| - mjw (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org) | 00:55 | |
| - jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@utdpat241106.utdallas.edu) | 02:41 | |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 04:00 | |
| - nsc (QUIT: Ping timeout: 240 seconds) (~nicolas@i5E861467.versanet.de) | 04:01 | |
| + nsc (~nicolas@89-48-142-46.pool.kielnet.net) | 04:02 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 04:21 | |
| - jacobk (QUIT: Ping timeout: 255 seconds) (~quassel@utdpat241106.utdallas.edu) | 05:57 | |
| violet | i have also had lockups after long periods of idle. but like, not consistently | 06:32 |
|---|---|---|
| violet | just, every now and then | 06:32 |
| violet | i think only twice now actually | 06:33 |
| violet | and i suspected it might be thermal related or something because i had it sitting on a surface that does not conduct heat well | 06:33 |
| doctorhoo | josch: thanks for the tip, I will try your workaround! | 06:36 |
| doctorhoo | violet: I have had a similar suspicion that heat could play a role; with the laptop sitting in the evening sun for example, it seems to happen quicker; however, it also locks up in the middle of the night with much cooler conditions and I have never had it lock up having it sit on my lap and with almost constant CPU load.. But still, I have been meaning to exchange the silver paste on the CPU at some point, will try that too : | 06:40 |
| violet | bodies *are* pretty good heat sinks in my experience | 06:41 |
| violet | lots of water, lots of thermal mass | 06:41 |
| doctorhoo | violet: Interresting points! Do you have a specific component in mind that you think is affected by overheating? | 07:09 |
| - doctorhoo (QUIT: Ping timeout: 255 seconds) (~doctorhoo@217-210-162-155-no600.tbcn.telia.com) | 07:29 | |
| + doctorhoo (~doctorhoo@2a00:801:717:bb7b:c16b:e94:9afe:38b5) | 07:50 | |
| josch | for me, i have my reform on 24/7 and it *never* locked up again since i switched from "dpms off" to "brightnessctl set 0" | 08:18 |
| josch | it can happily compile linux for six hours on all four cores while being placed on the couch cushion | 08:19 |
| vkoskiv | https://www.crowdsupply.com/cool-tech-zone/tangara | 09:19 |
| vkoskiv | I imagine it would look much like that if MNT decided to enter the music player space :D | 09:19 |
| minute | i only had lockups if i forgot to enable my swapfile and ran out of memory in unhappy ways | 09:43 |
| josch | minute: but how do you put your display to sleep? | 09:58 |
| josch | do you use dpms off or brightnessctl? | 09:59 |
| josch | minute: here are again images with reform-tools 1.33 together with the changes to u-boot to all three platforms that we have discussed. This includes the changes to the cmdline with cma=512 https://source.mnt.re/reform/reform-system-image/-/jobs/2386/artifacts/browse | 10:06 |
| minute | josch: ah. i do not | 10:15 |
| minute | josch: or, i manually do it with brightnessctl, no dpms | 10:15 |
| josch | then that explains :) | 10:15 |
| josch | i had my reform connected via serial when putting the display to sleep via dmps but nothing shows up in dmesg | 10:16 |
| josch | it just stops responding | 10:16 |
| josch | interesting how for eery it also happens with a311d | 10:16 |
| minute | weird | 10:25 |
| josch | this should've been something i should've tested with ls1028a... next time then | 10:31 |
| minute | bpi cm5 seems alive, but afaik there's no mainline support for a311d2 yet https://mastodon.social/@bananapi/111260738859393860 | 10:39 |
| josch | i worry that potential customers will again be put off by "only" having 4 GB of ram -- even more so in a year or so once support for it ends up in mainline linux and people expect even *more* ram in their machines... | 10:49 |
