| - mjw (QUIT: Quit: Leaving) (~mark@gnu.wildebeest.org) | 00:56 | |
| - Christoph_ (QUIT: Remote host closed the connection) (~Christoph@p4fe73ddc.dip0.t-ipconnect.de) | 01:01 | |
| - mtm (QUIT: Ping timeout: 240 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 02:02 | |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 02:31 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 02:33 | |
| NanoCodeBug | is the pi4 carrier board complete? was thinking about using one with a soquartz som module | 03:44 |
|---|---|---|
| NanoCodeBug | as in, is the design complete, I understand I would need to fabricate it myself | 03:44 |
| bkeys | I am not an authority but we were talking about this earlier. I believe a lot of the work was done but everything wasn't tested/validated with it | 03:49 |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 04:08 | |
| NanoCodeBug | I saw the work cooljqln was doing on twitter with her own custom build and would love to try similar. | 04:28 |
| NanoCodeBug | though without building a whole reform, i have one already :P | 04:29 |
| NanoCodeBug | OSHPark is near me too so getting a carrier board fabricated shouldn't be too bad, just gotta hope the chips are available. | 04:29 |
| NanoCodeBug | the carrier chips, I know a raspi som is rarer than a nvidia gpu | 04:32 |
| NanoCodeBug | probably will go with an soquartz module | 04:33 |
| NanoCodeBug | also! was messing around with the backlight controller. looks like adding a interpolution property works, to get a wider range of backlight levels. | 04:34 |
| - bkeys (QUIT: Ping timeout: 255 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 04:44 | |
| vkoskiv | NanoCodeBug: Actually nvidia gpus are starting to be available again. Crypto crash. | 07:59 |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 08:30 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 08:31 | |
| + MajorBiscuit (~MajorBisc@wlan-145-94-167-213.wlan.tudelft.nl) | 09:48 | |
| minute | NanoCodeBug: fully validating rcm4 is one of my next projects! currently finishing Pocket Reform D-1 | 10:38 |
| - GNUmoon (QUIT: Ping timeout: 268 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 10:42 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 10:52 | |
| * wielaard -> mjw | 11:12 | |
| + Christoph_ (~Christoph@p54bf6094.dip0.t-ipconnect.de) | 11:30 | |
| josch | I just managed to kill my reform by attempting to compile clapper from source -- what do you guys do to prevent excessive resource usage that would result in a lock-up of the system? | 12:40 |
| ex-parrot | hmm I've done a lot of compiling and didn't run in to problems | 12:49 |
| ex-parrot | do you know what resource ran out? or just a random lockup? | 12:49 |
| josch | interesting -- from how sluggish the system became (couldn't even move the cursor at some point) this totally looked like an OOM issue but i didn't have top open to be sure | 12:54 |
| josch | now i tried again and it worked fine | 12:54 |
| josch | then i guess this is the first problem i've run into -- i just got a random lockup :( | 12:54 |
| sigrid | i compiled clapper with no issues | 12:57 |
| sigrid | maybe having 16Gb swap helped | 12:57 |
| josch | i compiled clapper with htop open and the compilation didn't seem to consume any large amount of RAM so this is probably not related to clapper building at all but was triggered by something else | 12:59 |
| josch | and of course i cannot reproduce it which makes it even more frustrating | 12:59 |
| mjw | yeah | 12:59 |
| mjw | I do builds on my reform, but ssh into a larger server for really big compiles. Local builds never killed my reform though, they were just slow(er). | 13:00 |
| vkoskiv | I've never used distcc, but I'd be curious to try that out with reform | 13:04 |
| flowy | josch: any graphical glitches before lockup? | 13:30 |
| + chomwitt (~chomwitt@2a02:587:dc1c:800:5b3a:daea:252e:ac17) | 13:51 | |
| - mtm (QUIT: Ping timeout: 268 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 14:05 | |
| Boostisbetter | I totally recommend a swap that size of your ram is you are using an nvme drive. It really will help with stability and will make it easier to not worry about what you are running. | 14:05 |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 14:09 | |
