00:00 | Q_ | Thanks.
| |
00:02 | Bertl | np
| |
00:20 | intracube_afk | http://caffeineforge.com/2012/09/10/mind-the-gap-kickstarter-funding-trends/
| |
00:20 | intracube_afk | changed nick to: intracube
| |
00:26 | Bertl | interesting, thanks!
| |
00:36 | g3gg0 | ah interesting
| |
00:37 | intracube | it does seem like 60-70% of successful projects have that curve
| |
00:37 | intracube | random example: http://www.kicktraq.com/projects/1207064031/rise-of-the-kage-ninja-board-game-with-detailed-mi/
| |
00:38 | intracube | some look more.... unusual: http://www.kicktraq.com/projects/leviathaninteractive/project-lodus-a-cyberpunk-co-op-action-rpg/
| |
00:38 | g3gg0 | :)
| |
00:39 | g3gg0 | well, it mainly depends on publicity
| |
00:39 | intracube | 28K on the last day and -just- makes the target :P
| |
00:40 | g3gg0 | when it reaches the last week, people start to spread the word
| |
00:41 | g3gg0 | iirc alex made a referrer link for the direct donations when following his link
| |
00:41 | intracube | maybe around halfway through most projects go on a promotion blitz
| |
00:41 | g3gg0 | yeah
| |
00:41 | intracube | which then shows results in the last 1/4 of the project
| |
00:46 | g3gg0 | still impressive that there is such a coincidence that on the day the most video news sites reported about ML supporting axiom, the money flow went up that massive
| |
00:46 | g3gg0 | e.g. http://www.redsharknews.com/business/item/2060-breaking-news-the-axiom-camera-team-has-joined-up-with-magic-lantern
| |
00:46 | g3gg0 | or planet5d etc
| |
00:46 | intracube | g3gg0: yeah, it looks likely that ML promotion had a significant effect
| |
00:46 | g3gg0 | i guess that reached out much more people than before
| |
00:46 | Bertl | yeah, I still think this had a major part in it
| |
00:47 | intracube | having key people signed up matters and encourages others
| |
00:47 | intracube | ML, Philip Bloom, etc
| |
00:47 | g3gg0 | i remember alex saying that his direct referrer made ~4k after 4 days (iirc)
| |
00:48 | g3gg0 | but due to his internet problems, we didnt chat much since then :)
| |
00:48 | g3gg0 | nevertheless, the race isnt over yet
| |
00:48 | g3gg0 | 4 days, 5k€
| |
00:49 | g3gg0 | hitting the goal is the most important now :)
| |
00:49 | Bertl | yes, but it looks a lot better than 10 days ago :)
| |
00:49 | g3gg0 | i guess you mobilized also the sensor manufacturers?
| |
00:49 | intracube | that's about 4 €350 donations per day
| |
00:50 | g3gg0 | i.e. asking them if they are willing to support?
| |
00:50 | g3gg0 | i can imagine that they are interested in this project being successful too
| |
00:50 | intracube | Bertl: are the apertus° + AXIOM logos/text available as .SVG files anywhere?
| |
00:51 | Bertl | yes, I think so, but don't ask me where :)
| 00:52 | intracube | is asking Bertl where :)
|
00:53 | intracube | I didn't look enough: https://apertus.org/logo
| |
00:54 | g3gg0 | Bertl: http://thndl.com/how-dng-compresses-raw-data-with-lossless-jpeg92.html could make it into MLV format
| |
00:54 | Bertl | intracube: k. will try to remember that one, but I think it doesn't have the AXIOM one
| |
00:54 | g3gg0 | Andrew Baldwin made a really good MLV viewer and is trying to find a good RAW compression algorithm
| |
00:54 | intracube | Bertl: yep, I'm still looking for that
| |
00:55 | Bertl | g3gg0: yeah, the problem with lossless compression is that you still have to plan for full bandwidth
| |
00:55 | g3gg0 | yep
| |
00:55 | g3gg0 | you cannot rely on its success
| |
00:55 | Bertl | precisely
| |
00:56 | g3gg0 | this codec doesnt make it into ML firmware
| |
00:56 | g3gg0 | the cPU is too slow
| |
00:56 | Bertl | we probably want something small which can be put into the FPGA
| |
00:57 | g3gg0 | its meant for archiving footage
| |
00:57 | Bertl | that's the only way to make it reasonably fast
| |
00:58 | intracube | for recording 300FPS, would it be possible to distribute the frames across multiple hardware h.264 encoders
| |
00:59 | Bertl | define hardware h.264 encoders
| |
00:59 | intracube | Bertl: standardized h.264 encoder chips
| |
00:59 | intracube | I guess most of them will max out at 30 or 60fps
| |
01:00 | Bertl | got a datasheet for me?
| 01:00 | g3gg0 | is leaving to get some sleep. 4 hours of sleep is minimum ;)
|
01:00 | Bertl | sleep well then!
| |
01:01 | g3gg0 | gn8 altogether
| |
01:01 | intracube | Bertl: not currently. but 5 chips running at 60fps each = 300
| |
01:01 | intracube | g3gg0: night :)
| |
01:02 | Bertl | that probably wont work, as it would mess up all temproal coding
| |
01:02 | Bertl | you would want to split the image in sectors
| |
01:02 | Bertl | and run each sector at 300FPS
| |
01:02 | intracube | yes, the chips would have to be running intra-frame mode
| |
01:02 | intracube | so the final stream wouldn't have a GOP structure
| |
01:03 | intracube | Bertl: wouldn't that be more complex to recombine back to a single stream?
