rough osprey
Canon 6D (4.63um)
Rokinon 135mm (130mm)
13/04/2024 ddmm
following files:
rough osprey
@brazen walrus here it is if u still need it
brazen walrus
Thank
median sapphire
walls
rough osprey
balls
granite flicker
meteor?
granite flicker
very interesting but also annoying at the same time
rough osprey
i said at the top its in the canon raw sub
rough osprey
very interesting but also annoying at the same time
oh rly?
thought it added some balance
ill maybe try one without it too maybe
granite flicker
i would prefer the image to be without the meteorite but thats just my opinion
rough osprey
thanks for letting me know your thoughts
round dagger
@rough osprey took a crack at the data. i'm not entirely sure of this process, so let me know what you think.
crazy data btw
rough osprey
looks nice, you did a good job especially with the gradient
reds look a bit pushed for me, and greens look as if they were killed a bit
if its SCNR, id recommend going lighter, or only applying to a masked darker region
signal and background is clean
round dagger
reds look a bit pushed for me, and greens look as if they were killed a bit
alright, thanks for the feedback :)
sudden nexus
How does it look like ❓🙂
rough osprey
uh you could probably work on a few things
so your stars are very oversharpened, which when zooming in overwhelms the image
also, it washes out any fainter details that could otherwise be seen, such as the emission-dust region which was visible at the very top of the frame
colours have quite a warm tone, which is a bit hued compared to its true colour
the background also seems to have a orange-to-black gradient from left to right, but that could have been my misinterpretation due to the immense starfield
if we have a look at our processed cropped into one region
you can see the stars, together with the hued background mute out a lot of faint detail which would otherwise look quite nice
and the stars have a dark "ringing" around them
anyways, those are all pointers to work on, and i wouldve done a much worse processing a year ago :)
processing is difficult, and we all start somewhere, so i wish you luck!
median sapphire
eyjafjallajokull
polar tendon
Have never tried processing this target, looks like fun, having a crack at it. Thanks for sharing.
rough osprey
Have never tried processing this target, looks like fun, having a crack at it. T...
ur welcome, have fun and show me the result :))
polar tendon
So many stars
rough osprey
and yes this data is extremely fun (except for the gradient GL with that)
rough osprey
eyjafjallajokull
waiting for my processing ego to be inevitably kicked out the door when qqq drops his process
polar tendon
Lol even my starless is full of stars
median sapphire
waiting for my processing ego to be inevitably kicked out the door when qqq drop...
nahhh
I wanna see how much better your data is compared to my 5h of data but in b5 stock dslr
From just looking at your process I can see more detail and less noise and more signal
rough osprey
I wanna see how much better your data is compared to my 5h of data but in b5 sto...
is this mine?
looks good already
median sapphire
is this mine?
nah mine yours would be far better
rough osprey
nah mine yours would be far better
HUHH
thats B5???
holy cow
surely combine them or something right?
median sapphire
thats B5???
Same int time as you but I used a stock dslr and was indeed in b5 skies
Got this data like almost a year ago
median sapphire
I noticed when I looked at your data it had the same problem as mine
median sapphire
The bottom right of rho starless gets mega clipped idk why
rough osprey
wait like the dark bit?
median sapphire
This area
Super clipped compared to the rest just like how my data was
It’s a pain to fix
polar tendon
Oh it is lol
rough osprey
Super clipped compared to the rest just like how my data was
oh icic it is indeed darker
in my final proc i added a Lr mask to that corner 😭
(it should not have gotten to that stage lmao)
rough osprey
Oh it is lol
try DSNR on the starless thats what i did
it was too packed with stars to DSNR the pre-starremoval
polar tendon
Just putting the stars back now
rough osprey
yeah should look better then
i actually removed the first MLT layer because my star removal left artefacts
so rip a bit of detail
but whos peeping anyways
polar tendon
rough osprey
nice
a bit saturated for my liking
but its not bad
stars could be a bit softer too but thats just artistic lol
polar tendon
quick go at it, im definitely gonna put some time into a better process
beautiful data
rough osprey
quick go at it, im definitely gonna put some time into a better process
yeah bet
polar tendon
another quick go before i go to bed
and the starless before the final tweaks
well kinda starless lol
sudden nexus
2nd try 🙂
polar tendon
try DSNR on the starless thats what i did
I’ve tried installing this before and tried again tonight but it keeps failing
rough osprey
looks sick, thats an improvement id say
rough osprey
I’ve tried installing this before and tried again tonight but it keeps failing
ru on mac?
polar tendon
Windows pc
rough osprey
2nd try 🙂
looks way better than the first one, i would still work on the gradient correction and colour calibration, but the stars have a huge improvement, and the background is closer to neutral too
median sapphire
waiting for my processing ego to be inevitably kicked out the door when qqq drop...
