| - S0rin (QUIT: Ping timeout: 248 seconds) (~S0rin@user/s0rin) | 00:59 | |
| - mtm (QUIT: Ping timeout: 248 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 01:03 | |
| + S0rin (~S0rin@user/s0rin) | 01:05 | |
| - bluerise (QUIT: Ping timeout: 268 seconds) (~bluerise@user/bluerise) | 01:46 | |
| + bluerise_ (~bluerise@pc19f8f98.dip0.t-ipconnect.de) | 01:46 | |
| - oomono (QUIT: Quit: Connection closed for inactivity) (uid328183@id-328183.tinside.irccloud.com) | 02:17 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 02:35 | |
| klardotsh | oops. remembered this week my reform was in a box the past couple months after moving and not having time to fiddle with it. just got it back out. battery cells anywhere between 0.0V and 2.2V. RIP to all those cells :\ (bright side: I'm picking up my protected board from bnys this weekend in Seattle, woo) | 02:46 |
|---|---|---|
| klardotsh | what's the current recommended source of replacement cells stateside? looks like both JGNE links on the forum are currently sold out | 02:54 |
| bkeys | I go to a place called batterydepot.com | 03:01 |
| bkeys | klardotsh: You gonna be at the meetup? | 03:01 |
| klardotsh | possibly. more likely than the last one, but still not absolutely certain | 03:02 |
| bkeys | The battery boards have saved my battery set | 03:02 |
| klardotsh | this madlad laptop thinks it's charging the cells, somehow. I have the "0.0" ones reading 0.8 according to Circle-B now. wild. LFP shouldn't be able to do that, but then, I don't entirely know when they hit "truly drained". they're perma-degraded, at least. but who knows, maybe it'll charge somewhat. | 03:04 |
| - Guest824 (QUIT: Ping timeout: 260 seconds) (~nicolas@234-48-142-46.pool.kielnet.net) | 03:10 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 03:10 | |
| + nsc (~nicolas@128-49-142-46.pool.kielnet.net) | 03:11 | |
| * nsc -> Guest6226 | 03:12 | |
| bkeys | Yeah it's frustrating the battery boards are well worth having, I live in Idaho currently so I just had them mail him over to me | 03:15 |
| bkeys | But I'll be taking Amtrak over to Seattle for the meetup | 03:15 |
| sevan | foolishness of not doing updates, regularly. 600+ packages to update... | 04:04 |
| sevan | I'm still running the original battery board, but should be receiving the new board at some point. | 04:06 |
| sevan | just remember the UK post office got hacked end of December and they were recommending not getting things shipped over. erm | 04:07 |
| technomancy | yeeeeah not a fan of living on the edge | 04:09 |
| technomancy | I'm guessing once bookworm comes out it'll be possible to actually lock in and have a stable system finally? | 04:09 |
| sevan | I assume so, but I'm not sure what local change are carried if any. | 04:15 |
| sevan | just need the reform version of kernel and components related to that (lpc module?), stick that on stock install of upstream release? | 04:17 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:40) | 04:21 | |
| + bkeys1 (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 04:53 | |
| - bkeys1 (QUIT: Ping timeout: 260 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 04:58 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:40) | 05:32 | |
| chartreuse | klardotsh I bought my replacement JGNE cells from 18650batterystore.com | 06:03 |
| klardotsh | out of stock there and basically everywhere it seems :\ hopefully after Chinese New Year. | 07:05 |
| + bgs (~bgs@212-85-160-171.dynamic.telemach.net) | 08:05 | |
| - S0rin (QUIT: Ping timeout: 260 seconds) (~S0rin@user/s0rin) | 08:22 | |
| + S0rin (~S0rin@user/s0rin) | 08:22 | |
| - GNUmoon2 (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 09:31 | |
| violet | whats the deal with the GPU on the new 8GB module that's becoming available? far as I can tell it has half the shader cores- but is it just as simple as half the cores = half the GPU performance or what? | 09:43 |
| violet | idk anything about like clock rate for the GPU cores | 09:44 |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 09:49 | |
| violet | i think mainly i am just wondering about whether it will have the bandwidth to handle playing back video at 30fps smoothly with like mpv . which is a thing I was never able to get the rock64 to do within an X or wayland session | 09:49 |
| josch | violet: video playback is not so much relying on the gpu as it is relying on the processor (for software decoding) or on the hantro g2 (for hardware decoding) | 10:26 |
| josch | violet: the cpu can definitely do 30 fps h264 playback | 10:26 |
| josch | violet: if you use software that can do hardware decoding (like clapper+gstreamer) then you can playback h264 1080p@60 even | 10:27 |
| josch | violet: this is on the imx8mq -- as far as i understand it, the imx8mplus also has the hardware decoder so this should work similarly | 10:27 |
| josch | violet: the problem with mpv is, that it's ffmpeg based and ffmpeg can only do software decoding which (with the current cpu) means you can get up to 1080p@30 h264 but not much higher without frame drops | 10:29 |
| josch | there is an ffmpeg patch stack that has been rebased for several years now which allows hardware decoding but it never got merged in ffmpeg | 10:29 |
| josch | i managed to build ffmpeg with those patches and they do enable hardware decoding but the video is tinted red heavily | 10:29 |
| josch | so for more than 1080p@30 h264 you'd have to switch from mpv to clapper, for example | 10:29 |
| josch | minute: could you merge this? Thanks! https://source.mnt.re/reform/reform-tools/-/merge_requests/31 | 12:17 |
| josch | after that i'll re-run reform-debian-packages pipeline and then i'll send you another MR for reform-system-image | 12:18 |
| * wielaard -> mjw | 12:20 | |
