I recently purchased a Meade DSI III colour camera. Using Kstars/Ekos, no problem getting the camera recognized, however when I take an image a pop up note indicates that the Bayer Pattern is not recognized. As it is not a new camera I thought someone may have encountered the same problem and hopefully managed to find a solution./home/geoff/Pictures
When I press the OK tab on the pop up it does go to the FITS viewer. I think the result is in mono. (Need to get a bit more backfocus to get a c;lear image).
Any assistance appreciated
Geoff

You might find information in indilib.org/forum/ccds-dslrs/608-meade-d...rmware.html?start=18

-Eric

Thanks for that Eric. Unfortunately that didn't seem to be the answer for my problem. It is about Meade DSI versions I and II. I have version III which is quite a different chip.
I have searched all I can on the camera but cannot find exactly what bayer pattern it uses. Or if it is indeed a RGB pattern. As I am using Linux mint I can't download the Envisage program that Meade use for the camera.
At this point I am at a loss as to how to proceed.

From literature, it seems to be a CMYK pattern. Although you cannot use the FITS Viewer, you can still capture frames, and view them with an external tool I suppose. This is slightly annoying, but is it blocking for you?

-Eric

No, definitely, I'm wrong. It's not CMYK, it is RGB. It means the driver supporting that camera should not attempt to run a debayer, maybe that's a blatant bug.

-Eric

I am looking at this : www.hamiltonastronomy.com/ASTRO_ARTICLES/DSI_Pro_III.pdf

Does the INDI driver reports the right device version?

I think the issue is in the driver, dsi_ccd.cpp at line 136. The INDI glue considers there is a Bayer pattern as soon as the camera offers color frames. It is not the case for DSI III. Is it possible for you to rebuild the driver by changing this flag?

-Eric

Thanks again Eric, The little pop up that said "Unsupported Bayer pattern" was requiring me to press the OK button after each exposure, so I couldn't schedule a series of shots. However last night I found that by turning off the "auto debayer"in the FITS control panel it prevents the pop up and I can schedule a series of shots. So now to see if Siril will hopefully recognize the pattern, I tried it in GIMP however that still only gave me a mono image.

Changing the driver is not within my capabilities I'm afraid. I will ask around the Tasmanian astro society if there is any one that may assist me.
The Indi Cotrol panel gives me:
Name: DSI
Exec: indi_dsi_ccd
Version: 0.3
Interface: 2
It doesn't mention colour

Siril also just treats the image as mono also

It is possible that your dsi is not fully recognised by the driver. There should be a few lines of log when the driver starts, stating which device was detected. Do you get the "DSI Pro III" mention or the "unknown CCD" with the type of sensor?

On which computer are you running the dsi driver?

-Eric

My computer is a HP 6560B laptop running Linux Mint 20 Ulyana base: Ubuntu 20.04. On the log it only mentions Meade DSI no model.
I think that this is the part of the log yhatyou mentioned should indicate the DSI III.


[2021-01-20T21:08:22.055 AEDT INFO ][ org.kde.kstars.indi] - Meade Deep Sky Imager : "[INFO] Found a DSI with an unknown CCD: ICX285AQ "
[2021-01-20T21:08:22.057 AEDT INFO ][ org.kde.kstars.ekos] - "Meade Deep Sky Imager" Version: "0.3" Interface: 6 is connected.
[2021-01-20T21:08:22.057 AEDT DEBG ][ org.kde.kstars.ekos] - 1 devices connected out of 1
[2021-01-20T21:08:22.057 AEDT INFO ][ org.kde.kstars.ekos] - All INDI devices are now connected.
[2021-01-20T21:08:22.261 AEDT DEBG ][ org.kde.kstars.indi] - < Meade Deep Sky Imager >: < CCD_EXPOSURE >

The log excerpt matches my assumption. Your DSI is recognised, but not exactly as it should. I'll try to rebuild indi_dsi_ccd for your system to debug the situation. Could you open an issue on the INDI github?

-Eric