| 00:04 | gcolburn | left the channel |
| 00:10 | gwelkind | joined the channel |
| 01:10 | jucar | left the channel |
| 01:28 | gwelkind | left the channel |
| 01:33 | Bertl | off to bed now ... have a good one everyone!
|
| 03:33 | Lunastyx | left the channel |
| 03:33 | Bertl | left the channel |
| 03:33 | djp | left the channel |
| 03:33 | troy_s | left the channel |
| 03:33 | danieel | left the channel |
| 03:33 | FergusL | left the channel |
| 03:34 | PipeManMusic | left the channel |
| 03:34 | slikdigit | left the channel |
| 03:34 | mars_ | left the channel |
| 05:16 | mars_ | joined the channel |
| 05:16 | PipeManMusic | joined the channel |
| 05:16 | slikdigit | joined the channel |
| 05:16 | danieel | joined the channel |
| 05:18 | gwelkind | joined the channel |
| 05:18 | FergusL | joined the channel |
| 05:18 | Lunastyx | joined the channel |
| 05:18 | Bertl | joined the channel |
| 05:31 | troy_s | joined the channel |
| 06:04 | gwelkind | left the channel |
| 08:38 | rainer_vie_ | joined the channel |
| 08:39 | rainer_vie_ | seems like the blackmagic 4k camera uses also the cmv12000 http://forum.blackmagicdesign.com/viewtopic.php?f=2&t=20281&sid=99f3f806adfe8a0aa570ed0160df7fd9&start=50#p129148
|
| 09:18 | troy_s | left the channel |
| 09:19 | troy_s | joined the channel |
| 09:37 | se6astian | joined the channel |
| 09:38 | se6astian | good mornin
|
| 10:03 | Bertl | morning folks!
|
| 10:18 | jucar | joined the channel |
| 10:42 | rainer_vie_ | left the channel |
| 11:12 | jucar | left the channel |
| 12:04 | danhanes | joined the channel |
| 14:38 | jucar | joined the channel |
| 15:04 | gcolburn | joined the channel |
| 15:10 | gcolburn | left the channel |
| 15:14 | jucar | left the channel |
| 16:57 | jucar | joined the channel |
| 17:01 | troy_s | Bertl: It is where I suggest making the issues well known.
|
| 17:02 | troy_s | Bertl: Not fair to other groups; the BM folks that spent hard earned money on that crap sensor.
|
| 17:02 | gwelkind | joined the channel |
| 17:14 | gwelkind | left the channel |
| 17:50 | gwelkind | joined the channel |
| 18:39 | tyrone_ | joined the channel |
| 18:45 | tyrone_ | i just have seen on the archive that they say that bm is using the cmv12000.. from my point v view i don't think so because in the technical spec of the bm website they give in the sensor format current value: 21.12mm x 11.88mm and that's not a cmv12000... for me it's a custom design or a sensor that i couldn't find....
|
| 18:45 | danhanes | left the channel |
| 19:03 | troy_s | tyrone_: Good spot. Is that in that thread?
|
| 19:03 | troy_s | tyrone_: There has been much speculation that they are in fact using a CMV chip.
|
| 19:04 | Bertl | chances are good that it still is a cmosis chip
|
| 19:04 | Bertl | at least the images showing the bonding look very similar
|
| 19:05 | tyrone_ | well i think it's because some time ago someone said something like that but when you look the spec's and the price it doesnt make sens for this cmv12000 sensor. a custom one from cmosis yes that would make ore sens but not this one.
|
| 19:13 | troy_s | Well if Bertl's demos are any indication
|
| 19:13 | troy_s | There's a heck of a lot of data that is completely corrupted
|
| 19:13 | troy_s | (I really wonder how BM deals with that. I wonder if they clip it off too Bertl?)
|
| 19:14 | troy_s | (Any method you might be able to deduce?)
|
| 19:14 | tyrone_ | it's just a no to cmv12000... but not to cmosis.... so it could be....
|
| 19:14 | Bertl | I have no clue what BM does, I haven't even looked at their cameras (yet) and I'm not sure I want to ...
|
| 19:15 | Bertl | I think it is better to focus on our work and the feedback we get (and will get) from testing with e.g. the axiom beta
|
| 19:17 | tyrone_ | +1 for bertl.... your right...
|
| 19:23 | se6astian | somebody posted about us on reddit: http://www.reddit.com/r/photography/comments/22x9ry/axiom_alpha_camera_is_this_the_bridge_between/
|
| 19:25 | troy_s | Bertl: It is obviously not terribly interesting, but being an open project, I can only think that educated assessments of the sensor is fair to release.
|
| 19:32 | Bertl | yes, but there is nothing new we figured about the cmosis sensor, what cmosis hasn't already published in one of their annotations
|
| 19:33 | Bertl | (despite the fact that we found it before the ans were released :)
|
| 19:35 | danieel | tyrone_: the sensor size is not that.. it is the active area
|
| 19:35 | danieel | the 3840 x 5.5um =21.12 mm
|
| 19:36 | troy_s | Bertl: Does the ann speak of crufty data at the upper end?
