azonenberg changed the topic of #scopehal to: ngscopeclient, libscopehal, and libscopeprotocols development and testing | https://github.com/ngscopeclient/scopehal-apps | Logs: https://libera.irclog.whitequark.org/scopehal
<_whitenotifier-4> [scopehal-apps] azonenberg pushed 2 commits to master [+0/-0/±4] https://github.com/ngscopeclient/scopehal-apps/compare/1ff6d5a8dbcd...a4c46ce5a7ac
<_whitenotifier-4> [scopehal-apps] azonenberg da7294b - Y cursors: fixed rounding error. See #527.
<_whitenotifier-4> [scopehal-apps] azonenberg a4c46ce - Y cursors: added scopesession serialization. See #527.
Degi has quit [Ping timeout: 272 seconds]
Degi has joined #scopehal
cyborg_ar has quit [Ping timeout: 248 seconds]
cyborg_ar has joined #scopehal
<d1b2> <azonenberg> @josHua in general i've been leaning on tek for a while trying to get them to make their scpi backend less buggy and crashy
<d1b2> <azonenberg> but my main contact at tek left the company a few months ago
<d1b2> <azonenberg> and i never got in touch with his replacement
<d1b2> <azonenberg> Hmmmm
<d1b2> <azonenberg> So i have y axis cursors about 90% done
<d1b2> <azonenberg> the main thing remaining is to display position indicators on the line
<d1b2> <azonenberg> The question is where? my natural inclination would be at the far right, but then it'll step on the grid markings
<d1b2> <azonenberg> the two obvious options are a) somewhere else (left? center?) or b) on the right but with like a black background or something to make it easier to read, at the cost of covering up the y axis markings for the grid
<_whitenotifier-4> [scopehal-apps] azonenberg pushed 1 commit to master [+0/-0/±2] https://github.com/ngscopeclient/scopehal-apps/compare/a4c46ce5a7ac...cd2743948b9b
<_whitenotifier-4> [scopehal-apps] azonenberg cd27439 - Added text readouts to Y axis cursors. Fixes #527.
<_whitenotifier-4> [scopehal-apps] azonenberg closed issue #527: ngscopeclient: cursor support - https://github.com/ngscopeclient/scopehal-apps/issues/527
<d1b2> <azonenberg> This will do for now i think
<dingwat> azonenberg: I think that looks good, pretty similar to how some other software does it. Definitely don't want to cover up the y-axis
<azonenberg> dingwat: one open question is if we want to add any kind of data readout and if so, how
<azonenberg> for x axis it's easy as a waveform can have only one value at a given x coordinate
<azonenberg> but it will likely cross a given y axis value many times
<dingwat> hmm. closest y-crossing to cursor? that might get kinda distracting as it bounces around though
<dingwat> azonenberg: honestly it would be kinda cool if you could over your mouse near cursor-trace intersections and it would pop up the point.
<dingwat> s/over/hover/
<azonenberg> So that's the sort of thing i have in mind looking forward
<azonenberg> making it fast would be nontrivial
<azonenberg> but thats something i want to explore
<azonenberg> as you start getting e.g. multiple waveforms in one plot it would be even more fun
<dingwat> azonenberg: yeah, I imagine it's non-trivial to implement, but it seems like the most useful behaviour. Maybe you can compute all the intersections as the cursor(s) are adjusted, so then it's a matter of figuring out when the mouse is close to a given point? That could get really painful for a big waveform though
<azonenberg> yeah exactly the challenge is doing it at >60 FPS on potentially a >100M point waveform
<azonenberg> lots of things are easy to do by brute force on small waveforms
<azonenberg> but on big ones, much less so
<azonenberg> anyway food for thought
<azonenberg> Near term my priority is banging out the remaining... 17? issues blocking the v0.1 release
<dingwat> maybe it could be something that only shows up when the acquisition is stopped? So it's not something that has to happen every frame, and it could be an async task to find all the intersections
<azonenberg> it's more a quesiton of how to make the UX work
<azonenberg> this will tie into other things like having markers and cursors snap to data points on digital waveforms when you're dragging them
<dingwat> ah yeah, good point
<dingwat> saleae does a really good job on that IMO
<azonenberg> yeah
<azonenberg> right now my focus is on knocking out issues blocking the release and not merging any other PRs or doing any new feature dev
Degi has quit [Ping timeout: 260 seconds]
Degi has joined #scopehal