#Image Critique/Feedback
1 messages Ā· Page 2 of 1
?
yeah discussing about this lp
this thread now has over 1000 messages
and etc
do we get a prize
nice one
nice
give me money
jk jk
man light polution are baddd
very
VERy
hate it
use 70mm lens
u have apsc sensor
ull get somewhere around 2-3s shots i think
this is 30 mins exposure at 50mm
or something like that
but taken in bortle 5/6
taken like a year ago
and ive lost the photo with stars in it
it doesnt after you put the stars back in though
the stars cover up the ai generated parts
kindof
but sometimes like this example it can look a bit off
but it gives you an idea whats in your images
also this specific image has gone from beind a tif file to a jpeg to a screenshot so the quality is down the drain
yes
sometimes blueish thing
and i somewhat dont like it
the point is the using a wider lens with a faster aperture can get u bettwe framing
and get u better results
and is easier to do
and flexible
ok im gonna sleep now
gl
Thank you
on ur new mount?
yes
Iāve never had a problem with planes either
its really lacking intergration. that would be the main thing to improve on
Hm maybe it's just cuz live right underneath a flight path lol :p
Yeah
Planning to make a diy star tracker
Maybe 5-10 second exposure will help
Rather than 1s
Right
Right
right if bg was fixed, maybe overlay them to get best of both
maybe like this sorta thing
Did you do a dbe on it?
Oh lol
Yes
Mmm. I wonder how it tastes
if you have pixinsight, masking
background is clipped (perfectly dark, your losing detail), too much contrast, color of nebula is bad
I kinda like the color
Hmm
How would you process?
my processing method is scuffed and only I should do it
like in software or the end result?
i dunno watch some siril tutorials
I use siril to stacking
watch the tutorials to get better ig
I hate it when i use background extraction my color of nebula turns into orange orange thing
Maybe just lack of integration time
What to expect
Only 7 minutes only
Lmso
Lmao
yeah thats not enough
yeah thats not enough lol
š
could you share the data
Sure
huh
what was that code lol
its.. just python lmao
lol
processing critique none at all, but do get more data on Oiii and maybe broadband too?
Iām tryna get it all done by christmas š
Some clean nights ahead so maybe i can achieve more data
Broadband though idk
what
looks cloudy for a few days tho
yeah do narrowband first ofc
im just a stan of LRGB images š
Ill be that soon
why soon lmao
Cuz my stuff hasnāt arrived yet
Im mono now
Well not now, my cone is still osc
But from like 2025 onwards ill be mono
Yea
ah same that
It should all arrive by christmas
everyones going mono on this server basically simultaneously and its funny
like the meganerd OSC people
2" filters with ares pro why?
Cuz that ares pro is gonna become a peosdeion later down the line
Then itāll become a Zeus
Itās like pokemon evolution
nah but i feel like if u just get 36mm instead of 2" u can probably save enough to get the poseidon
Thatās why i only bought LRGB to start with aswell as the narrowbvad filters im looking to get range in the 2-3k$
ah ok
Ill hop to some b1-2 these holidays and revolutionise the world
as long as you arent too attached to Oiii, the OSC should be fine for most narrowband stuff
you can see how insane my Ha is with the Samyang
I remember mine had like crisp edges at f2 on my full framed dslr
get one that is like a filter drawer or wheel
its pretty nice
super simple OTA
nah thats mad
my corners with blurx
i dont have a raw example unless i dig out data
1.8 hours unmodded b7
Isn't that my process
no

?
wdym
Yeah this one
Depends on you tho
gimme some time im editing my own data rn
Kk no prob
Hi, I would workl a bit levels and curves and some background neutralization, its a bit on the purple side
Looks good, but Iād increase the contrast of the horse head and the stars but thatās just my preference
i like
it wins an astrology image/10 for me
also yea this region is hella beight for a stock dslr. i got 26ish mins on it and got some decent details on the taranntula
What could be improved with these images?
too yellow
needs a colour calibration/background neutralise
Sounds good, Iāll have to work on that tomorrow. I just used the Seetstar de-noising feature
Tried taking a few quick photos with 15PM 10s exposure, one captured andromeda (I think), images are edited and the location was not ideal due to light pollution
Dammnnn
good but bit too purple looking for my taste
but subjective
i would shift it more red tho
mabe you should try it...
