#What's missing from Siril?
6108 messages · Page 7 of 7 (latest)
I wouldn't call it fake. Its just another neural net tool among the ones we have, but it does more than it is asked to
But the way it does that is definitely strange
Like Prism deep that also sharpens or something
I'm having a problem with some python scripts. It falis with:
14:11:47: Begin: Sirilic()
14:11:47: Error initializing application: Unable to load tkdnd library.
14:11:47: Python Prozess (PID: 63970) mit Status 1 beendet
I've already tried resetting the Phython venv but still the same result.
This is not some python script. But only sirilic. You should write to the author. I think one import is missing
I would like to have some clarifications about temporary stretches in the image area and what rescaling does when opening any sort of dialog or stretch like GHS:
Most common thing I see is:
- I run Starnet on a linear image, check "Stretch linear image".
- The stretch seems to persist in the image area until i open any dialog that involves the "preview" checkbox
- The temporary stretch goes away
Can I better understand this behaviour? As it also happens with, for example, images from Darktable that I edited (tiff 16 bit, BP gets lifted), or pngs I download from the internet (that sometimes turn black after opening the dialogue)
I want to know as all this rescaling is confusing to me
Is this a setting in preferences?
Idk, what does that look like
No. With starnet a stretch is applied only for the input, then the inverse stretch is applied before display.
Ok. So my guess is that somehow that stretch momentarily stays on the image area?
Color management
Good news, my book is available in English!!
Please feel free to share your comments with me once you'll receive it
Where to buy?
Available as an E-Book too?
but having a book in his hand this is better 🙂
@stark ermine is it possible to only stack images below a certan Hfr value in your preprocessing script?
no. This script is made to be as simple as possible
No chance for a "advanced settings"?
you should be able to edit the script to add a filter at some steps
go to the seqapplyreg section
if you want something more advanced I think @proper forge scripts are the best
I don't want to code something similar
My opinion is that at least being able to choose weighting method would be great
Thanks @stark ermine! My script has 4 filters (FWHM, Roundess, Background, and Star Count).
No HFR option available in Siril so you'd have to filter those out beforehand. The script also allows you to choose weighting during stacking.
Can't post screenshots here but my channel has a demo from last month. I'm going to have a new version out for beta testing probably later today.
Tbh I would like to have an HFR measurement option in SIRIL just for the sake of consistency
so far there is fwhm measurement in Siril
you draw boxes around stars and it tells you
Oh yeah I use that all the time, it's just nice to have an alternative options.
Is there a recommended way to rescale with MAST HST/JWST data in Siril? It looks like the values aren't normalized to [0, 1], so that leads to the image looking weird
or wait, could this be due to NaN values screwing things up
Version 1.4.3 has been released
Whats new?
I wrote a release notes 🙂
it can now harvest organs for better astro rigs
What a clever feature
rip
On the website? At least I can not open it
why ?
Nothing opens up, when I click on it
Have you clicked "Bug fixes" ?
strage... which browser?
I tried switching, didnt work tho
Works for me on Arch Linux/KDE/Firefox
im making a sand dune sim
im gunna use it as a way to stretch imagws 
@stark ermine this is definitely due to NaN values in the FITS file; if I replace them with zeros the range of values changes
ok. FITS file can have NAN values. But where did you get this file?
Mikulski archive (MAST), it's IC 10 data from JWST
This occurs with and without the "allow FITS cubes to have different sizes" option enabled
Ah so this should be expected to be a little buggy for now?
yep. Especially with NAN values
Ah okay, this is good to know...I am not in a rush to work with this data
One thing I'll note though, is that star cores in this data are replaced with NaN values.
yes this is something I already saw in professional data
I got interviewed:
Today I'm sitting down with Cyril Richard - the lead developer of Siril, one of the most powerful free, open-source astrophotography tools out there. Cyril has just released Mastering Image Processing with Siril - a comprehensive 192-page, large-format reference covering everything from preprocessing and stacking to scripting, photometry, and re...
yes
Do you get it via dnf or must I build from source/flatpak
uts in the package manager and on flatpak
its more updated on the flatpak
But I dont get the absolutely latest version though do I?
Flathub says 1.4.2
Maybe I should just build from source still
I just haven't gotten around to updating Flatpak yet
Would you still recommend building from source even on Fedora?
as you want I would say.
If you build from sources, don't use master.
This is the dev branch
Yes I will use 1.4
@stark ermine thanks for your work on the sequence stretching bug! I am going to give it a test now (I don't check Gitlab enough sorry)
Anyone can create a script yes
would it be possible for you guys to add it in?
I'd much rather have someone else do it. I already have enough on my plate. Just creating a script isn't enough—you also have to maintain it.
But once it's created, it will be available to everyone.
Wait is this standalone DeepSNR?
Finally?
yep. theres a CLI for it now
no need for Wine. It runs on macOS, Linux and Windows
Oh ok. Its because the screenshot showed just windows
yooooo thats so cool!
holy advancments!
hey @stark ermine why isnt siriL able to save files as xisf?
ohh ok
I changed the URL for the English documentation