<_whitenotifier-e>
[scopehal-apps] azonenberg 9770856 - Middle click on Y axis now autoscales filter range/offset
<azonenberg>
ok so the refactoring went well but i'm out of time for the night
<azonenberg>
Next step is to add support for multimeters that are not part of a scope
<benishor>
should that also work with SDS2104X plus?
bvernoux has joined #scopehal
<azonenberg>
benishor: multimeter options for scopes are already supported
<azonenberg>
however i do not know if the siglent driver implements that feature
<azonenberg>
do the 2000x+ scopes have it?
<azonenberg>
i dont think the demo scope i have has it installed
<azonenberg>
@louis: So, thinking more about the multimeter stuff
<azonenberg>
iirc the old (WIP) multimeter dialog uses the GraphWidget code to display historical readings
<azonenberg>
i feel like using WaveformArea's and exposing the old measurements as a Waveform might make more sense
<azonenberg>
the question is how to fit this into the rest of the model because it's not coming on a trigger synced to anything else, we're recording single points rather than waveforms in sequence
<azonenberg>
which means we would end up basically having a single Waveform that updates periodically with new points and timestamps or something, idk how it would work
<azonenberg>
and how likely are we to want to use filter graph operations and such on historical multimeter data?
<azonenberg>
at the same time i feel like unifying it all will be a huge improvement for UX if we can figure out a good way to make it happen
<d1b2>
<louis> @azonenberg you don't have a MSO6x kicking around anymore, do you?
<d1b2>
<louis> I'm upstreaming my tek deep memory fixes; seems the chunking wasn't actually the fix, it was something else (either parsing malformed WFMO? responses, or manually setting DATA:START and DATA:STOP)
<d1b2>
<louis> But would be good to try on the MSO6 that was giving you trouble at deep depth
<azonenberg>
I do not have that scope, it was on loan from a friend who was out of town for the holidays of... christmas 2021 i think
<azonenberg>
i can ask if he'll let me borrow it again at some point?
<d1b2>
<louis> ¯_(ツ)_/¯
<d1b2>
<louis> If we un-limit the depth selector and someone opens a ticket for their MSO6 we can revisit it
<azonenberg>
Yep
<azonenberg>
makes sense
<azonenberg>
Right now I think you folks are the most active MSO6 users
<azonenberg>
so if you don't hit issues we're good :p
<d1b2>
<louis> wow, just crashed out MSO5 so bad i had to go physically reboot it
<d1b2>
<louis> something to talk about on the Tek call will be how to reset the #@ ($&@# ($ SCPI server when it gets jammed up
<azonenberg>
lol
<azonenberg>
i see you've had the pleasure :)
<azonenberg>
that's one thing i can say about LeCroy
<azonenberg>
i don't think i have ever hard crashed their scpi stack
<azonenberg>
no matter how much junk i've sent at it, sometimes the command works sometimes it doesn't
<azonenberg>
but it's always ready for the next command
<azonenberg>
i have also had hard crashes on Siglent that hung the UI and required a physical reboot as well
<azonenberg>
so far, none have been reproducible enough for me to file a bug report
<azonenberg>
I'm actually wondering about the potential of writing a scpi fuzzer of some sort
<azonenberg>
basically throw commands at the scope and check for either hangs or state being set wrong
<azonenberg>
i bet we could find a ton of firmware bugs if we actually tried, judging by how many we're finding by accident
<azonenberg>
but if we actively go and look for bugs we could probably greatly improve stability of the vendor firmware
<azonenberg>
(assuming they're willing to actually fix bugs once we find them)