Just a silly little nebula named after a spider
Looks really good imo. Get more int on the Ha
dont have any filters
Get more int overall
I was going to but my guiding was goofing up so I imaged carina
1 111 1
Where oiii
where is oiii arc
mid
ong bob dropped a good image
Lil bit of green noise
starless shows the sins
its a reprocess of this lol
the method of how i got it there is horrible and only I should be subject to it
Wait until you see my processing
you at least use like siril
(I may or may not use a very cursed photoshop filter that rhymes with ramera caw)
Why donāt you
haven't put aside the like 2 hours needed to learn it
Itās not that hard pls try
mine was running starnet on the old image and importing that into gimp, layering that, flattening it and stretching so the stars are the same brightness
i have watched 0 tutorials and have entirely reinvented the wheel
lol
it's missing the oiii arc (i'm late)
Fried to smithereens
My attempt at the heart nebula, reprocessed a few times. B5 skies. 750mm 6 inch Newtonian reflector. Captured with a canon dslr 650d. Just over 4 hours. (I know I got the framing wrong for this target)
Still looks really good, I love the vast amount of stars you can see along with the diffraction spikes
Appreciate the kind feedback, thank you!
Please be as harsh as possible, give me your strongest opinions.
Itās only 1 hour of data.
core cookesd
i think the first issue you should fix is the background, it stands out as very purple to me, as well as uneven
for the core, HDR is not too difficult these days, i personally iterate MMT masks, but there is a script which does a similar thing automatically called "iHDR"
i also notice your noise pattern looks super weird. eventually you want to get Pixinsight if you can afford, and use the tools that offers (DeepSNR is the best, MMT also works well)
it looks super plasticky and blob-ish
if you can get those 3, things down (colours, HDR, and noise management), you should be on the way to a pretty nice image
Thank you very much, I will work on that.
yeah good luck :)
Just reprocessed these two images again. For some reason when I tried colour correcting Orion it kept making the whole image red, maybe it is supposed to be red but it just looked off
Also both pictures are with a Seestar so theyāre not the greatest in terms or resolution
Iām a newbie by every stretch but I do like the Orion one. Background looks good, core isnāt too blown out. Could probably do with some longer exposures for the nebulosity which Iām going to try when i shoot Orion next. Short exposures for the core, long for the nebula
too much denoise
I'm sending over the data soon maybe ish so you can give that a run
I think thereās room for more lol
Ok Iāll keep going
I present the rosette nebula itās supposed to be color but it just baked the color out of it
Now that is deep fried
That ain't deep field that burnt
Fried
that is not merely burnt, that is reduced to smoke
Thatās supposed to be a color image
nah it's Atomized
This is the original picture only sharpened it and other stuff no color stuff tho
guessing this is an iphone pic, its completely average and just looks like phone astro. stacking and or using a lower mm eyepiece for more zoom would improve things (or just stacking if you dont have more zoom available or like that framing)
Not an iphone pic...
dangit
You know what
Its a stack pic
Its kinda worse than the single frame
Somehow
your image as a gradient
Damn 11 mins worth of data on m83
wheres the rest of the colours
Needs more data
Also needs more data
anything besides needs more data
Did you use calibration frames on this image it looks like thereās what Iām guessing is hot pixels and a lot of them
revive
Give me feedback
giga chad
traps are clipped š
This is my older 3h pleades attempt, how can i improve it?
try shooting in rgb 
wdym im shooting with a mirrorless
sorry im used to https://discord.com/channels/794642864218439681/1337566901496320101
can you up the saturation
yea im going to get more this and the other night
they have like 0 colours sadly
defo need to drop back on the very aggressive denoising
working towards getting a better background extraction
as for the noise, thats just more intergration
your stretching seems fine
Does using graxpert make the data look soft?
