<azonenberg>
lain, monochroma: btw, reading the datasheet it appears i was mistaken
<azonenberg>
the lower end SDA/WaveMaster 8 line (4/6/8/13 GHz) are 40 Gsps on all channels, and can only do 16 GHz with an external interleaving device
<azonenberg>
can only do 80 Gsps*
<azonenberg>
The 16 and 20 GHz models are also 40 Gsps on all channels but appear to also support an 80 Gsps mode
<azonenberg>
I'm curious how this plays with the serdes trigger
<azonenberg>
because the serial trigger is on channel 4
<azonenberg>
you cant do cdr triggering on 1-3
<azonenberg>
And with my other lecroy gear, when in interleaved mode you could only use channel 2 and 3
<azonenberg>
I wonder if channel 3 is usable as a trigger input in interleaved mode
<azonenberg>
And then the 25 and 30 GHz models have a "<25 GHz mode" for 4 channels at 40 Gsps and a ">= 25 GHz mode" for 2 channels at 80 Gsps
<azonenberg>
but then some of the other specs change in that mode
<azonenberg>
I wonder if instead of interleaving for oversampling, if they're doing DBI
<azonenberg>
like they do on the labmasters
<azonenberg>
(also yes doing a bit more reading it seems like channel 3 is the serdes trigger input in interleaved mode)
<azonenberg>
Found some older LeCroy datasheets. Looks like in the original WaveMaster 8Zi platform, they do indeed use DBI past 16 GHz
<azonenberg>
Which makes sense WRT why the upgrade cost from the 16 to the 20 is a lot more than the 20 to the 25
<azonenberg>
20 to 25 is probably just a software unlock
<azonenberg>
16 to 20 involves adding a lot of hardware
<azonenberg>
(it can be upgraded, it's a slide-in module)
<azonenberg>
Also hmmm, so it seems like they did some upgrades to the frontend between the base and the -B
<azonenberg>
actually sorry, between the -A and the -B
<azonenberg>
in particular, the base and the -A do not support 80 Gsps interleaving in the 16 or 20 GHz models
<azonenberg>
the -B does
<azonenberg>
it also seems like in the -B they don't start using DBI until 25 GHz, rather than 20 for the older ones
<azonenberg>
Adding support for all of these quirks to glscopeclient will be fun :p
<Bird|otherbox>
...DBI?
<azonenberg>
Digital bandwidth interleaving. It's the same thing they use in the LabMaster scopes >36 GHz
<azonenberg>
basically they use diplexers to split the incoming signal into multiple spectral bands, 2 in the case of SDAs or lower end LabMaster and 3 for the 100 GHz LabMaster
<azonenberg>
downconvert them to near baseband then digitize with a relatively low BW ADC
<azonenberg>
then upconvert in software to reconstruct the original signal
<azonenberg>
So everything past that diplexer only needs to support 1/N of the scope's full BW
<azonenberg>
there's a TSP video talking about the 100 GHz LabMaster from a couple years ago that goes into this in much more detail
<azonenberg>
So basically it looks like between the wavemaster 8Zi-A and -B families they managed to improve frontend performance to the point that 20 GHz BW was achievable without DBI
<azonenberg>
Which means they don't need that expensive RF deck until the 25 GHz model
<azonenberg>
it also means that all of the DBI-based scopes, if using DBI mode, are limited to 1/N of the actual hardware channels
<azonenberg>
since you use one ADC for low band and one for high band
<Bird|otherbox>
ah, yes
<azonenberg>
e.g. the "4 channel" 25/30 GHz SDAs are actually only 4 channel up to 20 GHz
<azonenberg>
and past there they're 2
<_whitenotifier-5>
[scopehal-apps] sam210723 forked the repository - https://git.io/JtMUE
electronic_eel has quit [Ping timeout: 246 seconds]
electronic_eel has joined #scopehal
Degi has quit [Ping timeout: 256 seconds]
Degi_ has joined #scopehal
Degi_ is now known as Degi
<d1b2>
<david.lenfesty> okay I'm close , getting attachment after floundering for a bit. Somehow broke the 5V and 7V supplies again, they seem to be extremely sensitive to something I'm doing, but I'm not sure what's killing them
<azonenberg>
Yaaay
<azonenberg>
meanwhile the AKL-PT2 boards are still in customs in LA
<d1b2>
<david.lenfesty> just realised attachment probably doesn't mean anything if you're not engrossed in the PD spec
<d1b2>
<david.lenfesty> not quite communicating yet
<azonenberg>
ah ok. Still, sounds like progress
<azonenberg>
I'm also running yet another sim of the AKL-PT1 with some different port configuration
futarisIRCcloud has quit [Ping timeout: 240 seconds]
futarisIRCcloud has joined #scopehal
Nero_ has joined #scopehal
Nero_ is now known as Guest28536
juli966 has joined #scopehal
<_whitenotifier-5>
[starshipraider] abowenk forked the repository - https://git.io/JtMRU