An Easier UniWB for EOS Digital Cameras (at least the XT)

E-K wrote in post #5492290
Basically UniWB is something used by people trying to expose to the right. The idea is that it will give you a more accurate view of whether something is really blown.

WB works by multiplying the various channels by a number (the coefficients being talked about). This adjusts the relative proportion of each channel.

Think of UniWB as the anti-WB. The coefficients are ideally all 1, so each channel is given the same priority.

e-k

Ahhhhh. Ok. So if Im shooting RAW, I set this uni-WB in the camera so I can look at my histogram and then just fix the WB later in post, right?

AdamLewis wrote in post #5492305
Ahhhhh. Ok. So if Im shooting RAW, I set this uni-WB in the camera so I can look at my histogram and then just fix the WB later in post, right?

Exactly.

There's no need to adjust this Uni-WB for the existing light conditions? Is there anything else that has to be adjusted (image style settings (contrast, sharpness, etc), perhaps) to fully utilize this?

mnealtx wrote in post #5492950
There's no need to adjust this Uni-WB for the existing light conditions? Is there anything else that has to be adjusted (image style settings (contrast, sharpness, etc), perhaps) to fully utilize this?

It's best to shoot in Neutral picture style, as that does not push certain colours and leaves saturation, contrast and sharpening flat. Do not boost these settings by modifying the default values of the picture style. I do not know whether it is more accurate or less accurate to slide them all as far to the left as they can go, but certainly do not move them to the right.

Once you're happy shooting with Uni-WB you just leave it there for all shots, day/night/indoors and out. You fix your WB in raw processing afterwards.

Been trying with a dark frame but can´t seem to get the coefficients close to 1. I tried a couple of times but get 2.2 1.0 1.5 or somthing around that. i´ll try with white.
I´m really over my head with this stuff but it´s fun to experiment.
What i found out is that the green cast makes the lcd almost useless, have to rely only on the histogram.

Interesting...thanks for the info! I may have to give it a try

ADAPTE wrote in post #5493169
Been trying with a dark frame but can´t seem to get the coefficients close to 1. I tried a couple of times but get 2.2 1.0 1.5 or somthing around that. i´ll try with white.
I´m really over my head with this stuff but it´s fun to experiment.
What i found out is that the green cast makes the lcd almost useless, have to rely only on the histogram.

I couldn't get dark frame to work with my 30D but had no trouble with my 40D. A saturated frame works well with both bodies.

If the one thing the green cast does is to make people use the RGB histogram to judge exposure instead of the preview image then that is a great step forward

ADAPTE wrote in post #5493169
Been trying with a dark frame but can´t seem to get the coefficients close to 1. I tried a couple of times but get 2.2 1.0 1.5 or somthing around that. i´ll try with white.
I´m really over my head with this stuff but it´s fun to experiment.
What i found out is that the green cast makes the lcd almost useless, have to rely only on the histogram.

What camera are you using?

e-k

I tried a Dark Frame with my 1d3.

I didnt put a lens on, just snapped a shot with the body cap on in manual mode.

Opened it directly from the card with UFRaw and it read 2.023, 1.000, 1.484.

I changed them all to 1.000 and saved it rite on the card.

I then set that shot as my white balance and took a few shots, it seems to have worked.

Edit..

I opened the test shots up with UFRaw and all of them read R=1.000 G= 1.000 B=1.000 when I change it to Camera WB.

wilvoeka wrote in post #5493874
I tried a Dark Frame with my 1d3.

I didnt put a lens on, just snapped a shot with the body cap on in manual mode.

Opened it directly from the card with UFRaw and it read 2.3, 1.3, 1.9 or something similar.

The dark frame image would just have WB coefficients for whatever your current WB setting was. You need to set this image as the CWB, then take another image. This image should then have WB coefficients of around 1.0, 1.0, and 1.0, assuming everything went well .

e-k

E-K wrote in post #5493919
The dark frame image would just have WB coefficients for whatever your current WB setting was. You need to set this image as the CWB, then take another image. This image should then have WB coefficients of around 1.0, 1.0, and 1.0, assuming everything went well .

e-k

So if I shoot a dark frame, regardles of what the multipliers read set it as the WB.

If I then was to set it as the white balance and then shoot a shot it should come out 1.0 1.0 1.0( or close) without doing anything?

Ok I answered myself by trying it.

Took a black frame, set the white balance with it and shot another black frame. The second black frame reads 1.000 straight across when set to Camera WB.

You got it. You don't have to take a second black frame, but I'm assuming that was just for testing (i.e. you can just use the first dark frame as your CWB).

e-k

E-K wrote in post #5493972
You got it. You don't have to take a second black frame, but I'm assuming that was just for testing (i.e. you can just use the first dark frame as your CWB).

e-k

Yeah I just wanted to see what the second shot read after setting the WB.

E-K wrote in post #5493625
What camera are you using?

e-k

Xti

ADAPTE wrote in post #5494631
Xti

Okay, just to confirm the steps you followed,

1. Shot a dark frame.
2. Set the dark frame as the custom white balance image.
3. Set WB to custom white balance.
4. Took an image using the custom white balance.
5. Looked at the WB coefficients in UFRaw with "Camera WB" selected of the image taken in step 4.
6. When you looked at the WB coefficients in step 5 you see they are not near 1.

If that is the case then it doesn't look like it works for the XTi . If instead you looked at the WB coefficients of the dark frame, then it's expected that they may not be anywhere near unity.

e-k