Can you show me how to get better at it?
with the red squares remove them off the nebula. if you get black splotces, either remove or add squares there
play with the smoothing
whys it 168MB
bro drizzled 4x
galaxies look a bit purple ish
Maybe it's my screen but to me it looks like it has a very very small green tint š¤
I see the Purple..
yeaa im still learning how to process galaxies color wise
online tutorials/advice has not been very helpful
Fair enough for me you just gotta get the hang off it..
Just mess around the the Buttons until it looks nice
So i'm just curious about this: From my instagram page, what do you think is my best image?
https://bit.ly/4dchke3
Your latest image. LMC + tarantula
LMC or one of the Orion constellation targets
orion neb
Hoping I can get some feedback on this? I just picked up my first monochrome camera and this is about an hour of combined LRGB data. Also my first time using ASTAP to stack since I was having trouble figuring out how to stack mono in DSS. I know I blew out the core but Iām not sure why there are the weird artifacts on the core of the nebula and the bright blue stars
How did you stretch your stars and nebula?
I used starnet in siril to seperate the stars and nebula. Nebula got a slight asinh transformation and then GHS. Stars got a histogram stretch while recombining the images. But the weird artifacts on the core and bright stars were there from the beginning after ASTAP stacked everything
@scenic nest @white elm @lusty dove
u guys all have similar problems.
-
background correction. if you have pixinsight use MGC its without a question the best way to correct your background. (ideally use own data, or just use the MARS files). if you dont have pixinsight, perform a careful dbe, and that means try the best you can to avoid touching even close to nebulosity, aim to ONLY put control points on true background. i prefer to undercorrect (have some bare residual gradients) than overcorrect (dark areas or weird colour splotches all over your image). i use graxpert for DBE, (the RBF model not the AI one) mainly because i like the interface
-
please correct your colours properly. its super easy to do it right. option 1: (preferred) SPCC. first platesolve your image so pixinsight/siril knows what its dealing with. then run SPCC with a background region defined. this region should be AS CLOSE AS POSSIBLE to neutral grey, and doesnt need to be unnecessarily big. a small box approximately 30x30 pixel is enough. option 2: standard colour calibration. simply open colour calibration, define a background region (as per the method 1), and draw a small box around the whitest star you can find as the white reference. how to tell its white? look at good images online of the same region, and compare you stars to theirs. this will set the neutral white point of your image. SPCC basically does this process but with scientific data + equipment based calibration
@white elm processing-wise, your image is probably the better out of the 3, but colours arent perfect (difficult with your fov), and there are horizontal lines across ur image (probably current or walking noise)
@scenic nest ur data looks best out of the 3 by far, but the processing needs a ton of work. try to do the points i said and you will probably be 70% of the way there. if you dont have pixinsight, i would highly recommend getting it and learning it, it is the best investment into astro for me behind camera/scope/mount
@lusty dove you do need more data, but thats an easy excuse a lot of people make. you can improve a lot simply through processing. try learn either pixinsight or siril, the former if you can afford it but siril is good on a budget.
oh and another thing:
until youre quite advanced in processing, just use autostretch/histogramtransform. it is so underrated. so many images blow out the core, but it is impossible to do so with the stretching methods aforementioned. i use GHS (must be very careful and know how it works), the histogramtransform, then very slight curves stretch. histogram is probably 60% of my actual stretching
amazing advice
I did use graxpert dbe but didnāt actually put the squares over the background. My fov is small (imx290) had to crop a bit cuz of artifacts, idk why I get that line in my images even tho I dither
Iāve been using ghs but ngl idk what it actually does, I just move sliders till it looks good.
