Weird master Flat from DSS

[philming]

Hi all !

I'm processing my images through DSS.

A friend of mine noticed something weird on ALL my master Flats, that one can find in the source folder of the images.

Just see by yourself on the screenshot underneath :

http://www.webastro.net/upload/images/10772-1331653501.jpg

The source flats are all just OK.

The way I process my images is as follows :

Open software, clic "add images", go get the subs then do the same for darks, flats, and offsets/bias. I never took care of the "dark flats", not knowing what they are.

I make my flats with a Geoptik flat box, and the offset/bias by just shooting as fast as I can in the dark. (like a dark, but as fast as the camera can shoot).

Then I just review the settings (flats are set to be calculated as "kappa sigma clipping median" 2/5) check the advised settings, and off it goes.

Can someone tell me if this is a problem ? Am I doing something wrong ? Is DSS screwing up my pictures ?

Thanx for your input !

BTW, I just checked my darks, and the master dark sort of has the same problem... as well as the master offset.

Edit : I guess some sort of "squaring" is normal, because of the Bayer matrix, but the variation between the greys seem very important when coming from DSS. A friend has showed me a master flat coming from Iris, and the level difference was much narrower (he gets a few percents difference, I get near 30%...)

[philming]

up ?

[dugpatrick]

It looks like the bayer matrix to me. I'm assuming you use a one shot color camera, such as a DSLR.

I haven't stopped to examine my master flat files but what you show is probably normal.

Doug

[dugpatrick]

Here's reference, and note the screen pattern in figure 2:

http://www.stark-labs.com/craig/articles/assets/Debayering_API.pdf

The screen pattern is what you actually get from the camera.

Doug

[martin_h]

I wouldnt stack the flats with cappa sigma...use median instead...dunno if it will help

[philming]

It's just that I tried this on Iris as well. If you zoom into the master in Iris, you just get thin lines, with a variation of a few percents for this matrix, on a single pixel row. With DSS, the matrix is at least 3 pixels with variations between the black area and the light area of about 30% difference in black level...