| |
01:04 | Bertl | well, it would require recoding, not sure what your plans are though
| |
01:05 | g3gg0 | left the channel | |
01:05 | intracube | to have a final stream that's h.264, I-frame only, might be flagged as 25fps (so motion would playback 12x slower than normal)
| |
01:06 | intracube | as I doubt many media players (software or hardware) would handle a 300fps stream
| |
01:07 | Bertl | I don't think that you will be able to simply mix 5 overlapping (with a phase offset) streams together, but I might be wrong
| |
01:07 | intracube | Bertl: http://www.soctechnologies.com/piece.php?Id=191
| |
01:08 | intracube | Bertl: yes, simply appending frames from 5 independent encoders to a single stream might not work
| |
01:08 | Bertl | that's proprietary IP
| |
01:08 | intracube | likely there would be flags + other data that need to be there to be a proper muxed video file
| |
01:09 | intracube | Bertl: I guess most hardware encoders would be IP :/
| |
01:10 | intracube | h.264 encoders at any rate
| |
01:10 | Bertl | that's not a hardware part
| |
01:10 | Bertl | that is a netlist you can build an ASIC with or program an FPGA
| |
01:11 | Bertl | and you have to sign an NDA for that and pay big bucks to be able to use it, but you won't get any source
| |
01:11 | intracube | sorry, I know virtually nothing about hardware design
| |
01:11 | intracube | hmm
| |
01:19 | intracube | Bertl: http://www.nxp.com/products/identification_and_security/security_and_surveillance_ip_camera/ASC8852AET.html
| |
01:19 | intracube | same IP issues there?
| |
01:21 | Bertl | nope, that is hardware, but seems to require an NDA for the datasheet
| |
01:21 | Bertl | so, can't say very much right now, the leaflet doesn't give away too much
| |
01:22 | Bertl | but it seems to be an soc solution with an arm core and specialized hardware
| |
01:41 | wescotte | joined the channel | |
02:29 | ItsMeLenny | joined the channel | |
02:32 | Bertl | hey, it's you lenny! :)
| |
02:45 | ItsMeLenny | it is me lenny
| |
02:45 | ItsMeLenny | it's i
| |
02:45 | ItsMeLenny | its not you bert though
| |
02:45 | ItsMeLenny | it is bert1
| |
02:45 | ItsMeLenny | how are you
| |
02:46 | Bertl | no, actually it is BertL (get a decent client :)
| |
02:46 | ItsMeLenny | lol
| |
02:46 | ItsMeLenny | #couriernew
| |
02:47 | Bertl | but I'm used to folks calling me Bert-i and Bert-1 on IRC :)
| |
02:47 | ItsMeLenny | lol, if i want to ping someone i type a couple letters and hit tab so that should work Bertl
| |
02:47 | ItsMeLenny | yeah
| |
02:47 | Bertl | yup, same here ...
| |
02:49 | ItsMeLenny | i'd make a joke about people also calling you bert-[the japanese character that looks like a l] but i dont know what its called, ive only seen it on the DS XL japanese version
| |
02:49 | ItsMeLenny | http://www.geeky-gadgets.com/wp-content/uploads/2012/10/Nintendo-3DS-XL1.jpg
| |
02:50 | ItsMeLenny | maybe i wouldve gotten better results if my google wasnt switched to german for some reason
| |
02:50 | Bertl | hehe
| |
02:52 | ItsMeLenny | axiom is nearly funded
| |
02:53 | Bertl | yes, it is looking good
| |
02:54 | ItsMeLenny | i like how axiom has put the sensor/lens mount in the centre of the camera, its one thing that drives me mad with both DSLRs and point and shoots
| |
02:54 | ItsMeLenny | as theyre uncentred, if that wasnt clear :p
| |
02:55 | Bertl | I can relate
| |
02:56 | wescotte | left the channel | |
03:04 | even | joined the channel | |
03:04 | even | hellos
| |
03:04 | ItsMeLenny | although i havent helped fund
| |
03:04 | ItsMeLenny | many hellos
| |
03:04 | Bertl | even: that's odd :)
| |
03:05 | even | i came to hang up paper streamers :D
| |
03:05 | Bertl | great! show them to us!
| |
03:05 | ItsMeLenny | can i have one?
| |
03:06 | even | hehe well you still have to wait 4 days
| |
03:06 | ItsMeLenny | tease us with these paper streamers and then not share!
| |
03:06 | Bertl | yeah, shame on you!
| |
03:07 | ItsMeLenny | permaban!
| |
03:07 | ItsMeLenny | :P
| |
03:07 | even | =)
| |
03:10 | even | bertl: even is my real name =)
| |
03:11 | even | i've been following the project for quite some time
| |
03:12 | Bertl | cool name, can't compete with that
| |
03:12 | even | looks great so far. can't wait to have it in my hands
| |
03:12 | Bertl | what are you going to do with it, if you don't mind sharing?
| |
03:13 | even | i'll put it on my drone and try to take over the world... just kidding =)
| |
03:16 | even | I'm looking forward to shoot some music videos.
| |
03:17 | even | and sorry for my bad english... i'm from germany
| |
03:17 | ItsMeLenny | people who speak english can't even speak it properly, so nobody noticed
| |
03:18 | Bertl | I'm not a native english speaker either, so no need to apologize
| |
03:18 | ItsMeLenny | i thought axiom was german
| |
03:20 | Bertl | no, not really :)
| |
03:25 | even | it's from these guys.. https://www.youtube.com/watch?v=Dm6TSVRi-BA
| |
03:25 | ItsMeLenny | that video is unavailable with safety mode enabled
| |
03:26 | even | don't worry haha it's nothing important just funny as a flying banana
| |
03:27 | ItsMeLenny | whens inspector rex come in
| |
03:38 | wescotte | joined the channel | |
03:49 | even | is there an estimated weight of the system already? *points at the beta concept model*
| |
03:50 | even | i know that question might be to early but i need some numbers to calculate with.
