| + mntreform2 (~root@p5B015AD5.dip0.t-ipconnect.de) | 01:33 | |
| mntreform2 | online with reform 2 | 01:33 |
|---|---|---|
| Jookia | sick | 01:34 |
| mntreform2 | case is pretty hacked up but it's a laptop, somehow | 01:39 |
| adjtm | it looks very good | 01:44 |
| mntreform2 | hehe | 01:45 |
| mntreform2 | thx adjtm | 01:45 |
| mntreform2 | the keyboard feels really good | 01:45 |
| mntmn | woops killed it while weighing | 01:49 |
| mntmn | (but it boots fine still) | 01:49 |
| adjtm | it looks like and old ibm thinkpad | 01:50 |
| mntmn | weighs 1.36kg but without trackball installed | 01:50 |
| - mntreform2 (QUIT: Ping timeout: 276 seconds) (~root@p5B015AD5.dip0.t-ipconnect.de) | 01:50 | |
| mntmn | adjtm: hm! | 01:50 |
| + B[] (~Thunderbi@122-61-176-31-fibre.sparkbb.co.nz) | 01:52 | |
| Jookia | mntmn: i think the reform has a LOT of promise and would be very attractive to hackers | 01:57 |
| mntmn | Jookia: i hope so :) | 02:02 |
| - chomwitt (QUIT: Ping timeout: 265 seconds) (~chomwitt@ppp-94-69-11-12.home.otenet.gr) | 02:08 | |
| marcodiego | photos? | 02:29 |
| adjtm | marcodiego, https://mastodon.social/@mntmn | 02:33 |
| marcodiego | beautiful! | 02:36 |
| marcodiego | if you are accepting suggestions for the next version, I think it would be good if the next version uses linaro's 96boards SOM specification: https://www.linaro.org/news/linaro-announces-launch-of-96boards-system-on-module-som-specification/ . That would allow a very simple and (most importantly) standard way to upgrade it in the future | 02:39 |
| Jookia | marcodiego: there's no nvme on that connector | 02:42 |
| Jookia | in general though standardized connectors are a bad idea | 02:43 |
| Jookia | except for very basic things like power and peripherals i suppose, but even then | 02:44 |
| marcodiego | Jookia, isn't it possible to put the nvme connector on the SOM? | 02:45 |
| Jookia | dunno where, it's like a pcie slot | 02:45 |
| marcodiego | Jookia, is there a nvme connector on the current reform prototype? | 02:50 |
| Jookia | marcodiego: yes | 02:52 |
| marcodiego | Jookia, is the nvme connector at 5 o'clock from the center of the blue thing? | 02:54 |
| Jookia | https://f2.tchncs.de/social.tchncs.de/media_attachments/files/004/688/847/original/72be8fe324be3cfd.jpeg | 02:54 |
| Jookia | it's to the right of mpcie | 02:54 |
| Jookia | says NVME under it | 02:54 |
| marcodiego | interesting... the specification mentions nvme: http://static.linaro.org/assets/specifications/96BoardsComputeSoMSpecificationV1.0.pdf | 03:00 |
| Jookia | which page? | 03:01 |
| marcodiego | 34 | 03:01 |
| Jookia | yeah, that's just boot selection | 03:01 |
| Jookia | are there any boards actually using this spec | 03:02 |
| marcodiego | Jookia, yes | 03:04 |
| adjtm | mntmn doesn't design or build the SOM, so it would need to find a SOM with that format and i.mx8m | 03:05 |
| marcodiego | Jookia, http://linuxgizmos.com/rk3399-compute-module-and-carrier-follow-96boards-som-spec/ | 03:05 |
| adjtm | and anyway there are also other standard SOM formats, I don't see why they should use specifically linaro's format | 03:05 |
| Jookia | http://linuxgizmos.com/files/beiqi_rk3399pro_devkit.jpg fancy | 03:06 |
| marcodiego | I think linaro's specs have some "weight" on the embedded space. So, on can think the standard will spread | 03:06 |
| Jookia | well, what advantages do you get by having a standardized som connector | 03:08 |
| marcodiego | Jookia, standardized upgrade | 03:08 |
| adjtm | Jookia, you can change to a different SOM | 03:08 |
| Jookia | yeah but what's the actual advantage there for me | 03:09 |
| Jookia | i still have to change all my kernels and bootloaders and dev env | 03:09 |
| Jookia | maybe my distro too if its another arch | 03:09 |
| adjtm | marcodiego, I don't think that linaro's SOM format has weight in the industry | 03:09 |
| marcodiego | adjtm, does the one adopted by reform have? | 03:10 |
| adjtm | I don't know, I suppose it doesn't | 03:11 |
| Jookia | it also means the board i'm using has to be compatible with the som | 03:11 |
| adjtm | marcodiego, but name me other i.mx8m SOM... | 03:11 |
| Jookia | so it can't do anything cool with the specific chipset | 03:11 |
| adjtm | I also like the modules with edge connector instead of socket | 03:11 |
| Jookia | having a generic interface is cool but it means you can't easily take advantage of chip specific features :( | 03:12 |
| Jookia | on the novena the FPGA is wired directly to the EIM on the imx6 | 03:12 |
| Jookia | how long do SOM standards live for anyway? | 03:14 |
