Polls 5.0

[tender hollow]

But almost no one is going to be making a huge mosaic to match the 571's sensor size

[rain void]

well you can also just compare a 571 iamging something that fits its FOV perfectly, and a 585 imaging something that fits its FOV perfectly

both objects of equal brightness

[tender hollow]

yes, the mono 585 wins

[rain void]

if you resample both images to the same resolution, the 571 wins...

and if you don't the 571 will look sharper because the effects of seeing are smaller relative to the image size

so no the 585 doesn't just automatically win at all

just imagine a hypothetical camera tha only 4 pixels of the same size as the 585, but have 100% QE, is that camera better? obviously not if you wanna make a good image

it's really the same comparison, just taken to the extreme

[tender hollow]

That's NOT the same comparison

[rain void]

yes, just more extreme numbers

[tender hollow]

That's completely different
The 585 is small but not THAT small

[rain void]

the perfect 4 pixel camera will have "better" data in exactly the same way the 585 has "better" data than the 571

[rain void]

ofc, i'm just saying it's the same principle taken to the extreme

[tender hollow]

This is subjective, but at a certain sensor size I would draw the line. 585 isn't that small
If it were like, a imx224 sensor size (pretend the 224 is good for a moment) then I would say it's too small

[rain void]

ok let's try something else then, compare a 2x downsampled 571 with the 585 imaging the same targets as described before

both images will be about the same resolution, but the 571 will have better SNR and be sharper in terms of pixel fwhm

that's why i'm saying if you choose your targets carefully, the 571 will win

[fallen obsidian]

chat what are we talking about

[rain void]

this

[fallen obsidian]

still the mono vs the osc

The 571 is not the best sensor rn i think too

it has amp glow

[rain void]

no?

[delicate lance]

The 571 OSC is, as per public vote, the most overrated colour sensor

[fallen obsidian]

the 585 has its mono, and has ir sensitivity also its got low noise

[fallen obsidian]

i was considering getting the 571

[tender hollow]

Wins in what way? Sharpness? Mono, due to the lack of a bayer matrix, should be sharper (although seeing still matters)

Resample? That would mean either upsampling the 571 which hurts its SNR even worse, or downsampling the 585 which would make its SNR advantage even larger
unless you mean crop to the same FOV, in that case the 585 has more pixels and (scope/seeing dependent) would be sharper

[delicate lance]

I’ve heard multiple times that the 571 Colour shouldn’t be bought

[rain void]

i'm saying same resolution (in pixels) not sampling

[tender hollow]

Okay, let's frame this a different way I need to understand what you're trying to say
Let's take two nearly identical sensors, the only difference is size: 533m vs 571c

In this case, why would the 571 win? (If you think it does)

[tender hollow]

The 571 would have a larger field then

[rain void]

yes that's why i'm saying you have the image the targets that fit your FOV

and the larger field is what makes the image sharper

(still meaning sharpness in terms of fwhm in pixels, not arcsec)

[tender hollow]

that also depends on the scope/FL

[rain void]

ofc

but assuming approximately critically sampled

[rain void]

I'll definitely say the 571c wins

basically more signal will almost always produce a better final image

[tender hollow]

I agree, however the 533m will absolutely eviscerate the 571c in SnR
Even while having to mosaic, it gets pretty close to the 571's resolution in two panels
But if you don't mosaic it's even stronger

[rain void]

the 571 has 2.884x the sensor area...

so idk about two panels

[tender hollow]

I think you're really underselling the ridiculous gains that mono gets you

[rain void]

and assuming a 2x signal loss from OSC vs mono (which in reality is probably less), the 571 gets more signal

[rain void]

there's no way the signal gain is >3x

[tender hollow]

It's actually 4x

[rain void]

show me how please

[tender hollow]

When shooting RGB it's less, but with Lum or narrowband it just ends all arguments for OSC

[rain void]

i guess if you factor in the ability to get luminance data

4x still seems like too much tho

[tender hollow]

(For the 585 vs 571, in Ha the difference is even more than that because 585 has better QE at that wavelength)

[rain void]

i wouldn't say the 533m "absolutely eviscerates" the 517c, but i agree the comparison is a lot closer compared to the 585m vs 571c

[rain void]

also that document doesn't really derive that 4x factor anywhere, it just says it...

[tender hollow]

If you're shooting in bad conditions (Bortle 8-9) like myself mono is a no brainer even if the 585 is smaller

[rain void]

now you're just stating things... the bad conditions don't change the comparison at all

but even if we grant that mono gets 4x the amount of data, the 571 still has more than 5x the sensor area compared to the 585 and thus still gets more total data.

[tender hollow]

We already established that this argument is down to personal preference. If you're content with the sensor size it will produce better images

[rain void]

that's what i'm saying it won't 😭

even granting the 4x which i still think is way too much, you will always be able to make an image of the same resolution as the 585, but with better SNR

with the 571 sensor

[tender hollow]

That's just not true

It'll be hard to set up the comparison, we need these sensors working with the same conditions

[rain void]

how not??? if it has 5.2x the sensor area, but 0.25x the QE, it will obviously collect 5.2*0.25 > 1 times the amount of signal

and if you resample to the same resolution as the 585, that directly means better SNR

[tender hollow]

Forget about the FOV, if you're content with the sensor size
so for the sake of comparison you'd be cropping the 571c down, the 585m wins

[rain void]

and that's assuming exactly same rig, and imaging targets of equal brightness that fit the sensor size

[rain void]

holy crap you're not cropping because i'm saying you should choose your target right

