PixInsight RAW or Fit?

[SkyExplorer]

Hi,

Is it worth converting all RAW dark and lights files to 32 bit fp fit files before doing an image intergration?

The reason I ask is that the fit files are huge and take up masses of disk space.

Cheers

Steve

[SkyExplorer]

Think I may have answered my own question, but it looks like you can't stack raw files in PixInsight, only fit files. I guess I can delete the fits once I've finished with them and just keep the finished articles.

[dlp]

When you open the image integration tool it only offers FIT as an option, more's the pity. Only just got used to converting raw to tiff for the other stacking progs. Can you actually have a 3 colour FIT image or do you have to split to R-G-B as well I wonder?

David

Edit - you got there before me...

[harry page]

Hi

Why don't you stack in 16 bit , the 32 bit really only comes in when you are processing your image

Harry

[Beyond_Vision]

DSS will stack raw or fits files

[dlp]

Harry, not my thread but I hope Steve won't mind me asking (or showing my ignorance!) - what do you mean? If you convert something to 16bit, surely it's then always 16bit? Or is this quantum theory

[Beyond_Vision]

More like quantisation theory

http://en.wikipedia.org/wiki/Quantization_(signal_processing)

[SkyExplorer]

Harry, not my thread but I hope Steve won't mind me asking (or showing my ignorance!) - what do you mean? If you convert something to 16bit, surely it's then always 16bit? Or is this quantum theory

No problem David, however another question to Harry.

Thinking about it, if my Canon takes images in 16-bit format then what benefit do I get by converting them to 32-bit at all.

[roundycat]

None. Once the image is in a bit depth lower than a (theoretical) maximum such as 16 straight out of the camera then changing to a higher bit depth is nothing more than chopping up your data into smaller bits. Nothing new is added.

I only ever shoot and save 16bit FITS and calibrate, align and combine in Maxim and then save as a 16bit TIFF. Be aware that once you open and then save an image in Photoshop it goes to 15bit. PI may be different in respect of the processing but you eventually end up with an 8bit jpeg for the web and very likely for printing as well. Both displays, monitor or paper can only really deal with seven bit.

Dennis

[cardconvict]

does anyone have a work flow on how to stack in pixinsight i have tried the instruction on pixinsight web sight but to behonest they are very confusing and i don't really get it, is that just me please be honest

[harry page]

Hi

During calibration and stacking you gain nothing as this is the way the data is captured

while you are processing there can be rounding errors causing posterization.

Now most people will not come across this error and a lot of people choose to use 16 bit only , but PI can process in 32 bit avoiding this problem and being it is easy to do in PI why not

Out interest this thread in PI shows the problem

New Tool: HDRComposition

Regards Harry

No problem David, however another question to Harry.

Thinking about it, if my Canon takes images in 16-bit format then what benefit do I get by converting them to 32-bit at all.

[harry page]

does anyone have a work flow on how to stack in pixinsight i have tried the instruction on pixinsight web sight but to behonest they are very confusing and i don't really get it, is that just me please be honest

Hi

I don't know if you have seen my vids on this subject might help

http://www.harrysastroshed.com/pixinsighthomenewbie.html

Harry

[cardconvict]

yeah harry i seen them vids and they are a great help for begginers and me, but one of the main reasons in try pixinsight is to see if i get better results calibrating and stacking dslr raw images, and at the min, that i have worked out is to batch format conversion all raw to fits. then from there to make bias and dark masters and flat masters but how do you calibrate the lights with all the masters thanks ian

[harry page]

Hi

I am going to a shamless cop out here , I use a sx camera and calibration is very simple with these and I use AA4 to calibrate my images

There is the calibration module to which you put in all your hard work, but I have not used it

Now you have made me feel guilty I will have to go and do it

I have found the intergration and pixel rejection in Pixinsight to be better than anything else I have tried , so keep going you will get your rewards

harry

yeah harry i seen them vids and they are a great help for begginers and me, but one of the main reasons in try pixinsight is to see if i get better results calibrating and stacking dslr raw images, and at the min, that i have worked out is to batch format conversion all raw to fits. then from there to make bias and dark masters and flat masters but how do you calibrate the lights with all the masters thanks ian

[cardconvict]

thanks harry i have been following this tutorial PixInsight — Master Calibration Frames: Acquisition and Processing but it might just be me and it does'nt seem to make sense what do you think

Thanks Ian

[harry page]

Hi

This is for creating the master frames only have you done this ?

harry

[cardconvict]

yeah i have done this, but when you have completed lets say the bias frames it comes up with three windows rejection high and low and intergration i take it you delete the low and high and keep the intergration as the master?

[harry page]

Hi

Right , got you know yes just apply a STF to the high and low to make sure you are not rejecting to much and keep the intergration for your master

Harry

[cardconvict]

right check, so i'am doing that right then you calibrate the master darks and bias to the flats with the calibration tool and then after that you imageintergration (stack them) but i get lost after that do you calibrate the calibrated flats and master darks and bias with the lights before stacking them

[harry page]

Hi

slightley lost me there but

create master flats / darks and bias then take these to the calibration tool and apply them all to you images creating calibrated images.

There is no need to calibrate the calibration frames ( I think thats what you were saying

Harry

[cardconvict]

sorry i think i know where you are coming from now in the tutorial it starts by creating bias and darks and then in the second part it says to calibrate the darks and bias to the flats and in the third part of the tutorial it says to stack them would you like to have a look or am i reading this wrong thanks ian

[harry page]

Hi

Yes I think you have it right after creating master darks and bias use these to calibrate your flats and then stack these flats to create your master flat (phew )

Harry

[cardconvict]

but this is where i get stuck as there is no more tutorials on pixinsight site to what to do next i would imagine that you now calibrate all the lights with the master flats bias and darks but how do you do that

[harry page]

Hi

This is where you open the image calibration module and insert your hard won calibration frames and lights , I think you incheck the calibration box and let it fly

Harry

Ok you have convinced me , tutorial needed

harry

[dlp]

Got lost in the dust here! I use a Nikon DSLR and therefore end up with .NEF (semi-raw) images. Is there a simple process I can get from there to FITS format? I want to try PI integration, so I don't want to stack in anything else first just to produce FIT files.

David