| 01:07 | anton_ | left the channel |
| 01:13 | noir_ | joined the channel |
| 01:17 | mithro_ | joined the channel |
| 01:18 | dmj7261 | joined the channel |
| 01:19 | cfelton_ | joined the channel |
| 01:20 | FergusL | joined the channel |
| 01:21 | dmj726 | left the channel |
| 01:21 | Celia | left the channel |
| 01:21 | seku | left the channel |
| 01:21 | mithro | left the channel |
| 01:21 | cfelton | left the channel |
| 01:21 | cfelton_ | changed nick to: cfelton
|
| 01:23 | aombk | left the channel |
| 01:25 | mithro_ | changed nick to: mithro
|
| 01:53 | cfelton | left the channel |
| 01:59 | danieel | left the channel |
| 02:01 | aPinky | left the channel |
| 02:10 | Topic | apertus° - open source cinema | www.apertus.org | Currently Hot: AXIOM Beta Crowd Funding is Live: http://igg.me/at/axiom-beta | IRC Logs available at: http://irc.apertus.org
|
| 02:10 | se6astian | has set the topic |
| 02:12 | troy_s | joined the channel |
| 02:20 | danieeel | left the channel |
| 02:20 | dmj7261 | left the channel |
| 02:20 | mithro | left the channel |
| 02:20 | noir_ | left the channel |
| 02:28 | dmj7261 | joined the channel |
| 02:28 | mithro | joined the channel |
| 02:32 | ThatCantBe | left the channel |
| 02:32 | intracube | left the channel |
| 02:36 | intracube | joined the channel |
| 02:36 | dmjweb | left the channel |
| 05:58 | ThatCantBe | joined the channel |
| 05:58 | danieeel | joined the channel |
| 05:59 | ThatCantBe | left the channel |
| 05:59 | ThatCantBe | joined the channel |
| 06:03 | cfelton | left the channel |
| 06:03 | cfelton | joined the channel |
| 06:33 | mithro | left the channel |
| 06:33 | cfelton | left the channel |
| 07:35 | derWalter | joined the channel |
| 07:39 | derWalter | left the channel |
| 08:15 | mithro_ | joined the channel |
| 08:18 | derWalter | joined the channel |
| 08:23 | Morethink | joined the channel |
| 08:23 | Morethink | left the channel |
| 08:24 | derWalter | left the channel |
| 08:24 | cfelton_ | joined the channel |
| 09:15 | troy_s | left the channel |
| 09:24 | alexML | joined the channel |
| 09:24 | alexML_ | left the channel |
| 09:27 | ThatCantBe | left the channel |
| 09:30 | ThatCantBe | joined the channel |
| 10:04 | philippej | joined the channel |
| 10:12 | noir__ | changed nick to: Seku
|
| 10:12 | Bertl_zZ | changed nick to: Bertl
|
| 10:13 | Bertl | morning folks!
|
| 10:13 | Seku | morning Bertl!
|
| 10:27 | philippej2 | joined the channel |
| 10:30 | philippej | left the channel |
| 10:58 | julian | joined the channel |
| 10:58 | julian | left the channel |
| 10:58 | juliancorrea | joined the channel |
| 10:59 | Bertl | hey juliancorrea!
|
| 10:59 | juliancorrea | thank you !
|
| 11:05 | juliancorrea | left the channel |
| 11:18 | cfelton_ | left the channel |
| 11:22 | cfelton_ | joined the channel |
| 11:34 | aombk | joined the channel |
| 11:41 | aombk | ibc?
|
| 11:41 | aombk | so what is going on there?
|
| 11:46 | Bertl | I'm pretty sure se6astian will report when he has time/comes back
|
| 12:05 | Celia | changed nick to: FergusL
|
| 12:24 | philippej2 | left the channel |
| 12:38 | derWalter | joined the channel |
| 12:45 | jucar | left the channel |
| 12:47 | jucar | joined the channel |
| 13:12 | derWalter | moin moin
|
| 13:27 | Seku | moinsen
|
| 13:27 | philippej | joined the channel |
| 13:28 | robin | joined the channel |
| 13:28 | Robin___ | joined the channel |
| 13:34 | mithro_ | changed nick to: mithro
|
| 13:39 | Robin___ | Hey, I'm @ IBC 2014. Where are you Apertus guys?
