<_whitenotifier-7>
[scopehal-apps] bvernoux 74ea5d2 - Fix crash with WindowSCPIConsoleMenu when SCPIInstruments m_nickname is empty
<_whitenotifier-7>
[scopehal-apps] bvernoux 2cccdfd - Revert "Fix crash with WindowSCPIConsoleMenu when SCPIInstruments m_nickname is empty" This reverts commit 74ea5d212c5778d52f11a327199d4715408d65b8. Use a default nickname for each instrument Force user to choose a nickname not empty (if it is empty)
<_whitenotifier-7>
[scopehal-apps] bvernoux 6bd32aa - Merge branch 'glscopeclient:master' into master
<_whitenotifier-7>
[scopehal-apps] azonenberg cf3e09c - Merge pull request #500 from bvernoux/master ngscopeclient Fix crash on Instrument empty nickname
<azonenberg>
ehntoo: just checking in re your power supply PR, seems it hasn't been updated with those fixes we talked about yet?
massi has quit [Remote host closed the connection]
<_whitenotifier-7>
[scopehal-apps] azonenberg 7199a0d - Continued work on channel drag and drop
<_whitenotifier-7>
[scopehal-apps] azonenberg 49a0c07 - WaveformArea: draw fancy drop targets when dragging waveforms
<d1b2>
<Mughees> How can I quickly generate a sparse sinewave...for testing? And possibly get some more examples of sparsewaveforms? Need them for testing........
<azonenberg>
mughees: hmmm
<azonenberg>
most of the measurement filters output sparse results, can you get something close enough with that?
<d1b2>
<Mughees> hmmmm
<d1b2>
<Mughees> 🙂
<azonenberg>
in general we aggressively avoid using sparse waveforms when we can
<azonenberg>
because uniform is so much more efficient to work with
<d1b2>
<louis> May be wise to have a filter that sparsifies a uniform waveform just for testing
<d1b2>
<Mughees> Yes...but for something that measures ike RMS we would need to support both waveform types
<d1b2>
<Mughees> For AC RMS measurement and sparse waveform, do you think this logic is correct..(Just want to clarify my understanding of sparsewaveform) for (size_t i = 0; i < length; i++) { temp += ((sadin->m_samples[i] - average) * (sadin->m_samples[i] - average)); } temp /= length; temp = sqrt(temp);
<d1b2>
<louis> Yes. Generally I think we want to support both sparse and uniform
<d1b2>
<Mughees> averaging for complete duration seems unecessary...
<d1b2>
<Mughees> For some measurements, it makes ife easy.
<d1b2>
<louis> Yes. Generally waveform transform filters are uniform input -> uniform output (if possible) otherwise sparse -> sparse. Because uniform is more efficient than sparse if the waveform is actually uniform
<d1b2>
<louis> But since the measurement waveforms are sparse they output sparse. But they must generally support both types of input, so do uniform -> sparse and sparse -> sparse
<d1b2>
<Mughees> wait a sec...I think for RMS measurement, a uniform analog waveform as output is more suitable? As it is just a scalar masurment?
<azonenberg>
all scalar outputs should be uniform with one sample
<d1b2>
<Mughees> yeah
<d1b2>
<louis> Oh I was assuming we would want one sample per cycle with the RMS for that cycle, then a user could average if they wanted it for the whole or some part of the waveform
<d1b2>
<Mughees> Hmmm....do we really wanna dive that deeper?
bvernoux has quit [Read error: Connection reset by peer]
<azonenberg>
louis: long term i'd like filters that do both
<azonenberg>
one that sparsifies a uniform waveform for testing
<azonenberg>
and one that resamples a sparse waveform at a user-provided uniform rate
<azonenberg>
for testing as well as for interop with a filter that may require sparse or uniform input
<d1b2>
<johnsel> Merged PR #500 into master. congrats everyone