azonenberg changed the topic of #scopehal to: libscopehal, libscopeprotocols, and glscopeclient development and testing | https://github.com/azonenberg/scopehal-apps, https://github.com/azonenberg/scopehal, https://github.com/azonenberg/scopehal-docs | Logs: https://freenode.irclog.whitequark.org/scopehal
<d1b2> <Steph> tried scopehal on my sds1104x-e (same as the supposedly once-working sds1204x-e except 100mhz) and nothing seems to be working
<d1b2> <Steph> all the vbs commands which seems to be used in the lecroy driver do not work
lethalbit has joined #scopehal
<monochroma> Error_404: ^
bvernoux has quit [Quit: Leaving]
Stephie has joined #scopehal
<azonenberg> steph: Yeah i think the lecroy and siglent drivers are beginning to diverge more
<azonenberg> i'm using more automation commands in the lecroy driver lately because they provide better control
<azonenberg> File a bug and we can work on it
<d1b2> <Steph> i think the automation commands are completely missing from my sds series, at least they're not listed anywhere in the programming guide
<d1b2> <Steph> i'll file an issue
<azonenberg> steph: yes
<azonenberg> i believe the lower end siglents do not have them because they're probably not windows ce based
<azonenberg> the higher end siglents run a custom fork of lecroy maui, the same as the wavesurfer 3000 family
<d1b2> <Steph> yeah no
<d1b2> <Steph> i've been shell on this scope
<d1b2> <Steph> it's linux
<azonenberg> ah ok. then yeah
<azonenberg> So there are 3 different classes of scope to consider then
<azonenberg> high end lecroy: designed in house, x86 based running full windows
<d1b2> <Steph> there's an old firmware image which left a SCPI command which runs arbitrary commands as root lol
<d1b2> <Steph> if you want to reverse engineer stuff thats a fun way to do it
<azonenberg> low end lecroy / high end siglent: collab between both companies, arm based running windows ce
<Stephie> also, hi
<Stephie> i'd probably better use this
<azonenberg> then low end siglent: linux based
<azonenberg> the top two should work
<Stephie> yeah
<Stephie> would you want 2 drivers or?
<azonenberg> the low end siglent are most likely compatible with the lecroy scpi command set other than the automation/vbscript stuff
<azonenberg> Right now the siglent driver is derived from the lecroy driver and implements only the functions where the commands are different
<azonenberg> given that the lecroy driver is gradually diverging due to using more automation commands
<azonenberg> it may make sense to split them off completely
<Stephie> mhmm
<Stephie> it seems a little weird to have drivers be per scope brand and not so much per scope line
<azonenberg> you would still use the lecroy driver for wavesurfer 3000 and sds3000 or whatever the siglent badged version is called
<azonenberg> It's totally fine to have more than one driver for a brand
<azonenberg> lecroy in particular has been extremely stable api wise
<Stephie> fair enough
<azonenberg> the driver i wrote for modern maui scopes still runs on 2005 era winXP/2000 based scopes
<_whitenotifier-f> [scopehal] RX14 opened issue #341: sds1104x-e doesn't work - https://git.io/JksiG
<azonenberg> But the "lecroy" driver is actually "lecroy maui"
<azonenberg> and, for example, will not run on the really old vxworks lecroys
<Stephie> I'm really tired and about to head to bed, so that issue isnt very good
<Stephie> but, thanks
<azonenberg> So if we wanted to support those we'd probably write a fully custom driver for it
<azonenberg> and give it a new name
<Stephie> yeah, makes sense
<azonenberg> right now, there's generally been enough commonality between devices from the same vendor that we can get away with just special casing the bits that are different
<Stephie> neat
<azonenberg> but when you have major architectural changes like that, it may not make sense
<Stephie> yeah, i think it's time to split lecroy and siglent at least
<Stephie> since soo much of it is automation commands
<Stephie> anyway, thanks for the help
<Stephie> nini
<Bird|otherbox> mubes: if you're still about -- why would the graphics card feed +5VSB to the video ports for DDC?
Degi has quit [Ping timeout: 256 seconds]
Degi has joined #scopehal
<_whitenotifier-f> [scopehal-apps] azonenberg pushed 1 commit to master [+0/-0/±1] https://git.io/Jksyt
<_whitenotifier-f> [scopehal-apps] azonenberg 740f886 - Removed debug message left in by accident
StM has quit [Quit: No Ping reply in 180 seconds.]
StM has joined #scopehal
<_whitenotifier-f> [scopehal] loxodes forked the repository - https://git.io/JkspR
_whitelogger has joined #scopehal
<loxodes> (I have a tektronix mdo3000 driver working somewhat for analog channels, it may also support mso/mdo/dpo4000 series)
<azonenberg> loxodes: oh cool
<azonenberg> send a PR when you've got something ready
<loxodes> will do (I may not bother with spectrum analyzer support initially because the MDO3000 can't run it simultaneously with digital and/or analog)
<azonenberg> loxodes: if it's only a software fft i don't normally bother supporting it
<azonenberg> because we can do that pc-side much faster
<azonenberg> whereas with mso5/6 you can run the spectrum views on the digital downconverter independently of the time domain views
<azonenberg> so it's effectively a completely separate channel sharing the same frontend
<loxodes> The MDO3000 has a separate frontend for the spectrum analyzer mode with more bandwidth
<azonenberg> oh interesting. in that case, we probably should support it eventually. but not a priority for initial support
_whitelogger has joined #scopehal
<d1b2> <Rogan> > mubes: if you're still about -- why would the graphics card feed +5VSB to the video ports for DDC? @Bird|otherbox#0000 might be misunderstanding the question, but afaik, the DDC eeprom in the monitor was historically powered from the pc, not from the display. Which is why pcs can still detect a monitor being connected, even if there is no power cord connected.