| josch | flowy: i have some graphical glitches every once in a while in the kde web browser konqueror but not before it locked up for me earlier today -- the screen just seemed to be frozen at some point (i'm on holidays so i couldn't check whether it was still alive via ssh or via serial) | 14:12 |
| josch | flowy: i'm currently building my own clapper because i don't like to solve problems just by adding more layers (aka flatpak) do you have a video for me that you can confirm works smoothly with clapper but fails to run smoothly with mpv? | 14:16 |
| flowy | yeah i've had exactly one freeze, it was preceeded by weird GPU stuff. i also didn't check remotely but i will if it happens again | 14:16 |
| josch | (i suspect the magic in tha clapper flatpak comes from their custom built gstreamer version) | 14:16 |
| flowy | haha my issues with clapper and mpv are so interesting | 14:16 |
| flowy | i've been using reform as my primary evening tv, so i've watched quite a lot. and i can say the following with certainty: | 14:17 |
| flowy | they both have issues playing back some files, but their problem sets are 100% mutally exclusive. so if i can't play something with mpv, it will always play with clapper. and vise versa. i have a strong feeling they are both encountering different sides of the same bug | 14:18 |
| flowy | so i need both players to have a reliable experience | 14:18 |
| flowy | but sure, i'll email you some links | 14:19 |
| josch | flowy: josch@debian.org -- thank you! :) | 14:19 |
| flowy | in fact i don't think i've found anything that can be played smoothly with both, or fail with both | 14:23 |
| flowy | over the weekend, i swapped my reform's original keyboard pcb out with the reform standalone keyboard (w/ kailh white switches for clicky goodness). interestingly, i immediately started encountering an issue that many in the community forums have been having- two days in a row the reform is unresponsive after being shutdown (properly) overnight. | 15:13 |
| flowy | the solution is to hit the keyboard's reset button. i can't easily reproduce the problem with quick shutdowns. i find it interesting because i never encountered this problem, having shutdown the reform hundreds of times before. | 15:14 |
| flowy | i wonder if the issue is with a change in components or the keyboard firmware | 15:15 |
| flowy | on the original keyboard i was running one of the first firmwares minute posted that does low power sleep, and the newest firmware (re-flashed) on the new board | 15:17 |
| flowy | and i'm not, like, not waiting when i hold down the circle button or whatever. it does not wake up, i could press the button a hundred times or hold it down for 30 seconds. | 15:17 |
| flowy | anyways, just saying in case this is familiar to anyone. | 15:18 |
| flowy | it's mysterious | 15:19 |
| josch | is anybody more familiar with gstreamer than me and has a hunch which of these options is the magic option that enables hardware accelerated playback on the reform: https://github.com/Rafostar/clapper/blob/master/pkgs/flatpak/testing/gstreamer.json | 15:26 |
| flowy | definitely -Dgst-plugins-bad:v4l2codecs=enabled | 15:27 |
| flowy | not sure about va or vaapi or gl stuff. | 15:29 |
| flowy | but v4l2codecs is essential, for sure | 15:29 |
| flowy | that's the only route the hantro can be accessed in the kernel i think, currently | 15:30 |
| josch | thanks! i'll try that one first | 15:31 |
| - bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 15:35 | |
| flowy | i see some potentially relevant notes here, at the bottom: | 15:41 |
| flowy | https://github.com/Rafostar/clapper/wiki/Hardware-acceleration | 15:41 |
| flowy | i think this means we'd probably want -Dgst-plugins-base:gl_api=opengl,gles2 too | 15:42 |
| josch | flowy: bingo! no need to recompile any gstreamer stuff, just setting GST_CLAPPER_USE_PLAYBIN3=1 when running clapper does the thing | 15:45 |
| josch | flowy: the video you sent me doesn't work with mpv and only works with clapper with that environment variable set | 15:45 |
| josch | minute: I already packaged clapper at https://salsa.debian.org/debian/clapper but we cannot built it as part of reform-debian-packages because clapper user gobject-introspection and that concept doesn't work when cross-compiling | 15:47 |
