2026-06-17.log

- sigrid (QUIT: Quit: Lost terminal) (~sigrid@ftrv.se)00:39
+ sigrid (~sigrid@ftrv.se)00:40
+ sigrid_ (~sigrid@ftrv.se)01:10
- sigrid_ (QUIT: Client Quit) (~sigrid@ftrv.se)01:10
minutejosch: hmm need to dig deeper. how about putting some printk()s into reform2-lpc.c and loading it manually?01:49
- mjw (QUIT: Ping timeout: 245 seconds) (~mjw@gnu.wildebeest.org)01:51
minutei really need to finish the new SC firmware. what's missing is support for mb1.0/selecting the mb revision.01:51
- whereiseveryone (QUIT: Read error: Connection reset by peer) (206ba86c98@2a03:6000:1812:100::2e4)02:10
+ whereiseveryone (206ba86c98@2a03:6000:1812:100::2e4)02:10
- jnerula (QUIT: Remote host closed the connection) (~jnerula@li1009-93.members.linode.com)02:13
+ jnerula (~jnerula@li1009-93.members.linode.com)02:13
- vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50)02:19
- talos (QUIT: Quit: The Lounge - https://thelounge.chat) (~talos@2600:6c5d:f0:1fd0:3e6b:b6d3:ab62:ca79)03:04
+ talos (~talos@35.151.128.31)03:30
- paperManu_ (QUIT: Ping timeout: 256 seconds) (~paperManu@204.244.197.237)03:32
- paperManu (QUIT: Ping timeout: 256 seconds) (~paperManu@204.244.197.237)03:44
+ spew (~spew@user/spew)04:23
+ kasim (~user@2601:644:8481:3750:2b8:fa0e:b10a:fb6b)04:56
kasim@minute: CrowdSupply closed the ticket I created to ask me to contact project creator. What should I do? 04:59
- Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.3.59)05:42
+ Ar|stote|is (~linx@149.210.3.59)05:51
- mhoye (QUIT: Ping timeout: 244 seconds) (~mhoye@li319-32.members.linode.com)06:16
+ mhoye (~mhoye@li319-32.members.linode.com)06:17
joschkasim: i'm just a community member. If crowdsupply ignores you, i think you could try to a) follow the steps on support.mnt.re to repair your unit (or figure out what's broken) or b) send your unit to MNT under warrenty to get it repaired by them or c) create an account in the forum and try to sell it (just yesterday somebody said they wanted to buy a imx8m+ pocket reform from crowd supply)06:30
- kasim (QUIT: Ping timeout: 246 seconds) (~user@2601:644:8481:3750:2b8:fa0e:b10a:fb6b)06:40
- tomenzgg (QUIT: Ping timeout: 252 seconds) (~tomenzgg@user/tomenzgg)07:04
gordon1josch: i guess you've been inside reform more often than i, is there a space for 2mm thick keyboard pcb? or is it already 2mm?09:50
gordon1ACTION slowly considering custom keyboard09:50
+ mjw (~mjw@gnu.wildebeest.org)11:43
minutegordon1: it's 1.6mm. i think 2mm should be fine, risk is keycaps touching the display, but you could test that by putting something 0.4mm thick on the keyboard and closing it, or putting some spacers/washers below the current one11:51
gordon1right, sounds good11:52
gordon1will try11:52
+ paperManu (~paperManu@204.244.197.237)12:38
- Ar|stote|is (QUIT: Ping timeout: 276 seconds) (~linx@149.210.3.59)12:53
+ Ar|stote|is (~linx@149.210.3.59)12:56
joschgordon1: yes, even more than 0.4 mm will be more than fine.12:57
joschI have the spacer between my mainboard and keyboard and when the reform is in my backpack it gets squeezed12:57
joschthat squeezing means that sometimes the keyboard touches the display so much that the display "presses keys"12:58
joschthis sounds bad but i've been living with my display pressing keys for more than 3 years12:58
joschso 2 mm pcb should be more than okay :)12:58
joschgordon1: what do you want to mod?12:58
gordon1hmm, i just worry about it leaving permanent mark on the display more than if it registers keypresses12:58