| |
03:50 | Bertl | not really, we did a few checks on the electronics side, but as the case/mount materials are still undecided, it's problematic
| |
03:52 | even | >1kg incl. lensmount?
| |
03:52 | Bertl | the MicroZed is about 45g, the Beta PCB with parts and socket will be about the same
| |
03:52 | Bertl | add two shields and the dual microSD, you'll get about 150g without case, sensor, mount and cooling
| |
03:54 | Bertl | mount and case can go between 50g and 500g I guess, depending on the materials used
| |
03:54 | Bertl | you need to add a lens system to that, but that can be rather light as well
| |
03:56 | Bertl | (the nikkor lens we used on the Alpha has about 170g)
| |
03:56 | even | nice thank you
| |
03:57 | Bertl | so if you want to use a quad copter or similar you probably want to avoid a heavy case and use light materials for lens mount and lens system, and of course, high density batteries
| |
03:58 | even | yap that was my thought. it's a self made cinestar x8
| |
04:00 | Bertl | nice! anyway, I'm off to bed now ... have a good one everyone!
| |
04:00 | Bertl | changed nick to: Bertl_zZ
| |
04:00 | even | sleep well
| |
04:40 | Nathan___ | joined the channel | |
05:05 | Nathan___ | left the channel | |
05:13 | ItsMeLenny | left the channel | |
06:38 | Hyuri | joined the channel | |
07:51 | g3gg0 | joined the channel | |
07:56 | daFred | joined the channel | |
08:05 | daFred | even: at the moment the stack of pcb's and the lensmount is planned to be held by a sturdy part including tripod thread. So for development it should work without housing. The lightest housing would be a plastic bag :-))
| |
08:09 | even | what about duck tape? :D
| |
08:09 | daFred | as the design is open source there could be lightweight versions in the community too. In the moment I'm thinking of a cast variant ranging from aluminium or transparent plastics or even ceramics...
| |
08:10 | dmjnova | even: you're interested in the axiom for drone use?
| |
08:11 | daFred | duck tape holds the world together :-)) btw is it open source??
| |
08:14 | dmjnova | what drone are you flying now?
| |
08:14 | even | daFred: i've seen the design. i have the possibilites to mill parts myself. got a 5 axis cnc and a 3 axis cnc at my work place so if anyone needs help. feel free to shoot at me. i can't use it everyday but weekends is np.daFred: i've seen the design. i have the possibilites to mill cases got a 5x cnc and a 3x cnc at my work place
| |
08:15 | even | whoops sorry
| |
08:15 | even | dmjnova: i built a cinestar x8
| |
08:15 | even | it's not completly done yet but mostly =)
| |
08:16 | even | i'm designing the gimbal at the moment
| |
08:17 | even | i was planning to buy a 5dm3
| |
08:17 | even | because of ml =)
| |
08:18 | even | but i'm pretty confident that this project will be awesome in a few years. the possibilities are endless
| |
08:18 | even | so i passed on the 5d for now
| |
08:21 | even | a live streaming module would be dope
| |
08:21 | even | 2,4 or 5.8 ghz
| |
08:22 | dmjnova | noted :D
| |
08:22 | dmjnova | designing your own gimbal, sounds familiar
| |
08:23 | dmjnova | axiom should be much nicer for drone flying than epic
| |
08:23 | even | hehe apart from the electronics
| |
08:23 | dmjnova | what will you be doing for gimbal controller?
| |
08:23 | even | alexmos 32bit 3 axis
| |
08:24 | dmjnova | was just tuning a gimbal with one of those
| |
08:24 | even | how is it? i haven't ordered it yet?
| |
08:25 | dmjnova | be very conscientious about IMU wire length
| |
08:25 | dmjnova | where were you going to order from?
| |
08:25 | even | sec have to look
| |
08:25 | even | http://flyduino.net/Brushless-Gimbal
| |
08:27 | dmjnova | Had some problems using frame IMU with the RCTimer one
| |
08:27 | dmjnova | so the onboard frame IMU sounds like a great idea
| |
08:28 | dmjnova | otherwise seemed to work great until it came to tuning the yaw axis
| |
08:29 | even | what motors and camera are you using?
| |
08:30 | dmjnova | that particular one was using the NEX, don't remember what motors
| |
08:30 | dmjnova | something brushless obviously
| |
08:30 | dmjnova | but that was just tuning an existing gimbal
| |
08:30 | dmjnova | designing a gimbal currently as well though
| |
08:31 | even | nice!
| |
08:32 | even | yeah it's tricky but best is learning by doing =)
| |
08:32 | even | i really like the mövis but the prices are insane
| |
08:41 | dmjnova | I happen to work at a multirotor manufacturer
| |
08:41 | dmjnova | so the axiom is pretty darn interesting
| |
08:41 | dmjnova | of course I was already into the axiom before that
| |
08:43 | dmjnova | hmm...the cinestar really could have more lifting capacity
| |
08:43 | even | conclusion: we need an apertus gimbal, an apertus timlapse slider and an apertus robot arm (kuka like) for slowmotion shots with micrometer precision. lol
| 08:44 | dmjnova | nods knowingly
|
08:45 | even | g3gg0: love you guys!
