Replying to @azonenberg@ioc.exchange
Like they don't openly state they're compatible (annoying) but the pins all have the same descriptions and numbers and the footprint drawings look the same
@azonenberg@ioc.exchange
Security and open source at the hardware/software interface. Embedded sec @ IOActive. Lead dev of ngscopeclient/libscopehal. GHz probe designer. Open source networking hardware. "So others may live"
Toots searchable on tootfinder.
Replying to @azonenberg@ioc.exchange
Like they don't openly state they're compatible (annoying) but the pins all have the same descriptions and numbers and the footprint drawings look the same
Spent way too much time agonizing about whether to rip out the VCCINT regulator on the switch board to upgrade it for a 20A module with more headroom, or keep the 12A module I had planned.
Looking at the datasheet it seems the MYMGK1R812 and 820 are pin and footprint compatible lol. It should be a drop in replacement possibly with some passive changes.
Gonna make a quick oshpark test board to validate ripple and thermal performance just to be safe but the swap seems a no brainer at this point.
Probably tagging ngscopeclient v0.2 RC1 soon. We're down to like four open tickets plus I need to do an updating pass over the docs.
Got the ILA working with digital vector signals, long overdue for ngscopeclient.
There's a lot more work to do (most notably trigger comparators) but as long as you can provide an external trigger input in gateware this is usable for debug.
Threw together a quick protocol decoder for APB to test it out.
Replying to @Lunalucardrose20@corteximplant.com
@Lunalucardrose20 no no when they say "clove" they mean a bulb. This is a common mistake.
Replying to @Lizette603_23@mastodon.social
@Lizette603_23 yeah my lab is near-cleanroom conditions most of the time so when the AQI gets above zero that's a notable event lol. But if I'm working there for a long time I have to turn on the fan to get the CO2 levels under control and the intake filtration isn't as good as I'd like (or junk is creeping in through leaks in the walls etc unfiltered).
I really want a proper HEPA filtered ERV or something but whatever, maybe one day. Actually not really an ERV because I want to reject heat (the lab is normally hotter than the outside except at the peak of summer days but a forced in-and-out ventilator rather than just an exhaust fan and makeup duct that leaves the lab negatively pressured so gunk can get in.
We have filters in the rest of the house too, although I don't currently have monitoring/logging there. I have some sensor units a friend built but until she 3d prints the cases to go with them I can't safely deploy or easily wall mount them, they're just fragile bare boards at the moment.
And she just had short hair back then, she's 4 now but I've kept the pic in part out of laziness and in part because she's pretty unrecognizable there (all babies look about the same even without the sunglasses lol)
Replying to @0h00000000@ioc.exchange
@0h00000000 Also check clocks going into your SERDES and make sure they're proper frequencies/phases. I've had all manner of problems when my slow and fast domains weren't the right multiple or misaligned or something
Replying to @Lizette603_23@mastodon.social
@Lizette603_23 I don't have outdoor air quality metering right now (on the list).
Inside my lab, there was a measurable spike but not enormous - on par with what I get when I run the ventilation fan to bring the CO2 levels down, but I wasn't running the fan (meaning the outdoor levels were much higher and I was just getting what snuck in through diffusion.
Replying to @azonenberg@ioc.exchange
Still hardcoded external trigger, comparators will come much later after I've tweaked more of the flow and maybe improved performance.
First light on the ngscopeclient ILA integration including digital vector support up to 32 bits (up to 64 will come soon and is straightforward to add, bigger will be a bit more work since there's no native int type bigger than that)
Next step will be to make a protocol decode for APB to prove out the decode flow on digital buses
@petrillic And this is why ngscopeclient still has an IRC dev channel (although it is bridged to discord because a lot of folks are there)
@0h00000000 is it bad that i can tell that big QFN is a TI part just from the font of the laser engraving?
@ireneista yeah i used cuda in the very early stages (c. 2009) when it was the only game in town. No more
@ireneista like the cuda driver api gives you everything but opencl, opengl, or the higher level cuda apis hide a lot of the complexity
Replying to @azonenberg@ioc.exchange
@ireneista of course I'm the guy who used to use the low level CUDA driver API instead of the "compile your source code with nvcc and use our magic syntactic sugar call flow" or the higher level API
@ireneista As someone who's done a lot of low level CUDA etc stuff, I really like Vulkan as a compute/acceleration API. So many things you used to have to just trust the driver would do right, you can now actually control
Replying to @azonenberg@ioc.exchange
@whitequark But the vivado UI was just barely not painful enough to force me to do something about it, so it just sat there until I started playing with efinix stuff and *their* ILA was enough of a disaster that I couldn't tolerate it
Replying to @azonenberg@ioc.exchange
@whitequark and ILA integration, to enable exactly this, has been on the wishlist for literal years.
Like the github ticket for REing the Vivado ILA and writing a driver for it was filed in April 2019
@whitequark I really do wonder how the T&M industry stagnated so much to the point that everything I'm doing is super groundbreaking.
Like this seems like an obvious way to do things
@whitequark well mostly it would also enable deskewing *arbitrary* LAs not just ILAs.
Or ILAs to analog scopes, etc. Basically just extending the existing cross-correlation based deskew logic to support non-analog signals