2024-10-01.log

joschit's metalic looking paste/pad -- i'll probably always be worried XD00:00
vagrantcanyways ... eagre to get the upgraded rk3588 running in my mnt/reform :)00:03
grimmwarePaste is stressful tho to be sure00:03
minuteif you stress the rk3588 it gets hottttt btw00:07
minuteafterwards it settles around 57 degrees00:10
minute(it's on charger)00:10
joschthat's colder than my a311d usually is00:10
minutereally? my a311d normally feels cooler in the pocket00:11
joschmy a311d is usually at around 64°C in classic reform00:11
minuteohh00:11
minutemy reform with rk3588 is idling at 52 degrees atm00:12
joschnice!00:12
grimmwareminute: I was wondering by the way, if the external copper design on the pocket were extended to have more like, filigree and shit coming off the MNT logo and extending out to the edges do you think it would make a significant improvement to its ability to dissipate heat?00:13
joschmust be a big relieve that the production rcore modules are actually working :)00:13
minutejosch: yesss00:15
minutegrimmware: actually maybe yes. it's weird how the logo is really hot. one could get a branding from it maybe00:15
minutetemp went down to 53-54 because i wasn't doing heavy things00:16
jfredSuppose I'll soon find out how a different backplate with less exposed copper on the top feels/how that impacts the thermals. I just installed the current revision of the custom plate I made00:19
jfredif I haven't completely wrecked wifi performance I'll keep it like this + finally install my wwan card (didn't want to permanently adhere the antenna to the backplate until I had one I knew I'd be sticking with)00:21
grimmwareminute: for a quick easy test I reckon something like https://shop.mntre.com/products/mnt-pocket-reform-t-shirt-pink-pcb?taxon_id=21 would look really cool and likely shed more heat00:21
minutejfred: did you make a solid metal plate?00:21
chminute: is the wifi fine for you with just the one antenna (+ back plate closed)?00:21
minutech: i've two antennas, just had only one in the photo at that moment00:22
chah00:22
jfredminute: no, still a PCB - but with exposed copper lettering: https://jawns.club/@jfred/113228639767886871 (so more copper over the edges where the antenna is, and not so much over the heatsink part)00:23
minutejfred: ahhh yes00:24
jfredmy previous revision had copper over the whole top layer, but I changed that for this one for fear of causing more wifi problems haha - now it's just copper where the silkscreen cutouts are00:24
joschjfred: oh no, i remember your custom backplate and i was looking for it on community.mnt.re but i remember them from irc i think!00:26
jfredironically though that may have helped spread the heat around had I left it that way00:26
jfredah yeah, I posted about it on the forum but not sure if I put pictures there00:26
grimmwareI’m considering fabbing some side plates (the one with the power switch) with a hole large enough for a plate-mount SMA so I can just have a honking great external antenna that unscrews when I need to put the thing in my bag00:29
grimmwareI bought some antennas for that purpose but migrating to nvme and installing the thermal pads improved the situation enough for me to get by at the time00:30
jfredI worry about external antennas a bit, I know you can damage a radio if you power it on without an antenna attached00:30
jfredbut I'm sure it would be pretty effective00:30
grimmwareOh. I should probably stop changing antennas with the radio turned on then :S00:31
jfredyeah my recollection is the antenna acts as a load and the RF signals bounce back at full power without it or something like that00:34
+ kensanata8 (~alex@user/kensanata)00:34
- kensanata (QUIT: Read error: Connection reset by peer) (~alex@user/kensanata)00:36
* kensanata8 -> kensanata00:36
+ Ar|stote|is (~linx@149.210.1.97)01:00
jfredSo far (unscientifically) wifi and thermals both seem fine with the new backplate. Will need to stress the CPU a bit and see if that changes01:10
