#avr Logs

Sep 08 2020

03:40 AM s-ol_ is now known as s-ol
12:51 PM TechChristoph: was on #avrs
12:51 PM TechChristoph: but now back on #avr
02:08 PM vmt: consider leaving #avrs
02:08 PM cehteh_: :D
02:09 PM cehteh_: well never been there, does that channel fulfill it promise to be better than here?
02:12 PM vmt: roughly on par with ##electronics. even when you take into account that this channel is mostly idling, #avr is still about 2.5x better
02:12 PM cehteh_: lol
02:13 PM cehteh_: well i only disliked the part that it/he/she/they tried to divert a small comminuty even more, otherwise i give a shit, would be best to put efforts into a single channel
03:25 PM cehteh_ is now known as cehteh
03:44 PM mindw0rk: hi, I haven't been really lurking here for a while. I was looking at one of those newer attiny402 for one of my designs.
03:44 PM mindw0rk: and I noticed avrdude hasn't seen a release in 4 years :/
03:45 PM mindw0rk: did everyone drop avrs and move on to something else?
03:48 PM nohit: yes
03:54 PM mindw0rk: alright, cool, I guess I will move on to stm8
03:55 PM nohit: i heard PICs are good
03:57 PM vmt: do you have some 8-bit fetish or why stm8 exactly?
03:58 PM mindw0rk: very low power consumption, I am not actually aware if there's anything as cheap and as power efficient as them
03:59 PM mindw0rk: also, I need a a tiny package, since I only need a spi interface and an extra i/o pin for my purpose
04:00 PM vmt: if you're talking about pin count, sure. but pretty much anything nowadays is a *tiny* package
04:00 PM mindw0rk: using an arm for this... feels wrong.
04:01 PM mindw0rk: well I want something I can still hand solder
04:01 PM vmt: you can hand solder qfps
04:02 PM vmt: also, if you don't want to go cortex-m i'd probably stick with avr. or pic
04:02 PM mindw0rk: I can, but are there any QFPs smaller than an SO8 ?
04:06 PM vmt: probably not. what's the space constraint though? bioimplants??
04:07 PM vmt: i'd probably stick with avrs though, the new ones come in small sizes and gcc can target them (i think?)
04:08 PM mindw0rk: you have setup a bunch of things for the toolchain
04:08 PM vmt: though i guess if you like sdcc you can go with stm8/pic, but personally i don't have much experience with it
04:11 PM mindw0rk: I haven't ever used sdcc, also the programming tools are not sure stable, but I'm willing to try both the attiny402 and the STM8L001J3 at this point
04:12 PM mindw0rk: actually I wasn't expecting to get real opinions, but thanks :D
04:18 PM vmt: there aren't real opinions, mere fakes
04:18 PM vmt: these*
04:19 PM mindw0rk: well then
04:19 PM mindw0rk: so is my thanks, I guess :)
04:19 PM vmt: hey!
04:19 PM mindw0rk: ¯\_(ツ)_/¯
04:20 PM vmt: at any rate, i'd suggest the new line of avrs. also, for a more real opinion i think i remember seeing LeoNerd saying good things about them and (if memory serves) has experience with them
04:22 PM vmt: he has been awfully awol lately though, but a shameless ping nevertheless
04:24 PM LeoNerd: "them"?
04:24 PM LeoNerd: Oh.. the new AVRs.. yah.. the tiny 1-series are nice :)
04:24 PM LeoNerd: I'm today working on an older chip and already missing quite a few features of them
04:24 PM LeoNerd: Currently: the lack of address-mapped eeprom
04:27 PM mindw0rk: LeoNerd, are you working on win? I am wondering how difficult it is to setup the toolchain for mac or linux.
04:28 PM LeoNerd: On Linux, it's dead-simple. At least on debian. apt-get install avr-gcc avr-libc
04:28 PM LeoNerd: and we're done :)
04:28 PM LeoNerd: http://leonerds-code.blogspot.com/2019/06/building-for-new-attiny-1-series-chips.html also might be useful
04:29 PM vmt: mindw0rk: if you want something out-of-the-box, atmel studio setup is also, dead simple
04:31 PM vmt: or then just download avr-gcc and avr-libc for windows
04:31 PM vmt: atmel studio uses them anyway
04:31 PM vmt: and avrdude or whatever you use nowadays
04:32 PM LeoNerd: avrdude still doesn't know how to talk UPDI though, so I use my own tool
04:32 PM LeoNerd: I don't know who is developing avrdude or how or what
04:32 PM LeoNerd: I will practically pay someone to teach me these things so I can add UPDI support for it
04:32 PM vmt: i think there's like two million python scripts written for updi upload right:D
04:33 PM LeoNerd: I've no idea. I only know of one
04:33 PM vmt: what about openocd?
04:34 PM LeoNerd: never heard of it
04:35 PM vmt: apparently it doesn't do updi. according to the website last updates are like from '18 or something
04:35 PM LeoNerd: Hmm
04:35 PM vmt: https://github.com/mraardvark/pyupdi but something like this. avr toolchain is dead simple to setup on any platform
04:35 PM LeoNerd: It's really not a difficult problem...
04:35 PM LeoNerd: I could write a C program for it in a couple of hours
04:36 PM LeoNerd: I just don't have any idea how to make it work with avrdude itself
04:39 PM mindw0rk: LeoNerd, thanks!
04:40 PM LeoNerd: Yah; mraardvark's is the one I'm thinking of. I used that as inspiration for mine
04:40 PM LeoNerd: At some point I'll get around to writing a sigrok decoder for UPDI, and then the fun will properly start
04:41 PM LeoNerd: because then I hope to reverse-engineer the OCD parts and document them
04:52 PM vmt: by the way, offtopic but LeoNerd is the "about"-section in your blog up-to-date? i.e. do people still get contract work with perl?:D
04:54 PM LeoNerd: That fills most of my day, yes
04:57 PM vmt: may i ask, what's your latest?
04:59 PM LeoNerd: I do loads of things...
04:59 PM LeoNerd: https://metacpan.org/author/PEVANS -- is a whole bunch of stuff
04:59 PM LeoNerd: I've added async/await syntax. I'm currently working on libuv bindings
04:59 PM LeoNerd: I do bits of teaching and style guidance/code review
05:00 PM vmt: libuv is that event loop lib?
05:02 PM LeoNerd: Yah.. it seems popular lately. all the cool kids are using it ;)
05:02 PM LeoNerd: Bit big for an AVR chip though
05:03 PM vmt: i figure it's mostly used for network (??) so probably... not avrs:D
05:04 PM LeoNerd: neovim uses it for the base
05:06 PM vmt: erhh, for what? disk io or thread sync or what?
05:07 PM vmt: or even input?
05:07 PM LeoNerd: event base
05:07 PM LeoNerd: For user input and allsorts
05:08 PM LeoNerd: I don't know if they have disk IO via it as well, but that wouldn't surprise me
05:10 PM vmt: interesting. i have been toying around with a text editor idea for some time. efficient data structures for the buffers, but it's mostly on the backburner as it doesn't really pay well and takes a considerable amount of time
07:00 PM kline is now known as kline\0
11:29 PM day_ is now known as day