Yea really good advice I really appreciate it š¤
thank you so much!!
i remember when selecting the background i also selected the very edges of the nebula which is probably why the colors are weird
I shall re edit and apply the advice and see how it goes
put the points manually. first try very little points like 5 or so, then increase them if u can based on the first iteration
yeah never touch the nebula with the points. also i recommend combining the RGB image first then correcting gradient
Do you change the grax model to do that or stick with the AI one ?
the RBF one. AI basically guesses where the background is, ive had like 1% of cases where the AI actually performs better than carefully selected RBF model points
Gotcha Iāll start doing that instead, can I select areas of the image that have faint dust/nebulosity or must it be a black background?
don't select it if you can avoid it. by selecting faint dust, the model thinks that area of the image has bright gradients and will overcorrect it, however, IF by selecting that area you can remove significant real gradients, then its probably worth doing. give different sample placements a try
Itās an alright start, if you aim to pull out ifn expect to be imaging for many more sessions, for whatās available now id make the bg darker so there remains more focus on the current thing shown which is the galaxies, another thing id adjust is adding a bit of curves or saturation for only m81 to add a lil blue to the galaxy. Now the bg itself looks decent as it appears pretty neutral coloured and the star colours seem decent (maybe a tad more saturation, remember to try and keep the stars ideally yellow/orange and blue for the most part).
I felt like the bg was a bit light so thanks for helping me decide on that haha. Also is there a way to only mask M81 so i can boost its saturation only? I have been struggling to pull much color out of m81 even though this is about 5 hours of integration.
thoughts?
Why is it purple?
m81 should not be purple, and m82 should not have that weird red shell to it. The Ha jets in M82 do not look like that
looks like some kind of selection/masked saturation or colouring. not great i prefer the first one. although, the darker background is nice. i disagree with qqq in that the stars are a problem. perhaps if the rest of your workflow was perfect, you can think about stars more but for now theyre very good imo
appreciate the feedback! back to processing i go š«”
Too washed out in the red, oiii needs a boost, eya carina is fried yet the other star colours need fixing
the blurx person put a halo around eta carinae sadly
pre blurx
also im struggling HARD with oIII
Nah the colours
should it be white or like orange or something
The dust
Red but not like lightsaber red
i got you
defo needs a redo (in pix)
Forgot to say colour noise as well
Nice
if you're ever struggling with colours, sometimes I find that it helps to process each channel individually and then re-combine the stretched images using different palettes. I find this works best for Foraxx
It could very easily be balanced in pix curves
S curve on the red channel would fix it
clipping simulator
not a particularly good method
I donāt have pix
part of me doesnt really like the galaxy brightness paird with the grey background.
use arcsinh stretch black point to make your background darker and that should be the trick for them to go well together
how's this?
better!
you can try going in for another crop down to make the galaxy more of the subject
depends on how many pixels youve got
true
there you go 
appreciate the feedback
What could make this better
Stars look a bit weird, perhaps bit out of focus or eyepiece vignetting?
Processing wise, could probs bring out more shadows or midtones - highlights are quite bright already
I believe the stars may have been a tad out of focus and the eyepiece might be doing something funky
Better processing, looks like was was done with an amateur rig so donāt know if thereās much you can really do in that department
Like better balanced background
Both colour and brightness wise
Yeah I did this untracked with my 114/500 and my Samsung s20
more contrast and saturation
Saturation is a personal preference
But yeah Iāll fix the contrast
how can i improve this with starnet? I tried streching with siril, rig is in my profile and i took 150s 200 iso frames in b2-3
Apart from the core of tarantula, this is a really nice image
Well rounded
Gyatt
@wispy yarrow this is the actual feedback
š
Revive
Critique one of my early images. (Low res screenshot, not really that bad normally)
@wild ocean you wanted it revived so be critical
Apart from the extreme lack of integration, the image itself is overstretched and oversaturated. Stars are oversaturated so having less saturation would help or even making the image starless would make it better.