|
| 13:48 | robin | left the channel |
| 13:56 | Bertl | Robin___: probably roaming around somewhere
|
| 14:55 | skinkie | left the channel |
| 14:55 | skinkie | joined the channel |
| 14:55 | Topic | apertus° - open source cinema | www.apertus.org | Currently Hot: AXIOM Beta Crowd Funding is Live: http://igg.me/at/axiom-beta | IRC Logs available at: http://irc.apertus.org
|
| 14:55 | se6astian | has set the topic |
| 14:56 | mithro | joined the channel |
| 15:19 | tyrone_ | joined the channel |
| 15:25 | ThatCantBe | left the channel |
| 15:25 | intracube | left the channel |
| 15:30 | ThatCantBe | joined the channel |
| 15:41 | philippej | left the channel |
| 15:58 | philippej | joined the channel |
| 16:00 | dmj7261 | left the channel |
| 16:08 | Meinereiner | joined the channel |
| 16:15 | dmj726 | joined the channel |
| 16:19 | philippej | left the channel |
| 16:29 | Robin___ | left the channel |
| 16:48 | Meinereiner | left the channel |
| 16:51 | Meinereiner | joined the channel |
| 16:51 | aombk | hey FergusL
|
| 16:52 | FergusL | hi aombk !
|
| 16:52 | aombk | long time no see
|
| 17:03 | intracube | joined the channel |
| 17:03 | derWalter | hey guys, i ve been looking for a good battery charger and found this german guy ( http://www.akku-selekt.de/ ) who is adapting, tuning and selling a battery charger. he is providing tons of information in german about batteries and their charging habits. as he is a big fan of eneloop batteries, maybe one could get him involved into development, so the axiom could get shippid with some kind of batterypack, where the people could
|
| 17:04 | derWalter | but recommended use of eneloops. i use them for years and have never ever used better batteries in my life. so you could exchange faulty batteries, without being forced to throw away all the other cells in the pack as well (as it is with proprietary solutions) and one would have a really beefy charger plattform as well. ... just my thoughts
|
| 17:13 | AndChat|176409 | joined the channel |
| 17:14 | Meinereiner | left the channel |
| 17:24 | Bertl | derWalter: go crazy, make contact, motivate him to participate in development ...
|
| 17:30 | derWalter | not wanting to go crazy :D
|
| 17:31 | derWalter | i just took out the battery from my nikon d800 and compared it to 4 AA eneloops
|
| 17:35 | derWalter | my nikon battery weights 105gramms, got a voltage of 7V and 1900mAh, so 105 gramms 14Wh. the 4 eneloops weigh 125gramms, got 1900mAh and 1,2Volts EACH, so in total 7600mAh 4,8 Volts and 36Wh
|
| 17:36 | Bertl | 4.8V is a little low, you want to get above 6V
|
| 17:37 | derWalter | so, the sustainabillity factor: the nikon en-el15 is reported to have very short life cycles, i bought three and cycle use them, never get them to low charged and one is already DEAD. the eneloops are reported to have 1800 cycles within 5 years while retaining 85% of their capacity
|
| 17:38 | derWalter | 8 of them in a case, you end up with a package around 110mm x 60mm x 17mm
|
| 17:38 | derWalter | this would weigh around 250-300 gramms and give you 72Wh at 9,6Volt
|
| 17:40 | Bertl | sounds good, you can buy an x8 box from your electronic shop and simply attach it
|
| 17:40 | derWalter | and if one cell dies early, open the case and change it. if you discharge charge discharge charge discharge charge (recycle) them 3-5 times a year and watch them to never get bellow 1Volt they will last at least 1800 cycles
|
| 17:40 | derWalter | yes, that was my idea
|
| 17:41 | derWalter | you would need the case for the batteries tough...
|
| 17:43 | derWalter | i also very much like the idea atomos realized with two packs attached at the same time, would be nice to be able to "stack" them, so you can attach em to each other. but if you need sooo much power you will end up with some external solution anyways i guess...
|
| 17:43 | derWalter | i mean... mhhh one could evel utilize the case the eneloops come in
|
| 17:43 | derWalter | /s/evel /even
|
| 17:43 | Bertl | http://www.ebay.com/itm/Battery-case-box-holder-for-x-8AA-size-cells-12V-3510-/281107209969
|
| 17:44 | Bertl | the come in all shapes and sizes, just search for battery holder xN (where N is the number of batteries)
|
| 17:44 | derWalter | woha, there you go.
|
| 17:45 | derWalter | i was thinking about something like this: http://www.ebay.com/itm/2x-8AA-Battery-Holder-Case-Sockets-Holding-8pcs-AA-2A-12V-R03-LR03-Box-plastic-/131271149055?pt=LH_DefaultDomain_0&hash=item1e905ec5ff
|
| 17:46 | derWalter | but to be honest, it looks quite cheap, better to get something good, which lasts and doesent need a replacement once in a while.
|
| 17:47 | niemand | joined the channel |
| 17:55 | derWalter | bertl have you thought about using something like a trello board? i came accross it, when i investigated a little bit on the upcoming unreal tournament and the development :) @ https://trello.com/b/Q07L4mSS/unreal-tournament-programming
|
| 17:56 | Bertl | what's that for?
