What could be the reason I'm getting a "CRITICAL WARNING: flat-field calibration can not be performed correctly!" message?
I'm getting this message even though the flat and dark flat files were taken with the same equipment on the same night. Is there a way to troubleshoot this error and does it have an influence on the final stack? I'm not using any Bias frames.
I am also seeing this message. It is true that I am not using a bias frame, HOWEVER I am using a flat frame generated by SharpCap with the "subtract bias" option selected during generation. In this case, SharpCap (presumably) already measured a corresponding bias frame and subtracted it when saving the flat. In this case, am I correct in assuming that no further bias frame is needed. Can this message be safely ignored, or is a bias frame required by APP to process correctly even in this scenerio?
Hi, I get this message all the time when using data from my ASI294MC but not the 533mc. My flats are fine according to other stacking programs and of course gain and time match with the darkflats and gain with the lights. I've never used bias frames for any astrocam- was under the impression bias was unnecessary for astrocams, especially the 294mc. Any ideas just what it is that APP is moaning to me about?! Thanks.
Even though other software may just use the flats (I know DSS does this), it can still have an issue, which could impact your data. That's what the warning tries to tell you so you can improve the end result. Are you using the latest APP version?
Even though other software may just use the flats (I know DSS does this), it can still have an issue, which could impact your data. That's what the warning tries to tell you so you can improve the end result. Are you using the latest APP version?
I wasn't. I am now! Actually I think there probably is an issue with my flats - they are way too underexposed. Which is unusual as NINA flats wizard normally won't allow a poor result. I might go back to Shaprcap for flats creation as at least I can see the histogram there. Thanks
I am also seeing this issue in the newest 22 release.
With both ICX 694 CCD and the ASI2600 containing IMX571 chips, dark frames are useful but not required when good dithering is performed.
Flats and bias together with dithering provide ideal calibration on chips like CCD with good gaussian background or the imx571 and related with no amp glows, but the latest version will not apply the use of bias corrected flats as it did in all previous versions. I get the calibration error, which never happened before. Right now, I cannot calibrate anything, which is quite annoying. I dont mind an error, but it seems the calibration is being prevented or not occurring at all.
At the moment, i have to use the older version of the earlier v2 betas. Any thoughts? Please dont suggest to take dark frames as a solution 🙂
This post was modified 3 months ago by Colm O'Dwyer
What could be the reason I'm getting a "CRITICAL WARNING: flat-field calibration can not be performed correctly!" message?
I'm getting this message even though the flat and dark flat files were taken with the same equipment on the same night. Is there a way to troubleshoot this error and does it have an influence on the final stack? I'm not using any Bias frames.
The warning that you are getting is there for good purpose and explains what the problem is, You are only subtracting the sensor offset from the flats with your dark flats. You also need the sensor offset to be subtracted from your lights to have flat-field calibration work as expected. So if you do not provide bias nor darks, you will get this warning because you are failing to have the sensor offset removed from your lights. This is really essential for high quality data calibration.
I am also seeing this message. It is true that I am not using a bias frame, HOWEVER I am using a flat frame generated by SharpCap with the "subtract bias" option selected during generation. In this case, SharpCap (presumably) already measured a corresponding bias frame and subtracted it when saving the flat. In this case, am I correct in assuming that no further bias frame is needed. Can this message be safely ignored, or is a bias frame required by APP to process correctly even in this scenerio?
If the bias were removed from the flats already by another process/application, then yes you can safely ignore the warning provided you do use either darks or bias for your lights. If not, then the warning is still valid.
Data calibration principles state that the sensor offset needs to be subtracted from both the flats and lights to be able to expect flat-field calibration to work correctly. That sensor offset subtraction is accomplished by subtracting bias/dark/darkflats from flats and lights.
Hi, I get this message all the time when using data from my ASI294MC but not the 533mc. My flats are fine according to other stacking programs and of course gain and time match with the darkflats and gain with the lights. I've never used bias frames for any astrocam- was under the impression bias was unnecessary for astrocams, especially the 294mc. Any ideas just what it is that APP is moaning to me about?! Thanks.
Also here it is vital to understand that you subtract the sensor offset from both the flats and the lights. So if you only supply darkflats for your flats and no bias/darks for the lights, the warning is justified because you will not get the results that you can actually get 😉 if the sensor offset is still part of your lights, you can never expect flat-field calibration to work perfectly.
I am also seeing this issue in the newest 22 release.
With both ICX 694 CCD and the ASI2600 containing IMX571 chips, dark frames are useful but not required when good dithering is performed.
Flats and bias together with dithering provide ideal calibration on chips like CCD with good gaussian background or the imx571 and related with no amp glows, but the latest version will not apply the use of bias corrected flats as it did in all previous versions. I get the calibration error, which never happened before. Right now, I cannot calibrate anything, which is quite annoying. I dont mind an error, but it seems the calibration is being prevented or not occurring at all.
At the moment, i have to use the older version of the earlier v2 betas. Any thoughts? Please dont suggest to take dark frames as a solution 🙂
"dark frames are useful but not required when good dithering is performed"
In my book this is a bit of a misconception. to be able to get optimal results, you really want to remove all sensor problems as soon as you can. Yes dither can solve a lot, but even then, the results will be better if you also provided the best calibration data, i know this from extensive testing. For sure, not using darks is easy and the result can be quite similar, if the sensor has a clean dark signal without non-linear signals.
We test a lot of data and alwyas use calibration data while doing this and we have no issues whatsoever with the latest beta's ourselves. Maybe you want to share your data so I can have a look at what is happening?