Such wholesome data for realz
polar tendon
wow
rough osprey
WOAH
hes crazy with it
wait do u reckon u can help process my crux image
i cant do it right for some reason
and i know the data is good cuz i could see so much dust in the subs
its 2h Bortle2
rough osprey
gradients and clipped blacks, but the rho colours look alright
median sapphire
wait do u reckon u can help process my crux image
Yea
rough osprey
alright bet ill upload it
granite flicker
gradients and clipped blacks, but the rho colours look alright
yeah any tips on how to remove gradients because i always struggle with that
median sapphire
yeah any tips on how to remove gradients because i always struggle with that
Do a dbe
granite flicker
but when i try it it always gives me shit results
median sapphire
but when i try it it always gives me shit results
Place your points on regions with gradients and no stars and not on nebulosity, do a few points, too many = usually worse results due to more sampling, not enough = worse results due to lack of the images gradient sampling
Basically any spots where there is noticeable brightness change
granite flicker
ah so like where the blue was on my image
median sapphire
Avoiding placement on stars helps usually
granite flicker
on the bottom
median sapphire
You don’t want the dbe points to be sampling stars
granite flicker
that i know
median sapphire
Just gradients
granite flicker
ok thanks for help
rough osprey
@qqq what dbe do u use
median sapphire
@qqq what dbe do u use
pixinsight dbe or Siril’s
rough osprey
i currently use graxpert, but the regular RBF model, cuz grax's own model is shi
median sapphire
Never liked graxpert
rough osprey
pix has so many parameters its actually like balls
rough osprey
Never liked graxpert
its so overhyped its not even good
median sapphire
Agreed
If you want a good free dbe platform just use Siril, it’s basically the same as graxpert but in my opinion easier to use
Use similar formulas
rough osprey
If you want a good free dbe platform just use Siril, it’s basically the same as ...
i just hate that i cant export xisf into siril, thats why i use grax's plateform rn
results are same
median sapphire
i just hate that i cant export xisf into siril, thats why i use grax's plateform...
I just save a copy as a tiff
dreamy crown
Canon 6D (4.63um)
Rokinon 135mm (130mm)
13/04/2024 ddmm
following files:
1. "c...
i cant download any of them ?
ah nvm now i got it working
i literally cannot get a better background then this
median sapphire
i literally cannot get a better background then this
so it is true
dreamy crown
starnet moment 
median sapphire
starnet moment <:monkaS:853356001221804073>
Do you know how to dbe ?
Well the right corner thing is normal but the image before isn’t
dreamy crown
starnet just makes it a black void if theres to many stars
median sapphire
starnet just makes it a black void if theres to many stars
Yeah ik
median sapphire
i literally cannot get a better background then this
This I mean
dreamy crown
idk i can crop it out
median sapphire
Crop what
dreamy crown
the huge black void on the right
round dagger
the huge black void on the right
does it do that when you use starnet on a stretched or linear image?
for me, linear starnet doesn't do that, but stretched does
round dagger
eh, or neh? i used scnr more sparingly and toned down the stretching a bit.
median sapphire
eh, or neh? i used scnr more sparingly and toned down the stretching a bit.
Nice
Only issues are the dbe and the clipping of the dark dust lanes near the centre of rho
Otherwise that I think it’s a nice process
rough osprey
oooh such a nice process
the only thing is that black clipped in the middle of rho @round dagger
rough osprey
nice process
rough osprey
yeah i take that back, was using the older release before due to compatibility issues, now im using the newest ai and its pretty impressive
no?
you shouldnt be applying DBE on stretched images anyways
background extraction should always be a linear processing step
tame ruin
@rough ospreyinteresting result from spcc lol
indigo mist
I'd love to get some insight on how people get this so nice. Especially the colours and the darker structures toward the bottom. @rough osprey @round dagger @median sapphire @granite flicker
rough osprey
<@443587757567377418>interesting result from spcc lol
maybe because i started the acquisition at a super low altitude? atmosphere affecting colour
rough osprey
I'd love to get some insight on how people get this so nice. Especially the colo...
can you tell me what your workflow is? maybe i can fix some points
indigo mist
can you tell me what your workflow is? maybe i can fix some points
I did DBE - SPCC - histogram transformation - slight curves (c) then (s) - mask with lower limit for darker structures - rgb curves again - then regular saturation for highlighting specific colours
tame ruin
It seems like the main issues occurred during your DBE and Color Calibration steps
indigo mist
Maybe my background selected was too starry.
However there's not much 'un-starry' space
tame ruin
maybe because i started the acquisition at a super low altitude? atmosphere affe...
@rough osprey This may be interesting for you https://pixinsight.com/doc/docs/SPCC/SPCC.html
That article shows you how to prevent the green cast as well
tame ruin
Both of these images have SPCC applied with the same settings, the only thing changed is the debayering
rough osprey
<@443587757567377418> This may be interesting for you https://pixinsight.com/doc...
that was interesting, however i did stack my image with WBPP drizzle, so i dont understand how im supposed to prepare the data?
lots of information in that article though, thanks for sharing
brazen walrus
that was interesting, however i did stack my image with WBPP drizzle, so i dont ...
I believe you have to select bilinear as the debayer method. In uhhhh, calibration, click on your lights, then on the right window thing select CFA then change (i think its default is VNG) to bilinear.
indigo mist
It seems like the main issues occurred during your DBE and Color Calibration ste...
