<azonenberg>
SI verification of MEAD logic pod at 1.25 Gbps including cable losses. This is what you would see at the input to MAXWELL, with another about 0.6-0.7 dB of loss from the OSHPark host board
<azonenberg>
Which is to say, this is about what the eye would look like all the way from probe to FPGA on an average length trace on MAXWELL
<azonenberg>
The longest trace, channel 1, will have about another 0.6 dB of loss past what you see here
<azonenberg>
The vertical eye opening here is about 200 mV and a kintex-7 IBUFDS is specced to go down to 100 mV differential. So I think we have a fair bit of margin
<azonenberg>
Hmmmm
<azonenberg>
actually there's plenty of vertical margin but horizontal margin is less good. Since we're doing async oversampling rather than hardware CDR you need at least two samples in the eye opening to get data recover
<azonenberg>
So this is a little iffy
<azonenberg>
Also threw together a quick and dirty diffprobe design
<azonenberg>
schematic done, need to do a little review on it then start layout
<azonenberg>
this is a naked diff amplifier board, takes +/- 2.5V on a pair of barrel jacks
<azonenberg>
no regulation, no offset calibration
bvernoux has quit [Quit: Leaving]
<_whitenotifier>
[starshipraider] azonenberg pushed 1 commit to master [+4/-0/±0] https://git.io/JLyn2