| 00:00.43 | lerc | Righty I'm gonna run http://glbenchmark.com/ on my aurora, see how it goes. |
| 00:03.26 | ZaEarl | http://limadriver.org/Hardware/ says 1 pixel per clock per core |
| 00:04.47 | popolon | on mli 200 |
| 00:04.53 | popolon | on Mali 200 |
| 00:05.15 | Gumboot | I'm not sure there's anybody who cites fill rates who actually takes them seriously enough to have a use case that can actually do them. |
| 00:05.39 | popolon | ok while the fill rate remains the same |
| 00:05.46 | ZaEarl | sure, fill rate is a sort of upper-bound. best case senario. |
| 00:05.57 | Gumboot | Well, not even. |
| 00:06.14 | Gumboot | Do all of these pixels have to land in memory? |
| 00:06.20 | Gumboot | Because you can cull at a nearly unlimited rate. |
| 00:06.31 | Gumboot | But some people count culled pixels. |
| 00:06.43 | Gumboot | And then make up an arbitrary "reasonable" cull rate. |
| 00:08.02 | Gumboot | Then they all stand around talking about how all their competitors are probably using much more optimistic values. |
| 00:09.22 | popolon | :) |
| 00:10.01 | Gumboot | Then there are the triangles whose area rounds to zero, but technically is not zero. |
| 00:11.41 | Gumboot | But to make that fly you have to run it on an infinitely fast CPU with infinite memory bandwidth. But since that's somebody else's problem we just assume they've solved that part and make projections accordingly. |
| 00:12.16 | Gumboot | (actually, I've never seen that done) |
| 00:12.30 | Turl | the CIA has 'em |
| 00:12.49 | Gumboot | Yeah, if they'd just dedicate more of their resource to GLES benchmarks.... |
| 00:12.59 | Gumboot | Maybe they do. |
| 00:14.44 | popolon | raspberry, as far I know, only demoed a teapot in opengl on their board |
| 00:15.07 | Gumboot | Did it have video on it? |
| 00:15.29 | popolon | yes |
| 00:15.56 | Gumboot | nostalgia. |
| 00:18.28 | popolon | http://jonmacey.blogspot.co.uk/2012/05/getting-started-with-egl-on-raspberry.html <= a torus |
| 00:19.27 | popolon | https://www.youtube.com/watch?v=A-vBbqamNBU <= Raspberry Pi, Qt 5, QML, Shaders |
| 00:19.53 | popolon | looks like some webGL demo |
| 00:20.06 | Gumboot | It's actually coloured a bit like a doughnut. |
| 00:20.14 | Gumboot | But there's no video! |
| 00:20.43 | popolon | ?? |
| 00:20.51 | popolon | which kind of video |
| 00:20.56 | Gumboot | Video like on a teapot. |
| 00:21.22 | popolon | ah, do you mean the version of teapot with video mapped on ? |
| 00:21.29 | Gumboot | I've disturbed myself a little, now. I can imagine a day in the future when you really can have video on your doughnuts.... and it'll be advertising. |
| 00:21.42 | popolon | :) |
| 00:21.46 | Gumboot | Yeah. |
| 00:22.01 | popolon | that was only a teapot with some texture |
| 00:22.06 | popolon | can't find it again |
| 00:22.09 | Gumboot | It's possible that only teapots can have video mapped onto them. Maybe it's impossible on other shapes. |
| 00:22.34 | popolon | ^^ |
| 00:23.00 | popolon | at they then can map 1080p on a rectangle |
| 00:24.41 | popolon | good night |
| 00:44.00 | penguin42 | Gumboot: Hmm video on donuts could be hard; stills on the top would be easy; perhaps something zoetrope like? |