| + mjw (~mjw@gnu.wildebeest.org) | 10:53 | |
| vkoskiv | It does say "Max 8G" on their wiki page about the A311D2 | 11:06 |
| - Boostisbetter (QUIT: Ping timeout: 255 seconds) (4a410829d7@irc.cheogram.com) | 12:06 | |
| violet | doctorhoo: i dunno, just the laptop in general. bottom was somewhat uncomfortable to the touch | 12:06 |
| - chomwitt (QUIT: Ping timeout: 252 seconds) (~chomwitt@2a02:587:7a17:8900:1ac0:4dff:fedb:a3f1) | 12:06 | |
| violet | ram is so overrated lol | 12:07 |
| violet | good to have but overrated | 12:07 |
| - doctorhoo (QUIT: Ping timeout: 258 seconds) (~doctorhoo@2a00:801:717:bb7b:c16b:e94:9afe:38b5) | 12:12 | |
| + doctorhoo (~doctorhoo@2a00:801:706:a0a0:31ae:9001:24a7:1df3) | 12:12 | |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 12:14 | |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 12:24 | |
| * mark_ -> mjw | 12:26 | |
| josch | Before i had the reform i had 12 g of ram. The only thing I'm missing is to have a huge scratch space in a tmpfs mounted /tmp. | 12:43 |
| josch | But I can also see how other people have workloads that are different from mine and where much more than 4 gigs are necessary. | 12:43 |
| minute | josch: yeah the interesting thing about a311d2 is the 8 gigs | 13:04 |
| minute | and 8 cores | 13:04 |
| josch | do the rockchip options that you are evaluating also have 8G+? | 13:05 |
| erle | regarding workloads, i recently had someone claim that having a single-digit of RAM was unbearable and then it turns out they were hugely addicted to electron apps | 13:07 |
| josch | speaking of 8G: the reform also works with the imx8mplus, no? should we not build a system image for that as well and add support for that to reform-tools? | 13:07 |
| erle | or, as some call it, the webkit vulnerability portability framework | 13:07 |
| erle | single-digit GB of RAM i mean ;) | 13:07 |
| josch | erle: yeah that's what i meant with "different workloads" above. I don't have any electron thing installed and even with firefox and 4.2k tabs open I never saw any problems... | 13:09 |
| josch | the other thing is compiling things with heavy c++ templating requirements | 13:09 |
| josch | which... incidentally... brings us back to chromium :D | 13:09 |
| erle | > 4.2k tabs open | 13:11 |
| erle | i think you beat me | 13:11 |
| erle | it crashes too often for me to get to that point | 13:11 |
| erle | i only manage hundreds | 13:11 |
| erle | how do you manage those anyway, tree style tab? | 13:12 |
| josch | most of them are never loaded into memory -- i think firefox only loads the few active ones | 13:12 |
| josch | i'm a heavy user of the %-search in the firefox address bar | 13:12 |
| josch | essentially i use tabs like other people use bookmarks | 13:13 |
| erle | to quote an acquaintance ”i am a turing machine going forwards and backwards through tabs” | 13:13 |
| sknebel | yeah, usually firefox is pretty good about only the tab-rendering processes crash when memory gets scarce | 13:13 |
| erle | on x86 i have this nasty thing that webrender manages to starve itself of memory at some point | 13:14 |
| sknebel | erle: ... tab-based esoteric programming language?! | 13:14 |
| erle | sknebel we have this already: https://en.wikipedia.org/wiki/Whitespace_(programming_language) | 13:14 |
| erle | > Commands are composed of sequences of spaces, tab stops and linefeeds. All other characters are ignored and thus can be used for comments. | 13:15 |
| sknebel | wrong kind of tabs ;) | 13:15 |
| erle | i know | 13:15 |
| erle | you should have made that comment context-free or regular! | 13:15 |
| erle | my take is actually that every joke requires a misunderstanding, i.e. context-sensitivity | 13:15 |
| erle | therefore, jokes can not be unambiguously parsed | 13:16 |
