopehal
<azonenberg>
driftingpancake: hmmm if you can find a 68k with pcie
<azonenberg>
you can probably run amd open source drivers on it :p
someone-else has quit [Ping timeout: 256 seconds]
bvernoux1 has joined #scopehal
bvernoux has quit [Ping timeout: 240 seconds]
<d1b2>
<mubes> So...back to the usbtmc issue. Looking at the Linux device driver. I can see funkiness, but I don't know why yet.
<d1b2>
<mubes> I think the problem is that Siglent are splitting their usbtmc transmission into usb-frame sized packets.