| 00:52.22 | L84Supper | looks like the R-Pi has been having USB problems due to the poorly handled USB power |
| 00:53.39 | Dandel_ | L84Supper: not that entirely... it's more that the power sources cannot give the R-Pi enough power often times |
| 00:55.30 | L84Supper | I haven't looked at the schematics yet but somebody posted that it's the polyfuses |
| 01:00.23 | Dandel_ | L84Supper: The problem appears to be more often then not related to driver bugs. |
| 01:00.50 | L84Supper | http://himeshp.blogspot.com/2012/07/raspberry-pi-usb-power-issues-ultimate.html |
| 01:01.39 | L84Supper | http://www.scribd.com/doc/102665984/94/Extra-Adapted-registers |
| 01:01.55 | Dandel_ | i know of those mods. |
| 01:02.17 | L84Supper | are there further issues? |
| 01:02.30 | Dandel_ | Yes, the issue even happens on Powered hubs |
| 01:02.55 | Dandel_ | it's somewhat improving, but that's what I would suggest to most end users anyways... mainly to go get a powered usb hub. |
| 01:25.55 | *** join/#arm-netbook avernos (~avernos@unaffiliated/avernos) |
| 01:51.23 | *** join/#arm-netbook orly_owl (~david@unaffiliated/orly-owl/x-3167833) |
| 05:19.59 | *** join/#arm-netbook bsdfox (~Bob@c-71-197-83-152.hsd1.ca.comcast.net) |
| 05:19.59 | *** join/#arm-netbook bsdfox (~Bob@unaffiliated/bsdfox) |
| 05:53.14 | *** join/#arm-netbook gimli (~gimli@xbmc/staff/gimli) |
| 07:03.02 | *** join/#arm-netbook popolon (~popolon@2a01:e35:2e1e:a6c0:6c20:9701:8bdc:332c) |
| 07:08.16 | *** join/#arm-netbook Quarx (~Quarx@46.233.202.239) |
| 07:49.34 | *** join/#arm-netbook bsdfox (~Bob@c-71-197-83-152.hsd1.ca.comcast.net) |
| 07:49.34 | *** join/#arm-netbook bsdfox (~Bob@unaffiliated/bsdfox) |
| 07:51.16 | *** join/#arm-netbook hehopmajieh_ (~hehopmaji@78.130.183.204) |
| 08:47.45 | *** join/#arm-netbook madmalkav (~Hamlet@188.86.253.69) |
| 09:28.55 | *** join/#arm-netbook Quarx (~Quarx@46.233.202.239) |
| 09:55.20 | *** join/#arm-netbook bsdfox (~Bob@c-71-197-83-152.hsd1.ca.comcast.net) |
| 09:55.20 | *** join/#arm-netbook bsdfox (~Bob@unaffiliated/bsdfox) |
| 10:49.52 | orly_owl | hmm a200 for 82 aud |
| 11:10.30 | *** join/#arm-netbook penguin42 (~dg@tu006.demon.co.uk) |
| 11:24.01 | *** join/#arm-netbook gimli (~gimli@xbmc/staff/gimli) |
| 11:54.51 | *** join/#arm-netbook w00p (6d681809@gateway/web/freenode/ip.109.104.24.9) |
| 11:55.48 | w00p | Hi, do you know what resolution the A10 Linux image on http://rhombus-tech.net uses by default? |
| 12:00.39 | zenitraM | i think 720p |
| 12:05.33 | w00p | Is there a way to change the resolution? |
| 12:06.04 | w00p | To 1080p or a 4:3 resolution like 800x600 |
| 12:12.32 | lundman | edit the script.bin i think |
| 12:18.21 | w00p | Thanks, I'll try |
| 12:27.44 | *** join/#arm-netbook QingPei (~qingpei@123.118.101.185) |
| 12:28.47 | rz2k | video review of that archos 101xs tablet with qwerty dock. http://www.youtube.com/watch?v=SIM9l_KJszQ&feature=g-u-u |
| 12:31.50 | orly_owl | a10 soc? |
| 12:31.59 | rz2k | omap4470 |
| 12:53.05 | popolon | oh, charbas has a gnagnacat tshirt :D |
| 12:54.56 | popolon | the little motherboard look like odroid-X |
| 13:02.54 | popolon | arf |
| 13:02.56 | *** join/#arm-netbook QingPei (~qingpei@123.118.101.185) |