<azonenberg> rogan: the question is, why is it feeding +5V *standby* power to the monitor
<azonenberg> rather than a rail that's only active when the computer is booted, aka normal atx +5V
<azonenberg> why would the monitor need power if the pc is off?
<azonenberg> it's not showing any video, does the monitor care there's a pc connected?
<d1b2> <Rogan> Ah, fair enough! That makes no sense at all. Unless it is to allow the monitor to go into sleep mode, rather than “no source detected”?
<azonenberg> That was kinda my thought
<d1b2> <Rogan> Like hey, I’m here, just not doing anything right now?
<azonenberg> +5VSB present, no sync pulses = sleep
<d1b2> <Rogan> Wonder what happens if you yank power from the pc, but not the monitor? Would it wake up again and display that message?
<azonenberg> Good question. I dont have any computers sitting around *not* doing something important that i can test this on :p
<d1b2> <Rogan> Haha
<azonenberg> well ok, at least that are connected to monitors
<azonenberg> I have a desktop sitting on the table a few feet from me that was built to show video on the projector in the classroom/conference room i was building pre covid
<azonenberg> the original intent was to be able to host trainings, have hackathons, etc so i set up a corner of this large family-room type room with a folding table, some chairs, whiteboards, a pull-down projector screen
<azonenberg> only thing missing was the projector
<azonenberg> Then covid hit and i suddenly had no reason to plan for people coming over the house
<azonenberg> so i never bought a projector :p
<d1b2> <Rogan> Haha. I envy the American basements.
<d1b2> <Rogan> Although not having to worry about sump pumps is a nice thing, I guess! 🤪
<azonenberg> If only
<azonenberg> The whole lower level of my place is below to at ground level. It's built into a hill, so depends on what part of the house you're in... where i'm sitting the top of my head is about at ground level
<azonenberg> in the lab, it ranges from knee high to floor level
_whitelogger has joined #scopehal
<d1b2> <Rogan> That’s neat! How are you measuring your sump flow rate? Rrd backend clearly. Just a simple Hall effect impeller?
<azonenberg> no. 4-20 ma submersible level sensor
<azonenberg> i get realtime depth, then use a conversion table to get volume
<azonenberg> (tank isnt an ideal cylinder, tapered walls and pipes inside etc)
<azonenberg> then dvol/dt = flow
<azonenberg> rainy season is just starting, it hits 30L/hr in winter
<d1b2> <Rogan> Yikes
<azonenberg> yeah we just had a very expensive excavation project on the other side of the house
<azonenberg> the sump only protects the uphill part of the house, the downhill was flooding last winter due to gravity-fed drainage pipes installed decades ago having failed and clogged
<azonenberg> of course the original installation had no cleanout access or anything so we couldnt fix it
<azonenberg> So we had to get a construction crew in to dig up that whole side of the house and put in new drain pipes, properly laid in a gravel bed etc, with cleanout access so that if they start to build up debris in the future we can actually clear them without digging again
<azonenberg> That project is the reason that I don't have hardware for the MAXWELL logic analyzer yet
<azonenberg> it ate up all my savings :p
<azonenberg> And then some
<azonenberg> the sad part is we had just finished a complete renovation of the inside of the house (we bought it 3 years ago) and were looking forward to settling in and actually living there
<azonenberg> and of course, the outdoor plumbing is like "hey, you didn't renovate ME! I feel left out"
<azonenberg> The good news is that although the construction was expensive, there was essentially no damage from the flooding. We knew the house had a history of moisture problems
<azonenberg> So the whole first floor was redone in liquid-tolerant materials like vinyl trim molding, epoxy flooring, etc so we could get a few mm of water on the floor with no long term ill effects
<azonenberg> Just a mess we had to vacuum/mop up before things went moldy
<d1b2> <Rogan> Yeah, damp / water problems suck, because they can fly under the radar for quite a while until it becomes obvious
<azonenberg> This was obvious immediately, it just hit when i was visiting family for christmas
<azonenberg> so it had been going on for a while before i got home
<azonenberg> But my foresight came in handy and there was no damage
<azonenberg> just a mess to clean up, then it came back again next time it rained
<azonenberg> i dug some temporary trenches in the yard by hand which eliminated further flooding but were a tripping hazard and mosquito breeding ground
<azonenberg> then called an excavation contractor to do a permanent job
<azonenberg> Now i have a water sensor cable hidden behind trim in one of those rooms so if i ever get more flooding an audible alarm will ring in seconds
<azonenberg> it's run by the same raspberry pi as the sump monitor
bvernoux has joined #scopehal
juli966 has joined #scopehal
bluecmd[m] has quit [Ping timeout: 260 seconds]
bluecmd[m] has joined #scopehal
<_whitenotifier-f> [scopehal-apps] azonenberg closed issue #255: With GDK_SCALE=2, trigger cannot be dragged - https://git.io/JTiJ6
bvernoux has quit [Read error: Connection reset by peer]