[delicate lance]

https://tenor.com/view/troy-community-room-fire-pizza-gif-5612111

[rain void]

assuming exactly same rig, imaging targets of equal brightness, both imaging a target that fills the FOV, the 571 will gather more signal

that target can even be that same target, if it's just a target that fills the 571 FOV

in that case you can't say that the 585 will simply produce a better image, when the 571 clearly has better SNR

it will obviously be different images, that's why i'm saying different sensor sizes allow different targets, but the 571 will without question have better SNR in that case

[delicate lance]

https://tenor.com/view/monkey-thinking-monkey-thinking-think-money-gif-9734025314392564509

[rain void]

and even if we grant that the 571 has 2x the FWHM (in arcseconds), after resampling by a factor of sqrt(5.2), the FWHM of the 571 (in pixels) will be lower than the FWHM of the 585 in pixels

[tender hollow]

I have to leave for now, this is not me giving up the debate however
I'm summoning @pallid pecan in my place

[rain void]

of course you can be interested in other things than just maximizing SNR and pixel FWHM, that's where the whole personal preference thing comes in. But in this scenario we must agree that there is a great argument for the 571 being better

[rain void]

i'll rather continue this with you at a later time than discuss with him lmfao

[zealous cloak]

i love engaging in online discussions

[rain void]

in case you haven't noticed.. me too :)

[pallid pecan]

Wha

[ivory mural]

What did octavio do

[rain void]

I had a discussion with him about SNR some time ago... That's when i found out it was impossible to have a productive discussion with him :(

so now i just avoid it ¯_(ツ)_/¯

[urban quiver]

I’ve got 5.6 hours of Broadband already

“What do I image tonight”

[zealous cloak]

Moons at 22% so id add more data to that m31

Im in fact adding more data to mine

[urban quiver]

No moon in the sky tonight

It’s bothering the southern hemisphere nerds ig

[zealous cloak]

For me it says it sets at 21:22

Im imaging rn, its 19:43

[urban quiver]

Yea it’s only out for maybe an hour or two of the night

And it’s far from andromeda

[zealous cloak]

Yurp

[urban quiver]

Il be getting HaOiii eventually regardless

I’m not sure if I should get more broadband though

Got a lot of detail already

[tender hollow]

Sure, that's because it has a larger sensor. It's naturally going to pick up more light because it has that area. That doesn't mean it's actually more efficient, we're talking about per-pixel SNR.
Objectively the mono 585 has superior noise performance, there's no debate to be had.

[low lance]

Not how that works but technically yes you are correct the 571 has bigger pixels so more signal but you are thinking of bigger sensor means more signal which is a misunderstanding you are collecting more overall light yes but it's also spread over more pixels so it ends up being the same amount

You mean up samples the 585 has smaller pixels

Tbh there are only a few target that i cant image with my 585 and 600mm fl which A i can simply do mosaic or B i can borrow my friends roki 135mm so there is no target that i cant image but if i had a 571 most galaxies would be out of my range and i could only do large galaxies and nebulas

Again what you are really saying is you like to be able to zoom in and out of your images which the 571 is perfect for it since it's not too expensive that it breaks your bank but the 455 or the 461 would be much better fit especially the 461 as it has a 100mp 44x33mm sensor

[rain void]

...but that's like the whole question. "better sensor vs larger sensor"
and while, yes, the 585 has lower read noise, if that's a concern you're really doing something wrong. The noise should always be extremely dominated by shot noise, so unless you have extreme read noise it's totally irrelevant

[rain void]

Also wdym "per-pixel SNR"? as opposed to what? i'm just using the smandard definition of SNR: SNR=μ/σ

[rain void]

huh?? signal just refers to the total amount of photons that hit the sensor... this is just plain wrong

or rather amount of photons that get converted to electrons to be precise

[low lance]

My point is it doesn't make your target brighter

It doesnt even make your scene brighter

[rain void]

of course it doesn't??

of course why would it

i'm not saying it is

[low lance]

So why not say the 571 gets more light per pixel because of its larger pixel

[rain void]

because it doesn't

that's like the whole debate... it gets less light per pixel but has significantly more pixels

[low lance]

Then i dont get it what do you mean by the sensor gathers more signal like what are you comparing it to

[rain void]

i'm just talking about the total amount of photons that the sensor registers

[low lance]

The 585 has 2.9um pixels the 571 has 3.75um bigger pixels means more light per pixel better snr

[tender hollow]

read noise is not even relevant in this discussion

[rain void]

why would it be?

[rain void]

bruh what. a 571c gather much less light per pixel because of the bayer matrix

[low lance]

Sure but it's offset by the larger pixels

Not fully but still

[rain void]

yes slightly, but not completely at all

[ivory mural]

then you agree its less light per pixel

[low lance]

Also the difference between mono and color cam isnt as big as people think it is you see how the red channel the green channel and the blue channel all overlap

If you look even red channel is capturing some blue light

[ivory mural]

yeah id say thats a pretty big difference in how much blue is in the red

[urban quiver]

I saw a comparison between mono and OSC…. You could barely see a difference

[ivory mural]

i worded this very bad

[rain void]

yeah that's what i was arguing too earlier, i just agreed for the sake of argument that we could say the difference was 4x, but you need to factor in the ability to get luminance data with a mono cam

so in reality i'd guess around 2-3x more signal with mono

and even less for narrowband because you can't get luminance

[low lance]

Which means the efficiency isn't as black and white as people think it is maybe 1.5x improvement depending on target but where mono shines the brightest is narrowband and better color accuracy and flexibility

[rain void]

are you arguing for or against me?