| |
08:46 | even | http://vimeo.com/94871761 =)
| |
08:51 | Hyuri | left the channel | |
08:51 | dmjnova | the look reminds me of The New World
| |
08:54 | dmjnova | quite steady
| |
08:54 | even | have a look at that one i love the flying shots http://www.youtube.com/watch?v=xQ_IQS3VKjA
| |
09:02 | dmjnova | that's some nice clear action shooting
| |
09:03 | dmjnova | beautiful gopro footage aside
| |
10:13 | philippej | joined the channel | |
10:27 | daFred | left the channel | |
11:38 | daFred | joined the channel | |
11:42 | daFred | left the channel | |
11:53 | philippej | left the channel | |
12:11 | Bertl_zZ | changed nick to: Bertl
| |
12:11 | Bertl | morning folks!
| |
12:14 | Bertl | 96.5k \o/
| |
12:20 | Gegsite | joined the channel | |
12:20 | Gegsite | left the channel | |
12:21 | Gegsite | joined the channel | |
12:26 | philippej | joined the channel | |
12:37 | Gegsite | left the channel | |
12:47 | nnailer | joined the channel | |
12:49 | danieel | Bertl / g3gg0 - i would rather see a side by side comparison of the two chips - but there is a lot of possibility to do software based corrections - the actual harware plays a minor role
| |
12:53 | philippej | danieel, I guess it's always better to have actual sample files, since you never know what a specific sensor will bring in reality. It might be interesting to see if the software side strategies can be reused for different sensors.
| |
12:57 | Bertl | of course I meant unmodified i.e. sensor raw data, with and without sensor side corrections enabled
| |
13:00 | wescotte | left the channel | |
13:12 | danieel | yes but of what subject - e.g. the sensor behaves very different when in complete darkness, and when even a tiny light hits it
| |
13:12 | danieel | we found a lot of strange things on the live 4k setup we have now
| |
13:13 | Bertl | dark frames, IIRC
| |
13:13 | Bertl | that's what alexML required
| |
13:13 | danieel | even after a fpn is removed (substractive) there is a fixed pattern remaining
| |
13:13 | Bertl | yes, that is why there are two components of FPN
| |
13:14 | danieel | unless the frames are taken under controlled conditions, the results are hard to compare to anything
| |
13:14 | nnailer | Thank you for adding a third batch of the super35's. I just heard about the project from Magic Lantern and made my contribution almost immediately after.
| |
13:15 | Bertl | you're welcome!
| |
13:15 | nnailer | Do you know if VAT will be added on to orders shipped to USA?
| |
13:16 | danieel | Bertl: the 1 min video is showing the kac with "super35" text in same time :) lol
| |
13:17 | Bertl | yeah, result from cutting down the video, we filmed the KAC but not the CMV at this time
| |
13:18 | nnailer | Never mind. I think I found my answer at http://www.brighton-accountants.com/blog/vat-uk-usa/. VAT is not added to orders shipped to USA.
| |
13:18 | philippej | nnailer : it shouldn't.
| |
13:18 | philippej | you might have to pay some import duties
| |
13:18 | nnailer | Cool. One less thing for my wife to freak out about :)
| |
13:18 | se6astian|away | changed nick to: se6astian
| |
13:18 | philippej | :-)
| |
13:19 | nnailer | I probably should've waited for her to wake up before I gave $440 USD up for a beta. I couldn't risk it though. Easier to ask for forgiveness than permission sometimes.
| |
13:19 | Bertl | nnailer: we still have to figure out the best way to get everything to the users
| |
13:19 | philippej | the good thing is that it's already a low risk proposal, since you fund only the development for now, you don't have to put the whole money on the table now
| |
13:19 | Bertl | for example it might be "better" to order all the MicroZed (or PicoZed) designated for the US, within the US
| |
13:19 | se6astian | good afternoon!
| |
13:20 | nnailer | Good morning.
| |
13:20 | philippej | afternoon !
| |
13:20 | Bertl | morning! :)
| |
13:21 | Bertl | I just got up an hour ago *G*
| |
13:22 | nnailer | Is there a getting started section for new developers to get up to speed on contributing?
| |
13:23 | Bertl | not yet, but if you are going down that path, maybe help creating one on the wiki or so?
| |
13:23 | nnailer | I don't know where to start. I am a developer by trade though and I would like to contribute.
| |
13:24 | se6astian | there are general pages
| |
13:24 | se6astian | https://apertus.org/contribute
| |
13:24 | se6astian | and https://apertus.org/join-the-team
| |
13:24 | se6astian | though non technical details related
| |
13:25 | nnailer | Are you guys tracking defects ?
| |
13:25 | Bertl | nnailer: what have you been developing so far?
| |
13:25 | se6astian | you mean a bugtracker, its planned yes
| |
13:25 | Bertl | nnailer: not yet, but a bug/feature tracker is in the works
| |
13:25 | nnailer | Anything that is requested ranging from low level touch screen and printer drivers to high level javascript webapps.
| |
13:27 | nnailer | Currently an IVR developer primarily working with C, VXML, Java, and Javascript.
| |
13:27 | Bertl | sounds good, how about embedded systems? microcontroller? FPGA?
| |
13:28 | nnailer | I've dabbled with ladder logic, stepper motors and such. Not much experience outside of that.
| |
13:28 | Bertl | okay, no problem
| |
13:29 | philippej | whare are we on the bugtracker side btw?
| |
13:29 | Bertl | nnailer: what platforms? Linux? MacOS? Windows?
| |
13:30 | nnailer | Primarily windows and linux.
