Not ure if this is the best place for it, but when I try to run analysis on a set of image files from SGPRo/ASI1600, when it's done analysing it pops up an error message:
" 416" is an invalid float
where the "416" is the value in the Star detections column for the first image (so the value changes if I try a different set of images) - tested on versions 345 & 346
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
All of them (I mean, I have to pick at least two before it wants to run the "analyse & organise", but the error appears whatever combination of images I choose.)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Could it be something to do with Windows Localisation? Belgium uses a comma for floating point separation, rather than the dot used by English and perhaps there's a conversion somewhere which doesn't like fractions using comma somewhere along the road to that column?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I can't explain it. Here no problem. I can only suggest some thing to check/try:
Clear the tab and browse for the files and try again.
Is the location a network drive?
Does it happen for other files?
Does restart of program fix it?
Does reinstall of the program fix it?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'll do some more testing this evening and let you know - I've got older FITS files from other cameras and software. Program restart doesn't fix it, upgrading from 345 to 346 didn't fix it, I'll try reinstalling if I get it on all files and I can install on a second PC also to see if I can reproduce there or not.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Allright, an uninstall/reinstall didn't help, but deleting the astap.cfg file then starting ASTAP again seems to have done the trick, no more error message. So I went and checked the differences between the original astap.cfg I had and the newly generated one - to cause the issue, the "After analyse untick worst images" option needs to be activated before launching the analysis.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Very good detective work. Unfortunately I activate the “After analyse untick worst images “ it still doesn’t happen here. I like to recreate the situation. Do you still have the astap.cfg file? If so does it happen with the four files, 0059,0060, 0061, 0062?. If so could you attach the astap.cfg to the next post? If it not could find out how to recreate the situation with other files and post them on your google drive.
I’m really puzzled at this moment. The current menu is stable it nothing similar has happened the last months.
An other question, is the decimal separator on your computer a dot or comma? You can quickly test it with the calculator and divide 10 by 3 and look to the result.
Han
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think I have fixed it in version 0.9.347. I used a strtofloat function for converting the number of stars detected to a number. That strtofloat function is now replaced by a strtoint function. That should fix the problem. The real cause, I'm not sure about. I suspect a bug in the strtofloat function possibly related to the Wondows regional settings.
Can you test 0.9.347 ? If still a problem, could you uncheck the “After analyse untick worst images “ and press the analyse button. Then copy all the data by CTRL+A and CTRL-C and paste it in a text file and send to me? The CTRL-C function only works in 0.9.347, an other problem I just found.
Han
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Not ure if this is the best place for it, but when I try to run analysis on a set of image files from SGPRo/ASI1600, when it's done analysing it pops up an error message:
" 416" is an invalid float
where the "416" is the value in the Star detections column for the first image (so the value changes if I try a different set of images) - tested on versions 345 & 346
Hello Bart,
Always interesting to test if the error is caused by a faulty FITS file or something else. I can als use it to make ASTAP more robust.
Could you provide me with the image file for testing? Via your google drive or upload it e.g to: https://uploadfiles.io/ en give me the link.
Han
You (and anyone who wants) can get this set of four files from https://drive.google.com/open?id=1RLCS1krrF4aroT4gn31e7z3vXeuy_fMI - I load them into ASTAP Stack mode, hit analyse & organise and get the error.
Hello Bart,
These files don't give an error here. Files 0059, 0060, 0061, 0062. Which one give the error?
Han
All of them (I mean, I have to pick at least two before it wants to run the "analyse & organise", but the error appears whatever combination of images I choose.)
Could it be something to do with Windows Localisation? Belgium uses a comma for floating point separation, rather than the dot used by English and perhaps there's a conversion somewhere which doesn't like fractions using comma somewhere along the road to that column?
I can't explain it. Here no problem. I can only suggest some thing to check/try:
Clear the tab and browse for the files and try again.
Is the location a network drive?
Does it happen for other files?
Does restart of program fix it?
Does reinstall of the program fix it?
I'll do some more testing this evening and let you know - I've got older FITS files from other cameras and software. Program restart doesn't fix it, upgrading from 345 to 346 didn't fix it, I'll try reinstalling if I get it on all files and I can install on a second PC also to see if I can reproduce there or not.
Allright, an uninstall/reinstall didn't help, but deleting the astap.cfg file then starting ASTAP again seems to have done the trick, no more error message. So I went and checked the differences between the original astap.cfg I had and the newly generated one - to cause the issue, the "After analyse untick worst images" option needs to be activated before launching the analysis.
Hello Bart,
Very good detective work. Unfortunately I activate the “After analyse untick worst images “ it still doesn’t happen here. I like to recreate the situation. Do you still have the astap.cfg file? If so does it happen with the four files, 0059,0060, 0061, 0062?. If so could you attach the astap.cfg to the next post? If it not could find out how to recreate the situation with other files and post them on your google drive.
I’m really puzzled at this moment. The current menu is stable it nothing similar has happened the last months.
An other question, is the decimal separator on your computer a dot or comma? You can quickly test it with the calculator and divide 10 by 3 and look to the result.
Han
It's happening even after zapping the entire .cfg file- it happens with any set of files I try here locally - but in attachment my initial .cfg file.
Our local settings use comma, not dot (I mentioned it a few messages ago) - I can try it on a different machine with other settings, but not today.
Hello Bart,
I think I have fixed it in version 0.9.347. I used a strtofloat function for converting the number of stars detected to a number. That strtofloat function is now replaced by a strtoint function. That should fix the problem. The real cause, I'm not sure about. I suspect a bug in the strtofloat function possibly related to the Wondows regional settings.
Can you test 0.9.347 ? If still a problem, could you uncheck the “After analyse untick worst images “ and press the analyse button. Then copy all the data by CTRL+A and CTRL-C and paste it in a text file and send to me? The CTRL-C function only works in 0.9.347, an other problem I just found.
Han
I can confirm it's working now on 0.9.347, thanks!