gordon1josch: ortholinearity (also want to re-route crystal for rp2080, maybe that will make it more stable)12:59
joschgordon1: what leaves a permanent mark on the display is not the keyboard but the trackball13:02
gordon1yeah, i noticed that but that's not too bad13:05
gordon1that's one more thing that stops me from trying SiN ball permanently13:05
- murphnj (QUIT: Ping timeout: 248 seconds) (~murphnj@user/murphnj)13:19
+ murphnj (~murphnj@user/murphnj)13:20
- mjw (QUIT: Killed (calcium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org)13:29
* Guest142 -> mjw13:29
* tisiphone -> onepict13:33
+ paperManu_ (~paperManu@204.244.197.237)14:37
+ mark_ (~mjw@gnu.wildebeest.org)15:25
- spew (QUIT: Quit: nyaa~) (~spew@user/spew)15:36
+ trypokarydos (~linx@149.210.3.59)16:08
- Ar|stote|is (QUIT: Ping timeout: 276 seconds) (~linx@149.210.3.59)16:10
- paperManu_ (QUIT: Ping timeout: 246 seconds) (~paperManu@204.244.197.237)16:24
- paperManu (QUIT: Ping timeout: 265 seconds) (~paperManu@204.244.197.237)16:27
- mark_ (QUIT: Ping timeout: 246 seconds) (~mjw@gnu.wildebeest.org)16:32
+ chrcav (~chrcav@user/chrcav)16:37
- chrcav (QUIT: Quit: leaving) (~chrcav@user/chrcav)16:46
+ paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca)16:47
+ chrcav (~chrcav@user/chrcav)16:48
+ tomenzgg (~tomenzgg@24.148.84.215)16:54
- tomenzgg (QUIT: Changing host) (~tomenzgg@24.148.84.215)16:54
+ tomenzgg (~tomenzgg@user/tomenzgg)16:54
- b0 (QUIT: Quit: adiós) (~b0@user/b0)16:56
- johl (QUIT: Quit: ZNC 1.10.1 - https://znc.in) (~johl@wikidata/Jens-Ohlig)16:57
+ johl (~johl@wikidata/Jens-Ohlig)16:58
+ b0 (~b0@user/b0)16:59
lidstaheeew, quasar 6490 module in the mnt shop, congrats :)18:52
lidstahbeen away a week, got a bad flu, coming back, boom, quasar module :)18:52
lidstahnot yet available for the classic reform, I'm wondering if a 16GB version will be available (and hi all!)18:53
lidstahand the gpu looks like twice as fast as the rk358818:54
+ amospalla (~jordi@user/amospalla)19:07
joschlidstah: nobody knew that all it took is for you to get sick -- sorry for that, hope you are better now XD19:16
joschlidstah: nobody knows about versions with more memory for the future — given the "ai" hype these days i wouldn't hold my breath though. I mean you cannot even get rk3588 with 32 gb anymore these days...19:17
joschlidstah: i think the main advantage of the gpu is that it can do more opengl and that means you can run blender without issues and probably even a bunch more games than was possible on the rk358819:18
joschi'm pondering another run of factorio on the rk3588 but it would mean that my time for foss would go into the gutter XD19:18
joschit has been a relatively long while in MNT land without major issues -- the next bad bug must come any day now :D19:19
- qbit (QUIT: Remote host closed the connection) (~qbit@user/qbit)19:30
+ mark_ (~mjw@gnu.wildebeest.org)19:34
- MartiniMoe (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~quassel@user/MartiniMoe)19:37
+ MartiniMoe (~quassel@user/MartiniMoe)19:41
lidstahjosch: ahah nm, yeah, the actual situation with the ai hype is... concerning to say the least19:49
^alexslop! slop! slop!19:50
lidstahfor e.g, machines we bought at 25K€ here are now around 55K€, meaning we'll not replace some critical parts of our infrastructure this year, and probably not next year either, even if they're around 8 years old and will be EOL soon.19:50
lidstahmeaning no support, and so on19:50
lidstah^alex: yeah :/19:50