| sknebel | hm... | 13:16 |
| sknebel | not sure that follows? | 13:16 |
| erle | well, you have to backtrack at last | 13:16 |
| erle | to reparse | 13:16 |
| sknebel | because with complete knowledge of the parser it could unambiguous what the ambiguity initially was? | 13:16 |
| sknebel | yeah | 13:16 |
| erle | but during the parse you are lead astray | 13:17 |
| erle | led? i don't know englishing | 13:17 |
| minute | josch: rockchip goes up to 32GB | 13:22 |
| minute | josch: we're kind of phasing out the plus-for-reform option. we are not able to buy these in small quantities | 13:22 |
| minute | josch: MOQ 400 | 13:22 |
| minute | josch: so we won't be able to order them later | 13:22 |
| minute | for reforms | 13:22 |
| josch | okay, makes sense | 13:23 |
| minute | and too much fragmentation for little gain | 13:23 |
| - iank (QUIT: Quit: ZNC 1.8.2+deb2+b1 - https://znc.in) (~iank@fsf/staff/iank) | 13:25 | |
| minute | josch: we found out today that the ls1028a display flashing can be remedied by just applying a few layers of isolation tape around the area of the flat cable that goes under the heatsink (until the origami section) | 13:27 |
| + iank (~iank@fsf/staff/iank) | 13:27 | |
| minute | josch: it looks like the most important thing is that the cable doesn't hug the hatsink nor the module too much. my theory is that it is about ground plane discontinuity | 13:27 |
| minute | s/hatsink/heatsink | 13:28 |
| josch | minute: could you post some photos of your setup so that i can replicate it the next time i install the ls1028a? | 13:28 |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 13:28 | |
| minute | josch: yep, i'll ask greta to make a photo of the next one | 13:28 |
| josch | thank you! <3 | 13:28 |
| minute | josch: https://chat.mnt.re/files/ns8auanantbz3j7dynnhq3both/public?h=wXs42GE59u5sH6hkrMife_z6iFgwtX0xduY6_dpshvs | 13:59 |
| minute | josch: https://chat.mnt.re/files/ns8auanantbz3j7dynnhq3both/public?h=wXs42GE59u5sH6hkrMife_z6iFgwtX0xduY6_dpshvs | 13:59 |
| minute | weird, these are the same link. | 14:00 |
| minute | josch: https://chat.mnt.re/files/obizdah5m3fydyfhmza37nm1tc/public?h=uqGMKItuYk7pcWw-mmFUVAoQesDbj3xXfQbd3WTV5Os | 14:00 |
| josch | minute: thank you! that looks like it's quite some wide tape -- does its type matter much? i don't recognize the type from the photos | 14:05 |
| minute | josch: the type doesn't matter. this one is called teraoka https://www.modulor.de/teraoka-klebeband-mit-gewebe-p-cut-nr-4140-50-mm-x-25-m-s-0-15-mm-gelb.html | 14:09 |
| josch | i'd wager it matters a little bit, given how my heatsink managed to get up to 84°C -- i'll try out what i have at home :) | 14:11 |
| josch | this is certainly an easier fix than having to keep the ferrite ring around it | 14:14 |
| josch | hah i see in the corner that this board already has the jtag boot-fix applied :D | 14:14 |
| josch | i'm so happy that you found that *and* that it was easily fixable like this! | 14:14 |
| minute | josch: can't tell you what a relief that was | 14:19 |
| josch | i probably cannot imagine anyways since i never had that much of my money depend on something like this XD | 14:20 |
| josch | it's quite the safe life if all you ever gamble is company money (without being the owner of that company) | 14:21 |
| Boostisbetter | you know with suspend working flawlessly now, I think it might be time to work on power efficiency improvements. Particularly when in standby. I have noticed that all of the ports stay powered when in standby. If we could get consumption during standby down to around 50 or 25 mah I think that would make the Reform lasting a full day off the charger a lot more possible. | 14:26 |
