#apertus IRC Channel Logs

good morning

is visible in the nonlinearity images after a contrast stretch: http://i.imgur.com/S8g32Eo.jpg

a diff of green with either red or blue channel + further contrast stretch: http://i.imgur.com/3Qvbd1x.jpg

yeah, I need a better clip frame (I'm experimenting with something that accounts for nonuniform illumination of the wall, but no meaningful results yet)

I'm also trying to recover the response curve with other methods; unfortunately, with existing programs like mkhdr or pfshdrcalibrate, I've got only bogus results

heh

oh ok

(I don't have a beta yet)

I also thought at something that would flip the exposure back and forth near the clipping point - if I let that run for a while, I guess it should give pretty exact per-pixel curves

you mean the granularity in the highlights (particularly saturated areas)?: http://i.imgur.com/qVfdkGs.jpg

that throws off the calibration, so if I want to get per-pixel curves from the existing dataset, there's a ton of noise in the data

if you repeat it 64 times => noise reduced by 3 stops

so in theory, it could be reduced a fair bit

so the noise would have zero mean (that is, so it won't introduce color casts)

and then there's the repeatability issue

and this profiling would have to be done for each individual sensor

yeah

same for dark frames

yeah, I want to have these calibration routines in the camera

I thought it was generally linear

on Canon, it seems to clip quite harshly

on 5D3, I turned down all the gains I could identify, so it no longer clips to white

and got some fixed-pattern noise in highlights

first, the CMOS charges a capacitor to a certain level

this process is not really linear per se

also, it has a roll-off when the capacitor reaches zero voltage

assuming that the actual sensel is operated in the linear range

you'll get a nice and proper voltage to digitize

i.e. you apply analog gain, which has nonlinearities at the rails

given that everything is optimal, you will only see a proper linear range

this

it just means that the full 12bit range will not map to a linear range in all gain settings

you're welcome!

off for now ... bbl

anyone currently here speak spanish?

or maybe better? ¡Lo siento, no hablo español!

LOL

Bertl_oO: intracube noticed that there's some odd glow in the lower right corner. Any clue as to what this might be? Light leak / splash off of an LED? Heat? Electronics?

Bertl_oO: http://i.imgur.com/S8g32Eo.jpg

troy_s: that corner was not completely overexposed, but the rest of the image was (obvious if check the other images in the set)

once you go past some overexposure level, the sensor output starts to decrease

And why.

Or heat, or magnetic magic fubar or who knows.

you will no longer blame cosmic radiation after you do that :P

check from 30 to 90ms for example

alexML: I don't blame anything. I'm just leaving everything open. Those shots are hooped.

So could very well be an LED leak off of one of the boards.

it's not visible in dark frames, so I doubt it's a LED leak

That's my point.

but please - see the entire sequence before judging

Does it flicker?

no, they are not

ooh

Dark frames are portal capped images then?

ok, that does change things slightly. but this issue only seems to be on the green channel

That's odd.

they all behave the same electrically, but green is usually receiving more light (hint: white balance)

This is off of the raw data.

yes, if you render raw data without white balance, what color cast do you get?

There's no colour.

just to confirm; I'm not seeing anything on a dark frame image

intracube: Then it must be across all data channels if it is a leak

but you'd expect to see something on the other channels, I guess

in the middle of the image, r=g=b (roughly, on raw channels, no colorimetry)

alexML: So it's a leak?

no

well, I was wondering about the corner as well (on the image intracube "corrected")

that's how a white paper appears when it's close to overexposure, and uniwb

try it on Canon, Nikon, whatever you want

It is a leak of some sort.

but then I saw the other image somebody visiting sebastian made from the setup

Bertl_oO: Difference?

Flare leak.

I didn't think twice when I first saw it and just assumed it was a glow from a lamp.

Slightly.

It has falloff like a lamp.

I mean as far as I am concerned anyways, it's all moot until the next iteration comes in with proper latitude.

Bertl_oO: Would you be able to get a camera set up such that it could be remotely controlled?

(God. Must go get Cintiq Blech.)

so don't read too much into the shape of it

upload the result here

with a bunch of remote controlled switches to change the lighting, it should probably work

Bertl_oO: should put a rotate data channels into the format so folks done erroneously call them colours

Very keen on PWLC

troy_s: we are looking at low-level sensor output, before the matrix gets applied

I know this

Notice how I never say colour when looking at data?

Data

troy_s: this is the data off the r, g, b sensels though. so you can legitimately call them that

no reference? the sensels have colour filters, no? :)

If you rotate those channels the data is equivalent

we don't have that here, so we can't rotate yet

that's it

can you show the raw (unmodified) one?

Say what?

and just debayered

convert \( -size 4096x3072 -depth 12 gray:input_raw_image.raw12 \) \( -clone 0 -crop -1-1 \) \( -clone 0 -crop -1+0 \) \( -clone 0 -crop +0-1 \) -sample 2048x1536 \( -clone 2,3 -average \) -delete 2,3 -swap 0,1 +swap -combine output.png

let's say from 1 to 100ms

in particular, compare the 75ms one with the 88ms one (since I didn't download past that)

alexML: I'll have a go later, yep

Did you rotate the channels and characterize yet? ;)

I expect the exact same results actually.

12:55 <alexML> troy_s: to be able to rotate the channels, you first need linear response (not clipped stuff)

proof please

Not the damn matrix

Hence why I keep repeating they ain't colors.

if so, please remind me when I'm back from my nap ...