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
<d1b2> <swdee2> In Pulseview you can add Markers and measure the time span between two points arbitrarily. Does ngscopeclient has this measurement ability?
<d1b2> <azonenberg> We have two different ways to describe points in time
<d1b2> <azonenberg> Cursors describe relative points in time, i.e. deltas from T=0 of the waveform
<d1b2> <azonenberg> Cursors are a property of the waveform group, i.e. you can have several different groups each with their own cursors at different points
<d1b2> <azonenberg> They describe time, so as you scroll and zoom they stay at the same timestamp rather than the same pixel position
<d1b2> <swdee2> That is nice.... ok i'll go through the hassle of compiling it to have a play 😉
<d1b2> <azonenberg> and they are a property of the view rather than the waveform, so as you acquire new waveforms they stay in the same place
<d1b2> <azonenberg> Markers, on the other hand
<d1b2> <azonenberg> describe points in absolute time
<d1b2> <azonenberg> So if you drop a marker and acquire a new waveform, the marker will disappear from your screen. it's still there, but it's attached to a specific timestamp in a specific waveform
<d1b2> <azonenberg> and you have to jump back in history to the old waveform to see it
<d1b2> <swdee2> i see
<d1b2> <azonenberg> Markers are also global, not for a specific view
<d1b2> <azonenberg> so if you have several viewports that can all see the same point in time, they all have the same markers shown
<d1b2> <azonenberg> Markers also show up as events in protocol decode views, so you can easily correlate between timeline and packet-list views
<d1b2> <azonenberg> See here, for example. markers "foo" and "bar" show up in the waveform view, in the history view, and in the protocol analyzer
<d1b2> <swdee2> Got it
<d1b2> <azonenberg> Note that the above only applies to vertical cursors
<d1b2> <azonenberg> horizontal cursors are one of the longest standing open todo items in ngscopeclient, the ticket is one of I think 18 total github issues blocking the v0.1 release
<d1b2> <azonenberg> I hope to get to it soon
Degi has quit [Ping timeout: 252 seconds]
Degi has joined #scopehal
_sgstair is now known as sgstair
Esierp has joined #scopehal
<Esierp> Hello,
<Esierp> will it be possible, to use the DPAuxFilter with VCDImported digital Signals?
<Esierp> and 2.) and to use an digilent Analog discovery kit, do i need to compile scopehal-waveforms-bridge and use them in order to get connection to the NGscopeclient or is it all included?
Esierp has quit [Ping timeout: 240 seconds]
Esierp has joined #scopehal
ALTracer has joined #scopehal
ALTracer has quit [Read error: Connection reset by peer]
ALTracer has joined #scopehal
Esierp has quit [Ping timeout: 240 seconds]
ALTracer has quit [Read error: Connection reset by peer]
ALTracer has joined #scopehal
<d1b2> <rapzak_> 2.) I think if you use the scopehal-apps rep, you will get all.. https://www.ngscopeclient.org/manual/GettingStarted.html
<d1b2> <azonenberg> @rapzak_ he's on IRC and dropped offline before anyone could reply. Lol
<d1b2> <rapzak_> oki.. tried to be nice 🙂
<d1b2> <azonenberg> yeah i was gonna message him earlier but he was gone before i saw the question
<d1b2> <azonenberg> one of the problems with the discord-irc bridge is that discord users have no visibility into whether irc users are online or not
<d1b2> <azonenberg> if an irc user messages the channel and a discord user is offline they'll see it when they come back
<d1b2> <azonenberg> but if a discord user messages the channel and the irc user is offline they'll never know
<d1b2> <rapzak_> yearh makes sense
<d1b2> <rapzak_> is it easy to change the userinterface for the control of ex. osciloscope - or is that fixed to the type of intrument?
<d1b2> <azonenberg> Explain what you mean by "change the user interface"
<d1b2> <rapzak_> could be like have an option to set filtering in HW
<d1b2> <rapzak_> dropdown somewhere with average 1x-2x-4x etc
<d1b2> <rapzak_> in my case that setting would be for all channels
<d1b2> <azonenberg> You mean for hardware averaging?
<d1b2> <azonenberg> We actually support hardware averaging already, you just have to expose the correct APIs to configure it
<d1b2> <azonenberg> For now, it's only supported per channel in the API
<d1b2> <azonenberg> but if you had that as a global instrument setting we could probably find some way to make that work
<d1b2> <azonenberg> (near term, make it so the get/set functions just ignore the channel number and set the global average config)
<d1b2> <rapzak_> where is that API used? in the instrument driver or UI part or how is it used? maybe link to an example in existing file?
<d1b2> <rapzak_> I am doing kind of low samplerate 4ch power meter... with graphs... and the V and I channel will be linked for settings like average.. and the sample rate for all channels has to be the same..
<d1b2> <rapzak_> could be used to measure efficiency of example tripple output buck converter... current in stepper motors... maybe with bigger shunt power consumption of mcu/sub systems in lowpower..
<d1b2> <azonenberg> Hmm, so I'm wondering if it would make sense for the V and I to be considered one channel with two streams
<d1b2> <azonenberg> or two channels
<d1b2> <azonenberg> Oscilloscope::CanAverage(), GetNumAverages(), SetNumAverages()
<d1b2> <azonenberg> CanAverage controls if the GUI should show the averaging settings or not (you don't want to show the setting if hardware/firmware doesn't support the feature)
<d1b2> <azonenberg> then Get/Set are self explanatory
<d1b2> <azonenberg> As of now, the API has averaging exposed per channel not per stream
<d1b2> <azonenberg> So it might make sense to treat {V, I} as a single logical channel
<d1b2> <azonenberg> For example, on a regular multimeter if you are measuring an AC signal
<d1b2> <azonenberg> and it has a secondary measurement capability
<d1b2> <azonenberg> we can show voltage and frequency or something as two streams on one channel
<d1b2> <azonenberg> Does that seem like it fits your architecture well?
<d1b2> <rapzak_> yearh it may... but still the 2 strams come out so it for an example can be multiplied for watt
<d1b2> <azonenberg> Yep. instead of it being two channels it'll show up as one channel with two output ports