[low lance]

I'm more neutral

[rain void]

great

anyways 2 sec my parents are calling brb

[low lance]

Also better resolution like actually quite a bit better resolution

[rain void]

i'm back

that took a while

[rain void]

on to the next point - i actually mean downsample here. I'm talking about resampling to same resolution, not sampling

and since the 571 is ~4x resolution, i'd 2x downample

[low lance]

Why not cropp

[rain void]

because that's just throwing out signal

[ivory mural]

why crop

[rain void]

as i mentioned, we were talking about a target that covers the FOV

[low lance]

Idk why resample

[rain void]

just for the sake of comparison

[delicate lance]

vro first the fullframe thing and now this?

https://tenor.com/view/robot-ai-china-robot-evil-robot-robot-malfunction-chinese-robot-gif-1847251548666509882

[low lance]

By resampling you wouldn't even match the same arcsecond per pixel

[low lance]

Idk its a comparison

[rain void]

basically if you follow my argument, you have the option to make an image that has better SNR and better pixel FWHM, that doesn't mean you have to

[ivory mural]

i dont think there is an answer to this argument

its subjective

[rain void]

you can also choose to not reseample in order to just have a high resolution image

but that's just yet another argument for the 571c's case

[ivory mural]

this poll is so far away now

[low lance]

Ig we both have just different astrophotography goals

Lets leave it that

[delicate lance]

wha

https://tenor.com/view/midnight-the-cat-stopped-working-midnight-the-cat-cat-cat-meme-funny-cat-gif-81113296414293415

[rain void]

sure, i specifically mentioned that some parts of the debate is up to personal preference

but Zegery was trying to argue that the 585 would always make a better image, which i was arguing against

[ivory mural]

how many 585 panels is 1 571 panel

[rain void]

5.2 i think

[ivory mural]

then the 585 is not always better

if mono gains is 4x and you want a 571 fov

[rain void]

specifically, if all you care about in an image is pixel FWHM and SNR, and you don't have a preference between small or large targets, it will be the other way around

[rain void]

exactly the point i made lol

5.2/4 > 1

and that's even granting the 4x increase in signal which wayyy overestimating imo

[ivory mural]

i honestly have no idea what to upgrade to

im using an uncooled 585 rn but a 571 is too expensive and it feels dumb getting another 585 but mono

and i heard 2 585 panels is better than a 533 panel

[rain void]

debatable

actualy more like... wrong

[delicate lance]

teeeechnically yes due to smaller pixels and higher efficiency but that also comes down to equipment and preference

[ivory mural]

i feel like id rather just save money and do a 585, even if its worse than 533

[rain void]

the 585 does not have 2x the QE

[ivory mural]

but again i dont like the idea of 2 585s

[rain void]

because that would be >100%

[delicate lance]

..obviously not

[ivory mural]

585 is just that good

[rain void]

lmao

[ivory mural]

photon generator

[rain void]

but 533 will gather more signal

[delicate lance]

this feels like something trump would say

"the new sensor has 1000% more quantum efficency, quite frankly"

[rain void]

what is your fl?

and also what is your FWHM in pixels normally

[urban quiver]

Sounds like you should just get a 533

lol I didn’t even notice someone else mention 533

[delicate lance]

why

just sell the uncooled 585

[ivory mural]

no

[delicate lance]

then use it as a guidecam

dumdum

[ivory mural]

ig

[rain void]

also nice to have for planetary

[ivory mural]

still nothing changes in image other than mono

[delicate lance]

mono planetary imaging exists

[ivory mural]

real question is wtf am i gonna do with my imx290

nobody's buying that

[delicate lance]

"still nothing changes in my image except 4x efficiency and better colours and no thermal noise im sobbing"

[rain void]

of course but if he's upgrading it probably can't do planetary amounts of fps

[delicate lance]

tru

[ivory mural]

i imaged mars at 1000 😼

[rain void]

but what is your fl and average fwhm in pixels

[ivory mural]

wow my laptops bootlooping again how pleasant!!!!!!

[ivory mural]

600mm and like 5 but i have very bad guiding and it was uncollimated

new mount will be sometime too

oh and no cc

but im going to try to buy a cc today

[rain void]

anything above 4 is oversampled, so the pixel size of a 533 will probably better than the small pixels on the 585

unless you can reduce it with a cc ofc

[ivory mural]

well it's a .95x cc so basically nothing

[rain void]

yeah but maybe it can reduce your fwhm because of cc stuff

not just reducing

[ivory mural]

the fwhm can absolutely go down

[delicate lance]

browi

[rain void]

?

[delicate lance]

i want to hear your opinion on the IMX492 sensor

[ivory mural]

i told you why it's high and what can be improved

[delicate lance]

(totally NOT because i know it will start an argument)

[rain void]

that's a very open question

it just has micro pixels i guess

[delicate lance]

do you believe its price is justified for being around 200 bucks more than a 533

and should it be in the upgrade lineup (585 -> 533 -> 571 -> 455)

[rain void]

wait it's only 300 bucks more?

crazy

hmm

the small pixels make it weird

[delicate lance]

if you dont buy from the evil zwo company its less

[delicate lance]

what me and rubolo agree on is

that its so nice because the pixels are so versatile since binning is an option

and binning it still gives you a 4k image

[brisk copper]

i think my 585 with 750mm fl is slightly oversapled

[delicate lance]

its a 180 USD difference

[tender hollow]

I will concede that's not always true?
I have no idea why but my question mark and comma got swapped on my phone keyboard and it's really annoying so just pretend I used a comma
I'll rephrase by saying that a larger sensor? technically collects more signal because there's more area? but assuming you aren't planning to mosaic the imx585 is going to be far more efficient due to mono benefits. Better efficiency = cleaner data = better image

[brisk copper]

my 0.9x cc should just barely fix it i think

[brisk copper]

(when it arrives)

[delicate lance]

zegery sounds horribly confused

[rain void]

lmfao

[lofty jetty]

just use the question mark button when writing commas 🥀 how hard is it

[brisk copper]

rate the image

[rain void]

i didn't say anything about needing to mosaic

[lofty jetty]

https://tenor.com/view/whyareyoublue-walterwhite-breakingbad-blue-jesse-gif-12242572209403578115

