| + sl (~sl@contrib.inri.net) | 00:06 | |
| sl | hello | 00:07 |
|---|---|---|
| + reading (~reading@c-174-164-41-109.hsd1.wa.comcast.net) | 00:26 | |
| reading | hey, so (flash was a success) and i wanted to come back just to say how impressed i am with this whole project | 00:29 |
| reading | i backed it when it first came out, just because it was exactly what i had been wanting/wishing the direction of things to go. i hadn't really intended it to be more than just playing around with and learning more of the linux underbelly | 00:31 |
| reading | life happened, and i didnt get a chance to use it much or keep track of the past few years. and so only a month ago or so i finally got some time to pick it back up again | 00:32 |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 00:34 | |
| reading | i was genuinely shocked when i saw that it had been linux mainlined to the point where i could just apt upgrade. seriously amazed | 00:35 |
| reading | i mean instead of atrophying like i expected it to have, im now set up for use as my backbone daily machine. just incredible, really | 00:36 |
| reading | i know the amount of work for that must have been crazy, so major props and a thank you to whoever deserves it. | 00:37 |
| reading | josch: seems like from the git history that goes to you especially heh | 00:39 |
| reading | ive only ever developed on linux, not for it. and so going from zero to successful build, and especially firmware upgrading which can be dicey, that fast with almost no hitches is just wild. | 00:40 |
| reading | so anyways, long story finished. just wanted to say major thanks. | 00:43 |
| minute | reading: thank you for the feedback, great to read it! | 00:49 |
| reading | minute: absolutely, thank you for the super fast help. heh and i just saw the log. you even pipelined an official build artifact already? awesome =p | 01:06 |
| - reading (QUIT: Ping timeout: 245 seconds) (~reading@c-174-164-41-109.hsd1.wa.comcast.net) | 01:17 | |
| sl | minute: is the new 3.0 keyboard layout going to be standard on the next batch of machine? i ordered another one, and i'd like to get the 3.0 keyboard (with kailh whites, like are available for the standalone keyboard) from the start if possible. | 01:51 |
| minute | sl: yes, it is | 01:59 |
| sl | excellent | 02:00 |
| - mtm- (QUIT: Ping timeout: 268 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:03 | |
| - kensanata (QUIT: Quit: Ping timeout (120 seconds)) (~alex@user/kensanata) | 02:04 | |
| - mtm (QUIT: Ping timeout: 240 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:04 | |
| + kensanata (~alex@user/kensanata) | 02:04 | |
| - sl (QUIT: Quit: Leaving...) (~sl@contrib.inri.net) | 02:13 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:10 | |
| + mtm- (~mtm@71.228.84.213) | 04:12 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 04:57 | |
| - S0rin (QUIT: Ping timeout: 250 seconds) (~S0rin@user/s0rin) | 05:35 | |
| + S0rin (~S0rin@user/s0rin) | 05:35 | |
| josch | minute: in reform2-lpc-fw/src/boards/reform2/board_reform2.c the default REFORM_MOTHERBOARD_REV is REFORM_MBREV_R3 -- why is FW_STRING2 set to R4 and why is it not depending on REFORM_MOTHERBOARD_REV? | 06:22 |
| josch | should it not? | 06:22 |
| josch | same with KBD_FW_REV which reads "R1 20230315" -- is this not incorrect when building with KBD_VARIANT_3? | 06:40 |
| + bgs (~bgs@212-85-160-171.dynamic.telemach.net) | 07:24 | |
| josch | minute: i chose this naming of the keyboard firmware hex files -- do you have another preference? https://source.mnt.re/josch/reform/-/jobs/1252/artifacts/browse/reform2-keyboard-fw/ | 07:30 |
