Color Matrix adjustments?
 

I tried to read what has already be posted on the camera "color matrix" adjustments and Canon's doc, and I also tried to experiment a bit, but I am hopelessly confused.

For a start: can someone please explain me why there are 2 adjustements per color? For example, on Canon's doc I read:

G-B matrix: This adjusts the color tint without affecting R (red).
B-G matrix: This adjusts the color tint without affecting R (red).

So what is the difference between G-B and B-G matrices?


Apparently, the "console" software is able to show the effect of R, G and B matrices in its "variation" window. Maybe someone could post a screen capture of this window?

See http://www.dvinfo.net/conf/showthread.php?t=86102

and http://www.dvinfo.net/conf/showpost....2&postcount=15

I'll look for a screen grab, but you can download a trial version of Console and see it for yourself.

Edit: see a grab of the variation window at the bottom of this page: http://www.usa.canon.com/app/html/XL...ec_panel.shtml

I found those posts already, but the tests I tried do not appear consistent with this explanations. Increasing the R-G matrix, a red apple should turn green (or brown= red + green), but it does not happen. And those posts do not explain why the matrixes have negative values (how can you remove green from a red apple?).

As soon as they have a mac version, I will.

That is the screen grab which made me aware of the "variations" window. What I need is someone to select the "Red matrix" (or green, or blue) on that window and to post THAT screen grab, thanks.

Jerome, increasing R-G will do nothing to a red apple. This adjustment is going to put red into the green channel. A red apple will stay red, but a green apple would start to turn brown. If you want to change the color of the red apple, then you would adjust G-R which would allow green to be added to the red object.

-gb-

These grabs from the Console variations screen are from an XL H1, however the XH A1 will be very similar, but with a wider range of adjustment. The target is a DSC Labs CamBook. Sorry it's so quick and dirty (the dark blob at the one o'clock position of the test chart is simply a reflection of the XL H1 stereo microphone). Color, R Matrix, G Matrix, B Matrix. Click to see them bigger. Hope this helps,

Increasing R-G does change the reds, try it for yourself. (edit: or try the "convar-rmatrix.jpg posted above)

Exactly what I needed! Thanks a lot!

So from the screen grabs Chris was kind enough to post, it seems that the answer to my question is:
G-B and B-G both affect the saturation of the G and B colors without affecting R colors. The difference appears to be that G-B affects the saturation of B more than G and conversely B-G affects G more than B.

blue red problem
 

Hey guys, reading your post almost 3 yrs later, I have a problem with theatre shows. The dresses in this picture are purple(except for the center gray one) and they're supposed to be burgundy. They are missing the red tint. Any suggestions about fixing this? It's about a 40 minute show and each time the girls come out in these dresses, they are the "wrong" color. Other colors are resolving seemingly well. It appears to be red/blue issue.
Seems like I want to try the R-B matrix? Add red to the blue? I'm also going to lower the blue gain about -10 and increase the red gain about the same.
thx
Canon XH-A1

David, it's not something to do with coloured stage lighting is it?

I think it'll be easier if you load a custom preset and experiment with that, get as close as you can and colcorrect it in post.

I'd use VIVIDRGB and increase the red. The problem will be assessing it on site,
HTH Cheers.

not sure what u mean by colored? Gels? the lighting guy assured me that the color temp and gels etc... were all tungsten family. He said there were no daylight elements involved with the lights at all. so......

I meant the colour of the house lighting, coloured spots and stage fill lighting etc.

Cheers.