[brisk copper]

duochrome infrared

[rain void]

it applies if each camera imaging a single panel panel for the same amount of time. Each imaging an equally bright target that fills the FOV of the camera

[tender hollow]

The question mark is in the wrong spot

[brisk copper]

i do have the rgbir image tho

[tender hollow]

The comma goes there

[delicate lance]

your keyboard is probably just confused

[tender hollow]

I don't know why this happened

Or how to fix it

[delicate lance]

press the language button

[brisk copper]

its evil

[delicate lance]

i meant zeg

so he can change it back

[brisk copper]

sorg

[delicate lance]

and see if the keyboard becomes un-confused

[tender hollow]

Yeah that didnt work

[brisk copper]

[delicate lance]

guess youll stay confused

[rain void]

so basically you're calling the image with lower SNR and lower pixel fwhm "cleaner = better"

[delicate lance]

how does a mono 585 with more sensitivity have lower SNR

sorry if this has been answered since im not reading allat

[rain void]

because i set up a scenario where you resample the 571 to have the same resolution as the 585

so more data in the same amount of pixels = better SNR

[delicate lance]

does that really account for the advantage mono and higher sensitivity gives you

[rain void]

yes we factored that in too

[delicate lance]

https://tenor.com/view/curious-monkey-curious-monkey-thinking-thinking-monkey-scary-gif-7331100376461803246

[tender hollow]

Who is we

[delicate lance]

oh its the weirdass gif

[rain void]

me

[delicate lance]

browi
bro wi
bro we

[rain void]

i have split personality fu

[rain void]

if we grant that mono is 4x as efficient (which is imo a high estimate, but lets just grant it for the sake of argument). The 571 has 5.2 the amount of sensor area, and thus collects 5.2/4 times the amount of signal

and if we then resample to the same resolution the, the 571 will have sqrt(5.2/4) times the SNR

(which is greater than 1)

[tender hollow]

We've been over this? Just because the sensor has a larger area doesnt mean you're getting more signal per pixel

[rain void]

that's why i specifically set up a scenario where we downsample to the same resolution

[delicate lance]

also

[rain void]

because then we do indeed have more signal per pixel

[delicate lance]

yall are talking about numbers when yall are disregarding that mono images look cooler

[tender hollow]

What exactly do you mean by downsample

[rain void]

???

you know what downsampling is

now you're just going full on jordan peterson on me lmfao

[brisk copper]

[delicate lance]

i mean if we resample the 585 by the same factor it's SNR stays superior

realistically no one in their right mind would resample a 571 to 4k

[brisk copper]

really???????

[tender hollow]

For comparison you would be downsampling the 585 to match the pixel size of the 571?

[rain void]

of course, but then we don't have images of the same resolution, making them harder to compare

[tender hollow]

Not the other way around

[rain void]

holy moly.... once again... same RESOLUTION not SAMPLING

[tender hollow]

I know what downsampling is? You gotta explain what you're using it to do

So downscaling?

[brisk copper]

😰

you scaring me

[delicate lance]

thats what he means

[brisk copper]

wont downscaling improve the snr still tho?

[rain void]

exactly!

[tender hollow]

No?

[rain void]

what

now you're really spitting ooga booga on me

[brisk copper]

[rain void]

SNR is the mean pixel value divided by the standard deviation of the pixel values

by definition

[rain void]

so downsampling will absolutely increase SNR, just as drizzling will decrease it

[tender hollow]

Hold on

[brisk copper]

i make sure to do 100* drizzling to ensure theres no signal left

[rain void]

yeah that's in regular signal processing. In image processing you use this

[brisk copper]

u o

a e i o u

am sleepy sory

[rain void]

because from this definition we can actually measure SNR from the image alone

[rain void]

like, please define where how to calculate P_signal and P_noise from an image you know nothing about how was obtained

[brisk copper]

send a formal letter to the taker of the image and ask them

[rain void]

but surely that's not what matters for the quality of an image

if we cannot measure the quality of an image by only looking at the image, what are we even measuring?

[brisk copper]

to measure the image you must first know what the image is not, by subtracting what the image is not by what the image is you get the measurement

[rain void]

let me make a great example from some of my data

[brisk copper]

the second image has better snr

but looks god awful

[tender hollow]

alright I think I understand what you're trying to say now, but downsampling still isn't enough for it to match the efficiency of the 585

[brisk copper]

is 585 or 571 better

[tender hollow]

the 571 is better if both are mono

this debate is between a color 571 and a mono 585

[brisk copper]

o yea!

[rain void]

nope, still osc vs mono

5.2/4 > 1

these two images have the same SNR if we ask you guys

[brisk copper]

i would rather have mono 585

[rain void]

right is bin4

both autostretched

[brisk copper]

cant you have it look like theres more signal with worse snr?

wouldnt a nb filter do that

[rain void]

huh

[delicate lance]

yes but the x4 is for osc 585 vs mono 585 which disregards the QE gap between 571 or 585 or am i dumb

[brisk copper]

bc anything that isnt the wavelength u want is noise right?

[rain void]

idk, i just granted the 4x because my argument still holds

in reality im sure the difference is smaller

[brisk copper]

yeaa, 571 not worth it

[rain void]

and compared to RGB (and not LRGB) the difference is smaller yet

[brisk copper]

i took luminance with my andromeda image, it defo made the data sharper

[tender hollow]

if you're downsampling the 571's resolution roughly 67% to match the 585s, you'll get an signal improvement per pixel, but it's not 5.2x and certainly not enough to match or beat the 585
And you also aren't going to downsample your 571

but even if you do

[low lance]

Because of trail right

[brisk copper]

you could actually 2x bin and make it mono

[rain void]

that's only because you're not factoring in the pixel size difference

it cancels out if you do the maths

[brisk copper]

it still wouldnt be as efficient but it would work

[brisk copper]

weh?

