electronic_eel_ has quit [Ping timeout: 265 seconds]
<azonenberg>
Degi: that is a 1gbit 8b10b stream, you're looking at 1000base-SX idles measured right off the SFP
<azonenberg>
it is enig so there's some loss
<azonenberg>
but there was already 6 dB of insertion loss from going through the splitter, i wasnt too worried about an extra 0.1 dB or so
<_whitenotifier-f>
[scopehal-apps] azonenberg opened issue #123: Add "halt conditions" to freeze acquisition when interesting things happen - https://git.io/JJvPf
<_whitenotifier-f>
[scopehal-apps] azonenberg labeled issue #123: Add "halt conditions" to freeze acquisition when interesting things happen - https://git.io/JJvPf
<_whitenotifier-f>
[scopehal-apps] azonenberg labeled issue #123: Add "halt conditions" to freeze acquisition when interesting things happen - https://git.io/JJvPf
<_whitenotifier-f>
[scopehal-apps] azonenberg commented on issue #43: Add dialog for changing protocol decoder field colors - https://git.io/JJvPY
<azonenberg>
monochroma: hey, what would you call a filter that integrates waveforms to find the min/max/avg vs time over many acquisitions?
<_whitenotifier-f>
[scopehal] azonenberg opened issue #165: Refactoring: change "protocol decoder" to "filter" in all class/function names etc - https://git.io/JJvPc
<_whitenotifier-f>
[scopehal] azonenberg labeled issue #165: Refactoring: change "protocol decoder" to "filter" in all class/function names etc - https://git.io/JJvPc
<_whitenotifier-f>
[scopehal] azonenberg opened issue #166: 8b/10b decode: allow display of Dx.x characters and Kx.x characters independently as dotted or hex form - https://git.io/JJvP4
<_whitenotifier-f>
[scopehal] azonenberg labeled issue #166: 8b/10b decode: allow display of Dx.x characters and Kx.x characters independently as dotted or hex form - https://git.io/JJvP4
<_whitenotifier-f>
[scopehal-apps] azonenberg opened issue #124: When adding a filter with X axis units that don't match the current group, split the current group horizontally - https://git.io/JJvPS
<_whitenotifier-f>
[scopehal-apps] azonenberg labeled issue #124: When adding a filter with X axis units that don't match the current group, split the current group horizontally - https://git.io/JJvPS
<_whitenotifier-f>
[scopehal-apps] azonenberg opened issue #125: glscopeclient sometimes hangs with 100% CPU during shutdown rather than cleanly exiting - https://git.io/JJvP7
<_whitenotifier-f>
[scopehal-apps] azonenberg labeled issue #125: glscopeclient sometimes hangs with 100% CPU during shutdown rather than cleanly exiting - https://git.io/JJvP7
<_whitenotifier-f>
[starshipraider] azonenberg pushed 1 commit to master [+0/-0/±19] https://git.io/JJfI7
<_whitenotifier-f>
[starshipraider] azonenberg 4e6b62e - Continued MAXWELL layout. Rough-placed power and ESD diodes for all probe channels. Placed and routed power for right 3 channels.
bvernoux has joined #scopehal
<bvernoux>
hello
<bvernoux>
just received my electronic microscope 42MP ;)
<azonenberg>
miek: what'd they cost you? a new sma gauge set from minicircuits is almost $3K and i see a maury microwave one on ebay for $800ish
<miek>
£80 on ebay :p
<azonenberg>
o_O
<azonenberg>
you really got a deal
<miek>
indeed!
<azonenberg>
any way to check if they're working/properly cal'd?