I think itās an hour lol
No wonder
And I processed it before I knew about GraXpert and doing Star/non Star seperate
F/1.8 and stock dslr. Got a decent amount of Ha
My opinion is that it's too red
But my monitor isn't colour calibrated so on mine it looks more pinkish before I view it on a phone screen
Other than that I think I've brightened up the core of the statue of liberty nebula
red are overstretched. this region is heavy star dense so bring them back - dont run too strong of a star reduction
Ye I was thinking the same thing with the reds
I'll do a reedit and send back
Eventually
dear lord
You like the spikes! 
- remove those star spikes. theyre abyssmal.
- running man is compleletly blown out. either save it or crop it out as its distracting.
- i can tell your background has been SCNR'd. Revert that
Whatās SCNR
Whereās there green noise?
I only used GraXpert Denoise
theres no green noise in your image and bc of that your background colours are imbalenced
So I shouldnāt have removed green noise?
correct
And the spikes are amazing btw so
you do you
Vintage lens things
Rate it out of 10
5
Insanely good dust for 1 hour though nice!
My turn now. First time doing galaxies, but please lemme know how i can improve
Cigar Galaxy looks odd
maybe its cuz i focused more on m81 while processing
Yea looks like thereās no Ha, red, or whatever itās called there very strange
there seemed to be some of that when I was procesing the image. but at that time M81 seemed to dim and underprocessed. Maybe processing m81 blew out m82. thanks for pointing that out tho
Bodeās colors look good. I think maybe some HDR and sharpening could be helpful
alright thanks!
Reminds me of the DSS on Stellaruim Web
becuase... of how blown out they are
This is good work Bob.
Your background noise could be neutralised better and not too clipped which can be fixed by getting more integration
Apart from that, run some blurx. Will sharpen the galaxy

very obvious around the buildings, also were these at a different fl?
Same fl, different regions of sky
I'll do generative fill
Apart from that how is it?
grrrr
šŖ
Thanks
im just saying you can fix the gradient
it wasnt meant to be a roast
like if this region was more orange
hmm, maybe do a light mask before making the bg blue
ideas?
its quite cropped and i did some fast processing
im still not great at processing
the background is also a little weird
Do you dither?
i believe so yeah
at least i had it on and it mightve failed the last half an hour or so
critique wise I think the image would benefit from less stretching to make the backround a bit less busy also your stars appear to be a little dim overall and some contain this weird artifact within them
are those stars clipped?
i dont think so, but all of the stars appear almost pale and flat. I would assume that something went wrong when you were stretching them
a couple yes
I suffer from ca
I use an old lens
my condolences
it cost me only a 130 bucks though
nice small upgrade
im interested to see how the esprit performs on the heq5
stars having a fiesta
Disco disco
Hell yeah
guiding is swaying from .5 to 1. but that could be bc I have some high clouds
I do have a belt mod though
I think more integration time and a longer fl
thats not processing advice.
need more integration on the LMG(?).
the 2nd one looks blurry idk.
stars are really big in Orion, maybe make them smaller. The dust is also not very clear, very bright orion. the stars look fuked
Lol they look great Iām just kidding
i think orion stuff is bc its a tight crop in, the slightly wider fov makes the stars look better
its been blurx'd so it shouldnt be having fat issues
its not clipped btw. ive seen the histogram
It looks a tad bit too blue from my eyes I think
hurts my eyes
I would just reduce the intensity of the blue channel as it kind of looks like youāve decreased the image temp. For ref, hereās The image if you just increase temp slightly
Stock dslr moment
I'll probably go somewhere in between mine and this
orion is kinda overly blue in the bg, the core colour is good but bg is leaning towards purple, carina is too warm, and the lmc panel could be stretched harder
Quite blue imo
Idk if that was by choice but maybe less blue could balance things out
for which image
All of them - lmc I understand is blue that's fine, carina looks magenta/purple, orion's reds are simply overshadowed by the blues
carina is a magenta target
surprised about orion tho. stock dslr doesnt pick much ha and youre saying ive got too much of it
Too much? Ha is red and I'm saying you've got too little red..?