[low lance]

Well worst fwhm caused my trail and wind and seeing ect ect

By* not my

[brisk copper]

the conditions in the right image were better

its just that its like 60% crop

[low lance]

Oh lmao

[brisk copper]

misalignment issue

i now use the screw on my focuser to align my cam

hmm, i think i should get plides in infrared

[tender hollow]

where are you getting 5.2x from

[rain void]

area

[brisk copper]

space

[rain void]

like i can write up the math if you really want me to

[brisk copper]

dewit

[rain void]

grr

[brisk copper]

or fake

[rain void]

ok

[tender hollow]

why is area relevant here?
i've acknowledged that yes, because a sensor has more area, it will collect more signal, but you're not getting any more signal-per-pixel unless the pixels are larger
And yes, the 571's pixels do get a bit more signal for being larger, but the QE loss from OSC negates it entirely

[rain void]

uggghhh imma just write up the math

[brisk copper]

what if they both osc, which better

[tender hollow]

571

[brisk copper]

darn

i need to get a mono 585 at some point

using my 585 like a mono cam does suprisingly get me better image quality than if i didnt

its weird

[delicate lance]

dont phrase it like that please 😭

it sounds like you took a colour camera and captured LRGB with it using filters

[brisk copper]

idk how else to phrase it

[brisk copper]

it does

i did kinda do lrgb actually

[tender hollow]

https://tenor.com/view/facepalm-really-stressed-mad-angry-gif-16109475

[brisk copper]

its lrgbir

i didnt use this image bc the misalignment tho

[brisk copper]

i should probably try and explain maybe

[brisk copper]

its uvircut as luminance, quadband filter as rgb and my 2 ir filters as ir

idk if that makes sense to u

[brisk copper]

gif-finder.com

-# im tired, please forgive me

[brisk copper]

hybrid osc+mono?

[rain void]

as i said, cancels out ¯_(ツ)_/¯

[brisk copper]

grrr snrrr

[rain void]

indeed

maybe even grrr math too

[brisk copper]

zeggyry dissapeared

[rain void]

https://tenor.com/view/bruh-zone-bruh-idolmaster-gif-19321120

[brisk copper]

buh

zegyry is bac :D

[tender hollow]

need a moment

[rain void]

p means total amount of pixels if that was unclear

[brisk copper]

pickled

[rain void]

and s means pixel size

[brisk copper]

pickled pixels

[rain void]

basically the intuition is just that if the pixels are k times bigger (by area) you would only downsample 1/k times as much, but the pixels would get k times the amount of signal from the beginning

so it doesn't matter

https://cdn.discordapp.com/attachments/697450809847316540/1396841981589586033/attachment-17.gif

[tender hollow]

i will give my thoughts after i polar align my rig real quick (I'm imaging )

[rain void]

https://en.wikipedia.org/wiki/Signal-to-noise_ratio_(imaging) this btw

[urban quiver]

mono cult not gonna win

[rain void]

i'm all on the mono side tho, just not in a case this extreme

[urban quiver]

I think mono definitely is better than OSC. But it’s not a crazy amount

I’d much rather have a 571 OSC than a 585 mono

I literally wouldn’t be able to do mono if I had the money for the filter, EFW, and camera

My mount cannot handle it

[rain void]

Yeah my guess is somewhere around 2-3x as efficient, and even less for narrowband

But probably closer to 2x assuming an LRGB ratio of 3:1:1:1

[brisk copper]

sell 571 osc and buy 533 mono

[urban quiver]

My mount cannot handle the extra weight for mono

[rain void]

narrowband has the advantage of being able to choose the ratio between filters yourself tho

[urban quiver]

It would die

[brisk copper]

you can manually change filters

[urban quiver]

It’s already crying and pleading my for help. And you want to add an EFW?

[brisk copper]

one day amma have narrowband ir filters

[urban quiver]

Uh no thanks

[rain void]

[urban quiver]

585 is too small sensor

Il get a 533m one day

Once I buy pix and my new mount arrives

[brisk copper]

585 is good for ir tho

[urban quiver]

I couldn’t give less of a crap about IR

[rain void]

this is also completely neglecting that you reduce your fwhm in pixels by a factor of r when downsampling

[brisk copper]

buh

[rain void]

once again a win for bigger sensors

[urban quiver]

If I had like an SQA55 I’d be open to a 585. But with my Askar 71f, 533 is the smallest i want

[rain void]

or maybe more precisely fwhm / field radius is sqrt(A1/A2) times bigger

which sounds like a great measure of how sharp an image is

[rain void]

https://tenor.com/view/my-dishes-give-me-my-dishes-front-door-cam-gif-19916823

[rain void]

https://cdn.discordapp.com/attachments/794651916646350859/1403496284798586910/image0.gif

me rn

waiting for @tender hollow to return

[tender hollow]

Math checks out. I think our main disagreement is on definition of SNR, or rather the type of SNR we care about.
Continuing down this path, it would quickly be turning into the same thing as the "Great SNR Argument of Broken Biscuits". I don't want to participate in that honestly, it's not going to go anywhere
I thank you for having a respectful conversation

[rain void]

Thank you too :)

[tender hollow]

i don't even have a mono cam myself. i'm just a poor man with an uncooled 585. I'm no cultist

[rain void]

its almost 2am and i have lectures tomorrow morning so i should probably go to bed

[brisk copper]

yooo, im a poor woman with an uncooled 585!!!

uncooled 585 tha best

-# in winter

[tender hollow]

winter is crucial

[brisk copper]

if you hibernate summer you wont have issues

[rain void]

Would love to hear you guys' definitions of SNR sometime

Tho actual good night this time lol

[cerulean elm]

