*** tpb has joined #timvideos | 00:00 | |
*** futarisIRCcloud has joined #timvideos | 00:08 | |
micolous[m] | Tumbleweed: email me an Android bug report ZIP | 00:25 |
---|---|---|
*** Thaelim has joined #timvideos | 02:21 | |
*** Thaelim has quit IRC | 03:38 | |
*** rohitksingh_work has joined #timvideos | 03:50 | |
*** samsagaz has quit IRC | 04:41 | |
*** samsagaz has joined #timvideos | 04:41 | |
*** nick00000 has joined #timvideos | 04:55 | |
*** futarisIRCcloud has quit IRC | 06:48 | |
*** futarisIRCcloud has joined #timvideos | 07:29 | |
tumbleweed | micolous[m]: I did the feedback email thing. does that include the zip or must I generate that with the developer menu? | 07:46 |
futarisIRCcloud | _florent: I made those changes, added the pins and "export CPU=or1k PLATFORM=arty TARGET=video" / "make gateware" fails with litex.build.generic_platform.ConstraintError: Resource not found: oled:None | 08:06 |
futarisIRCcloud | https://www.youtube.com/watch?v=_LnZrXrdC00 | 08:43 |
futarisIRCcloud | cnlohr was using a FX3 as a logic analyser... | 08:44 |
futarisIRCcloud | If I get a cheap FX2 devboard, I should be able to do most of what Opsis does (minus HDMI capture) on an Arty board, right? | 08:44 |
futarisIRCcloud | _florent_: I made those changes, added the pins and "export CPU=or1k PLATFORM=arty TARGET=video" / "make gateware" fails with litex.build.generic_platform.ConstraintError: Resource not found: oled:None | 08:45 |
*** sb0 has quit IRC | 09:15 | |
*** sb0 has joined #timvideos | 10:04 | |
_florent_ | futarisIRCcloud: just answered you mail, it should be fixed now | 10:12 |
futarisIRCcloud | _florent_: thanks | 10:34 |
micolous[m] | tumbleweed: I found the report. Another fscking CEPAS error... i wrote a patch that I think can solve it, the code for handling CEPAS isn't great. | 11:58 |
micolous[m] | Though I think it's actually calypso protocol, just that's xobs code for a Singapore card ;) | 12:00 |
tumbleweed | :) | 12:01 |
xobs | CEPAS is the Contactless E-Purse... application system? I forget. But it's the protocol name of the Singapore metro card. | 12:01 |
tumbleweed | well, if you want dumps and balances - I can probably get several from other FOSDEM attendees | 12:01 |
micolous[m] | Heh | 12:02 |
tumbleweed | it seems to have the concpt of tickets rather than balances | 12:02 |
micolous[m] | So for some reason it says it's CEPAS, because it does Android.nfc.tech.nfcB | 12:02 |
tumbleweed | https://www.dropbox.com/s/0e0a4es5khdp2vu/2018-02-03%2010.08.50.jpg?dl=0 | 12:02 |
tpb | Title: Dropbox - 2018-02-03 10.08.50.jpg (at www.dropbox.com) | 12:02 |
micolous[m] | I suspect the application protocol it's trying to use is totally wrong | 12:03 |
micolous[m] | It also looks like it doesn't hump whole tags | 12:03 |
futarisIRCcloud | So that patch that I wrote, doesn't add vga as an output in the hdmi2usb firmware on arty. Anyone? | 12:04 |
xobs | TOC of the specification (which costs $50): http://www.siongboon.com/projects/2012-03-03_rfid/doc/Singapore%20Standard,%20specification%20for%20contactless%20e-purse%20application.pdf | 12:04 |
tumbleweed | yeahthe website is also o_O - apparently there's a reader you can use to add things from home: https://www.stib-mivb.be/mystib/?l=en#mybootik | 12:04 |
tpb | Title: www.stib-mivb.be (at www.stib-mivb.be) | 12:04 |
futarisIRCcloud | I'll look at it again tomorrow. Is there an easy way to hard code a buffer into the vga output stuff? | 12:04 |
micolous[m] | tumbleweed: can I order cards by post? | 12:05 |
tumbleweed | sure. They cost €5 each - you want some? | 12:06 |
tumbleweed | ther are also single journey NFC (I presume) tickets | 12:07 |
micolous[m] | 1 of the reloadable and one of the single tickets would be fine. | 12:07 |
tumbleweed | sure | 12:08 |
xfxf | he means 1 of each with subsequent dollar values on them | 12:08 |
xfxf | so that's like 50 tickets | 12:08 |