<miek>
so that's the neat thing - as long as the dial gauges themselves are good, and the tips aren't completely mangled, they'll be fine
<miek>
there's no calibration adjustment, or critical lengths in the tips that are added to the gauges - you zero out the fixed/moving planes against each other with the setting gauge
<miek>
i can check the dial gauges are sane with a gauge block stack or something, but they seem fine
<azonenberg>
interesting
<miek>
i had always assumed there was some critical step length on the tip, or something like that - or maybe a critical height on the setting master. but the way SMAs are referenced/toleranced makes it much simpler
<miek>
so this is the MP (male, pin position) gauge: https://i.imgur.com/z2OBAbl.jpg - the innermost ring sits on the step in the pin contact, and the end of the SMA connector sits on the shoulder inside the fixed outer part of the gauge
<bvernoux>
miek, ha yes very nice MauryMicrowave a must
<bvernoux>
miek, it is for 2.92mm ... ?
<miek>
when you use the setting master, it aligns those two plans and you zero it out. then you just directly measure the distance from the SMA end (ref. plane) to the connector step
<miek>
bvernoux: it's SMA, though i think it would probably be applicable to the other compatible-ish standards
<miek>
* to the pin step
<bvernoux>
nice
<miek>
this also means that idea of making some DIY ones is totally viable :)
<azonenberg>
also reminder, we currently have DVI and TMDS decoding but not HDMI
<azonenberg>
or displayport
<azonenberg>
lain: also i started working on a feature that's a must for compelx protocols that we don't have triggers in hardware for
<azonenberg>
e.g. ethernet
<azonenberg>
stop the trigger when certain conditions are met in post
<azonenberg>
i.e. continue acquiring waveforms, decode each one, and if you see say an 8b10b error, stop triggering
<lain>
ah nice
<azonenberg>
Right now it's simple string matching on the protocol decode, you can trigger on a single protocol symbol whose text decode exactly matches, or does not match, a string
<azonenberg>
but we can add more options later on
<azonenberg>
not committed yet as i was trying to trigger on K27.7, the 8b10b code for 1000base-X start-of-frame
<azonenberg>
and it never triggered
<azonenberg>
in my attempts to investigate, i discovered the 8b10b decoder parsed that bit sequence wrong :p
<lain>
hehe
<azonenberg>
and now that i'm trying it with more compelx packets i'm finding more decode errors
<azonenberg>
so i'm fixing that first. Lol
<lain>
unit tests!
<azonenberg>
I'm working on acquiring a corpus of test data we can use to develop proper testxs
<_whitenotifier-f>
[scopehal-apps] azonenberg pushed 1 commit to master [+2/-0/±3] https://git.io/JJfWe
<_whitenotifier-f>
[scopehal-apps] azonenberg 5801996 - Initial implementation of conditional halting. Right now only supports string matching on protocol decodes, but easily extensible. Fixes #123.
<_whitenotifier-f>
[scopehal-apps] azonenberg closed issue #123: Add "halt conditions" to freeze acquisition when interesting things happen - https://git.io/JJvPf
<_whitenotifier-f>
[scopehal] azonenberg pushed 1 commit to master [+0/-0/±1] https://git.io/JJfWv
<_whitenotifier-f>
[scopehal] azonenberg a9a644a - 8B/10B: Fixed handling of K27.7, Dx.A7, and a few other edge cases. Fixed horizontal position of decode overlay to line up with UI boundaries, not recovered clock edges.
<azonenberg>
note that this works even on scopes without a hardware pattern trigger, since it works on the output of the filter graph
<azonenberg>
so it can do arbitrarily complex trigger conditions, however it only works on stuff that was within the acquisition window obviously
<azonenberg>
in the immediate future i'll also support greater/less than and other stuff on analog channels
<azonenberg>
So you can, say, trigger when frequency of a signal drops below 1.15 GHz
<azonenberg>
or rise time of a signal is more than 2o00ps
<azonenberg>
200ps*
kbeckmann has joined #scopehal
<_whitenotifier-f>
[scopehal-apps] azonenberg pushed 2 commits to master [+0/-0/±2] https://git.io/JJfln
<_whitenotifier-f>
[scopehal-apps] azonenberg 72784b8 - Improved default config for halt conditions dialog
<_whitenotifier-f>
[scopehal-apps] azonenberg 2906048 - Conditional halt now supports analog channels and a much wider range of filters