Yeah... you'd say that. But it's funny how I manage to get 1h worth of 533 data in only 10min with the 585. QE in IR does matter, because it pulls up the Hα/SII lines as well

[brisk copper]

ir is hot af

[low lance]

Higher QE in ir means higher QE in the rest of the spectrum

Specifically the visible range and Specifically the Ha and SII lines

[low lance]

Also this

If you have bad transparency IR is your friend when visible light gives garbage data (which for me is most of summer) IR will give nice clean images

Also proto stars need i say more

[urban quiver]

I don’t care

[low lance]

Then why are you considering an astro cam

[urban quiver]

I’m not considering one

I own one

[low lance]

Either way why

[urban quiver]

its minimal improvement being IR sensitive imo

[low lance]

It adds up in longer integration time

Like 10% improvement is 6 extra minutes of data in one hour of imaging time in 10hrs of data you get a free extra hour

[undone sinew]

its so over for this thread bro

What a shame

[ivory mural]

Why

The argument was respectful

[undone sinew]

What argument

[ivory mural]

Or are you talking about callo

[undone sinew]

Im saying that all the polls have been used up

[ivory mural]

oh i thought you meant it was gonna be shut down like the others

[undone sinew]

Ever since you guys executed thomas this thread has been really unmoderated so idk there are no polls left to make

Actually wait i have an idea

Sure! Could you clarify where you want to post the poll (e.g., Instagram, X/Twitter, Reddit, YouTube, Discord, etc.) and what the topic should be — for example:
• Favorite type of astrophotography (e.g., nebulae vs. galaxies)
• Preferred gear (telescopes, cameras, mounts)
• Editing techniques
• Night-sky targets

Once I know that, I’ll tailor the poll format, tone, and options for that platform.

But here’s a general example to get us started 👇

⸻

🌌 Astrophotography Poll: What Do You Love Shooting Most?
1. 🌠 Milky Way landscapes
2. 🔭 Deep-sky objects (nebulae, galaxies, star clusters)
3. 🌙 The Moon
4. ☀️ The Sun (solar imaging)
5. 🌍 Planetary imaging

⸻

Would you like me to make:
• One poll like this, or
• A series of themed polls (gear, techniques, targets, editing)?

Theire we go

This is the new poll

enjoy

[rain void]

Absolute clanker

[undone sinew]

i wrote it myself

[rain void]

you must be a clanker then

[undone sinew]

hdjshssdjjsjdbskjdskjdkhdksjs

[rain void]

¯_(ツ)_/¯

[rain void]

catadioptric above 6", reflector at 6" and below

maybe 8" depending on seeing

and yes that is solely because RASA exists lmfao

[undone sinew]

What even is rasa bro

[rain void]

it's like an SCT but with the perfect amount of focal length at 620mm and also not shit

[zealous cloak]

Rowe-Ackermann Schmidt Astrograph

[swift heart]

an inferior Ritchey–Chrétien

[undone sinew]

im sorry for being a protostar

[pulsar hazel]

Everyone starts somewhere, you’ll get there eventually l

[brisk copper]

rowe ackerman shit astrograph

[rain void]

then get better data yourself

[zealous cloak]

Literally this

I love how much people in here talk shit about various things/criticise other people

Yet their astrophotography is...

[brisk copper]

mfw the clouds

[brisk copper]

mine isnt even that bad ;-;

[brisk copper]

ive had 2 clear nights over the last month

[brisk copper]

@swift heart

[brisk copper]

aragoscope

[cold moon]

This is something that you always say but genuinely makes no sense. It's going to gather more signal but also more noise. Per-pixel SNR is what really matters in a comparison between two sensors, otherwise you can just ignore everything else and call the bigger sensor the winner because it gathers more signal. The real advantage of a bigger sensor is that you can frame more stuff, that's it. And it comes with cons too obviously, such as needing much better corrected optics.

[brisk copper]

hai rubo :3

[cold moon]

You can obviously bring it to the extreme and argue that a single photon multiplier is, by my logic, better than a 571 but we both get what I mean

[cold moon]

Evening

[brisk copper]

gud eevning

[brisk copper]

have you started assembling the new rig?

[cold moon]

Planning stuff, but I do know that I'm getting the cem70 as soon as it's out of maintenance

[rain void]

... I really wanna get into this but it's almost midnight and i have lectures at 8am 😭

[brisk copper]

niice

[brisk copper]

go eep >:(

[cold moon]

But... ap is more important

[rain void]

hmmm

you make a fine point

gimme 5 min to let me finish reading lmao

[cold moon]

I'm tempted to get a fat RC

But I also like the idea of a F/3 newt

[brisk copper]

noots are cheaper

[cold moon]

I would try to get higher end stuff

[brisk copper]

oouk :3

get a dall kirkham

[cold moon]

Too expensive lol, and also a bit of a gamble

[rain void]

ok done

sooo...

[brisk copper]

sadge

[rain void]

I completely agree. SNR is what really matters. This is essentially what i discussed with Zegery yesterday. Comparing SNR of two images that are not the same resolution is unfair. If we resample both images to have the same resolution, the bigger sensor will have better SNR, assuming everything else is the same.

also you guys keep saying "per-pixel SNR".... as opposed to what???

there is only one type of SNR

https://en.wikipedia.org/wiki/Signal-to-noise_ratio_(imaging)

This also explains why SNR changes when you resample, even tho lots of people in here keep saying it doesn't

[brisk copper]

them people are goobers

[rain void]

it's a standard definition + simple addition of noise. I don't get why it's so hard to understand

[brisk copper]

yea

ig its due to people not actually looking into it

and just hearing someones incorrect info and believing it

[cold moon]

Well, nothing, it was just for clarity

[cold moon]

Wait let me think about that a sec

[ivory mural]

this isnt off topic polls

[cold moon]

It's not off topic, it's very relevant to this discussion

[ivory mural]

youre right, my bad

[brisk copper]

get out :(

[rain void]

as a logical conclusion the poll must be very on topic

[cold moon]

It does, I can definitely tell with my mosaic

[rain void]

that reminds me

[brisk copper]

i did a moon mosaic once and i got traumatised

[rain void]

image coming soon

(8x downsample because obviously)

[brisk copper]

b&w

[rain void]

yeah it's just like that irl trust me

[brisk copper]

i knew it.