|
| 17:58 | derWalter | project organisation and gives a clue whats been done and who is working on what right now
|
| 17:58 | derWalter | i can instantly read into topics i am interessted in
|
| 18:00 | Bertl | well, I'm pretty sure there is an open source pendant for this, I don't think it is wise to depend on proprietary services
|
| 18:01 | Bertl | even if they offer it for free, all kind of things can happen, like sudden charges, termination of service, drastic feature changes, etc
|
| 18:02 | derWalter | thats right ofc
|
| 18:03 | Meinereiner | joined the channel |
| 18:05 | AndChat|176409 | left the channel |
| 18:06 | AndChat|176409 | joined the channel |
| 18:06 | niemand | left the channel |
| 18:08 | Meinereiner | left the channel |
| 18:11 | derWalter | am i right that 2,5" formfactor is 10mm oversized compared to the axiom beta case? so if you want to design a storage module the only possible form factors for sata/ssd media would be 1,8" or msata? everything else gets extremely expensive and urges for a special adapter to be connected to a pc
|
| 18:14 | Bertl | yes, 2.5" SSDs are slightly larger than the planned AXIOM Beta
|
| 18:15 | Bertl | weilcome AndChat|176409!
|
| 18:16 | Bertl | mSATA MO-300A and Half Slim MO-297 would work as well
|
| 18:17 | AndChat|176409 | changed nick to: Meinereiner
|
| 18:17 | Meinereiner | Hey there. Sorry for the connection issues.
|
| 18:18 | Bertl | no problem
|
| 18:18 | Meinereiner | Quick question, do you already have a 3D lookup table implemented within the camera?
|
| 18:19 | Bertl | no, the AXIOM Alpha uses a 4x4 Matrix combined with 1D gamma LUTs
|
| 18:20 | niemand | joined the channel |
| 18:20 | Meinereiner | Is this something you would like to have?
|
| 18:21 | Bertl | do you write 3D lookup tables for FPGAs?
|
| 18:21 | Meinereiner | Because maybe I could contribute to that. I've written a 3D LUT for an altera cyclone v fpga for my bachelor's thesis.
|
| 18:22 | Bertl | ah, yeah, that would indeed be very helpful
|
| 18:22 | Bertl | is it written in Verilog or VHDL?
|
| 18:22 | Meinereiner | VHDL
|
| 18:23 | Meinereiner | It is optimized for 4k resolutions, throughput is at 600 million pixels per second.
|
| 18:25 | Meinereiner | I don't have any xilinx experience so I'd have to look how much effort it takes to port to a zynq
|
| 18:25 | Bertl | hmm, okay, we have a maximum of roughly twice the amount of pixels per second, so given that it doesn't use up all FPGA resources that could be parallelized
|
| 18:26 | Bertl | is there a pdf available of your thesis? is the code already public? if so, what license?
|
| 18:27 | Bertl | my experience is that the various FPGAs are quite similar at what they can do, usually the tools are trying to be as different as possible :)
|
| 18:27 | Meinereiner | I'm not quite done writing the thesis, but it will be entirely public once I'm done with my degree in a few weeks.
|
| 18:28 | Bertl | okay, anything to look at before that happens or do we simply wait till you're finished?
|
| 18:30 | Meinereiner | I can give you a preliminary PDF by tomorrow if you like. I just stumbled over your Projekt and wanted to ask if you are interested.
|
| 18:30 | Bertl | okay, great! and yes, of course we are interested :)
|
| 18:30 | niemand | left the channel |
| 18:32 | Meinereiner | Do you really require twice the amount of throughput? Because I'm wondering how you're getting that amount of data out of the fpga :D
|
| 18:33 | Bertl | yeah, we wondered for some time as well, but there are actually several ways in the AXIOM Beta
|
| 18:33 | Meinereiner | Is that raw Bayer?
|
| 18:34 | Bertl | yes, the sensor does raw bayer images at up to 300 FPS
|
| 18:34 | Bertl | (at lower bitdepths)
|
| 18:36 | Meinereiner | Well, after a 3D LUT you would probably want to stick to a 3 Compotent color space. And 1,2 giga pixels per second at 30 bit per pixel or more is quite a lot.
|
| 18:37 | Bertl | indeed, nevertheless, debayering increases the amount of data
|
| 18:37 | Bertl | so with a good debayer, we might end up with 4096x3072 pixels at up to (realistically) 150FPS
|
| 18:38 | Bertl | (which is even more, about 1.8G pixel/s)
|
| 18:39 | Bertl | but on the other hand, if we can't get it to work with more than 600M pixel/s, that would still be enough for the typical 4K @ 30 FPS applications
|
| 18:40 | Meinereiner | For that kind of throughput I'd have to implement a different interpolator. Trilinear interpolation uses a lot of hardware multiplicators.