| + stites (~stites@130.44.147.204) | 14:29 | |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 14:49 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 14:50 | |
| josch | i just finally migrated both my desktops at $work to the wayland+greetd+sway+waybar setup from my reform using the sway config from reform-tools. Much easier to switch between machines now. :) | 14:54 |
| minute | josch: nice! which greeter for greetd? | 14:56 |
| josch | just the default agreetd | 14:57 |
| josch | i do not need it to look nice for the three times a year that those machines get rebooted and i have to log in :D | 14:57 |
| josch | though the list of available greeters on https://sr.ht/~kennylevinsen/greetd/ did grow considerably since the last time i visited | 14:58 |
| - doctorhoo (QUIT: Ping timeout: 272 seconds) (~doctorhoo@2a00:801:706:a0a0:31ae:9001:24a7:1df3) | 15:09 | |
| + doctorhoo (~doctorhoo@2a00:801:725:ef2b:d821:b669:fa6b:f6e) | 15:11 | |
| vkoskiv | josch: I also use sway + the reform config in my VM on my work mac | 15:24 |
| vkoskiv | Though I never fully copied the needed configs, so some things don't match | 15:24 |
| Boostisbetter | the greeters basically make it so that you log in similar to gnome instead of manually via the CLI? | 15:42 |
| vkoskiv | So the Reform charges at 48 watts with 24V @2A. I can put in up to 32V @ 1.5A, but if I were to plug in 9V @ ~5.3A, wouldn't that blow the 4A fuse at F1? | 15:43 |
| minute | Boostisbetter: yup | 15:44 |
| minute | vkoskiv: do you have a power supply that can do 5+A? | 15:45 |
| vkoskiv | I probably don't :D | 15:45 |
| vkoskiv | I forget the max of my old bench supply, probably like 2.5A | 15:45 |
| vkoskiv | But if I were to just plug a 12V solar panel directly in the barrel jack, what would happen then? | 15:45 |
| vkoskiv | (Not going to, just wondering) | 15:46 |
| minute | don't know! | 15:46 |
| vkoskiv | I know enough to know big amp bad, if wire small | 15:46 |
| vkoskiv | Would be a cool off-grid kinda thing to just plug a solar panel into this thing :D | 15:47 |
| vkoskiv | ACTION says, being extemely on-grid, living in a large city. | 15:47 |
| Boostisbetter | minute: any plans to move this into the official image, or do we feel that this is guiding the direction of Reform use too much? I don't have a problem with using the TTL to launch sway after logging in manually | 15:47 |
| jfred | Yeah I hope that can be made to work without frying the Reform 😅 | 15:47 |
| vkoskiv | If I understand correctly, the D13 diode is there if one were to hook the polarity up wrong | 15:48 |
| vkoskiv | So then it'd either overload the power supply, or worst case, blow F1 | 15:48 |
| josch | vkoskiv: you read about the other people in this chan who directly plugged their solar panel into their reform, right? | 15:51 |
| vkoskiv | Nope | 16:00 |
| vkoskiv | I can grep the logs though | 16:00 |
| vkoskiv | Oh, there is me pondering this same thing in june 2022 | 16:01 |
| josch | ah found it! noam was the one who powered their reform off of a solar panel by directly plugging it into their reform | 16:37 |
| josch | ah and vagrant | 16:38 |
| vkoskiv | I'm not sure if the boost converter knows to draw lower current if it detects the voltage dropping further | 16:39 |
| vkoskiv | Is there any downsides to having a hidraw command like 'xJTBL' that makes the mcu jump into the bootloader? | 16:47 |
| vkoskiv | I added that for the trackpad, and made flash.sh send that command, and it seems like a nice way to develop the firmware | 16:47 |
| vkoskiv | i.e. no need to open the back to reflash | 16:47 |
| vkoskiv | Same for the keyboard, I've been using circle+x to kick it into the bootloader to flash it, no need to remove the bezel | 16:48 |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@130.44.147.204) | 16:58 | |