Oh whopps I interpret thag wrong
Yes the lack of red is stock dslr thing. Nothing I can do about thag
Fair enough lol
I really need feedback on this because my school wants to hang it on the wall but I can't seem to get the hang of processing M51 property
honestly good process. Maybe the black point is a bit low, but other than that it's good
would benefit from more data to get the tidal streams more visible
if you really want you can experiment a bit more with your colours more as they look maybe a bit generic
Thanks, yes that's true more integration would help for sure. Black point is low because the background would look too blotchy for my eyes
you can somewhat fix this by using masks
apply your stretched m51 with a mask onto a less stretched image where your bg isnt blotchy
You can also get yours to be a bit less yellow (which at least imo doesnāt look too great) and get more of a blue tone since m51 has a lot of star forming regions
Okay I'll try to do that. Nice image and thanks for the quick reply. I appreciate it!
do you use pixelmath for that?
Yes
share what
its not hard pixelmath
you apply a mask to image A, go to pixelmath, write "B", and apply that to image A, where B is your other image
oh oke, I thought you had some sneaky formula
perhaps a more realistic version
this looks much nicer, the tidal stuff less pushed
perhaps the HDR is a bit too much, the brighter parts shouldn't be darker than the tidal dust
I agree, the HDR is a bit too much
im ready to be critiqued, its my first good (i think) processed image
bright nebulae are a lil clipped
remove some of the faintest stars as it kind of makes it look more noisy than it is
this isnt #1337566901496320101
that was feedback my bad if it sounded harsh
it might be bc this is a compressed screenshot i think
thats possible
send ful res
not a screenshot, this is the png
where Oiii
Unable to bring it out š
Very nice and sharp
thoughts on this ? was my first time shooting a real wide milky way, 24mm iirc
not super happy with the processing, i know some stuff is done wrong i just don't know what/how to improve on that next time
wrong channel lmao
No he put this here cuz wants image critique
there was something else posted and then removed
yeah haha
thats mb should have clarified
i reccomend dithering as it looks like you have walking noise
I dither once every 3 frames and i dont see it on my side
it may be discord compression
How long are your exposures and how many pixels do you dither?
120s not sure how many pixels but i think like 3ā
Not sure, i use asiair lol
3 pixels every third frame is not enough dither
how much should i dither?
Also im using ra only mount so that might be why idk
Not sure about RA only, I dither 10 pixels every frame
i do 5 or 10
it died 
Attention thread lurkers, your thread has been selected in the champion of thread comps. Go vote! #1369694419963613295 message
Yes
it looks like theres mottling on my screen with the background?? not too sure what could be causing that
I searched up motteling and its disgusting
wtf is mottling
nvm idk why think its bc i uploaded to my phone and went thru lightroom a few times
Do you think the background is too bright?
Looks good to me
Can do a slight arcsinh black point adjustment
I would say it's a good amount of lighness, though I can also agree with Sword that you could adjust the black point a bit more.
top>bottom for sure
top better?
yeah, i can see more of the gas structure easier
i like the colours on bottom except its filled with gradients
did you use blurx +adjust halo's?
no idea. i just gave it to someone to blurx
yeah, alright. Just wondering since in blue horsehead the blue star has the halo around it deleted (which would also be nebulosity)
I think adjust Halo was probably used? If you want, I can run Blurx on it without it to see what happens, later?
alr, ill get the PC booted up
i stand corrected, its a result of the stellar sharpening
ah
This looks good on my phone, just need to clear up the gradients
Thanks, I have to find a good compromise between contrast and naturalness. I played around a bit and I think this is not too contrasty. Maybe a very slight black point adjustment would be fine, like you guys said.
thoughts?