tumbleweed | :P | 12:09 |
micolous | lol | 12:09 |
micolous | The single use tickets in a lot of systems are pretty useless | 12:09 |
xfxf | micolous: if you tell me where the secret 'write free pass onto card' option is in metrodroid i won't tell anymore, promise | 12:09 |
micolous | I spent today rewriting a bunch of the sqlite3 code in it to use protobufs instead :| | 12:10 |
micolous | I saved a whole 500KiB in the APK! | 12:10 |
tumbleweed | now you just need to implement sql queries on top of that | 12:10 |
micolous | hah | 12:10 |
micolous | I only need 1 type of query in the app | 12:11 |
micolous | https://github.com/micolous/metrodroid/tree/proto-stationdb/extra/mdst | 12:11 |
tpb | Title: metrodroid/extra/mdst at proto-stationdb · micolous/metrodroid · GitHub (at github.com) | 12:11 |
micolous | And this is the problem code and why it was crashing: https://github.com/micolous/metrodroid/commit/e5a9a28c09cb24cce3ed05c6dc3383dab3dbb42c#diff-40970344909643a227533ead930e31e9R90 | 12:12 |
tpb | Title: cepas: Handle error state in CardHWDetailFragment better · micolous/metrodroid@e5a9a28 · GitHub (at github.com) | 12:12 |
micolous | NullPointerExceptions | 12:12 |
tumbleweed | yeah, that looks like it was quite fragile | 12:13 |
micolous | It looks like it might try to dump all the purses on the card | 12:13 |
micolous | But if your card doesn't talk the same protocol, it may have issues (which is likely also) | 12:14 |
tumbleweed | yeah | 12:14 |
micolous | I'll need to dig up some non-NDA specs for the Calypso card | 12:15 |
micolous | But looks like they did a similar thing to SG and implemented an application protocol ontop of some other stack | 12:16 |
micolous | So your transit card is more similar to a Mastercard than a dumb storage device | 12:16 |
micolous | *dumb-ish | 12:16 |
tumbleweed | cape town's system is like that too (actually a mastercard) | 12:17 |
tumbleweed | I'll poke at that when I get home | 12:17 |
micolous | Yeah, the ones that are _actual_ Mastercard may be problematic | 12:17 |
micolous | Because they don't really store transactions on the card | 12:17 |
tumbleweed | yeah, I assumed as much | 12:17 |
tumbleweed | there is stored value somewhere | 12:18 |
micolous | Also found it somewhat funny that News Limited ran with three stories about the different transit card replacement projects in AU about 3 days after my talk... | 12:23 |
tumbleweed | heh | 12:23 |
*** rohitksingh_work has quit IRC | 12:26 | |
micolous | http://demo.calypsostandard.net/ lol | 12:26 |
tpb | Title: Calypso ticketing in action (at demo.calypsostandard.net) | 12:26 |
tumbleweed | poor access.php | 12:27 |
tumbleweed | clearly someone enjoyed transport tycoon | 12:27 |
futarisIRCcloud | CAD and open hardware at #fosdem seems to have a lot of FPGA related talks | 12:28 |
micolous | http://demo.calypsostandard.net/common/js/ClientCardAccess.class.js search for "sendCommand" | 12:35 |
micolous | If you use MSIE on that site, it has an activex control to talk to a real reader | 12:35 |
micolous | And all the code is implemented in javascript | 12:35 |
tumbleweed | wat | 12:36 |
micolous | All the control commands are documented nicely | 12:36 |
tumbleweed | yeah | 12:36 |
micolous | :) | 12:36 |
*** sb0 has quit IRC | 13:03 | |
*** rohitksingh has joined #timvideos | 13:23 | |
*** sb0 has joined #timvideos | 13:53 | |
*** rohitksingh has quit IRC | 14:39 | |
*** rohitksingh-demo has joined #timvideos | 14:42 | |
*** rohitksingh-demo has quit IRC | 14:44 | |
*** futarisIRCcloud has quit IRC | 14:48 | |
mithro | CarlFK[m]: ping? | 17:45 |
mithro | cr1901_modern: ping? | 17:45 |
cr1901_modern | mithro: pong. | 18:13 |
cr1901_modern | I have previous engagements this weekend, won't really have time to work on anything. I'll try to check in tomorrow night. | 18:13 |
CarlFK[m] | mithro: pong | 18:13 |
mithro | cr1901_modern: I wanted to find out the status of things | 18:13 |