[rain void]

also no stars in cygnus like you can clearly see

[brisk copper]

i think bob ate them all again

[cold moon]

I mean, why is it unfair? When comparing two things everything needs to be taken into consideration. I can argue that the improved QE of the 585 is also unfair so let's just pretend it's less sensitive

[cold moon]

Hell yeah, but I think you need 200% more panels

[brisk copper]

you should do the entire north and south hemisphere

[rain void]

well it's not unfair to compare, but it's unfair to campare by only looking at SNR

[zealous cloak]

Cygnus? More like GOAT

[rain void]

otherwise, you would always conclude that downsampling an image makes it better

[brisk copper]

cygnus is okay but theres more intresting stuff

[cold moon]

I agree, everything is important, size, qe, pixel size etc.

[cold moon]

It's definitely in the top 3 of constellations, easily

[zealous cloak]

Say Vela please

[rain void]

exactly, that's why i'm making a scenario with many of the parameters equal, to make the comparison easier

[brisk copper]

i havent actually seen it

i dont really look at southern targets

[rain void]

so that's essentially my point, a sensor that gathers more total signal will always have better SNR when you resample the sensors to have equal resolution

[cold moon]

I should make a website where you can compare sensors ngl, kinda like those CPU ones if you know what I mean

[rain void]

true lol

[cold moon]

If they have identical dark current/read noise and QE

[rain void]

and then the second part of my argument is: if you have the option to make an image of equal resolution, the image before resampling must be either that good or better

[brisk copper]

i dont really like big ha regions :(

[rain void]

no? i'm not saying just larger sensor, but gathers more signal

so it will always be true independent of the many specs, as long as the sensor gathers more data

which of course depends on the specs

[brisk copper]

pickle size

[cold moon]

That it gathers more signal sure, but not necessarily higher SNR even after resampling, I don't see how noise is not relevant. Ignoring the QE differences and assuming it does gather more signal the noise "gathered" matters too. If everything else is equal tho I agree

I'm talking about extreme cases here because it's very hard for just noise differences to overcome the size difference of, say, a 571 vs a 585 lol

[rain void]

noise is relevant 99 times out of 100, because you're almost certainly doing something wrong if the noise is not extremely dominated by shot noise

for long exposure DSO that is, but i assume we agree that's what we're discussing

[brisk copper]

lucky imaging has different rules

[rain void]

i'm not ignoring QE differences at all. gathers more signal very much takes the QE into account

and a quick calculation will show you that the sensor area is nearly twice as large, meaning twice the amount of photons hitting the sensor, while the QE loss of a 533 compared to a 585 is not at all a factor of two

[cold moon]

Yeah that's what I meant by "extreme cases", I'm just being pedantic

[rain void]

wdym?

[brisk copper]

not always twice the amount, 533 has worse ir qe so theres a loss there

oh, damn

i cant fkin read

my bad, ignore me

[cold moon]

I'm saying that if QE is different then a bigger sensor doesn't equate higher total signal. Obviously in practice the QE difference barely matters compared to the size. When talking about a 571 vs a 585 this is clear, but it could make a difference between sensors with a similar size

[rain void]

yeah but that's what i'm saying is very easy to take into account: S1/S2 = A1*QE1/(A2*QE2)

[brisk copper]

https://tenor.com/view/toby-cat-washed-soggy-wet-gif-16755473866192073463

[cold moon]

We really need a latex bot

But yeah

[rain void]

and that's essentialy what we discussed yesterday with the 585 mono vs 571 osc

[zealous cloak]

This guy is goofy as hell

[rain void]

where we just said that the "effective QE ratio" when taking into account you can get L data with a mono camera is 4x

[brisk copper]

drowned rat

[rain void]

tho that number is just made up, and i suspect it to be lower, I just granted it for the sake of argument

[brisk copper]

add a ~

[rain void]

?

[brisk copper]

idk im tire

[cold moon]

Roughly 4 is what they mean

[rain void]

we assumed it to be exactly 4 for ease of calculation

basically an upper bound

[cold moon]

I would compare, say, red filters

In that case 4x seems absolutely believable no?

Maybe even higher considering there's no bayer matrix

Assuming pixel size and everything else is identical

[rain void]

you can't just compare one filter say you have to throw out the rest on the OSC lol

you gotta compare it all

so x hours of OSC, vs x hours of mono LRGB with a 3:1:1:1 ratio or something

[cold moon]

Yeah but even then. Let's say 1h per RGB filter on mono (4x, 2x and 4x signal respectively, so I'll just say that it's the equivalent of 4h of red, 2h of green and 4h of blue). 3h on OSC are only better with green. Very simplified but I guess you get what I mean. I ignored lum but it depends on the ratio so I won't bother.

[rain void]

yeah but you have to compare that to 3hrs of OSC

since you used 3 hours

[cold moon]

I did?

[rain void]

yeah, 1h per filter on 3 filters is 3 hours

[summer turret]

gotta like show each other the math lmao

[zealous cloak]

Im more of a meth guy

[rain void]

you can't compare 1hr of osc to 1hr in each channel lol

that's cheating

[cold moon]

I literally said "3h on OSC"

The differences seem evident to me lol:
R: 4 vs 3;
G: 2 vs 3;
B: 4 vs 3.