Oversaturated
overcooked
Nah I think its just fine
Nah it is medium rare
Oversaturated and clipped
First time processing RGB and Ha data using Continuum Subtraction. Critique please
im not sure what to do with the background noise
RGB feels very weak
This may be a result of compression, but your background has some funky stuff happening.
should be in #1337566901496320101
Yea I realised that a tad too late lol
Pls fix the blue fringing
I get it aswell
This sounds wierd but I use Narrowband Normalisation in pix and I set it to HOO mode 1 which removes the bloating blue channel and replaces it with a fake one
Still gets me decent star colors and basically removes the fringe
Do this on a starmask
Or use Siril remove purple filter
But that doesnt work as well imo
Can anyone point anything out on this that may need improving?
Used a Seestar S50 to image
Stack more frames, is this processed?
And if you are using the base seestar app to do processing it might be your time to change
Yeah I processed this through siril and I think I got around 40 minutes of 10s frames
nice, u can def do better with processing
Do you know of a more advanced tutorial that I can follow?
Whats your workflow?
I copy the siril Seestar processing page, Iāll link it:
This tutorial explores the pre-processing and processing of images obtained from
ZWO Seestar. Pre-processing will be carried out using a dedicated script.
Preprocessing
1. Download the script to preprocess your images
2. Create the folder to hold your images
3. Put your RAW images in the directory create...
@sharp marten
Pretty good
But for color calibration use the new SPCC
and instead of using histogram use GHS
And make sure to use starnet to nit stretchur stars too much
Ah okay Iāll have to look into that and reprocess it, do you know of any good tutorials?
nope i js do stuff and see what works best lol
Fair enough, ill have to do that lol
idk about this but I like to hear what you guys think about it
it looks off to me but idk how to fix that
stars kinda feel off on the darker areas imo, everything else is grat
Red saturation is very strong, I'd cut back on jt
From my screen it was so desaturated
TBF my phone night filter ua on
no i agree
I dont really get what you mean, like their desaturated?
idk, it kinda looks like theyve been halo reduced and just looks kind of off, also i meant great not gray
I get what you mean, idk how to fix it though
Sadr is overstretched, thereās no real colour either
Donāt save as JPEG because the NA is super compressed
Hey just got around to reprocessing my carina nebula data: how does it look
Before and after
Really good improvement!
Maybe run SPCC or some sort of colour calibration to get rid of that blue tint
Will do
i agree with badunkie, the star cores are too bright and cutoff suddenly, like they are clipped
how were they stretched?
I stretched them using the stf thingy
admiral octavio taught me that
don't think so, could be I upped the blackpoint after I added them
shouldnt matter. maybe its your data, do the stars looked clipped in the linear state?
are these star cores really too bright? they look fine to me tbh
or like the black linear image very starry
its just that in thise darker areas, you would expect more stars to be fainter
yet they all look super "peaky"
overall, image very well processed
colours are nice, are how i would do them
hm yea true
thankyou!
this is an older process though
idk I made it a little bit more intens
is this better or worse?
nah not a fan of this
too cotton candy?
too contrasty
cant see whats going on within the blacks, and the bright areas are overwhelming
the blacks are exactly the same
in the original one, the dark areas have so much going on, like different layers of dust etc
uhm
i would maybe push the sat very slightly for the original image
but not the contrast
is this better then?
I only have the png, I'll try rq
btw im on phone so all this might look very different on a good screen
so take my advice with a grain of salt haha
eh it isn't really working
I have 4 versions of this process
I personally like the contrast on the bottom right, but the top left is the most realistic ig
I think top right or bottom lef might be the best, they are the ones inbetween
much helpful
I'd get rid of the green at the very bottom
It's MW glow
Noticeable enough for 2 ppl to see
on my screen and on my phone i barely see it
Also many stars appear purple
camara lens moment
Should be easy to fix
Take the extractred stars and run Correct magenta stars script
Or invert + scnr
thats for purple stars the invertscnr trick
That's what I'm saying š
you need more data
dslr?
this looks either way to compressed or pushed too hard idk
yeah i pushed quite hard
Yea camera raw colour balance
im trying to use it less
But it makes tweaking colours so easy
can you drop the dataset for this?
is this cfa drizzled?