| josch | if we build the firmware on CI, it should be offered for all hardware versions in all variations | 07:30 |
| - jjbliss (QUIT: Ping timeout: 240 seconds) (~jjbliss@1464766-static.elnsmiaa.metronetinc.net) | 07:42 | |
| - bgs (QUIT: Remote host closed the connection) (~bgs@212-85-160-171.dynamic.telemach.net) | 07:56 | |
| + jjbliss (~jjbliss@1464766-static.elnsmiaa.metronetinc.net) | 08:28 | |
| - S0rin (QUIT: Ping timeout: 240 seconds) (~S0rin@user/s0rin) | 08:32 | |
| + S0rin (~S0rin@user/s0rin) | 08:38 | |
| vkoskiv | Are the rpi/atmel firmwares for the keyboard and trackball two distinct firmwares or a single one with shared code? | 08:52 |
| vkoskiv | I understand the libraries are distinct for both platforms | 08:52 |
| vkoskiv | Would be ideal to merge them as much as possible, having shared code for all the MNT-specific stuff and then just two targets for the different platforms | 08:53 |
| - mlarkin (QUIT: Ping timeout: 240 seconds) (~mlarkin@047-036-074-225.res.spectrum.com) | 09:12 | |
| + mlarkin (~mlarkin@047-036-074-225.res.spectrum.com) | 09:14 | |
| + mjw (~mjw@gnu.wildebeest.org) | 10:26 | |
| josch | vkoskiv: can you propose a MR with that change? | 10:27 |
| josch | sharing code would also make Debian packaging easier | 10:27 |
| vkoskiv | Would be a fun project to merge the two and figure out the abstraction! | 10:27 |
| vkoskiv | I will investigate that. | 10:28 |
| + chomwitt (~chomwitt@2a02:587:7a17:7500:1ac0:4dff:fedb:a3f1) | 10:54 | |
| + eery (~eery@5.29.16.232) | 11:01 | |
| - S0rin (QUIT: Ping timeout: 250 seconds) (~S0rin@user/s0rin) | 11:20 | |
| + S0rin (~S0rin@user/s0rin) | 11:22 | |
| - chomwitt (QUIT: Remote host closed the connection) (~chomwitt@2a02:587:7a17:7500:1ac0:4dff:fedb:a3f1) | 11:24 | |
| minute | vkoskiv: there is no more reform rp2040 keyboard because the atmel became available again. for the trackball we will stick with rp2040. | 12:38 |
| josch | i have the subjective feeling that the mouse speed of the trackballs is different between v1 and v2 with the latter being much slower -- can this be true? | 12:42 |
| + chomwitt (~chomwitt@2a02:587:7a17:7500:1ac0:4dff:fedb:a3f1) | 12:53 | |
| - mjw (QUIT: Ping timeout: 248 seconds) (~mjw@gnu.wildebeest.org) | 12:58 | |
| vkoskiv | minute: Did you ever ship keyboards with rp2040? | 13:55 |
| minute | vkoskiv: no, but we have it in the pocket reform | 14:03 |
| minute | josch: yes, the older versions had acceleration built in, but i figured it would be better to leave this to the OS | 14:04 |
| - mtm- (QUIT: Ping timeout: 250 seconds) (~mtm@71.228.84.213) | 14:04 | |
| - mtm (QUIT: Ping timeout: 268 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:04 | |
| josch | minute: thanks! do you also have some quick feedback for me concerning REFORM_MOTHERBOARD_REV and KBD_FW_REV from above? | 14:06 |
| minute | josch: regarding FW_STRING2, that's a bug indeed | 14:08 |
| vkoskiv | Would still be ideal to share as much code as possible. | 14:09 |
| minute | josch: ideally during the build process we would build several revisions | 14:09 |
| vkoskiv | So the keyboard-fw would have a makefile with targets for rpi2040 and atmel | 14:10 |
| minute | josch: i mean targeting several motherboard revisions, and for keyboard, US vs non-US | 14:10 |
| josch | minute: i've already set this up for the keyboard (see the link above) | 14:10 |
| minute | vkoskiv: there is no rp2040 keyboard-fw. | 14:10 |
| vkoskiv | Except on the pocket, no? Or did I misread your message | 14:11 |
| minute | vkoskiv: but it's very different and still in development. right now merging those firmwares would not help me. | 14:11 |
| minute | maybe at a later point when things are more stable | 14:11 |
