02:32 <JonathanDickinso> can someone review+merge this? There are more improvements that I would like to make: https://github.com/embassy-rs/embassy/pull/2169

11:18 <JamesMunns[m]> I'm thinking about poking around no_std CoAP. I'm aware of the coap crate (on std), and coap_lite (on alloc), does anyone here have any notes or opinions on CoAP?

12:50 <cr1901> what is CoAP?

12:56 <JamesMunns[m]> "What if HTTP, but for constrained devices and transports"

12:57 <JamesMunns[m]> https://en.wikipedia.org/wiki/Constrained_Application_Protocol, also https://datatracker.ietf.org/doc/html/rfc7252

12:58 <JamesMunns[m]> probably the second most popular "IoT protocol" after MQTT, also the basis of LwM2M used by mobile/cell operators for certain devices

13:00 <Jan-Henrik[m]> <JamesMunns[m]> "I'm thinking about poking around..." <- There is also this library on which I've worked a couple of years ago, although I don't know how the current state of it is: https://git.openlogisticsfoundation.org/silicon-economy/libraries/serum/coap-zero

13:01 <JamesMunns[m]> > Fraunhofer IML Embedded Rust Group

13:01 <JamesMunns[m]> Oh?

13:01 <JamesMunns[m]> this is very relevant, thanks Jan-Henrik!

13:02 <Jan-Henrik[m]> Maybe Tobbe: has some updated information about this?

14:38 <waveguide[m]> Has anyone tried using pyomo for clock optimization?

14:38 <waveguide[m]> I tried scipy optimize (non integer results meh) and gekko with apopt which did ok

14:38 <waveguide[m]> Like I modeled the clock tree for audio on imxrt, kind of neat

17:39 <Farooq> I just watched a small soldering tutorial. It says I should be using wide tips rather than pointy ones.

17:39 <Farooq> Is it true?

17:40 <diondokter[m]> Farooq: For small components that usually easier yeah. For bigger stuff you don't need a chisel

17:44 <Farooq> Thank you

18:13 <fu5ha[m]> <thejpster[m]> "20 MHz is pretty fast. What’s..." <- Hm, it is a 2040. Would that just be relevant for the cs pin I guess? I'm successfully driving a display over the same spi at 62.5Mhz. Also, what's high speed mode? A cursory search didn't turn anything

18:14 <fu5ha[m]> <thejpster[m]> "Also do you have CRCs enabled..." <- CRC is enabled, I think it supports it? I'll try to confirm. And I haven't tried on Linux yet but I formatted it to fat32 mbr and wrote the data u have on it from macos. Can try on Linux too

18:23 <thejpster[m]> The 2040 pins have adjustable drive strength which also affects how sharp the edges are on your transitions and what speed you can reach. But if you have other devices on the bus that work that’s probably not the issue. You could try physically removing the other device though in case it’s causing an issue?

18:24 <thejpster[m]> It could also be anbe incompatibility with the card (maybe I didn’t follow the spec correctly) so you could try another card.

19:33 <fu5ha[m]> Aha, interesting. Will see. Yeah I have a couple others to try. Will do some debugging over the next couple days and see. Unfortunately I can't physically remove the other device as it's soldered on to the PCB. Really I should have used the other SPI peripheral and different pins for it, regret making the design decision to use only one bus between them, but gotta deal with it for now :P

21:41 <jannic[m]> There's the drive strength, but there is also the SLEWFAST bit that selects between a slow and a fast slew rate. Default is slow.