| 13:02.56 | popolon | bybye |
| 13:04.55 | orly_owl | x86 pc, but what is this connector called? http://i186.photobucket.com/albums/x267/bradmax57/fx160a.jpg |
| 13:22.30 | hno | orly_owl, looks identical to the UART connector in Mele. It is a standard type connector, but have not seen this used for SATA Power before. |
| 13:23.11 | orly_owl | yep, its a dell thin client/net top |
| 13:23.20 | orly_owl | dell optiplex fx160 |
| 13:25.47 | orly_owl | http://forums.overclockers.co.uk/showthread.php?p=14622633 |
| 13:27.41 | orly_owl | pic here http://www.itcsales.co.uk/acatalog/360s/large/l_H224H-itc.jpg |
| 13:30.30 | simonckenyon | orly_owl: that is a floppy drive power connector |
| 13:30.53 | orly_owl | pc has no room for floppy drive |
| 13:31.09 | orly_owl | brb |
| 13:33.04 | hno | It's not a floppy drive power connector. Those are 2.54mm raster width and much higher and closed above the hole in the side. |
| 13:33.29 | hno | A data sheet I found from JST is titled PH Connector. |
| 13:35.09 | hno | JST Crimp 2.0mm (0.79") pitch, PH Connector, Disconnectable Crimp style connector. |
| 13:37.32 | *** join/#arm-netbook orly_owl (~david@unaffiliated/orly-owl/x-3167833) |
| 13:37.43 | hno | <hno> It's not a floppy drive power connector. Those are 2.54mm raster width and much higher and closed above the hole in the side. |
| 13:37.43 | hno | <hno> A data sheet I found from JST is titled PH Connector. |
| 13:37.43 | hno | <hno> JST Crimp 2.0mm (0.79") pitch, PH Connector, Disconnectable Crimp style connector. |
| 13:38.18 | orly_owl | finally |
| 13:43.55 | hno | B4B-PH-K-S |
| 13:44.10 | orly_owl | ? |
| 13:44.40 | hno | part number of the JST connector. |
| 13:45.10 | orly_owl | ok |
| 13:56.18 | *** join/#arm-netbook gimli (~gimli@xbmc/staff/gimli) |
| 14:09.41 | *** join/#arm-netbook tuliom (~tuliom@186.214.89.120) |
| 14:21.50 | *** join/#arm-netbook orly_owl (~david@unaffiliated/orly-owl/x-3167833) |
| 14:36.02 | *** join/#arm-netbook tuliom (~tuliom@186.214.89.120) |
| 14:44.58 | *** join/#arm-netbook gimli (~gimli@xbmc/staff/gimli) |
| 14:51.21 | *** part/#arm-netbook QingPei (~qingpei@123.118.101.185) |
| 15:17.24 | *** join/#arm-netbook L84Supper (~ly@unaffiliated/l84supper) |
| 15:17.47 | *** join/#arm-netbook gimli (~gimli@xbmc/staff/gimli) |
| 15:25.59 | *** join/#arm-netbook bsdfox (~Bob@c-71-197-83-152.hsd1.ca.comcast.net) |
| 15:26.00 | *** join/#arm-netbook bsdfox (~Bob@unaffiliated/bsdfox) |
| 16:00.40 | *** join/#arm-netbook tuliom (~tuliom@186.214.89.120) |
| 16:02.21 | *** join/#arm-netbook L84Supper (~ly@unaffiliated/l84supper) |
| 16:21.49 | *** join/#arm-netbook popolon (~popolon@2a01:e35:2e1e:a6c0:d810:c98b:2769:92e2) |
| 17:17.39 | *** join/#arm-netbook j1nx (~j1nx@163-62-223.ftth.xms.internl.net) |
| 17:31.34 | popolon | Cortex A7 is more energy efficient than Cortex A15, and serveral SoC builder mix 1 Cortex A7 core with several Cortex A15 cores to better manage energy efficiency, the CCI-400 used for this big.Little technique is already supported by Linux :) |
| 17:46.13 | *** join/#arm-netbook RITRedbeard (~redbeard@c-68-37-165-37.hsd1.nj.comcast.net) |
| 18:11.51 | *** join/#arm-netbook amamarabane (pjp@84.6.88.109) |