Yes
yes
buih
its not that bad
i meant the google drive link was funny
anyway I can tell you a few things alreayd
whatever youre doing to the stars isnt good
they look super weird
were you out of focus?
ahhh thats me waighting for my 10mm spacder as im not in the corect back focus
I mean, that shouldn't impact the central stars
its just defocused
incorrect spacing will affect the image corners
it was also very windy wile having a sail of a telescope
your stars in the sub still generally look much better than the final image
especially in the center here
your stars seem like theyre all over the place
Yeah wind
hmm
honestly i shouldve asked for the stack
what does that look like
jus screenshot abn autostretch
this is no stretch
Bc I was just goofing off in light room
This is the one I actually care about
Give me your worst
stars are very desaturated
and the core of the galaxy is green...look into getting more colour into the galaxy
Needs more colour
You might be colour blind bc that aināt green
your image is clipped as well
I know where youāve been
?
youre meant to say hi to him
hi to both of yall
Heās hiding secrets that I know
I know where youāve been
š§ šļø

HDR multiscale did something
Whats that
Yes my sire
One of the most difficult targets to process in astrophotography are ones with extremely large differences in brightness. These can include galaxies with very faint halos, or nebulae with very bright cores, such as the Orion Nebula; in all of these examples, the target includes both a very bright inner ācoreā area, and a faint outer āhalo...
Big fan of the right image
Eta car regions is overstretched. Remember to use iHDR on bright regions
nice colors but it is overdone
Yeah its mb with the corr
core
but starxterminator really fcked up the nebula
Heres a better core but when i did this the redt of the nebula looked wierd
first time doing serious processing on milky way shots, thoughts on these?
yea thats the one thing i was iffy on how to do right, thanks š
Gradients on edge of image, bright star on bottom is wierd, mw glow slightly purple
i love it tho
Saturate the Ha more
Blotchy bg
Thats all ngl
I love this image tho
tis gonna be my next tatget
btw what do you use to denoise
and how do you make your stars have glow around em
DeepSNR
no idea how im getting glow...might be from a little dew even thoive got a dew heater
Oh
I think im gonna throw away my dew heater
I love that glow
youll need a adew heater to prevent dew getting on the csop
better?
Very
Ngl i would use dbxtract and extract the ha, then turn it into a mask, clone stamp the mw glow and do curves to make the nebulae pop more
To make a āHaRGBā image
kind of a HaRGB
seems like a lot more effort than what ive currently dojne
I guess feedback shouldn't be an issue for competition submissions? (If it is someone let me know)
How did you calibrate this image? Are you after a wider variety of colors or do you want something that looks more monochrome?
i use a duoband filter and process both the ha and the oiii seperately
really struggling to bring out the oiii colours without mottling the edges of the nebulosity
so in the end im left with a very single-coloured image
here are a few more processes i did while experimenting with different pallettes
the noise and the mottling is super noticeable i think
this is about 7hrs data btw
Thought so
Need some tips for the next time I process and add duoband data
StarX and Starnet work very poorly on this data for some reason
It makes the right side of the image have wierd lines that look like walking noise and black spots
But the stars were very cooked from tilt
Tried 2x upsample?
How!
? Mb
i think its because of the very dense starfield
Starnet litterally gives you that option. Pixinsight or siril?
Im using StarX
But starnet worked better
With 2x
thinks?
I do not thinks therefore I do not ams
1st one
Any feedback/critique?
DITHERDITHER DITHER DITHER
Dither or die
how many frames did you take?
75
Then I think you should dither more often
I'd say dithering every 5 frames is not enough
critique please
damn stars are tiiiny
the image is slightly out of focus, so that helps hide them 
But yes youāre right. Theyāre a bit small. Iām not too used to doing star removal and recomposition yet
wow ty shiv
New process
Trying to be a bit less over processed
Like for example here are my old edits
Worse
In what way
Core lost contrast
how did you reduce the brightness
idk maybe try histo or HDRmultiscale