| josch | minute: okay, thank you. I'll file a merge request once i'm done then you can see whether how i interpreted things is how it should be. :) | 14:11 |
| vkoskiv | Yeah, makes sense. | 14:11 |
| minute | josch: the naming is cool, but there are actually 2 variables, so more permutations. internal/external X language/layout | 14:12 |
| minute | josch: i mean standalone/laptop X layout | 14:13 |
| josch | ah whoops, indeed | 14:23 |
| josch | it is KBD_VARIANT_STANDALONE X {KBD_VARIANT_QWERTY_US, KBD_VARIANT_NEO2, KBD_VARIANT_V} | 14:24 |
| josch | KBD_VARIANT_3 doesn't need KBD_VARIANT_STANDALONE because it has this switch that allows it to be both, right? | 14:24 |
| josch | what confused me was that the variables were all named KBD_VARIANT_* so i assumed they were the same type | 14:27 |
| josch | ah no, KBD_VARIANT_3 can also be KBD_VARIANT_STANDALONE | 14:27 |
| josch | (at least according to the code) | 14:27 |
| josch | does that match reality? | 14:27 |
| josch | minute: can i rename KBD_VARIANT_STANDALONE to KBD_MODE_STANDALONE to indicate it's a different type of setting? | 14:28 |
| josch | what is actually the "V" variant? | 14:30 |
| - ysionneau (QUIT: Ping timeout: 256 seconds) (~fallen@tools.sionneau.net) | 15:14 | |
| + mjw (~mjw@gnu.wildebeest.org) | 15:33 | |
| minute | josch: V variant was an experiment by a friend, not necessary to build | 15:36 |
| minute | josch: yes, renaming to MODE sounds good | 15:37 |
| minute | josch: and yes, VARIANT 3 is the new layout (keyboard v3), and it can be standalone as well | 15:37 |
| minute | josch: the keyboard fw can't recognize the switch. so the standalone variant just influences the menu, i.e. what options are shown there, for example battery status doesn't make sense on the standalone | 15:38 |
| josch | okay, thank you! | 15:40 |
| josch | this looks better https://source.mnt.re/josch/reform/-/jobs/1254/artifacts/browse/reform2-keyboard-fw/ | 15:45 |
| - XYZ (QUIT: Ping timeout: 240 seconds) (~XYZ@89-24-41-33.nat.epc.tmcz.cz) | 16:02 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:09 | |
| mtm | minute: just ordered a caseless keyboard 3 w/whites. Is ETA for | 16:10 |
| mtm | shipping still around 3 months? | 16:10 |
| + mtm- (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:13 | |
| + ysionneau (~fallen@tools.sionneau.net) | 16:23 | |
| + XYZ (~XYZ@89.24.41.33) | 16:24 | |
| minute | mtm: yeah, hopefully less | 17:29 |
| - chomwitt (QUIT: Remote host closed the connection) (~chomwitt@2a02:587:7a17:7500:1ac0:4dff:fedb:a3f1) | 18:07 | |
| josch | minute: maybe the NEO2 variant can be killed as well and users can instead use an evdev entry in their /etc/udev/hwdb.d for the same effect? | 18:22 |
| josch | minute: did somebody else other than me and erlehmann order the NEO2 layout? :) | 18:22 |
| minute | josch: i can check! | 18:23 |
| minute | josch: there was one more person, but that order was canceled | 18:25 |
| josch | (i'm not even using KBD_VARIANT_NEO2 myself because even though i am a NEO2 user, i need to add more changes to have one spacebar be "return" and some other changes) | 18:25 |
| minute | josch: ohh ok | 18:25 |
| minute | josch: then we don't need a build for that | 18:25 |
| josch | sweet -- i can contact erlehmann and ask then :) | 18:25 |
| minute | josch: cool! | 18:26 |
| josch | minute: but your attention to a niche layout like NEO2 totally was one of the hooks that got me when i decided for the reform :D | 18:29 |
| minute | josch: was totally worth it! | 18:29 |
| josch | XD | 18:29 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 18:44 | |
| josch | the correct FW_STRING2 for motherboard 2.5 is "25_R1" and not "R4"? | 18:47 |