minutenice01:30
- Ar|stote|is (QUIT: Ping timeout: 260 seconds) (~linx@149.210.1.97)01:36
+ Ar|stote|is (~linx@149.210.2.103)01:40
- mtm (QUIT: Ping timeout: 276 seconds) (~textual@c-71-228-84-213.hsd1.fl.comcast.net)02:02
+ mtm (~textual@c-71-228-84-213.hsd1.fl.comcast.net)02:06
+ jacobk (~quassel@2603:8080:b200:7b02::b71)02:08
- Ar|stote|is (QUIT: Ping timeout: 252 seconds) (~linx@149.210.2.103)02:10
- Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66)02:11
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)02:12
+ Ar|stote|is (~linx@149.210.1.237)02:14
- jacobk (QUIT: Ping timeout: 276 seconds) (~quassel@2603:8080:b200:7b02::b71)02:55
- paperManu (QUIT: Ping timeout: 260 seconds) (~paperManu@198.16.214.40)03:10
- mjw (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org)03:12
digitalraneminute, i think i missed a ping from you whilst my bouncer was messed up about the replacement reform screen? (the one I got is an IVO M125NWF4). I don't recall having the same problems you described about it switching off at low brightness, but I did noticed the brightness curve (PWM values vs brightness level) is different to the original inolux03:14
digitalrane*did notice03:15
+ jacobk (~quassel@47-186-105-237.dlls.tx.frontiernet.net)03:28
- nsc (QUIT: Ping timeout: 276 seconds) (~nicolas@87.122.72.114)03:31
+ nsc (~nicolas@i5C74DEDC.versanet.de)03:32
- sts-q (QUIT: Ping timeout: 276 seconds) (~sts-q@2a11:fb80:3a6:8301:3453:af1b:2a:ed74)03:59
* aloo_shu -> CandidCadaver04:00
- kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de)04:03
+ kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de)04:05
- murphnj (QUIT: Quit: Leaving) (~murph@user/murphnj)04:18
* Sario -> Scario04:33
+ Gooberpatrol_66 (~Gooberpat@user/gooberpatrol66)04:45
- Gooberpatrol66 (QUIT: Ping timeout: 276 seconds) (~Gooberpat@user/gooberpatrol66)04:46
jfredinterestingly after stressing the machine some I've only had one wifi drop that gave me trouble and it was when I moved to a spot where I know wifi coverage is poor. took it a while to reconnect after that but since then it's been solid05:04
jfredhighest temp I've seen so far is 70 degrees on the soc sensor, it's sitting around 67 now05:04
jfredpotential confounding factor, I think the thermal pad I put over the CPU on this backplate may have been slightly larger than the stock one05:06
jfredthe center above the CPU module is still very hot, but it's not as immediately unpleasant to touch without the exposed copper directly in the center05:08
- CandidCadaver (QUIT: Ping timeout: 265 seconds) (~aloo_shu@85.51.17.35)05:36
- spew (QUIT: Quit: spew) (~spew@201.141.99.170)05:41
+ CandidCadaver (~aloo_shu@85.51.17.35)05:45
+ tuxella (~textual@89.207.171.108)06:00
- tuxella (QUIT: Client Quit) (~textual@89.207.171.108)06:01
+ sts-q (~sts-q@89.58.73.0)06:14
- vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50)07:05
minutedigitalrane: cool, thanks!09:34
kensanatajfred: I don't use wifi often (got my Pocket Reform on ethernet) but the one time wifi kept failing was also extraordinarily warm (room temp around 28°C). So now I keep thinking if people from the northern hemisphere report that their wifi problems are solved, perhaps that has to do with the end of summer.09:46
kensanataWe need to control for ambient temp, I suspect. Also very unscientifically speaking, of course. :D09:46
+ jaume (~user@user/jaume)10:04
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)10:04
- Gooberpatrol_66 (QUIT: Ping timeout: 260 seconds) (~Gooberpat@user/gooberpatrol66)10:04
+ mjw (~mjw@gnu.wildebeest.org)10:24
- wakest (QUIT: Remote host closed the connection) (m-a7d6fe@45.77.48.108)10:30
+ wakest (m-a7d6fe@45.77.48.108)10:30
+ erle (~erle@user/erle)10:32
joschminute: also some wifi feedback from me: since you changed a311d wifi, i did only have connection drops every couple of days, so the situation greatly improved10:47
minutejosch: oh that's awesome11:01
- jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@47-186-105-237.dlls.tx.frontiernet.net)11:03
