Notices by Graham Sutherland / Polynomial (gsuberland@chaos.social), page 2

@foone this is where KSP has spoiled me, by making me pay attention to delta-V numbers. "I could alter my velocity relative to this object by 800m/s, or I could expend 20m/s of dV to alter my approach angle by 2 degrees for the same result".

@mntmn protection always takes the longest to add lol

@mntmn btw, do you have any existing documentation published about your power domains and filtering? I always love digging through that stuff. (no worries if not, it's just academic curiosity, don't do any extra work on my behalf!)

the #emfcamp swap shop has new policies

if you find payment bypass bugs in for-profit rent-a-basic-utility services, the ethical response is to help low income people get free access

two students about to become incredibly unpopular for being narcs

https://techcrunch.com/2024/05/17/csc-serviceworks-free-laundry-million-machines/

@foone iirc one of the tricks was to add some sort of clear oil (like baby oil) to stick it to the PCB, which made the white paper go kinda clear.

the actual home etching process is super messy though. ferric chloride stains eeeeeverything, it's bad for your skin, and you have to neutralise it before disposal which also creates a lot of mess. I used to own a kit and used it only a couple of times because the cleanup is horrible.

@mntmn I really need to sit and dig through your PCB designs at some point. I bet there are a ton of interesting little details in there.

@mntmn out of interest, what was up with the HDMI on 4k60 and what was the fix?

so as for *why* the PuTTY P-521 bug happened: they wrote the implementation in September 2001, which is a month before Windows XP was released. Win9x had no good random number generator APIs, so they came up with an alternative trick using SHA512 to generate deterministic but non-predictable nonces. but, of course, SHA512 outputs are 512 bits long, not 521 bits, and they just left the other 9 bits at zero, which resulted in this problem. the code was not reviewed since, so it never got fixed.

without digging into the full details of how the determinism stuff interplays with the protocol stack, their solution actually might have worked out ok if they had generated a second hash to fill the last 9 bits.

if they had happened to write this implementation just a month or two later, they could've written it to use CryptGenRandom on XP or later, and fallen back to the deterministic approach on Win9x, and this bug would've been avoided.

@mntmn that's one way to save having a power button :)

@foone mine contains a bunch of leaked laptop motherboard schematics so that could get pretty awkward pretty fast

@foone someone at revision mentioned getting one and said that the SDK was not a lot of fun to use at the moment, so they didn't bother making a prod for it. (no idea on the specifics)

@mntmn the alignment on that one to the north isn't great either, but that's a surprising level of error for a single misalignment.

if you check other boards, are they correct, or are they also offset and just happen not to be shorting?

@foone might be calibration storage for the touch screen controller if that's nearby

@foone honestly the board is a really common culprit for this kind of random crash, and it sounds like you've already eliminated everything else. I know it's annoying to RMA but right now it's by far the most likely culprit.

@foone (honestly at this point I'd just RMA the board given that you've tried two lots of RAM and CPU failures are super rare)

@foone I really don't think it's thermals. Sounds like a bad motherboard or PSU. Does memtest return clean?

there's a saying in electronics that there are two types of PCB designers: people who make antennas on purpose and people who make antennas by accident.

and having looked at a bunch of designs made by big companies recently, I think more PCB designers should go be antenna designers instead, because they've clearly got some innate skill at it which is currently untapped.