| + chomwitt (~chomwitt@2a02:587:7a17:7500:1ac0:4dff:fedb:a3f1) | 19:06 | |
| minute | josch: yep | 19:16 |
| minute | josch: wait | 19:16 |
| minute | josch: it's actually 25_R2, R1 was not released | 19:17 |
| josch | okay, then i'll change that as well | 19:34 |
| josch | minute: so this is just a naming issue and i can just do s/REFORM_MBREV_25_R1/REFORM_MBREV_25_R2/ in the codebase? | 19:35 |
| josch | or is an additional define called REFORM_MBREV_25_R2 missing? | 19:35 |
| minute | josch: just a naming issue | 19:47 |
| josch | okay, replacing | 19:47 |
| vagrantc | to build the various keyboard firmware variants ... requires tweaking the source code definitions? | 19:53 |
| josch | vagrantc: not anymore soon | 19:54 |
| vagrantc | oh good | 19:54 |
| vagrantc | ACTION is just doing a quick refresh of the packaging but sounds like worth waiting a bit :) | 19:54 |
| vagrantc | or ... just going ahead and pushing what i have so far :) | 19:55 |
| josch | vagrantc: soon you can build like this https://source.mnt.re/josch/reform/-/blob/versioning/.gitlab-ci.yml?ref_type=heads#L23 | 19:56 |
| josch | vagrantc: if i were you i would wait until that gets merged and a tag has been put | 19:56 |
| vagrantc | josch: that looks much nicer, yes. :) | 19:57 |
| vagrantc | although what i did was just update to the latest git ... i haven't even tried building variants on different things | 19:58 |
| vagrantc | ACTION is really eyeing up the more conventional keyboard layout board... | 19:58 |
| minute | vagrantc: it's pretty nice! | 19:59 |
| josch | but the new layout is completely replacing the old, right? | 20:01 |
| vagrantc | ACTION struggles to figure out where to put pgup and pgdn key | 20:01 |
| JC[m] | * jc struggles in general | 20:02 |
| josch | vagrantc: i'd just replace AGR and ... with pgup and pgdn | 20:07 |
| vagrantc | if i didn't use AGR, that would make sense | 20:09 |
| vagrantc | might just use the middle and left spacebars for pgup and pgdn | 20:09 |
| vagrantc | the default firmware doesn't have a way to implement a FN-key or something? | 20:10 |
| vagrantc | ACTION wonders if shift+up/down arrow would work | 20:11 |
| minute | josch: yes, i don't think we wanna produce the old layout again | 20:23 |
| minute | vagrantc: there's hyper up/down planned for pg up/dn, and hyper left/right for home/end | 20:23 |
| minute | vagrantc: hyper is that diagonal arrow symbol now (this and "AGR" are harmonized with pocket reform legends) | 20:24 |
| vagrantc | minute: that sounds quite workable! | 20:24 |
| minute | ("AGR" is alt gr if that wasn't clear) | 20:24 |
| vagrantc | i assumed | 20:24 |
| vagrantc | hyper didn't do anything out of the box on the original layout ... if i recall? at least i never figured out what it did | 20:25 |
| josch | minute: oh sad, i really enjoyed the four-way split spacebar instead of the 3-way split with version 3.0. I also enjoyed the symmetry that only two types of keycaps give the layout but i am aware that i am the minority. :) | 20:27 |
| josch | minute: how is this old keyboard versioned? The new one is 3.0 -- which ones are 1.0 and 2.0? | 20:27 |
| minute | josch: the keyboard shipped in mnt reform and standalone so far is 2.0 | 20:38 |
| minute | i.e. 1.0 shipped only in Reform 1.0 of which only 13 units exist | 21:03 |
| vkoskiv | Do the serial numbers start from those units? i.e. #13 would be the first Reform 2? | 21:11 |
| vkoskiv | Mine is #182! | 21:12 |
| minute | vkoskiv: #1 is the first. | 21:16 |
| vkoskiv | Reform 2? | 21:16 |
| - Gooberpatrol66 (QUIT: Ping timeout: 240 seconds) (~Gooberpat@user/gooberpatrol66) | 22:07 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 22:10 | |