+ jacobk (~quassel@47-186-105-237.dlls.tx.frontiernet.net)11:05
grimmwarekensanata: that's a good observation about the end of summer11:31
grimmwareI think a backplate with more exposed copper would be sensible, even if it just disperses the heat from the MNT logo a little more to make it less unpleasant to touch11:32
minutecurrently trying to source some glass.12:02
joschglass to touch? :)12:04
- mjw (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org)12:12
grimmwarewhat for???12:19
chglassy models confirmed!11 ;)12:30
+ paperManu (~paperManu@198.16.214.40)12:36
grimmwareminute: just wanted to check, is the fpc connecting the enh-Ob00910003 on the OLED module the one that appears to come with the enh-Ob00910003 or a custom job? Starting to think about the feasibility of this OLED module with a qwiic connector on it.12:41
+ mjw (~mjw@gnu.wildebeest.org)12:42
+ jaume` (~user@85.51.226.170)12:44
- jaume (QUIT: Ping timeout: 252 seconds) (~user@user/jaume)12:45
- mjw (QUIT: Ping timeout: 272 seconds) (~mjw@gnu.wildebeest.org)13:13
* Guest2216 -> mjw13:14
minutegrimmware: if you mean the flex pcb attached to the display itself, it is standard13:24
+ murphnj (~murph@user/murphnj)13:29
grimmwareI do13:54
grimmwarethanks :)13:54
grimmwaretrying to order a sample from alibaba :P13:54
grimmwarefuckin lol, they've bumped the order price from $2.40 to $44.40 for one unit13:56
grimmwareNOPE13:56
minutegrimmware: huh13:57
minutegrimmware: shipping maybe?13:57
minuteor sampling cost13:57
grimmwarenfi, they didn't say13:57
minutemaybe aliexpress?13:57
minutealibaba makes more sense for big volumes and then for single qtys there are sample fees13:58
grimmwareyeah I just haven't seen the same item anywhere else13:58
grimmwaresearch results are alibaba, the manufacturer's website and MNT13:59
grimmwarehmm, unless I can scavenge it from a breakout board14:01
minutehmm yeah i assume there are identical/compatible ones14:01
grimmwarehow much do you wanna bet this is it from the shape of the fpc https://shop.pimoroni.com/products/0-91-inch-white-oled-display-module-i2c-interface?variant=3997562634248314:02
grimmwareyeah same thing on my keyboard by the looks of things!14:02
- mtm (QUIT: Ping timeout: 246 seconds) (~textual@c-71-228-84-213.hsd1.fl.comcast.net)14:03
grimmwareyeah these are all over eBay and in fact if I do that I bet I'll get most of the resistors/capacitors for free!14:06
+ mtm (~textual@c-71-228-84-213.hsd1.fl.comcast.net)14:06
minuteyeah14:08
grimmwareokay cool, I think I've convinced myself that this is doable with my tools and skillset14:11
grimmwareI just need to look at the clearance on how the module is mounted in both pocket and big reform so I don't put the qwiic connector somewhere stupid and muster the emotional resilience to open KiCAD.14:13
grimmwareI have a hot plate and scope that I can use to do the side with all the resistors etc on and then I think I'll probably have to try to do the chisel-tip sweep method with a soldering iron on the FPC side which I managed exactly once with a surface mount mini-USB socket14:16
grimmwareunless someone has a better suggestion because ihavenoideawhatimdoing.jpg14:18
Zabajust add flux 14:25
Zabaand remove extra solder with braid or a pump if it becomes necessary14:26
Zabayou generally need miniscule amounts of solder for smd 14:26
Zabaso the solder residue on a normal-sized soldering iron tip is often all you need for a small smd component 14:29
grimmwareI meant with the strategy of doing the sweep for the FPC, like I say I'm using the hot plate for most of it14:30
Zabayeah, as long as there’s flux and no excess solder to create shorts, surface tension takes care of it all14:34
Zabait might also be tricky to properly heat things like big ground planes or other copper fills with an iron 14:35
Zabamy solution to that is usually to cuss and to try to reposition the iron tip to enlarge the contact area14:36