| josch | but i can see to get this into Debian asap and then it can be installed normally | 15:47 |
| josch | minute: we could add to GST_CLAPPER_USE_PLAYBIN3=1 to the other environment variables we set for each user to make clapper work hardware accelerated out of the box | 15:47 |
| flowy | ah nice! i didn't understand the playbin ref. as i don't know what that is. | 16:05 |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 16:07 | |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 16:10 | |
| - Nulo (QUIT: Read error: Connection reset by peer) (~Nulo@user/nulo) | 16:54 | |
| + Nulo (~Nulo@user/nulo) | 16:56 | |
| minute | flowy: there is a patch that might help with your keyboard-wakeup-problem | 17:57 |
| minute | flowy: ah, you've already found the community thread | 17:58 |
| flowy | yeah, i might try it eventually. i don't want to change too many things at once. even if that patch fixes it, it's sort of unsatisfying that i didn't encounter this previously | 18:11 |
| flowy | or it's sort of unexplained, is what i mean | 18:13 |
| minute | yeah, i agree | 18:24 |
| minute | https://liliputing.com/2022/06/mnt-pocket-reform-is-7-inch-mini-laptop-with-a-modular-open-hardware-design.html | 19:00 |
| flowy | if you offer the pocket in that purple i may be forced to come up with a reason to get one | 19:07 |
| minute | hehe | 19:07 |
| flowy | or rather... lilac? | 19:07 |
| flowy | i was thinking it could be my kitchen computer that i use to select music and podcasts. might be able to replace my rpi4/stereo (sh)airplay target at the same time | 19:10 |
| flowy | i'm curious what the prce will be. maybe this chan can do a guessing game | 19:11 |
| flowy | (sorry lol) | 19:11 |
| kfx | still holding out for high-gloss chrome. | 19:18 |
| minute | kfx: heh! | 19:18 |
| minute | now finally testing the LPC battery status SPI driver things by NanoCodeBug https://source.mnt.re/reform/reform/-/merge_requests/28 | 19:21 |
| flowy | i have an LE precursor(betrusted), they ended up using high gloss steel. it is wonderfully hefty | 19:23 |
| minute | nice | 19:44 |
| minute | ok, reform2_lpc driver loaded | 19:44 |
| Boostisbetter | minute: those pocket reform renders have me biting at the bit. Really can't wait. | 20:08 |
| NanoCodeBug | yay! let me know if there are any issues with the battery driver. | 20:19 |
| + mark_ (~mark@gnu.wildebeest.org) | 20:21 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mark@gnu.wildebeest.org))) (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 20:22 | |
| * mark_ -> mjw | 20:22 | |
| + wielaard (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 20:22 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 20:22 | |
| minute | NanoCodeBug: i've merged the DTS changes to reform-debian-packages, and the driver probes now and shows some stats (trying with gnome-power-statistics) | 20:40 |
| minute | i need to let it run out to learn battery capacity again though | 20:40 |
| NanoCodeBug | in testing a full charge does it, though for accuracy a discharge too | 20:45 |
| - MajorBiscuit (QUIT: Ping timeout: 272 seconds) (~MajorBisc@wlan-145-94-167-213.wlan.tudelft.nl) | 20:50 | |
| minute | NanoCodeBug: testing with KDE + gnome-power-statistics https://mastodon.social/web/@mntmn/108550944307510706 | 20:53 |
| - Christoph_ (QUIT: Remote host closed the connection) (~Christoph@p54bf6094.dip0.t-ipconnect.de) | 21:56 | |
| ex-parrot | oh awesome | 22:41 |
| ex-parrot | does this mean the mystery of the "no SPI packets" is resolved?? | 22:41 |
| jfred | Ooh, USB-C on the Pocket Reform… any chance it’ll support DP Alt-Mode? | 23:02 |
| minute | ex-parrot: spi works now, yep | 23:11 |
| ex-parrot | amazing | 23:12 |
| minute | jfred: not planned | 23:12 |
| ex-parrot | what was the issue in the end? | 23:12 |
| ex-parrot | (if it's easy to succinctly explain) | 23:13 |
| minute | not sure, probably missed something in the dts originally, and NanoCodeBug's solution works | 23:13 |
| ex-parrot | great | 23:13 |
| - tadeus_brick (QUIT: Quit: Stay Cheeki Breeki) (tadeustad@psifactor.pl) | 23:22 | |
| - bkeys (QUIT: Ping timeout: 272 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 23:26 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 23:28 | |
| + klardotsh (~klardotsh@172.56.104.172) | 23:44 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!