Nice program! (and a few suggestions)

[SwissCheese]

Hi,

I was also a little bit frustrated with that other 32 bits free stacking program , where I had either to crop my images or to convert them to 8 bits to handle them (and which is otherwise an excellent Software).

I just tried your Software, it worked nicely, thank you very much!

I have two suggestions:
- it would be very practical to estimate and display the number of detected stars before launching the full process
- it would be really nice to be able to select the reference image that will define the region of interest (without renaming all images) (or something more elaborate to select the region of interest)

and a question:
- does the Software use all images? or does it throw away automatically some of them? (low quality images, stars not round, ...)

and finally... what do you think about open source code? It might interest people to complement your already nice program.

I'm also saying this because of all these nice, free programs that are at some point abandoned with no possibility to keep them up-to-date. A pity to have to code from the beginning the same stuff again and again.

[asornoso]

(18-02-2016, 12:32 AM)SwissCheese Wrote: and finally... what do you think about open source code? It might interest people to complement your already nice program.

I'm also saying this because of all these nice, free programs that are at some point abandoned with no possibility to keep them up-to-date. A pity to have to code from the beginning the same stuff again and again.

I agree with SwissCheese. Your program is already free, why not build up the community to keep it up-to-date as well as possible performance boosts. I'd love to try and make this program run faster using multithreading if it isn't already parallel.

[Andrew Cool]

(18-02-2016, 12:32 AM)SwissCheese Wrote: Hi,

I was also a little bit frustrated with that other 32 bits free stacking program , where I had either to crop my images or to convert them to 8 bits to handle them (and which is otherwise an excellent Software).

I just tried your Software, it worked nicely, thank you very much!

I have two suggestions:
- it would be very practical to estimate and display the number of detected stars before launching the full process
- it would be really nice to be able to select the reference image that will define the region of interest (without renaming all images) (or something more elaborate to select the region of interest)

and a question:
- does the Software use all images? or does it throw away automatically some of them? (low quality images, stars not round, ...)

and finally... what do you think about open source code? It might interest people to complement your already nice program.

I'm also saying this because of all these nice, free programs that are at some point abandoned with no possibility to keep them up-to-date. A pity to have to code from the beginning the same stuff again and again.

[asornoso]

@Andrew Cool, I think you may have made a mistake when you replied and left out your actual message lol

[Andrew Cool]

(11-03-2016, 08:17 AM)asornoso Wrote: @Andrew Cool, I think you may have made a mistake when you replied and left out your actual message lol

Doh! Actually, with the benefit of hindsight, I'm not that pleased with this forum software and its interface. Anyway, after the dummy reply, I did write another lengthy reply, but I have no idea where that ended up!

The guts of what I said was :-

SkippyStak aligns and stacks images. It does what it says on the tin. Any filtering/removal of images is up to the user before you invoke SkippyStak.

But having said that, I'm happy to make improvements around the edges, such as :-

- display the number of detected stars before launching the full process

- select the reference image


Open Source is a problem though. I code in a language called Interactive Data Language (IDL), which is the language of professional Astronomers.
Oddly enough there are some very nice routines available for handling astronomy images that made SkippyStak a doddle to write.

The big "BUT" is that IDL is a proprietary language that costs big bucks for a developers licence. So I can put the source code out there, but very few, most probably no one, will ever be able to modify it.

Major PITA, huh?!

I also have a version of SkippyStak that links to Python routines to directly read RAW files, avoiding the intermediate step of creating TIFF files via a call to DCRAW. Now that speeds things up nicely. Alas, setting up the Python environment is like choosing to make love to a rose bush, and its is not something I could release into the wild without spending my Life supporting something that I can barely get to run on the computer in front of me, much less your computer in some far corner of the planet. Not gunna happen...

As to multi-threading, IDL will automatically multi-thread if the memory load is sufficient. If you want multiple frames being aligned/stacked,
then you could try running multiple instances of SkippyStak, sourcing input files from different folders, and then finally aligning and stacking the
stacked image from each folder. I haven't tried it. The code may break!

Hope that explanation helps.

Hope this post actually gets posted!

Andrew