- nsc (QUIT: Ping timeout: 265 seconds) (~nicolas@i5C74DEDC.versanet.de)14:48
minutei'm now reading a bit of HDMI driver sources of rk3588 to see if we could patch in hdmi1 support more cleanly in the latest collabora series14:56
grimmwareHmm, would doing the FPC first with the hot plate and then sticking it down with kapton tape whilst I do the other side be sensible?14:56
minutegrimmware: the fpc can be easily hand soldered with soldering iron14:57
minutegrimmware: it has quite long exposed pins/tabs and little holes that take up solder14:57
grimmwareOh okay14:58
minutegrimmware: you can see that on the one that's in the reform14:58
minute(we don't fold it behind the display)14:58
grimmwareYeah that’s the reference I was going from.14:59
- ptrc (QUIT: Remote host closed the connection) (~ptrc@ptrc.gay)15:04
+ ptrc (~ptrc@ptrc.gay)15:05
- jaume` (QUIT: Remote host closed the connection) (~user@85.51.226.170)15:09
- Ar|stote|is (QUIT: Ping timeout: 276 seconds) (~linx@149.210.1.237)15:26
+ Ar|stote|is (~linx@149.210.0.75)15:30
* jackhill -> KM4MBG15:35
minutehere's my plan for adding hdmi1 support back to the collabora patch stack for rk3588 https://source.mnt.re/-/snippets/1215:41
+ stephano (~stephano@71.238.14.13)15:47
+ mark_ (~mjw@gnu.wildebeest.org)16:30
jfredkensanata: indeed, that seems possible :)16:34
minuteinteresting error > [drm] HDMI Sink doesn't support RGB, something's wrong.16:51
minutebut hdmi1 works! so the required patch is pretty smol16:54
- stephano (QUIT: Quit: ZZZzzz…) (~stephano@71.238.14.13)16:59
- mark_ (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org)17:02
minuteaha, fixed by fixing EDID which was fixed by fixing HPD for hdmi117:19
- qbit (QUIT: Remote host closed the connection) (~qbit@mail.suah.dev)18:08
+ qbit (~qbit@mail.suah.dev)18:10
+ nsc (~nicolas@i5C74DEDC.versanet.de)18:33
ryukazouI found a shop that welling to re-andoizing for me, but I kinda worry the de-anodizing process before re-andoizing would hurt the screw hole. They seem to use alkali to de-andoizing.18:48
grimmwareryukazou: what are you doing this for, a different colour? :O18:52
ryukazougrimmware: Yeah, I plan to make my pocket reform this colour https://i.imgur.com/ny8zE61.jpeg18:54
minuteryukazou: yeah i'm not sure about the effect on screw holes... maybe they could be plugged or sth? maybe they have a solution for it?18:55
+ mark_ (~mjw@gnu.wildebeest.org)18:56
ryukazouPurple is kinda grow on me too, so I’m not really quite sure. If the de-andoizing treatment will hurt the shell, I would probably order purple one and give the original to my dad.18:57
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50)18:58
- Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66)19:02
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)19:03
Boostisbetterdoes anyone have a suggestion on where you can get 18650 cells in Germany? I am looking for normal LIPO ones as I finally found a battery pack that uses 18650s that you can replace. 19:04
hramrachany random elecrtonic part store https://www.tme.eu/de/en/katalog/rechargeable-batteries_100076/?queryPhrase=18650https://www.tme.eu/de/en/katalog/rechargeable-batteries_100076/?queryPhrase=1865019:10
hramrachhttps://www.conrad.de/de/search.html?search=1865019:11
* mjw -> Guest189819:15
- Guest1898 (QUIT: Killed (tungsten.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae)19:15
* mark_ -> mjw19:15
+ Guest1898 (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae)19:16
- cow321 (QUIT: Ping timeout: 246 seconds) (~deflated8@user/meow/deflated8837)19:40
- mjw (QUIT: Ping timeout: 246 seconds) (~mjw@gnu.wildebeest.org)19:41
+ chomwitt (~chomwitt@ppp-94-67-130-214.home.otenet.gr)19:44
+ mjw (~mjw@gnu.wildebeest.org)19:46
joschminute: just saw your mail to Chris from collabora -- thank you for working on this! I cannot wait holding my rk3588 reform in my hands. :)19:48
- chomwitt (QUIT: Ping timeout: 248 seconds) (~chomwitt@ppp-94-67-130-214.home.otenet.gr)19:55
+ Gooberpatrol_66 (~Gooberpat@user/gooberpatrol66)19:59