| |
13:30 | nnailer | Current day job is linux.
| |
13:31 | Bertl | okay, sounds good ... my suggestion would be to get a Zynq based development board which can be reused for the Beta later and get comfortable with that first
| |
13:32 | Bertl | unless you want to work on the "other side" i.e. outside the camera on applets or other stuff required to communicate with the camera
| |
13:32 | nnailer | I've worked with FFMPEG code in the past and wrote FLIC rendering into SDL among other things.
| |
13:32 | nnailer | Not sure that there is much crossover there.
| |
13:32 | Bertl | yeah, SDL is nice, I've been using it for many things, but I do not see an application there right now
| |
13:33 | nnailer | I would like to work on both sides if possible but I think my immediate experience lends more towards the software side..
| |
13:33 | Bertl | yeah I was referring to software on both sides
| |
13:34 | nnailer | Ahh. I meant the fluffy side.
| |
13:35 | se6astian | current status on bugtracker: we used buggenie in the past, good user experience but it unfortunately was not really ready for production sites - got flooded with spam and has pretty much no countermeasures so we had to close it again; I am hesitant to deploy something like bugzilla because its a total overkill for our target audience as it will consists of a lot of non-software developers so we need to make sure the system is easy to navigate and re
| |
13:35 | nnailer | Are you guys using the Zynq 7020 for beta?
| |
13:35 | se6astian | we should test buggenie though maybe they fixed it by now
| |
13:36 | Bertl | Zynq Z-7010/7015/7020 yes
| |
13:37 | Bertl | se6astian: I had good experiences with bugzilla, it is rather straight forward
| |
13:37 | regnirps | left the channel | |
13:38 | nnailer | Any specific model that I should get Bertl?
| |
13:39 | Bertl | we will definitely do a Beta for the MicroZed 7010/7020 (in this case I'd go for the 7020)
| |
13:40 | Bertl | but we also plan a Beta variant using the new PicoZed 7015
| |
13:40 | Bertl | (which might be a little further down the road)
| |
13:40 | Bertl | and we plan to make another variant based on the Parallella board
| |
13:40 | Bertl | which features an Epiphany core in addition to the Zynq
| |
13:41 | danieel | what will the PB do?
| |
13:42 | Bertl | PB?
| |
13:42 | danieel | parallella board
| |
13:42 | Juicyfruit | se6astian, for a bugtracker / issue tracker / patch tracker the blender project uses phabricator and I am very happy with it in terms of features and interface and use patterns
| |
13:43 | Bertl | danieel: mostly processing image data I guess
| |
13:43 | philippej | we could try to compare bugzilla, mantis and phabricator
| |
13:45 | Juicyfruit | philippej, it mostly depends on what you as a project want and need to get out of the software .
| |
13:45 | nnailer | Any vender recommendation Bertl? digilentinc.com has a 7010 dev board for $189 USD.
| |
13:45 | Juicyfruit | phabricator has really nice patch review and in is a lot more then just a bugtracker
| |
13:46 | Bertl | nnailer: no recommendations, wherever it is readily available and the price acceptable you can go for it
| |
13:46 | Bertl | (but consider the 7020 version for the MicroZed, it has more I/Os)
| |
13:47 | Bertl | and get one of those Xilinx programming cables, they simplify interfacing the board till you get a Beta Board
| |
13:48 | nnailer | Thanks Bertl
| |
13:48 | philippej | Juicyfruit
| |
13:49 | Juicyfruit | philippej, yes
| |
13:49 | philippej | Juicyfruit, it will probably depend of the ease of use for non developpers
| |
13:49 | philippej | (sorry, pressed return too soon :-) )
| |
13:49 | Juicyfruit | philippej, well i would just look around on https://developer.blender.org/ :)
| |
13:49 | philippej | since developpers usually have some special abilities to use non straightforward tools
| |
13:50 | philippej | It looks really nice btw
| |
13:50 | Juicyfruit | philippej, it was made by facebook people i think they even use it
| |
13:51 | Bertl | one important thing is that there is some kind of email interface
| |
13:51 | Bertl | i.e. notifications about updates at least and email replies would be highly appreciated
| |
13:52 | Bertl | from the interface, the developer.blender.org looks fine to me
| |
13:53 | nnailer | I have to take off for now. I will place my order for the 7020 and stop back in soon. Thanks again for this awesome work. I hope to help in the near future.
| |
13:53 | Bertl | nnailer: you're welcome! and looking forward to it!
| |
13:53 | philippej | inbound email documented here : https://secure.phabricator.com/book/phabricator/article/configuring_inbound_email/
| |
13:54 | philippej | See you nnailer !
| |
13:57 | even | gn8 =)
| |
13:58 | even | left the channel | |
13:58 | mars_ | over 97k!
| |
14:03 | Bertl | \o/
| |
14:03 | Gegsite | joined the channel | |
14:09 | se6astian | ok gotta run
| |
14:09 | se6astian | see you later
| |
14:09 | se6astian | changed nick to: se6astian|away
| |
14:47 | Bertl | 98k \o/
| |
14:49 | philippej | 2k to go
| |
14:51 | Gegsite | will make
| |
14:55 | Gegsite | dont the mega :P
| |
14:55 | Gegsite | Done the Mega :)
| |
14:56 | Bertl | great!
| |
14:57 | Gegsite | just for the T-shirt lol
| |
15:04 | comradekingu | wont a lens mount in aluminium be lighter and better for heat dissipation?
