Save files as .fit or .tiff 32bits

[JAD]

Hello, why can't files be saved in .fit or .tiff 32bit format?
Sometimes processing images takes several days and it is necessary to save images from one day to the next.

Thanks and regards,
Juan

[Carles]

Hello Juan,

Even if you could it wouldn't matter as tracking would be lost as soon as you close startools.

Why does it take you days ? PC performance? Mosaic?

What are your system/computer specs ?

Regards
Carles

[JAD]

Hi,
Well, actually the problem is not really my PC, I have an Intel core i7-3770 CPU with 16Gb of RAM. I don't always have enough free time during a day to completely finish processing an image, work, family, etc. require time too.

Regards,

[Mike in Rancho]

Hi Juan,

Boy and I thought my i7-6700 (desktop) was falling behind the times. Glad you are making it work.

The matter has been discussed quite a few times, and may be something that could be implemented in future versions of ST, thus allowing a full bit depth save-state that could be reopened later for continuation, and includes all ST's tracking (vitally important). May also require some re-jiggering of how ST does its log files, but that's just all details.

When that does happen, it may push computer hardware requirements along too - at least as far as storage capability.

In the meanwhile, there is ST Replay made by user Guy, which can step along through a log of previous processing and bring your file back to the place you left off, or any other spot along the way. I believe that's current through stable release 1.8. You can also manually follow an ST log file. Depending on how much you've done and if you are pretty good with ST's UI and controls, it might even be faster than using ST Replay.

Rather than one giant continuing log (it's in the distro folder), I always rename that file after finishing every processing I do, that way I have one log per workflow attempt (mostly). Makes it easy to find if I need to check something or replicate previous workflows. ST will then create a new log file with the default name upon next opening.

[Carles]

Hi,
Well, actually the problem is not really my PC, I have an Intel core i7-3770 CPU with 16Gb of RAM. I don't always have enough free time during a day to completely finish processing an image, work, family, etc. require time too.

Regards,

OK, I'm asking because depending on the size of your datasets, it will take longer as Track makes temporary copies of the image through many processes. Typically, for one of my images with qhy8L, around 70mb size stack, won't take me more than 15 or 20min unless is narrowband, splitting channels and so on.. then maybe a bit longer. But never 1hour even. And testing some friend's datasets from a more modern and bigger dataset size, around 160mb, might take me a little bit more, depending on the processes.

that's why I'm surprised about "days". Now, if you count different processing attempts...that's a different story, but a single processing should be quite fast, unless the file is really big because processes might take longer.

So what sizes are we talking about ? What sort of workflow do you follow?

Regards,

[JAD]

Hello,
I am using a QHY 168C camera that generates 32MB .fit files and an integrated image (with stacking software) of 192MB
My workflow basically follows what is indicated in the StarTools manual.

Regards,