- Gooberpatrol66 (QUIT: Ping timeout: 260 seconds) (~Gooberpat@user/gooberpatrol66)20:00
+ gustav28 (~gustav@c-5834524e.019-141-67626730.bbcust.telenor.se)20:02
minutejosch: yeah i'm working on the frequencies now... would be nice to be able to just plug in any monitor and go20:16
- jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@47-186-105-237.dlls.tx.frontiernet.net)21:01
+ jacobk (~quassel@47-186-105-237.dlls.tx.frontiernet.net)21:02
minutenice, 1440p @ 120Hz works21:03
chyay21:09
+ jaume (~user@user/jaume)21:10
+ jaume` (~user@139.47.115.207)21:11
minutelets see if this goes through CI... https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/58/diffs?commit_id=ab344169899a7e627ab9ea837a65a15ad34b693921:40
minutethis will also be interesting to get even more bandwidth, but something for later https://gitlab.collabora.com/hardware-enablement/rockchip-3588/linux/-/commit/b90e5bd5beb974de975aea562ca6dce1ea96d5f921:45
+ cow321 (~deflated8@user/meow/deflated8837)21:47
minutepipeline failed :D21:54
minutejosch: strange, there are panthor patches that fail now, were they backported?21:55
- jaume (QUIT: Remote host closed the connection) (~user@user/jaume)21:57
- jaume` (QUIT: Remote host closed the connection) (~user@139.47.115.207)21:57
+ chomwitt (~chomwitt@2a02:587:7a07:9500:1ac0:4dff:fedb:a3f1)21:58
minuteanyway, removing 'em22:02
+ skf (sid272774@id-272774.hampstead.irccloud.com)22:07
skfo/22:11
- gustav28 (QUIT: Quit: Quit) (~gustav@c-5834524e.019-141-67626730.bbcust.telenor.se)22:15
minutewat > The next patch would create the file drivers/gpu/drm/rockchip/dw-mipi-dsi2-rockchip.c, which already exists!  Applying it anyway.22:19
minutebut how22:19
minuteah it's because of merging in main and some clashing patch file names22:20
- jacobk (QUIT: Ping timeout: 276 seconds) (~quassel@47-186-105-237.dlls.tx.frontiernet.net)22:36
joschminute: as far as i see in MR 58 you rebase on branch "main" and my backporting is in branch linux6.11 unless you mean something else?22:42
- murphnj (QUIT: Ping timeout: 260 seconds) (~murph@user/murphnj)22:42
joschoh you are on 6.10 but still had 0068-drm-pantor-...?22:45
joschminute: that got merged in 6.10.10 as a6b268fce7bf522:46
joschminute: see https://source.mnt.re/reform/reform-debian-packages/-/commit/0568bf58940076bf4cf44be11de1b36e9800240a22:46
minutejosch: aha, thx22:57
minutefunny > can't find file to patch22:59
minuteanyway, am in feierabend now, tomorrow more22:59
joschminute: i think you want to apply that patch against drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c and not against drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c23:11
joschminute: what's the base of that patch?23:11
minutejosch: no, i need to first import the patches from collabora that create _qp-rockchip.c23:21
minutei mistakenly assumed they were already in there23:21
joschah kk23:22
joschyeah, i see now that all the functions you patch are missing XD23:22
minutethis stuff and related https://gitlab.collabora.com/hardware-enablement/rockchip-3588/linux/-/commit/a320500222e2ea8707d3735ac77246dcad22f6d923:23
joschwe should also find a nicer way for you to manage these patch stacks23:23
joschdo you have some ideas what would make it easier for you?23:24
joschminute: that commit you linked is from the rk3588 branch which is the branch that collabora already based on linux 6.1123:26
joschminute: but you base your work on 6.1023:26
minutejosch: yeah it doesn't matter in this case, that driver should also work on 6.10 and then we'll have an easier time bringing it to 6.1123:27
- cobra (QUIT: Ping timeout: 265 seconds) (~cobra@user/Cobra)23:32
+ cobra (~cobra@user/Cobra)23:37
- chomwitt (QUIT: Ping timeout: 272 seconds) (~chomwitt@2a02:587:7a07:9500:1ac0:4dff:fedb:a3f1)23:40
joschminute: maybe we should switch to maintaining the patch stack in multiple git branches of the stable linux git -- would that make it easier for you?23:53

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!