| |
15:05 | comradekingu | with some ridges on the outside it doubles as a capable heatsink and grip for holding the camera
| |
15:05 | seku | joined the channel | |
15:06 | seku | hiyas! congrats on reaching the 98k!! and random greetings from canada. taking the plane home tonight and looking forward to see how the campaign ends :)
| |
15:06 | mars_ | thx seku!
| |
15:07 | seku | coming home with 1.5 tera of ML raw video ... i wonder how i will cope with 4k raw with the axiom ^^;
| |
15:09 | seku | now .. packing up. ill be joining the chat more regularly again when im home.
| |
15:13 | seku | cya tomorrow
| |
15:13 | mars_ | cya
| |
15:13 | seku | left the channel | |
15:20 | Gegsite | left the channel | |
15:22 | mars_ | 98.6k, getting closer
| |
15:25 | Bertl | I think we might make the goal tonight
| |
15:25 | jucar1 | left the channel | |
15:27 | jucar | joined the channel | |
15:27 | philippej | left the channel | |
15:29 | mars_ | that would be awesome
| |
15:39 | comradekingu | sent mail to my other videographer aquaintance
| |
16:03 | philippej | joined the channel | |
17:09 | g3gg0 | Bertl: if you want to provide a1ex raw images he can use to determine sensor quality, there are various methods to do so
| |
17:10 | g3gg0 | a) static scene with deepest shadows (light trap) and overexposed areas, totally out of focus (blurry) so you have gradients. this shot with every ISO the sensor has, at least two images per ISO. (fixed position, no movement at all)
| |
17:12 | g3gg0 | b) total uniform gray shots without any visible pattern. doable with e.g. a tissue over the lens pointing into sky (of course no focus at all) also at different ISOs
| |
17:12 | g3gg0 | a) is for determining DR and noise and b) is for FPN measurement. (plus a few dark frames of course)
| |
17:14 | g3gg0 | i think if he can get hands on these raw images, he can run his statistics tools and present you DxO-Mark like results :)
| |
17:14 | jucar | left the channel | |
17:15 | Bertl | yeah, that was the idea, danieel can you do that?
| |
17:16 | jucar | joined the channel | |
17:22 | Q_ | I was thinking about doing some stats too, but didn't get around to it yet.
| |
17:23 | intracube | €99,000 :)
| |
17:24 | comradekingu | alluring 99%
| |
17:41 | se6astian|away | changed nick to: se6astian
| |
17:41 | se6astian | back
| |
17:44 | se6astian | 99% - I think now is a good point to consider if we really want to build a camera..... :P
| |
17:45 | Bertl | nah :)
| |
17:46 | se6astian | I have always been really fund of cookery you know
| |
17:47 | se6astian | erm I mean *fond :)
| |
17:48 | Bertl | let's make an open source hot-plate then :)
| |
17:54 | se6astian | I read on heise there is great demand for it!
| |
17:55 | philippej | left the channel | |
17:56 | troy_s | Congrats se6astian / Bertl
| |
17:57 | Bertl | thanks! congratulations/thanks to the community!
| |
17:57 | troy_s | g3gg0: You can profile without the full sensor diffuse white and a light trap for black.
| |
17:57 | troy_s | You basically add two synthetic points into your profiles for generating after you have done the first pass in processing them.
| |
17:58 | troy_s | This allows the resultant LUT or matrix to span the full range, with proper correct values.
| |
17:59 | troy_s | If you use real-world white / black, you are A) likely to never get the true values of black and white B) are subject to the worst regions of sensor capture etc.
| |
17:59 | troy_s | The resultant profile will almost always likely be a skewed mess.
| |
18:08 | wescotte | joined the channel | |
18:21 | Legendin | joined the channel | |
18:21 | danieel | troy_s: do you know how to do a 3x3 matrix math in Baselight?
| |
18:21 | danieel | seems that there is no such functionality easily available
| |
18:22 | troy_s | danieel: You can do a transform in Baselight. Any reason you are using Baselight for it?
| |
18:22 | troy_s | Just use OCIO and OIIO's tools. Lean and weaponized.
| |
18:22 | Q_ | So I'm looking at one of the darkframe images. Does it have known dead pixels?
| |
18:22 | troy_s | (And utterly configurable)
| |
18:22 | danieel | we had our matrix written on paper, and wanted to apply to the reels we got
| |
18:22 | troy_s | danieel: Then use OCIO / OIIO. Want me to set up a config for you?
| |
18:22 | danieel | we found just a rather complicated profile setup for in/out conversion
| |
18:23 | danieel | no.. i wanted to do it as an effect
| |
18:23 | danieel | like realtime :)
| |
18:23 | troy_s | Say what?
| |
18:23 | troy_s | Snag Resolve then.
| |
18:24 | se6astian | Q_: yes there are 8 dead pixels AFAIR
| |
18:24 | danieel | ok, i thought so that different sw would be better for the task
| |
18:24 | Q_ | I think I see 5?
| |
18:24 | Bertl | you won't see black sensel on the dark frame I guess
| |
18:25 | Q_ | I guess so too.
| |
18:25 | troy_s | danieel: Honestly though, you can do it with GPU and OCIO.
| |
18:26 | troy_s | danieel: Hell, you could do it in Blender if you wanted.
| |
18:26 | troy_s | danieel: Dead easy. Just set the transform as a spimtx file (a 4x3 matrix with far right column being 16 bit offsets (0))
| |
18:27 | troy_s | danieel: Then set up a view. Done.
| |
18:28 | troy_s | If you want help, it would take me four minutes.
| |
18:28 | troy_s | Dead simple.