| 18:14.43 | *** part/#arm-netbook amamarabane (pjp@84.6.88.109) |
| 18:17.23 | *** join/#arm-netbook j1nx (~j1nx@163-62-223.ftth.xms.internl.net) |
| 18:48.37 | xxiao | yawns |
| 18:49.33 | xxiao | wonders why married ones eager to show their pets instead of raising some real stuff, the kids |
| 18:50.37 | Marex | xxiao: fucking is fun |
| 18:50.46 | Marex | xxiao: you can always get rid of a pet fast ... not so much of a kit |
| 18:50.48 | Marex | *kid |
| 18:51.22 | xxiao | Marex: this is someone keep sending me their pig pet video |
| 18:51.27 | xxiao | s/this/there/ |
| 18:53.38 | Marex | xxiao: you can always filter emails ;-) |
| 19:06.14 | hno | only have fishes as pets, not much to show off. |
| 19:08.26 | specing | is the cortex-a7 64-bit too? |
| 19:09.00 | specing | Or are they still 32? |
| 19:09.57 | specing | Ah, still v7 |
| 19:13.40 | hno | A15 & A7 is v7a, |
| 19:15.44 | hno | don't know what the reoadmap for cores implementin v8 is. |
| 19:17.02 | popolon | I've some flies and mosquitos at home |
| 19:17.50 | popolon | all with 6 legs and 2 wings |
| 19:19.06 | hno | popolon, what CPU do they run? |
| 19:20.33 | popolon | small one with 1 core to fly and one core to detect blood or fructs depending on unit (fly or mosquito |
| 19:24.43 | *** join/#arm-netbook hehopmajieh_ (~hehopmaji@78.130.183.204) |
| 19:32.15 | Turl | popolon: where's the core to do damage control? |
| 19:41.09 | Marex | aaaaaaaa |
| 19:41.17 | Marex | specing: aarch64 |
| 19:41.29 | Marex | specing: that's the name of the 64bit arm ... it won't be arm anymore |
| 19:41.35 | Marex | it iwll be aarch64 |
| 19:41.46 | Marex | such a sexy name :) |
| 19:41.50 | xxiao | arm64 |
| 19:42.00 | Turl | awhat |
| 19:42.30 | specing | :S |
| 19:42.43 | specing | why not arm64? Is it too similar to amd64? |
| 19:43.06 | specing | It could be simply a64 :) |
| 19:43.32 | Turl | aarch64hf distros are gonna kick ass! :P |
| 19:43.58 | rm | I believe someone has beat some sense into them |
| 19:44.08 | specing | Intel is going to get its ass kicked sooner or later |
| 19:44.46 | specing | AMD is going to go bankrupt soon after aarch64 takes over the low-mid end market |
| 19:44.58 | Marex | rm: into guys who underwent gender changes and such ? :) |
| 19:45.05 | rm | https://lkml.org/lkml/2012/8/14/380 Kernel port directory and related functions renamed to "arm64". |
| 19:45.10 | Marex | xxiao: aarch64 |
| 19:45.18 | Turl | specing: doubt it, AMD is the goto cheap x86 |
| 19:45.48 | Marex | Turl: now that there's market for it |
| 19:46.50 | specing | Turl: aarch64 is going to be cheaper |
| 19:46.53 | xxiao | aarch64, ouch, i think ARM starts to lose its head |
| 19:47.13 | specing | as the cpus will be integrated into the motherboard and wont require *bridges and stuff |
| 19:48.04 | Turl | Marex: the market will still be there, windows home PCs aren't going anywhere |
| 19:49.16 | xxiao | hates udev |
| 19:50.30 | specing | Whats wrong with udev? |
| 19:52.53 | xxiao | specing: it's a mess when i have 8 SSDs on two HBAs and 3 ethernet ports |
| 19:53.22 | xxiao | devtmpfs worked fine without issues, but it took me a few hours to remove udev from rootfs |