| minute | https://www.aries-embedded.com/system-on-module/cpu/rzfive-renesas-riscv-msrfive-osm-dual-ethernet-can | 12:45 |
| - mtm (QUIT: Ping timeout: 260 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 13:03 | |
| - qbit (QUIT: Quit: WeeChat 3.7.1) (~qbit@h.suah.dev) | 13:22 | |
| pandora[m] | The biggest issue with software video decoding is the high cpu usage what results in high energy consumption and short battery runtime | 13:26 |
| + qbit (~qbit@2602:ff16:3:0:1:3a0:0:1) | 13:33 | |
| flowy | josch: to workaround the red tint, did you happen to try a different vo driver, like vo=dmabuf-wayland? random guess, do not revisit if that build is not on your system any longer | 13:44 |
| josch | flowy: i did not! | 13:48 |
| josch | flowy: do you want to try out the ffmpeg deb packages and have a look? | 13:48 |
| flowy | i'm also slightly concerned about the imx8mplus having worse GPU performance. i'm really enjoying moonlight on my reform @60fps. but there seems to be inadequate data available to know in advance what it will mean for various applications | 13:49 |
| flowy | josch: like, just whatever apt pulls down? | 13:49 |
| flowy | i don't mind trying | 13:50 |
| josch | flowy: i'll let the source.mnt.re pipeline build the debs for you :) | 13:51 |
| flowy | ah, nice | 13:51 |
| josch | minute: if you have latest u-boot (with extlinux, lcdif and usb keyboar support) installed and want to try out how it would look like with a boot menu, you can install the u-boot-menu package which will auto-generate /boot/extlinux/extlinux.conf with all currently installed kernels. Currently, if you try this, your system will become unbootable because the kernel cmdline the reform needs isn't added. I | 13:56 |
| josch | created this MR for reform-tools to fix this: https://source.mnt.re/reform/reform-tools/-/merge_requests/32 So if you merge this and then wait for another reform-debian-packages pipeline run, you can "apt install u-boot-menu" and see if you like it. :) | 13:56 |
| josch | flowy: ffmpeg should show up in the artifacts of this job: https://source.mnt.re/reform/reform-debian-packages/-/jobs/1118 | 14:03 |
| josch | flowy: you should be able to use the artifacts url in your sources.list directly | 14:04 |
| josch | just don't forget to put [trusted=yes] because nothing is signe | 14:04 |
| josch | d | 14:04 |
| - mjw (QUIT: Quit: Leaving) (~mjw_@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 14:11 | |
| + mjw (~mjw_@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 14:24 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 15:10 | |
| flowy | josch: i'm intimidated. some of these libav* packages are named differently than the ones i would have to replace on my system. i'd have to hunt conflicts it seems | 16:02 |
| flowy | and it probably won't work anyways | 16:03 |
| flowy | i don't have the dpkg foo to manage this mess | 16:03 |
| flowy | (mess of my own system) | 16:04 |
| - sevan (QUIT: Quit: need to shutdown) (~sevan@user/venture37) | 17:38 | |
| rafostar[m] | <josch> "violet: video playback is not so..." <- Hardware accel decoding is not the answer on its own. When you get HW decoded data on the GPU, app still needs to find a way to display it in efficient manner. Clapper provided display sink relies on GTK4 GL functionality which along with GStreamer GL elements can display decoded frame without downloading/uploading it back and forth from system memory to GPU. Software decoding always | 18:00 |
| rafostar[m] | needs an upload to GPU. Also, colorspace conversion with GL might require different GL version depending on format (for e.g. 10bit video you might need GLES 3.1 or later to take the most efficient code path). So what GPU can do is also partially relevant in some cases. | 18:00 |
| - qbit (QUIT: Quit: WeeChat 3.7.1) (~qbit@2602:ff16:3:0:1:3a0:0:1) | 18:09 | |
| + qbit (~qbit@h.suah.dev) | 18:12 | |
| Ar|stote|is | Now I need to hunt the shop for the wifi antennas and then bam! | 18:45 |
| Ar|stote|is | minute: I got my reform since I was one of the first "pre-orderers" from the EU without it going to the States first. Will there be a similar way to preorder the Pocket? Will you sell preorders outside of Crowd Supply or only after the preorder phase ? | 18:47 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:40) | 19:11 | |
| + bkeys1 (~Thunderbi@157.201.98.1) | 19:32 | |
| Boostisbetter | minute, in case I didn't already tell you, I was able to use the MNT camera to make a dino video call using the Librem 5. As such I think making a video call on the Reform is a foregone conclusion at this point. | 19:38 |
| - bkeys1 (QUIT: Ping timeout: 246 seconds) (~Thunderbi@157.201.98.1) | 20:32 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:40) | 20:38 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by markw!~wielaard@gnu.wildebeest.org))) (~mjw_@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:39 | |
| + wielaard (~mjw_@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:39 | |
| + ajr (~ajr@user/ajr) | 20:42 | |
| + bkeys1 (~Thunderbi@157.201.98.1) | 21:08 | |
| * bluerise_ -> bluerise | 21:19 | |
| - bluerise (QUIT: Changing host) (~bluerise@pc19f8f98.dip0.t-ipconnect.de) | 21:19 | |
| + bluerise (~bluerise@user/bluerise) | 21:19 | |
| - bkeys1 (QUIT: Ping timeout: 252 seconds) (~Thunderbi@157.201.98.1) | 21:21 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 21:28 | |
| minute | Boostisbetter: oh, that's interesting! you could just plug it in over usb-c? | 23:01 |
| minute | Ar|stote|is: not sure tbh | 23:01 |
| Boostisbetter | minute: I like as a mini x laptop dock with it and used a USB c port on it. | 23:24 |
| Boostisbetter | Worked great! | 23:24 |
| minute | that is very good to hear | 23:32 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!