Page 1 of 1

Problems with Dither and Debayer

Posted: Fri Jun 18, 2021 6:36 am
by TigerClaw35
Hi,

yesterday I tried the SharpCap 4.0 32bit Beta. First I notice that if Dithering is running, in PHD2 there is no notice for this. With other tools PHD2 shows me "Dithering" string.

Second I used "debayer on" and also Force RGGB for my QHY 268C. Bot for both it's GBRG if I try to load it in APP. With SharpCap 3.2 I have no problems and it runs as RGGB. I am sure it's a bug.

Re: Problems with Dither and Debayer

Posted: Fri Jun 18, 2021 2:10 pm
by admin
Hi,

First on the guiding - I have never noticed a full notification in PHD2 when guiding, but I always see a 'Dither' mark in the guiding graph when SharpCap requests a dither, like this.
Capture.JPG
Capture.JPG (26.46 KiB) Viewed 268 times
Do you see that? Are you expecting something more than that? Maybe I should update my PHD2 (on 2.6.9 test).

Now, FITS headers.... Aaarrghh. :)

Unfortunately, although there is a correct way round for the FITS headers, lots of software gets it wrong. You can load the same FITS file into PixInsight and AstroArt and find that it requires different bayer patterns in the two applications to debayer correctly.

SharpCap 3.2 was definitely getting it wrong - it was writing out the FITS headers and image data as if the image started top left, but the FITS standard says it should start bottom left. That wasn't too much of a problem because most of the processing software out there ignores the FITS headers anyway and about 1/2 of it is also wrong in the interpretation...

In SharpCap 4.0, I tried adding a new fits keyword that processing applications are starting to support - ROWORDER - that allows SharpCap to specify that its images are top-down. With software that understands that the current SharpCap 4.0 output is correct.

What I didn't consider in enough detail was the effect of this change on people used to SharpCap 3.2 - basically the headers have flipped to try to be more correct but have broken your workflow. I think I will need an option for this :(

cheers,

Robin

Re: Problems with Dither and Debayer

Posted: Fri Jun 18, 2021 3:50 pm
by admin
Hi,

could you test this image in APP please?

https://drive.google.com/file/d/1YM4Lws ... sp=sharing

The word 'STATION' should be the right way up and in red.

I'd be interested in the answers to the following questions

1) Is the image right way up or upside down
2) Does the application auto detect the bayer pattern on this image?
2a) Did it get it right?
2b) If wrong, what pattern did it use?
3) What bayer pattern is needed to give the correct colours?

So far in my testing, the bayer patterns needed for correct colours are:

RGGB : Siril, PixInsight, Astap, DeepSkyStacker
GBRG : PIPP, AstroArt

The following applications show the image upside down : Astap, AstroArt, PiPP, Astap
Right way up : PixInsight, DeepSkyStacker

If anyone else has any other processing software that they can test, I'd be interested in the results.

cheers,

Robin

Re: Problems with Dither and Debayer

Posted: Tue Jun 22, 2021 4:53 am
by TigerClaw35
Hi,

I have no "Dithering" mark in PHD2. This is what I would expect.With other tools it's working fine.

I tested the image.

1) it's correct oriented
2) Yes, it autodetect it. But if I use tools like Sequence Generator Pro they create "16b GRAY" instead of "16B RGB" so I have to force bayer pattern. But it works.
2a) no, it autodetects "16b RGB
3) If i select: Force Bayer than it shows me correct colors and for GBRG

But for my camera it's RGGB so the result of a SharpCap image is still GBRG and I it will not work.

Re: Problems with Dither and Debayer

Posted: Tue Jun 22, 2021 5:37 pm
by admin
Hi,

well I am still a bit confused about the APP colour issue at the moment, but it seems that you have a workaround for that by overriding the default, so let's concentrate on the guiding issue.

Please can you capture a log file from SharpCap showing the dithering issue? When I test I see the following

Code: Select all

Info   	18:33:57.653074	#33	Live stacking paused : PHD2 is currently dithering.																		
Info   	18:33:57.653074	#33	Requesting dither from PHD2 1, False, 2, 5, 60																			
Info   	18:33:57.653907	#33	Guiding SettleState changed to Dithering																				
Info   	18:33:57.654740	#3 	Event received from PHD2 : LockPositionSet																				
Info   	18:33:57.656404	#3 	Event received from PHD2 : GuidingDithered																				
Info   	18:33:57.657235	#30	Event received from PHD2 : SettleBegin																					
Info   	18:33:57.657235	#30	Guiding SettleState changed to Settling																					
Info   	18:34:02.656903	#32	Event received from PHD2 : Settling																						
Info   	18:34:02.660227	#32	Event received from PHD2 : GuideStep																					
Info   	18:34:02.660227	#32	Event received from PHD2 : Settling																						
Info   	18:34:02.661079	#32	Event received from PHD2 : GuideStep																					
Info   	18:34:02.661079	#32	Event received from PHD2 : Settling																						
Info   	18:34:02.661079	#32	Event received from PHD2 : GuideStep																					
Info   	18:34:02.661079	#32	Event received from PHD2 : Settling																						
Info   	18:34:02.661920	#32	Event received from PHD2 : GuideStep																					
Info   	18:34:02.661920	#32	Event received from PHD2 : Settling																						
Info   	18:34:02.661920	#32	Event received from PHD2 : GuideStep																					
Warning	18:34:02.723129	#33	Frame not stacked because : Live Stacking is paused - PHD2 is currently dithering.										
Debug  	18:34:02.723129	#1 	Notification (Status=Warning): Warning: Last 1 frames have not been stacked, because: Live Stacking is paused - PHD2 is 
Info   	18:34:03.406900	#3 	Event received from PHD2 : SettleDone																					
Info   	18:34:03.406900	#3 	Guiding SettleState changed to Settled																					
Info   	18:34:03.407759	#32	Event received from PHD2 : GuideStep																					
Warning	18:34:03.628376	#33	Frame not stacked because : Live Stacking is paused - PHD2 is currently dithering.										
Debug  	18:34:03.628376	#1 	Notification (Status=Warning): Warning: Last 2 frames have not been stacked, because: Live Stacking is paused - PHD2 is 
Info   	18:34:03.659990	#33	Live stacking unpaused																									
This shows the communication between SharpCap and PHD2 including the request for dithering and the updates from PHD as the dither happens. This was captured during live stacking, but similar should show up if dithering using the sequencer.

cheers,

Robin

Re: Problems with Dither and Debayer

Posted: Tue Jul 20, 2021 1:45 pm
by TigerClaw35
Hi,

Dithering seems to work. I tested it yesterday with SharpCap 3.2 and it worked fine. Ok, I lose the current frame during dithering, but itÄ working.

Normally I use exp=60 seconds and dither after 185 seconds. Normally I can also set to 180 but maybe it will be to close to the end of the third frame and start dithering during the third frame.

So my problem at the moment is only the Debayer mode: I used "debayer on" and also forceed RGGB for my QHY 268C. But the frames are in GBRG. This problem occures with SharpCap 4.0 . With SharpCap 3.2 I have no problems and it runs as RGGB.

Best regards