lidstahI've not even watched the price to replace our two 1PB netapp (prod) and 2PB (backup)19:53
lidstahprobably 3x times the price they were last year19:53
lidstahand on the homelab front, clearly not time for a ram or disk failure >.>19:56
+ wielaard (~mjw@gnu.wildebeest.org)19:56
+ spew (~spew@user/spew)19:57
- mark_ (QUIT: Ping timeout: 264 seconds) (~mjw@gnu.wildebeest.org)19:57
- mjw (QUIT: Killed (copper.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:4600:5952:3a9:6e0e:555a)20:53
* wielaard -> mjw20:53
+ Guest2946 (~mjw@2001:1c06:2486:4600:5952:3a9:6e0e:555a)20:53
- MartiniMoe (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~quassel@user/MartiniMoe)20:55
- spew (QUIT: Read error: Connection reset by peer) (~spew@user/spew)20:55
+ MartiniMoe (~quassel@user/MartiniMoe)20:56
+ qbit (~qbit@user/qbit)20:59
+ siviq (~siviq@user/siviq)21:05
+ spew (~spew@user/spew)21:14
siviqlidstah welcome back! :) One thing to do for the systems going EOL and avoid spending now fortune on new hardware is to check for extended warranty options. But extending the warranty usually comes with limited support and hw replacement options. There are news for China based companies producing memory chips21:20
siviqhttps://www.tomshardware.com/pc-components/ram/chinese-memory-vendors-snub-industry-giants-in-favor-of-homegrown-ram-chips-samsung-micron-and-sk-hynix-face-a-chinese-supply-chain-revolt It's matter of time the companies outside China to begin negotiations for chips/memory purchase.21:20
- siviq (QUIT: Quit: Client closed) (~siviq@user/siviq)21:22
+ siviq (~siviq@user/siviq)21:23
- siviq (QUIT: Client Quit) (~siviq@user/siviq)21:27
- spew (QUIT: Read error: Connection reset by peer) (~spew@user/spew)22:15
- paperManu (QUIT: Quit: WeeChat 4.9.0) (~paperManu@modemcable141.205-200-24.mc.videotron.ca)23:08
joschso, patched mpv+ffmpeg can do hardware decoding on rk358823:21
joschbut i cannot get it to work with gstreamer -- does somebody have a recipe?23:21
joschhttps://mister-muffin.de/p/9VsA.txt23:24
+ siviq (~siviq@user/siviq)23:27
- siviq (QUIT: Client Quit) (~siviq@user/siviq)23:27
+ paperManu (~paperManu@204.244.197.237)23:29
+ paperManu_ (~paperManu@204.244.197.237)23:30
sigridjosch: is your gstreamer patched in any way?23:38
joschnot at all23:38
joschshould it?23:38
sigridI don't think so. mine works fine here23:39
sigridon a random video I have laying around23:39
sigridgstreamer 1.28.323:39
joschsigrid: what version are you running? 1.28.4 here23:40
joschah23:40
joschi tried with big buck bunny23:40
sigridlet me try the same video23:41
joschsigrid: do you use gst-play or a different frontend?23:41
sigridI've no idea. just running the same command line as yours23:42
joschokay23:42
joschthanks a lot for trying -- at least that means i didn't make any silly mistakes in my pipeline for example23:42
sigridI have a pretty bad internet here, so downloading bbb will take ~15 minutes :')23:43
joschno worries :)23:44
joschi downloaded the 4k@60 version to really make sure that hardware decoding did its thing XD23:44
sigridthat's a good way that I didn't think about. small videos and htop is not a good indicator of whether it's doing hw decoding or not23:45
sigridalso slightly on the topic, I've tried libva patched for 6.18 kernel. unfortunately it did not work for any of the videos23:46
sigridno matter what, mpv/ffmpeg insist on creating a NV12 surface which fails23:46

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!