| 20:04.23 | specing | What is a mess? |
| 20:05.32 | specing | xxiao: use LVM |
| 20:05.49 | specing | xxiao: add those into LVM and make it manage that |
| 20:06.29 | specing | LVM can do striping, mirroring or anything else |
| 20:08.47 | *** join/#arm-netbook mikey_w (~mike@va-76-6-223-29.dhcp.embarqhsd.net) |
| 20:09.05 | *** join/#arm-netbook n6pfk (~mike@va-76-6-223-29.dhcp.embarqhsd.net) |
| 20:11.22 | xxiao | specing: i need benchmark a system, just doing basic mdadm ATM |
| 20:11.30 | xxiao | raid5/6 |
| 20:12.27 | xxiao | for a 10G link the best I can do is about 6Gbps RAID5, which is not good enough |
| 20:12.41 | xxiao | i mean iscsi |
| 20:13.07 | xxiao | excluding overheads i think the best i can get should be around 8Gbps |
| 20:14.21 | specing | 0.o |
| 20:14.30 | specing | hardware or software raid? |
| 20:14.46 | specing | I think you should go hardware raid in your case... |
| 20:15.08 | xxiao | the issue is that, i'm trying to be _the_ hardware raid |
| 20:15.18 | specing | uhm |
| 20:15.36 | specing | you are doing it wrong |
| 20:15.37 | xxiao | the stuff i work on is shooting for a possible hardware RAID |
| 20:16.10 | xxiao | while running software RAID but with accel-raid-engine in hardware |
| 20:16.54 | xxiao | specing: how so? |
| 20:17.43 | specing | oh |
| 20:18.43 | xxiao | comparing to ASCI-hw-raid this will be fully open source |
| 20:18.49 | xxiao | ASIC |
| 20:19.22 | specing | There should be no firmware on an ASIC |
| 20:19.52 | specing | Also, the firmware running on the disks is not opensource |
| 20:19.56 | specing | :) |
| 20:22.18 | Marex | Turl: I dunno ... we don't have a windows PC at home nor at work |
| 20:23.26 | xxiao | i think for hardware RAID cards you do need binary firmware? |
| 20:23.36 | xxiao | yeah the firmware on SSD is another story |
| 20:24.48 | Turl | Marex: yeah I don't either, but I still use x86 |
| 20:25.05 | Turl | what I meant is 'typical x86 box' :P |
| 20:55.47 | popolon | no need to access to firmware to use hw raid |
| 20:56.06 | popolon | perhaps on some desktop motherboard with integrated half harware raid |
| 20:56.19 | popolon | hardware |
| 20:56.47 | popolon | most raid hardware is well managed on linux as far I know |
| 21:16.03 | *** join/#arm-netbook tuliom (~tuliom@186.214.89.120) |
| 21:24.21 | hno | popolon, hardware raid controllers have their firmware in a flash chip. |
| 21:25.15 | popolon | as most controllers isn't it ? |
| 21:25.54 | hno | SATA controllers do not have much firmware. |
| 21:26.00 | popolon | but I don't understand why to change it, well, on some cards that have problems and where support is discontinued |
| 21:27.04 | hno | popolon, xxiao is not trying to change, he is building one. |
| 21:27.10 | popolon | oh ? |
| 21:27.46 | popolon | an open raid controller ? |
| 21:49.42 | *** join/#arm-netbook rsalveti (~rsalveti@177.99.134.35) |
| 21:49.43 | *** join/#arm-netbook rsalveti (~rsalveti@linaro/rsalveti) |
| 22:04.34 | *** join/#arm-netbook tuliom_ (~tuliom@186.214.75.36) |
| 22:34.36 | *** join/#arm-netbook lerc (~quassel@121-74-246-157.telstraclear.net) |
| 22:54.41 | *** join/#arm-netbook gimli (~gimli@xbmc/staff/gimli) |