mithro | CarlFK[m]: How goes the GSoC tasks I gave you? | 18:13 |
mithro | cr1901_modern: IIRC You were going to try some prjxray work too - I assume that didn't go anywhere? | 18:14 |
cr1901_modern | mithro: Not yet. The issue is still open and I'm in contact w/ digshadow tho. | 18:14 |
CarlFK[m] | I looked over the ideas list and it looks good. what was next? | 18:15 |
cr1901_modern | mithro: Current status is: working on modifying _florent_'s firmware to handle old SD cards, and then I'll create a small wishbone bus peripheral for litex-buildenv to talk to the SD card. | 18:15 |
cr1901_modern | mithro: I've confirmed mimasv2 can actually read/write just fine | 18:15 |
mithro | cr1901_modern: So I also wanted to support very low access via bitbanging to the SD Card using SPI mode | 18:16 |
mithro | cr1901_modern: Does it make sense to just use that for old SD cards? | 18:17 |
*** xobs has quit IRC | 18:17 | |
mithro | CarlFK[m]: Make the hdmi2usb.tv news items look less dead | 18:17 |
mithro | CarlFK[m]: You have a photo of the post-it notes IIRC? | 18:17 |
cr1901_modern | mithro: We could include that as an option, but it doesn't really make my job easier. Having read the spec, there's not much reason to support SPI mode if we already have an SD protocol handler | 18:19 |
cr1901_modern | (and in addition using SPI wouldn't solve the issues I've been seeing wrt READ/WRITE_MULTIPLE) | 18:20 |
mithro | cr1901_modern: Okay -- the bitbanging SPI mode however is very efficient in terms of FPGA resources (since it basically uses none) right? | 18:20 |
cr1901_modern | Oh _bitbanging_ SPI | 18:20 |
cr1901_modern | I should read more closely | 18:20 |
cr1901_modern | Erm, yes that is also doable. But slow of course. | 18:21 |
mithro | cr1901_modern: With the MimasV2 being a SLX9 I'm unsure if it makes sense to try and use hardware accelerated SD card? | 18:23 |
mithro | cr1901_modern: Of course we probably do want to use the hw accel on the Arty and Opsis | 18:24 |
CarlFK[m] | mithro: i have the post-its - I should type them into something - any suggestions? i'm thinking a goog doc is fine. | 18:24 |
*** suhdood has joined #timvideos | 18:24 | |
mithro | CarlFK[m]: Don't care | 18:25 |
cr1901_modern | mithro: Putting the SD card on the FPGA SPI bus is prob a good compromise for mimasv2. But I want to get the SD protocol working anyway before working on that (write my wishbone peripheral once, deploy everywhere) | 18:27 |
mithro | okay | 18:36 |
CarlFK[m] | mithro: is the Arty fast enough to decode an hdmi in, discard the video data, look for errors and light an LED or two for pass/fail ? | 18:56 |
mithro | CarlFK[m]: That is a complicated question | 18:56 |
CarlFK[m] | mithro: is this simpler: can it generate a 720p color bar test pattern like the hdmi2usb does ? | 18:58 |
mithro | CarlFK[m]: That is probably yes | 18:59 |
mithro | CarlFK[m]: I need to finish my pmod hdmi board for that to work | 19:11 |
CarlFK[m] | mithro (IRC): could someone else "finish pmod hdmi board" as a gsoc project? | 19:14 |
CarlFK[m] | having a small device that can be used for diagnostics would be handy. and maybe also for unit testing the hdmi2usb | 19:15 |
mithro | CarlFK[m]: Not really | 19:15 |
CarlFK[m] | shucks. | 19:16 |
mithro | CarlFK[m]: It just needs someone to put together and test | 19:41 |
mithro | CarlFK[m]: It should be pretty easy to do but needs time to do it | 19:41 |
CarlFK[m] | mithro (IRC): oh cool. I'll try to find a PS1er to do that - I know 2 that might be interested | 19:42 |
CarlFK[m] | http://whiteboard.debian.net/tvgsoc2018.wb post-its and a few more | 21:07 |
tpb | Title: tvgsoc2018 - whiteboard (at whiteboard.debian.net) | 21:08 |
*** suhdood has quit IRC | 22:27 | |
*** suhdood has joined #timvideos | 22:29 | |
*** suhdood has quit IRC | 23:44 |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!