| + stites (~stites@130.44.147.204) | 16:58 | |
| - GNUmoon (QUIT: Ping timeout: 256 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 17:01 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 17:02 | |
| - stites (QUIT: Ping timeout: 258 seconds) (~stites@130.44.147.204) | 17:03 | |
| + stites (~stites@2607:fb91:de5:5d85:fe03:1608:5962:585c) | 17:03 | |
| + IchikaZou (~IchikaZou@27-53-25-142.adsl.fetnet.net) | 17:04 | |
| - IchikaZou (QUIT: Remote host closed the connection) (~IchikaZou@27-53-25-142.adsl.fetnet.net) | 17:19 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 17:54 | |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@2607:fb91:de5:5d85:fe03:1608:5962:585c) | 18:03 | |
| + stites (~stites@2607:fb91:de5:5d85:fe03:1608:5962:585c) | 18:04 | |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@2607:fb91:de5:5d85:fe03:1608:5962:585c) | 18:05 | |
| + stites (~stites@155.33.134.28) | 18:06 | |
| - doctorhoo (QUIT: Ping timeout: 264 seconds) (~doctorhoo@2a00:801:725:ef2b:d821:b669:fa6b:f6e) | 18:49 | |
| + doctorhoo (~doctorhoo@2a00:801:7a7:5c1a:9472:da2e:8edf:b925) | 18:50 | |
| * deflated8837 -> cow321 | 19:03 | |
| - cow321 (QUIT: Remote host closed the connection) (~deflated8@50.39.247.54) | 19:03 | |
| + cow321 (~deflated8@50.39.247.54) | 19:04 | |
| + ajr (uid609314@user/ajr) | 19:29 | |
| - doctorhoo (QUIT: Read error: Connection reset by peer) (~doctorhoo@2a00:801:7a7:5c1a:9472:da2e:8edf:b925) | 19:32 | |
| + doctorhoo (~doctorhoo@217-210-162-155-no600.tbcn.telia.com) | 19:33 | |
| - doctorhoo (QUIT: Ping timeout: 255 seconds) (~doctorhoo@217-210-162-155-no600.tbcn.telia.com) | 19:43 | |
| + doctorhoo (~doctorhoo@2a00:801:7a7:5c1a:9472:da2e:8edf:b925) | 19:44 | |
| - doctorhoo (QUIT: Read error: Connection reset by peer) (~doctorhoo@2a00:801:7a7:5c1a:9472:da2e:8edf:b925) | 19:53 | |
| + doctorhoo (~doctorhoo@217-210-162-155-no600.tbcn.telia.com) | 19:53 | |
| + jacobk (~quassel@129.110.242.224) | 20:07 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 20:31 | |
| + Ar|stote|is (~linx@149.210.32.15) | 20:39 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mjw@gnu.wildebeest.org))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:40 | |
| * mark_ -> mjw | 20:40 | |
| + mark_ (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:40 | |
| - pandora (QUIT: Quit: Connection closed for inactivity) (uid585533@id-585533.ilkley.irccloud.com) | 21:14 | |
| - jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@129.110.242.224) | 21:21 | |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 21:26 | |
| + IchikaZou (~IchikaZou@27-53-25-142.adsl.fetnet.net) | 22:10 | |
| - jacobk (QUIT: Ping timeout: 240 seconds) (~quassel@utdpat241106.utdallas.edu) | 22:31 | |
| - IchikaZou (QUIT: Remote host closed the connection) (~IchikaZou@27-53-25-142.adsl.fetnet.net) | 22:36 | |
| + jacobk (~quassel@129.110.242.224) | 23:03 | |
| noam | josch: Hi yes I have done that | 23:17 |
| josch | noam: you posted photos about that but those are offline now? | 23:18 |
| noam | ...probably | 23:25 |
| noam | If I post a photo, it's generally temporary | 23:25 |
| noam | I toss those in /tmp | 23:25 |
| noam | [on the server] | 23:25 |
| noam | but yeah, 12V nominal-100W panel plugged directly in through a DIY barrel jack adapter :P | 23:25 |
| noam | worked like a charm | 23:25 |
| - stites (QUIT: Ping timeout: 240 seconds) (~stites@155.33.134.28) | 23:36 | |
| + stites (~stites@2607:fb91:de8:94ca:cd75:d180:a559:e9f9) | 23:36 | |
| - stites (QUIT: Ping timeout: 272 seconds) (~stites@2607:fb91:de8:94ca:cd75:d180:a559:e9f9) | 23:44 | |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 23:58 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 23:59 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!