| |
18:28 | danieel | no need to, offline i can do it too
| |
18:30 | troy_s | danieel: OCIO can be online ;)
| |
18:31 | troy_s | danieel: Use a quickie python script and GPU.
| |
18:31 | troy_s | You won't get tetrahedral interp, but you can demo it (and CDL) using the GPU.
| |
18:31 | danieel | just noticed that 4.4 release got a matrix :) we were on a 4.3 setup
| |
18:32 | troy_s | But a matrix is kind of crap.
| |
18:32 | troy_s | You probably want a 3D LUT for a camera profile demo.
| |
18:32 | danieel | no i did not, i wanted to mess with the channels with a matrix
| |
18:36 | aombk | joined the channel | |
18:36 | aombk | hi everybody
| |
18:36 | troy_s | danieel: Mess? There is a right and a wrong, no real inbetween.
| |
18:37 | troy_s | danieel: Unless it is a creative twist, but that would slug in after the camera transform.
| |
18:44 | philippej | joined the channel | |
18:45 | philippej | 964 euros and we'll finaly be able to work on the pong appliance
| |
18:47 | se6astian | philippej: damn how did you figure out where the entire budget would go?
| |
18:48 | troy_s | se6astian: How is your plan looking? 100k € doesn't buy a lot of time for certain.
| |
18:48 | Q_ | Their seems to be a weird pattern to the samples from that dark frame.
| |
18:48 | danieel | what is the weirdeness?
| |
18:49 | danieel | g3gg0: is somewhere that evaluation script? or it is hand work?
| |
18:50 | Q_ | The first 2 rows have all samples zero (except 2), the other rows have around 100 pixels zero at the start and end of each row.
| |
18:51 | g3gg0 | danieel: a1ex uses matlab and a bunch of selfmade scripts. i dont have them, but he posted some of them in forums
| |
18:51 | Q_ | I guess you could also see that on images that a1ex made.
| |
18:52 | se6astian | 99.4%
| |
18:52 | Legendin | left the channel | |
18:53 | se6astian | troy_s: we built the AXIOM Alpha with less than 10K€ - for the Beta this funding will open up a new world of possibilities! :)
| |
18:53 | troy_s | I know
| |
19:00 | daFred | joined the channel | |
19:02 | Bertl | evening daFred!
| |
19:03 | daFred | evening, happy crowd funding!!!
| |
19:03 | Q_ | Any idea where that pattern comes from?
| |
19:05 | Bertl | Q_: no idea, there might be up to 16 columns black (the so called black columns) but not up to 100, and certainly the rows at the top are not special in any way
| |
19:12 | danieel | se6astian: material cost for my kac model is ~ 4000e, and it can record 4k video and has sdi link :) so i wonder what did cost 10k on the alpha
| |
19:13 | Bertl | we built two of them, no?
| |
19:13 | Bertl | besides, not everybody is as efficient as you are :)
| |
19:14 | Bertl | to compensate for that, we are doing FOSS/OH not proprietary cameras :)
| |
19:16 | g3gg0 | whose F5 button already broke?
| |
19:17 | Bertl | what does F5 do for you?
| |
19:17 | g3gg0 | reload the indiegogo page
| |
19:17 | philippej | they probably have a cache of at least one minute to compensate for people like you (and me) :-)
| |
19:20 | se6astian | danieel: mainly booze
| |
19:26 | daFred | I fear I need a new Safari window when the campaign is over.... and a couple of new reload buttons...
| |
19:27 | Bertl | just take the 600 EUR perk and do a single reload :)
| |
19:29 | Bertl | but we still have 3 days left :)
| |
19:31 | daFred | Se6astian I think it's time to make special offer for the Platinum Deluxe pay one, take two
| |
19:33 | philippej | latest update is nice !
| |
19:35 | philippej | (talking about that : https://vimeo.com/107801527 )
| |
19:37 | se6astian | daFred: I was thinking about just giving away everything for free from now on :)
| |
19:38 | wescotte | philippej: reminded me of Upstream Color
| |
19:38 | wescotte | se6astian: I want my 5 euros back then! :)
| |
19:45 | Gegsite | joined the channel | |
19:47 | aombk | there is an auto reload addon for firefox
| |
19:49 | se6astian | wescotte: no refunds, we already used your 5€ for buying rocket fuel for our escape!
| |
19:50 | intracube | €99,786!
| |
19:51 | se6astian | woah, so close!
| |
19:51 | philippej | (no need to reload anything, just have an irc client open :-) )
| |
19:51 | Bertl | goal reached!
| |
19:51 | Bertl | \o/
| |
19:51 | philippej | !!!!!!!
| |
19:51 | se6astian | HURRAY!!!
| |
19:51 | intracube | woooooo! :)
| |
19:52 | intracube | congratulations se6astian, Bertl and the other Apertus folk :)
| |
19:52 | aombk | congratulations people
| |
19:52 | Gegsite | YEEEAAAHHH
| |
19:52 | intracube | \o/
| |
19:52 | aombk | very nice and clever campaign
| |
19:56 | wescotte | congrats guys
| |
19:58 | Bertl | thanks to everyone for making this possible!
| |
19:59 | se6astian | you guys rock :D
| |
19:59 | Gegsite | could be a stat in a map what country did the support, to money and amount :P
| |
19:59 | beat_b | joined the channel | |
20:00 | beat_b | Congrats team!
| |
20:00 | beat_b | €100,136EUR raised of €100,000 goal !!!!!!!!!!!!!!!!!!!!!!!!!!!!!
| |
20:00 | Bertl | thanks!
