05:35 <marcan> florolf: yeah there was some faffing around with the pinout on the sync connector, I'm not surprised I got it wrong

05:35 <marcan> pretty sure I *tried* to get it right but... :-)

05:35 <marcan> I guess we should swap the pins

05:37 <marcan> that does involve a production change but... it's so stupidly trivial I don't even think it's worth calling that revC2

05:38 <marcan> whitequark: thoughts?

05:38 <marcan> we'd be breaking the a priori "it's fine to push to the branches if it doesn't change the PCB" rule but... the change is trivial (pad shape and silkscreen pin1 indicator)

05:39 <whitequark> the schematic doesn't change, right

05:39 <whitequark> then i *guess* it's fine

05:39 <marcan> it does, we swap the pins

05:39 <marcan> we swap the pins everywhere basically, so that the physical layout does not change

05:39 <whitequark> well then it has to be revC2 i think

05:39 <whitequark> it'd be pretty confusing otherwise

05:39 <marcan> would it?

05:39 <whitequark> if someone is wiring up the connector

05:39 <whitequark> i mean, i'd probably look at the connector on the schematic when populating the mating part

05:39 <marcan> the silkscreen on the bottom is correct, and that's way easier to look at than the schematic

05:40 <marcan> and that will not change with this

05:40 <marcan> I mean the way it is right now, if you look at the schematic and try to populate the connector following the pin1 designator in the housing, you'd get it wrong

05:40 <marcan> in revC1

05:40 <marcan> we'd be fixing that

05:41 <whitequark> yes, and that warrants a schematic bump, i think

05:41 <whitequark> precisely because you'd get it wrong

05:41 <marcan> but if we call that revC2, then someone with a revC1 will try to follow the revC1 schematic

05:41 <marcan> and get it wrong anyway

05:41 <whitequark> :/

05:41 <whitequark> ok

05:41 <whitequark> point

05:41 <whitequark> i guess we're fucked either way so i don't object

05:41 <marcan> yeah it's just stupid

05:42 <marcan> but really what we're saying here is "oops the revC1 schematic and board were wrong all along, we're fixing them and just ignore the silkscreen pin1 (can't get it wrong as long as you have the right connector anyway) and you'll be fine"

05:42 <marcan> and then nobody will get confused

05:43 <marcan> like revC1 is mechanically and electrically okay, the schematic and silkscreen pin1 indicators are just incorrect descriptions of it

05:43 <marcan> so we can just fix those and call it a day

05:43 <whitequark> fine

05:43 <marcan> (was revB right? I guess I should check that...)

05:44 <marcan> well that one had the wrong part anyway for that connector

05:44 <marcan> I'm pretty sure I kept the physical part the same as revB, but I'll double check

05:44 <marcan> if the pin#s are wrong for revB then that's worth an equivalent fix

07:47 <florolf> marcan: afaict, revB has the same bom line for J4 as revC1. revA uses another connector (640455-2 from TE) though, which *does* number the pins differently

07:48 <florolf> all follow the same idea (angled, latch on top) though. there seems to have been an issue with a connector that has the latch on the bottom side? i didn't come across that yet.

08:00 <whitequark> i might be misremembering

16:22 <kc8apf> whitequark: I can give you my Facedancer patches if you want them before Kate has a chance to merge them

16:25 <whitequark> kc8apf: it's probably fine, the project I need this for is very slow anyway