What’s your usual workflow?
for a quick process
tame ruin
Depends on the specific data, however, it's more about how you use those processes rather than the workflow. Your DBE can be improved, the same goes for color calibration. Colors are very off in your version.
indigo mist
Agreed. However i can’t really find background spots that don’t have stars lol
tame ruin
Agreed. However i can’t really find background spots that don’t have stars lol
You should always do DBE on the starless though
indigo mist
so run colour calibration first, then deep snr, then starnet and then run DBE?
tame ruin
Yeah that could be a good workflow
indigo mist
Alright man thanks, i’ll give this a go when I get back from work
Also, are there any pix processes to bring out the darker structures?
tame ruin
so run colour calibration first, then deep snr, then starnet and then run DBE?
Typically, you run StarNet first to remove the stars and THEN apply DBE on the starless image to ensure that the stars don't interfere with the creation of the background model during DBE. However, if the gradients are too noticeable in the original image, doing it this way can introduce gradients in the star mask itself.
To fix this, temporarily duplicate your original image and apply DBE to this duplicate first, then run StarNet to generate a more gradient-free star mask. Use this star mask on the original and delete the duplicate.
This method requires a bit more work, and it's not always necessary
tame ruin
Also, are there any pix processes to bring out the darker structures?
Just stretching I guess
indigo mist
Typically, you run StarNet first to remove the stars and THEN apply DBE on the s...
So I colour calibrate and stretch before this?
indigo mist
Typically, you run StarNet first to remove the stars and THEN apply DBE on the s...
In the second paragraph, you say run starnet on the original image, meaning the starless?
tame ruin
No lol, running StarNet on a starless image wouldn't make sense
You can just forget about it if you feel like it's too complicated. It's just an additional step
tame ruin
No worries, glad I could help
median sapphire
I'd love to get some insight on how people get this so nice. Especially the colo...
Looks better then mine tbh
indigo mist
so run colour calibration first, then deep snr, then starnet and then run DBE?
trying to do this workflow
tame ruin
No idea, it worked fine for me. Try
92.4mm
4.63μm
@indigo mist
tame ruin
this fine?
idk lol I didn't memorize such a name. But you can try rho ophiuchi
you typed ophuici
indigo mist
indeed it is
it dosent even load it just says failed right away
ill restard the application
ill reduce star detection
oh it worked
@tame ruin Just one more question, are there specifc settings to use in PCC for best results (im on an older version so no SPCC)
tame ruin
Not sure, haven't used PCC, only in SiriL, back when I didn't have pix yet
tame ruin
oh it worked
For future issues like this, try using SiriL's PCC first. It's less strict on the settings than PixInsight's PCC. After that, try PCC again in PixInsight. SiriL will have saved the exact astrometric solution in the metadata, so you won't need to figure out the right settings anymore.
If that makes sense
tame ruin
god deep snr gonna take ages lol
This is how short it'll approximately take with GPU acceleration
indigo mist
This is how short it'll approximately take with GPU acceleration
Oh wow
I have ryzen 7 5800x3d with a 3060 how can I get this?
tame ruin
Not sure if it's going to be as fast as this since I have a 4080 rtx, but it will still significantly speed it up.
I'll send you a link on how to configure it
tame ruin
https://rikutalvio.blogspot.com/2023/02/pixinsight-cuda.html bit complicated but worth it
tame ruin
depends solely on the specific image, I usually experiment with the settings until I have the best result
indigo mist
@tame ruin please can you combine my stars and starless
pixel math is making the stars weird
tame ruin
sure
indigo mist
its just upladoing to wetrasnfer
its so bad though i cant get it looking clean idk how people do it
tame ruin
I use Jimmy's pixelmath expression
indigo mist
I use Jimmy's pixelmath expression
~((~starless)*(~stars)) - ive had problems with pixel math on my version so this is the only one that works. doisent work for this image idk why
like this, so clean and saturated well
tame ruin
How did you create this starlayer?
The stars already look clipped
Idk what happened, but you might need to start over entirely tbh lol
The starless version looks all messed up as well
tame ruin
Anyway, here's the stars added https://we.tl/t-v95NFa7Iql
rough osprey
oh yeah the equipment is 6.5um, 130mm
and make sure uu spell ophiuchi right
i spelt it wrong in the title
i'll change it rq
Bortle 2 Rho Ophiuchi
gray jolt
granite flicker
not the ckeanest process
summer wedge
very quick edit
just did a colour calibration, background extraction, star removal then a very intense curve stretch
using ps+siril
pliant wagon
hey awesome data how should i edit the meteor and combiine it with the image?
acoustic bridge
@rough osprey nice data
rough osprey
nice processing!
dust is very much popping out
the background seems to have a little bit of colour imbalance, but its a pretty good effort given the complex region, i assume you just used an old background correction algorithm?
acoustic bridge
the background seems to have a little bit of colour imbalance, but its a pretty ...
probably
wide onyx
dull gorge
humble light
hearty plank