| Jookia | in the PC market usually you have to upgrade your motherboard and CPU at the same time | 03:14 |
| adjtm | yeah, for example Q7 have 8 usb 2 but not usb3... | 03:16 |
| - B[] (QUIT: Ping timeout: 276 seconds) (~Thunderbi@122-61-176-31-fibre.sparkbb.co.nz) | 03:16 | |
| adjtm | well, it seems that theobroma Q7 modules have usb3 | 03:17 |
| Jookia | but then again, the reform's SOM hits the same problems i think unless there's like a new upgrade for imx8m quad plus or something | 03:18 |
| Jookia | the idea of having a risc-v module would require addressing these issues too :\ | 03:19 |
| Jookia | i wonder what mntmn has to say on this topic | 03:21 |
| + kaleun (~kaleun@2001-48F8-9015-158C-0-0-0-3-dynamic.midco.net) | 03:41 | |
| - kaleun (QUIT: Remote host closed the connection) (~kaleun@2001-48F8-9015-158C-0-0-0-3-dynamic.midco.net) | 05:11 | |
| + erlehmann (~erlehmann@46.114.32.15) | 05:29 | |
| + kaleun (~kaleun@2001-48F8-9015-158C-0-0-0-3-dynamic.midco.net) | 06:14 | |
| + B[] (~Thunderbi@122-61-176-31-fibre.sparkbb.co.nz) | 08:59 | |
| + chomwitt (~chomwitt@ppp-94-69-11-132.home.otenet.gr) | 09:15 | |
| - marcodiego (QUIT: Quit: Leaving) (~pi@177.37.134.14) | 09:31 | |
| - erlehmann (QUIT: Remote host closed the connection) (~erlehmann@46.114.32.15) | 11:01 | |
| + erlehmann (~erlehmann@46.114.32.15) | 11:03 | |
| - erlehmann (QUIT: Remote host closed the connection) (~erlehmann@46.114.32.15) | 11:36 | |
| + erlehmann (~erlehmann@46.114.32.15) | 11:37 | |
| - erlehmann (QUIT: Remote host closed the connection) (~erlehmann@46.114.32.15) | 14:29 | |
| + erlehmann (~erlehmann@46.114.32.15) | 14:29 | |
| - erlehmann (QUIT: Remote host closed the connection) (~erlehmann@46.114.32.15) | 14:33 | |
| - B[] (QUIT: Remote host closed the connection) (~Thunderbi@122-61-176-31-fibre.sparkbb.co.nz) | 14:43 | |
| + B[] (~Thunderbi@122-61-176-31-fibre.sparkbb.co.nz) | 14:46 | |
| + erlehmann (~erlehmann@46.114.32.15) | 14:58 | |
| mntmn | sorry, i was asleep already | 15:11 |
| mntmn | i’m very happy with connectors like so-dimm compared to these board to board connectors in terms of soldering, reliability, ease of plugging in and out | 15:14 |
| mntmn | there are some standards like COM and MXM for this as well, but the modules are usually quite big | 15:15 |
| mntmn | also none of these modules are open so it doesn’t really help | 15:15 |
| mntmn | what i find more interesting is to make a kicad template for the so-dimm pinout and let people (or us) design SOMs for reform with their favorite chip | 15:16 |
| mntmn | for example i’d like to see xilinx zynq ultrascale because we could do open IP for displayport etc | 15:16 |
| mntmn | also the main upgrade path that i expect for reform will be the successors to i.MX8M | 15:17 |
| mntmn | i believe there will be some pin compatible ones, and boundary devices are committed to i.MX, so i think it’s a pretty safe bet we’ll see at least one upgrade | 15:18 |
| - erlehmann (QUIT: Remote host closed the connection) (~erlehmann@46.114.32.15) | 15:25 | |
| + erlehmann (~erlehmann@46.114.32.15) | 15:26 | |
| adjtm | mntmn, Q7 looks nice | 16:05 |
| adjtm | I have not seen any with i.mx8m tough | 16:06 |
| mntmn | adjtm: yep according to the wikipedia entry it doesn’t have usb 3.0 though? | 16:08 |
| adjtm | I thought that, but theobroma's rk3399 Q7 SOM has usb3 | 16:09 |
| adjtm | maybe they are reusing some of the 8 usb2 to route one usb3 | 16:09 |
| adjtm | and probably that makes standard carrier boards to loose some usb2 without being able to use the usb3 | 16:10 |
| adjtm | I agree with you about the so-dimm style (edge) connectors | 16:11 |
| mntmn | ok got it | 16:12 |
| adjtm | some other SOM use several connection that requiere one (or several) plugs with connectors on both sides | 16:12 |
| adjtm | in both sides (connectors in both the carrier board and the SOM) | 16:12 |
| adjtm | I guess that it will be in the future more SOM with M2 connectors | 16:13 |
| mntmn | hm that would be interesting | 17:10 |
| - B[] (QUIT: Ping timeout: 265 seconds) (~Thunderbi@122-61-176-31-fibre.sparkbb.co.nz) | 17:58 | |
| - erlehmann (QUIT: Remote host closed the connection) (~erlehmann@46.114.32.15) | 21:02 | |
| + erlehmann (~erlehmann@46.114.32.15) | 21:03 | |
| - wiedi_ (QUIT: Read error: Connection reset by peer) (~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de) | 21:48 | |
| + wiedi (~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de) | 21:49 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