| |
20:01 | beat_b | I am sooooooooooo happy for all of us :-)
| |
20:01 | daFred | yeah baby yeahhh, all the best!!!!
| |
20:02 | g3gg0 | congratulations
| |
20:05 | wescotte | Now get the Horizon 2020!
| |
20:09 | KurtAugust | joined the channel | |
20:09 | KurtAugust | Apertus: Ok guys, get to work! Congratulations...
| |
20:10 | Bertl | thanks, but we are already working on the Beta :p
| |
20:12 | wescotte | You were clicking refresh 24/7 for the past 26 days!
| |
20:13 | Bertl | hehe, no, I was just clicking reload at the right time, i.e. right after we passed the goal :)
| |
20:13 | KurtAugust | left the channel | |
20:20 | jucar | left the channel | |
20:20 | jucar | joined the channel | |
20:24 | Legendin | joined the channel | |
20:26 | Bertl | wb Legendin!
| |
20:27 | troy_s | Et voila.
| |
20:27 | philippej | une bonne chose de faite.
| |
20:31 | se6astian | moved up the stretch goals in the campaign site
| |
20:34 | se6astian | WORLD NEWS brings the latest update: https://www.youtube.com/watch?v=8Q5wGasoiXE
| |
20:34 | aombk | i like the idea of backers funding the development, not prepaying for a future product
| |
20:34 | wescotte | I think you guys did a great job with your stretch goals. Often they have one serious one and all the rest are just wishful thinking. All of these feel very feasible both in the ability to deliver and the amount of extra funds to raise.
| |
20:34 | aombk | and gaining a discount in the process
| |
20:40 | comradekingu | se6astian: i think the changes to the explanation page make sense while the campaign is live too.
| |
20:41 | comradekingu | Those who already know are onboard, but it might win over more people who are only videographers atm
| |
20:41 | comradekingu | 100% :))))))))))))))))))))))))))))))))
| |
20:42 | g3gg0_ | joined the channel | |
20:43 | aombk | when will we get our buttons?
| |
20:43 | comradekingu | buttons?
| |
20:44 | g3gg0 | left the channel | |
20:45 | se6astian | Roberto Schaefer, ASC just emailed me: "right on! I'll chill one here too." (champagne)
| |
20:46 | Bertl | nice!
| |
20:50 | aombk | the buttorn perk
| |
20:50 | aombk | button perk
| |
20:51 | Legendin | left the channel | |
20:52 | comradekingu | i think the people who arent quite sure are still on the fence
| |
20:52 | Bertl | aombk: I guess as soon as possible, right when the campaign is over or so ... involves some logistics
| |
20:58 | troy_s | OK... Where's the camera?
| |
20:59 | Bertl | right beside me ... a few months in the future :)
| |
21:00 | philippej | See you all "tomorow" for the rest of the story!
| |
21:00 | Bertl | yeah, have a good night philippe!
| |
21:05 | philippej | left the channel | |
21:11 | beat_b | left the channel | |
21:21 | wescotte | troy_s: you didn't get yours yet? Strange... Mine shipped out the same day I donated!
| |
21:28 | Gegsite | left the channel | |
21:43 | troy_s | Bertl: I need a series of raws (the bayered footage) for a short.
| |
21:43 | troy_s | Bertl: Are there any of the shots in raw format out there? I need it for testing memory and such.
| |
21:45 | Bertl | testing memory in what way?
| |
21:46 | Bertl | I mean, why not simply take a raw or two and repeat them (forever)
| |
21:52 | troy_s | Bertl: Because I'm spending a teeny bit of time on a lab app and I need a whole shot.
| |
21:52 | troy_s | I could repeat them
| |
21:52 | troy_s | but I'm more interested in a real-world shot
| |
22:00 | troy_s | Bertl: Not fussy, but something with a human in it would be great.
| |
22:00 | Bertl | well, we only have time lapse in 4k raw, and it takes ages to upload
| |
22:01 | Bertl | because transferring a 4k raw image takes about 10 seconds on the Alpha
| |
22:04 | Bertl | and no human in it of course, just a garden from sunrise to sunset
| |
22:05 | Bertl | (maybe sebastian has done other time lapse tests)
| |
22:11 | troy_s | Bertl: I'd take anything at this point honestly.
| |
22:11 | troy_s | Bertl: Anything than a fricking still of an IT8 LOL
| |
22:28 | Bertl | a small sequence will show up here: http://vserver.13thfloor.at/Stuff/AXIOM/ALPHA/TL/
| |
22:28 | Bertl | takes about 10 minutes per frame to upload
| |
22:31 | daFred_ | joined the channel | |
22:32 | g3gg0_ | left the channel | |
22:34 | daFred | left the channel | |
22:52 | troy_s | Bertl: Love you. Thanks.
| |
22:52 | troy_s | Bertl: If I gave you an IAM share at S3, would that be faster?
| |
22:53 | troy_s | Or is the crimp at the upload side?
| |
22:54 | Bertl | it's the upload which is the limiting factor here
| |
22:54 | Bertl | but in a few seconds, you have already 3 frames :)
| |
22:58 | Bertl | they will be quite similar, as they are in short sequence, i.e. 60 seconds apart each or so
| |
23:04 | wescotte | left the channel | |
23:07 | __anton__ | joined the channel | |
23:07 | __anton__ | Congrats, great day :)
| |
23:08 | Bertl | thanks! and yes!! we made it!
| |
23:14 | se6astian | ok long day already
| |
23:14 | se6astian | good night!
| |
23:14 | se6astian | changed nick to: se6astian|away
| |
23:18 | Bertl | troy_s: do you need more than seven frames?
|