Current Server Time: 12:53 (Central Europe)

#apertus IRC Channel Logs

2013/10/28

Timezone: UTC


22:58
Sasha_C
left the channel
00:35
Sasha_C
joined the channel
01:14
FergusL
left the channel
07:30
se6astian
joined the channel
07:35
se6astian
good morning
07:36
Sasha_C
Good morning Se6astian. How are you today?
07:41
se6astian
good good, many things to do but thats normal :)
07:41
se6astian
and you?
07:43
Sasha_C
I'm well thanks, but like you, I'm very busy too
07:43
Sasha_C
I won't have a chance to continue writing another article up until a couple of days from now
07:44
se6astian
ok, thats fine
07:44
Sasha_C
I watched Oscar's short film (Spring of Cesar) on the weekend with my girlfriend. It's really good! Great imagery and performances.
07:44
Sasha_C
I only wish things were a little more resolved at the end :(
07:53
se6astian
ah great, you got the dvd in geneva?
07:55
Sasha_C
yep
07:56
Sasha_C
I was surprised with how nice the footage looked, seeing that it was shot on a 5d Mk II.
09:17
mars_
joined the channel
10:12
se6astian
hi mars_, what brings you to us?
10:13
mars_
your mail from friday ;)
10:14
mars_
pcb/fpga engineer
10:15
se6astian
ahh, Markus?
10:15
mars_
yeo
10:15
mars_
yep
10:15
se6astian
great!
10:16
se6astian
Bertl, should be awake soon, he will be able to give you an update on the latest tech development progress
10:16
se6astian
I can in the meantime answer any general project questions you have
10:17
mars_
thx, i'll just lurk here for a while
10:17
se6astian
good :)
10:38
Bertl
morning everyone!
10:39
Sasha_C
Good morning Bertl. I think someone is here to see you
10:39
se6astian
Ah the man of the hour, hello
10:39
Bertl
Sasha_C: yep, thanks :)
10:40
FergusL
joined the channel
10:48
se6astian
ok I am off to pick up the acryl glas for the prototype enclosure
10:48
se6astian
they replied that order is finally ready for pick up :)
10:48
Sasha_C
Sounds awesome! Good luck
10:57
FergusL
cool !
11:20
FergusL
left the channel
11:21
se6astian
left the channel
11:22
Sasha_C
left the channel
11:22
FergusL
joined the channel
11:42
Sasha_C
joined the channel
12:21
Sasha_C
left the channel
13:04
se6astian
joined the channel
13:04
se6astian
back
13:05
se6astian
the acrylic glas is ours!
13:05
mars_
pix or it didn't happen!
13:06
se6astian
I see you already fitted in perfectly ;)
13:07
se6astian
I left it in my car, doesnt look that spectacular they way it is packed
13:14
Bertl
\o/
13:53
Bertl
mars_: are you subscribed to the development mailing list?
13:53
mars_
not yet
13:55
FergusL
left the channel
13:58
Bertl
check with se6astian, he will probably invite you to the relevant mailing lists/forums
13:59
se6astian
will do right away
13:59
mars_
thx
16:21
Bertl
mars_: so any comments on the sensor adapter yet?
17:08
se6astian
how to not raise money: http://www.kickstarter.com/projects/725991125/open-source-graphics-processor-gpu :)
19:54
[1]se6astian
joined the channel
19:57
se6astian
left the channel
19:57
[1]se6astian
changed nick to: se6astian
20:01
dmj_nova
joined the channel
20:01
dmj_nova
Bertl: excellent job!
20:02
dmj_nova
About that sensor, are such defects to be expected or is it because it's a particularly low-grade sample?
20:12
[1]se6astian
joined the channel
20:15
se6astian
left the channel
20:15
[1]se6astian
changed nick to: se6astian
20:19
Bertl
dmj_nova: the ones on the right I'd say can be expected, at least for lower grade
20:19
Bertl
the one on the left seems extreme
20:20
dmj_nova
yeah, the ones on the right can likely be smoothed out without much notice too
20:20
dmj_nova
the one on the left...not without severely limiting the dynamic range
20:21
Bertl
the funny thing is, when scaling down, as in the commentary picture, the effects on the right basically disappear without even removing them
20:22
Bertl
so I'd say, if you know that a particular column is bad, it's rather trivial to remove
20:23
Bertl
from the way the sensor works, I don't think that you can get any 'row' failures, only bad pixels and those column failures
20:26
dmj_nova
Yeah, you could probably even retain some of the detail observed by the row while color/brightness correcting it based on adjacent rows
20:27
dmj_nova
(to avoid aliasing issues)
20:27
dmj_nova
actually, no, not quite
20:28
dmj_nova
if an entire row is stuck, it won't observe anything
20:28
Bertl
I guess there will be different errors
20:28
Bertl
if you check the effects on the raws (right side) you already see a bunch of different effects
20:28
dmj_nova
but the adjacent rows may be part of its RGB pixel
20:29
dmj_nova
so perhaps part of the process will be to characterize the effects.
20:29
dmj_nova
and build a program for it.
20:29
Bertl
yep, and I guess we will see many different ones over time :)
20:30
dmj_nova
Will these defects change over time?
20:31
Bertl
no idea
20:32
Bertl
but I guess it would help to have some kind of calibration routine which detects changes and maybe even allows to address them
21:26
se6astian
good night!
21:26
se6astian
left the channel