|
| 18:49 | Bertl | or maybe just optimize the pipeline structures
|
| 18:49 | Bertl | but I agree, you won't be able to go over a certain limit with a single pipeline
|
| 18:50 | Bertl | as I said, we can start simple, we do not need to have every feature at the beginning
|
| 19:00 | Meinereiner | The module is fairly optimized for speed already. The limiting factor is DSP and block rams fmax at 300 MHz for the cyclone V. If you need more throughput, I'd have to double the resources. Which means an additional 84 Hardware multiplicators and 200 KByte worth of block RAM.
|
| 19:02 | Meinereiner | Thanks for now, I'll be back tomorrow.
|
| 19:05 | derWalter | interessting!
|
| 19:06 | AndChat|176409 | joined the channel |
| 19:09 | Meinereiner | left the channel |
| 19:12 | AndChat|176409 | left the channel |
| 20:06 | Bertl | off for a nap now ... bbl
|
| 20:06 | Bertl | changed nick to: Bertl_zZ
|
| 20:50 | rexbron | joined the channel |
| 20:50 | rexbron | left the channel |
| 20:50 | rexbron | joined the channel |
| 21:04 | dakorl | joined the channel |
| 21:05 | derWalter | left the channel |
| 21:07 | derWalter | joined the channel |
| 21:10 | tyrone_ | left the channel |
| 21:11 | dakorl | @derWalter just read your calculation about the eneloops. Am I right that the batteries in serial have the sum of the voltages but the same capacity as one cell?
|
| 21:22 | pppp | joined the channel |
| 21:25 | ray__ | joined the channel |
| 21:29 | zealee | joined the channel |
| 21:30 | ray__ | dakor1 is right about the eneloops, in series the voltage add up but capacity stay the same, in parallel capacity add up but voltage stay the same, 4x eneloops only have 9.6Wh
|
| 21:35 | zealee | left the channel |
| 21:40 | pppp | left the channel |
| 21:41 | zealee | joined the channel |
| 21:45 | ray__ | left the channel |
| 21:47 | zealee | left the channel |
| 22:25 | dakorl | left the channel |
| 22:37 | ThatCantBe | left the channel |
| 22:37 | aPinky | left the channel |
| 22:45 | Topic | apertus° - open source cinema | www.apertus.org | Currently Hot: AXIOM Beta Crowd Funding is Live: http://igg.me/at/axiom-beta | IRC Logs available at: http://irc.apertus.org
|
| 22:45 | se6astian | has set the topic |
| 22:46 | ThatCantBe | joined the channel |
| 23:05 | derWalter | hahaha, yeah, right :D
|
| 23:14 | Topic | apertus° - open source cinema | www.apertus.org | Currently Hot: AXIOM Beta Crowd Funding is Live: http://igg.me/at/axiom-beta | IRC Logs available at: http://irc.apertus.org
|
| 23:14 | se6astian | has set the topic |
| 23:17 | derWalter | well, it wouldnt be a problem to go from 8 to 16 eneloops, as thickness wouldnt be a problem for the module, but... hell, that 9,6V 15200mAh monster would weigh more than 500gramms, would say around 600-750 to stay halfway real, which is a lot. but on the other hand, that would be a lot of power... @bertl how long would the axiom B be able to run with such a battery attached? lets add 5 watts for writing to an ssd ( according to
|
| 23:18 | derWalter | thats allready a lot, the lowest when writing was about 1 watts by an intel msata ssd drive... average ~3watts, so 5 watts seems to be fair enough :)!! )
|
| 23:21 | alexML_ | joined the channel |
| 23:21 | derWalter | am i right that the whole system would use 10-12 watts while recording to an ssd?
|
| 23:21 | derWalter | hey alexML_
|
| 23:27 | alexML | left the channel |
| 23:27 | danieeel | left the channel |
| 23:29 | danieeel | joined the channel |
| 23:30 | intracube | joined the channel |
| 23:45 | aombk | left the channel |
| 23:49 | aombk | joined the channel |
| 23:50 | aombk | left the channel |
| 23:50 | aombk | joined the channel |
| 23:53 | skinkie | left the channel |
| 23:56 | Bertl_zZ | changed nick to: Bertl
|
| 23:56 | Bertl | back now ...
|
| 00:01 | skinkie | joined the channel |
| 00:12 | Bertl | derWalter: it really depends on what the camera is doing, the typical minimum is about 2 Watt for the MicroZed, which can go up to roughly 12W max
|
| 00:14 | Bertl | and for the sensor it is typically around 1W, but can go up to 2.5W
|
| 00:14 | Bertl | the remaining electronics should stay below 1W
|