|
| 19:37 | _rifoRim | joined the channel |
| 19:43 | tyrone_ | danieel: well i find it's not the cmv12000 but it could be a custom one... that safes you cost on the silicon and driver electronics.... i don't say it's not from cmosis....
|
| 19:45 | Bertl | troy_s: as I tried to explain, there is no 'crufty data' it is just that the sensor doesn't use the full range with the settings we used so far
|
| 19:45 | Bertl | it is hard to tell if it is a configuration problem or a sensor design problem
|
| 19:46 | danieel | if you have a BM, take a picture of the sensor so we do not have to judge it by the renders on the website
|
| 19:46 | danieel | then we can tell if it is that or not
|
| 19:46 | danieel | Bertl: both ES and production does the same thing?
|
| 19:48 | Bertl | yes
|
| 19:48 | troy_s | danieel: I don't think that is good enough to reveal clipping.
|
| 19:48 | troy_s | Bertl: It doesn't use them because of some secret magic register code or?
|
| 19:51 | Bertl | there could be a multitude of reasons for that, a flip of an undocumented bit in the register set could fix it
|
| 19:51 | Bertl | even setting certain known register values to 'better optimized' settings could fix the range issue
|
| 19:52 | Bertl | for now, I don't want to spend any more time on figuring out stuff which should have been documented by cmosis in the first place
|
| 19:53 | Bertl | we might be able to setup a remote test platform with the axiom alpha though
|
| 19:53 | Bertl | so if somebody wants to play around with the cmosis register settings, I think that could be arranged
|
| 19:54 | Bertl | (i.e. place the camera somewhere with a led lighting facing the old chart with ssh access to the axiom)
|
| 20:01 | tyrone_ | i have got here just the 2.5k version no 4k :-/
|
| 20:01 | danieel | Bertl: is the 2.5k version similar in programming interface / pixel architecture ?
|
| 20:02 | danieel | you could get an ES of that, put it in alpha and test if you still get the issue
|
| 20:03 | gwelkind | left the channel |
| 20:03 | troy_s | Bertl: Amen on the documented part.
|
| 20:03 | troy_s | Bertl: Web based tweaking? LOL
|
| 20:03 | Bertl | danieel: 2.5k version?
|
| 20:04 | danieel | Bertl: sorry, that is another sensor maker (fairchild, not cmosis)
|
| 20:04 | Bertl | troy_s: yeah, why not
|
| 20:04 | danieel | troy_s: you can log the commands sent and images getting out... and if you detect it is okay, then look back the log :)
|
| 20:09 | troy_s | danieel: That would be pretty damn cool.
|
| 20:10 | troy_s | Cute way to 'Here... play with the camera folks" too.
|
| 20:10 | troy_s | Put it on a cheapo USB controlled pan unit... LOL
|
| 20:12 | Bertl | let's see what se6astian thinks and if he finds the time to set something up (e.g. at the university)
|
| 20:22 | danieel | Bertl: with tyrone we have confirmed partly this article: http://www.eoshd.com/content/12053/successor-blackmagic-cinema-camera-cmos-sensor-announced-possible-hardware-hdr-mode
|
| 20:23 | danieel | the sensor in 2.5k blackmagic cinema camera is Fairchild (BAE) CIS2521F - http://www.fairchildimaging.com/catalog/focal-plane-arrays/scmos/cis-2521f
|
| 20:24 | danieel | the pocket camera is FHD, while the sensor size matches (6.5um), but the bonding and the alignment mark does not (or the image from BM is render/photoshop)
|
| 20:29 | Bertl | still not seeing how that helps us in any way :)
|
| 20:31 | danieel | looking at those models would not help at all - now we know for sure ! :)
|
| 20:33 | danieel | can you notice the flaw in real picture? or it is only a calibration issue?
|
| 20:33 | danieel | could we detect it when somebody would have the BMPC4K ? (i am not sure if they actually have it solved)
|
| 20:42 | _rifoRim | left the channel |
| 20:58 | gcolburn | joined the channel |
| 21:13 | gcolburn | left the channel |
| 21:13 | se6astian | time for bed
|
| 21:13 | se6astian | good night
|
| 21:13 | se6astian | left the channel |
| 21:24 | troy_s | danieel: rexbron has a BMC 4k
|
| 21:34 | danieel | ok, I ask him for the sensor bonding photo
|
| 21:35 | danieel | it would be nice to confirm that BM had solved the issue.. then it would be possible to take the risk and snoop the SPI - but that would either solve it or confirm that the issue is fixed in processing
|
| 21:35 | danieel | gn
|
| 22:00 | troy_s | danieel: See ya.
|
| 22:29 | intracube | joined the channel |
| 22:58 | gwelkind | joined the channel |
| 23:04 | jucar | left the channel |