<ZipCPU>
Ah ... using his github name, I'll be forwarding your link to daveshah1
<ZipCPU>
Looks like I got it! or ... at least the package unzipped successfully ... ;)
<strfry>
just so you know, i already abandoned the idea to use UP5K for this project, so it's not directly relevant to me
<ZipCPU>
Ok. <sniff> Sorry we couldn't help you faster then.
ar3itrary has joined #yosys
digshadow has quit [Ping timeout: 240 seconds]
FabM has quit [Quit: ChatZilla 0.9.93 [Firefox 52.5.0/20171114221957]]
dys has quit [Ping timeout: 240 seconds]
<daveshah>
For the record, I've solved it anyway.
<daveshah>
The problem is, DSP cascading with ACCUMCI and SIGNEXTIN isn't supported in arachne-pnr yet
<daveshah>
The example doesn't use DSP cascading, but still ties ACCUMCI and SIGNEXTIN to FALSE (0)
<daveshah>
This causes arachne-pnr to fail, as it doesn't find these signals in the chipdb - because they're not really signals but tell the placer how to place the DSPs
<daveshah>
In the short term, I will fix this in arachne-pnr by ignoring these signals when hard tied to 0, in the long term hopefully I will add DSP cascading support
<daveshah>
Until even the short term fix, it can be fixed in the Verilog by commenting out the ACCUMCI and SIGNEXTIN connections
digshadow has joined #yosys
dys has joined #yosys
nrossi has quit [Quit: Connection closed for inactivity]
<daveshah>
PR for the simple fix:https://github.com/cseed/arachne-pnr/pull/90
m_w has joined #yosys
ar3itrary has quit [Remote host closed the connection]
ar3itrary has joined #yosys
daveshah has quit [Quit: daveshah]
<strfry>
ZipCPU: got nothing to do with the arachne issue, i'm sure that could have been resolved
<ZipCPU>
Do you have any remaining issues?
<strfry>
oh, it already has been :)
dys has quit [Ping timeout: 240 seconds]
dys has joined #yosys
<shapr>
ZipCPU: do you happen to know anything about high voltage serial to parallel chips? I need 400v output
<shapr>
and I think I needto drive 128 pins at 400v
<ZipCPU>
Ahm, yeah, I know something about them: stay away from high voltage!!!!! That's about all I know, though. ;)
<shapr>
haha!
<ZipCPU>
At one time I looked into controlling a 120V power line. Most of the solutions I found involved relays.
<ZipCPU>
I never tried any of those solutions, though.
<shapr>
I bought a nixie 64x64 pixel display from the ukraine
<qu1j0t3>
shapr: got a datasheet?
<shapr>
qu1j0t3: only in russian
<qu1j0t3>
that's okay, i know an electronics genius who is russian
<shapr>
oh really?
<qu1j0t3>
definitely. I used to work for him. I can enlist his help...
<qu1j0t3>
i'm actually curious what this nixie looks like!
<shapr>
but this discussion is entirely off-topic on this channel, so ...
<sorear>
How much current do you need to source or sink at 0.4kV?
<shapr>
sorear: 25 uA for each pixel
<shapr>
I got the red and green panel, so I figure that's about 80 watts to fire up 64x64 times two
<awygle>
shapr: do your control at low voltage, switch an NPN? (idk much about nixies)
<shapr>
yeah, that's the plan
<shapr>
SPI display protocol to a microcontroller that drives a pile of transistors
<shapr>
I found a chip that does 300v serial to parallel, but I need 400
<awygle>
i guess what i meant is why do you need high voltage and serial to parallel simultaneously? low-voltage serial-to-parallel, driving high-voltage transistors
<shapr>
that would be perfect, I agree.
<shapr>
I hope I can find something like that
<shapr>
I'm now looking for high voltage PWM
<shapr>
someone suggested a transistor array, but .. do they come in high voltage and blocks of 128?
<awygle>
you should be able to do better than 128 lines, right? isn't there a clever thing people do for these kinds of displays?
<shapr>
awygle: oh hey, low voltate serial to high voltage parallel is a thing! and I found a youtube video where someone drives the single color version of that panel at 200v