| - Gooberpatrol66 (QUIT: Ping timeout: 250 seconds) (~Gooberpat@user/gooberpatrol66) | 22:15 | |
| Boostisbetter | man Nheko takes forever to open up initially on the Reform. Once it is up it works great. | 22:18 |
| + [tj] (~tj]@188.166.156.159) | 22:21 | |
| minute | http://dyn.mnt.re:8000/cgi-bin/hello.sh | 22:26 |
| minute | i guess it will need a fan, or conduct heat to the chassis | 22:27 |
| josch | Hello from MNT Rack Reform. | 22:31 |
| minute | (LS1028A with a very small kepler nvidia gpu) | 22:31 |
| josch | MiB Mem : 7872.7 total | 22:32 |
| vkoskiv | minute: lspci printout would make sense here too? | 22:32 |
| josch | ah because the pre-massproduction units came with 8 gig and not the final 16 gig | 22:32 |
| minute | vkoskiv: ok, let me do that | 22:33 |
| minute | josch: yep | 22:33 |
| minute | i love that this is a cgi script tbh | 22:33 |
| minute | it's like 1997 | 22:33 |
| minute | just a lot faster | 22:33 |
| josch | it's awesome :) | 22:33 |
| vkoskiv | usually the approaches of old are the most efficient way still | 22:34 |
| josch | currently, this is with the reform heatsink? | 22:34 |
| minute | josch: it's with the "generic heatsink" but the problem is more the closed case | 22:34 |
| josch | vkoskiv: usually the approaches of new are the approaches of old but "now with internet" | 22:34 |
| josch | recently this became "now with ai" | 22:35 |
| vkoskiv | with more abstraction, for abstraction's sake. | 22:35 |
| josch | minute: the reform also has a closed case though, no? | 22:35 |
| minute | josch: well, that is true. i guess it's different because the heatsink is connected physically to the case, that really helps | 22:36 |
| vkoskiv | NVIDIA card shown! | 22:36 |
| josch | blender | 22:37 |
| josch | my arch nemesis | 22:37 |
| josch | finally whoever wants me to rebase these damn patches again, can just get the LS1028A and a kepler gpu | 22:38 |
| minute | haha yes | 22:41 |
| - chomwitt (QUIT: Remote host closed the connection) (~chomwitt@2a02:587:7a17:7500:1ac0:4dff:fedb:a3f1) | 22:53 | |
| josch | if anybody wants to speed this up, feel free to try out the firmwares built by this merge request https://source.mnt.re/reform/reform/-/merge_requests/42 | 22:57 |
| josch | otherwise i do that myself tomorrow | 22:57 |
| josch | minute: if we are not building V and NEO2 variants, I think it makes sense to name the keyboard firmwares like this: https://source.mnt.re/josch/reform/-/jobs/1255/artifacts/browse/reform2-keyboard-fw/ | 22:58 |
| josch | with those changes it should now be easy to do the versioning correctly once a git tag is put | 22:59 |
| - S0rin (QUIT: Ping timeout: 240 seconds) (~S0rin@user/s0rin) | 23:24 | |
| + S0rin (~S0rin@user/s0rin) | 23:27 | |
| minute | josch: ah, well there is one more complication, i'm so sorry. US vs non-US | 23:41 |
| minute | i regret that i am spoonfeeding this information | 23:41 |
| minute | 2 and 3 is good though | 23:42 |
| minute | just to explain: US means there is DEL on the key left to Z. | 23:42 |
| josch | oh no worries, i don't know how this could've worked differently than it did :) | 23:42 |
| minute | on other layouts, there's another scancode there, i think for example on qwertz it's the < > key | 23:43 |
| josch | aha, so it's KBD_VARIANT_QWERTY_US versus it nothing being set | 23:43 |
| minute | yes | 23:44 |
| minute | so then there is 6 firmwares in total | 23:44 |
| minute | 8 | 23:45 |
| josch | i'd name those keyboard-2.hex and keyboard-2-US.hex then I guess | 23:45 |
| josch | right, because keyboard 3.0 also has US and non-US | 23:46 |
| minute | yes | 23:51 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!