In total you can argue that mono is 1 extra hour

[rain void]

oh right lol

well, no

let me show you

[cold moon]

In some very arbitrary and made up sense

[rain void]

say mono gets 4 units of R, G and B each in a span of 3 hrs, so 1 hr per filer.

in that same time, OSC will get 3 units of R, 6 units of G, and 3 units of B

both totaling 12 units total

this is just saying that you will get equal amounts of data with OSC and mono, assuming equal QE(!)

you mistake comes from counting green units as half the value of the other units

[brisk copper]

get triband filter that has 6nm bandpass for r and b and 3nm for g

green is genuinely evil btw

[rain void]

it would of course be outrageous to claim that OSC and mono gets equal amounts of data because of this, because the QE is very much not equal

[brisk copper]

you have to remember it lumpy bumpy on osc

[cold moon]

Yeah I see that, very silly of me

[rain void]

but for pure RGB, the QE difference is not even close to 4x on mono vs OSC, as OSC QE is obviously more than 25%

[cold moon]

For some reason I thougth that halving the mono signal instead of doubling the osc signal was equivalent

[cold moon]

Sho disagrees

[rain void]

most of the advantage on mono comes from being able to get L data, because assuming equal QE, that still gather 3x the amount of data!

[brisk copper]

sh better

[brisk copper]

you can do luminance with osc

its just like so much worse

[rain void]

????

but you can only do it once or something???

by stripping away the bayer matrix lol

[cold moon]

Turn the RGB image b/w

[rain void]

or i guess, when you do, you can never stop

[brisk copper]

[cold moon]

There's also processing advantages, I find it easier to manage colours with mono

[brisk copper]

it works for my ir data

[rain void]

oh yeah there's definitely many other advantage to mono, but this is just from a pure signal / SNR viewpoint

[cold moon]

Just go high enough that the bayer matrix is transparent smh

[brisk copper]

[brisk copper]

i want to get a mono cam to rid myself of the bayer matrix bc it still effect the ir transmittance :(

i already got all the filters so its gonna be cheaper than if i did sho

or rgb

[cold moon]

Buy my ares-m fr, I want to upgrade to aps-c

[brisk copper]

what type sensor that?

[rain void]

....

[brisk copper]

doesnt it have ass ir

[rain void]

bigger sensor better

[swift heart]

SNR per unit of solid angle

[cold moon]

Precisely

[cold moon]

No but yes

[brisk copper]

????

NO

[rain void]

lmfao i'm just joking

i meant

more signal better

[cold moon]

Tbh I would get a 455 if I could

[brisk copper]

:3

[cold moon]

But I'm too poor

And I don't feel like stealing people's kidneys tonight

[rain void]

that doesn't make sense

[cold moon]

SNR per bald eagles

[brisk copper]

smoke pot and it will eventually

[brisk copper]

snr per cubed burger squared

[rain void]

if it's not a joke i would love to understand

[cold moon]

Very american

[rain void]

but you can't define SNR as "SNR per solid angle"

that's just circular

[brisk copper]

yup

[cold moon]

SNR per SNR

[brisk copper]

how to you measure sgr?

-# signal to grain ratio

[rain void]

SNR = 1 in all cases 🔥 then we have none of this math crap to deal with

[brisk copper]

for film

[brisk copper]

yooo, my cam is now so much better

i already feel the amount of signal increase

[cold moon]

It can be if we really want it, let's just redefine noise as being just the signal

[brisk copper]

good idea, time to make sure i have more noise, wouldnt want them pesky stars being visible

[rain void]

and here we thought we were actually imaging real objects smh. Turns out that all our images just turned out like that by pure rng

[swift heart]

Each pixel subtends some amount of solid angle in the sky, so you can just reinterpret your SNR independently of the sensor's spatial discretization

[rain void]

..but how are you calculating the SNR

because SNR per angle is a very circular definition

[rain void]

like gimme an actual formula, i'd love to see it

[swift heart]

Would you call SNR divided by the pixel scale something different?

[rain void]

... you're still defining SNR in terms of SNR

[swift heart]

Then it's not SNR as you define it, but how else would you compare the images created by sensors with different pixel scales (all else being the same)?

[rain void]

wdym? my definition of SNR is this

has nothing to do with pixel size or sampling

[brisk copper]

r u sure?

i see a +4 on the end

/j

[rain void]

of course in the case of an image taken with a telescope and astro cam, the specific values of μ and σ will depend on many factors including pixel size, sampling, specs of the scope, etc.

but fundamentally it has nothing to do with those

[brisk copper]

do astrophoto with a pinhole

do narrowbad ap on a pinhole

[swift heart]

How would you compare the quality of two images taken with the same setup except for sensor pixel size?

[rain void]

i'd use the formula

[swift heart]

How are you measuring µ and σ

[rain void]

it's just the mean and standard deviation of the pixel values in my image

of couse we can easily show that σ = sqrt(μ) for a poisson distribution, and μ is proportional to the area of a pixel

that way we can get SNR ~ pixel side length

and equally easily we get the well known SNR ~ sqrt(t)

and SNR ~ 1/F ratio

and SNR ~ aperture diameter

where each proportionality is ofc assuming all else constant

but notice how SNR is independent of sensor area

because the mean and standard deviation will still be the same if you just add more pixels

i guess i should say independent on pixel count rather

[rain void]

so that's how i'm measuing μ and σ

but once again i'd love to hear other definitions

[swift heart]

I guess my question for you is: how would you quantify the quality of an image under the assumption that you will not be guaranteed to see it at a particular scale?

[rain void]

Then I'd just use the definition directly

Like there's nothing hard about it, just look at a bunch of pixels and calculate mu and sigma from the standard definition from statistics

That's what makes this definition truly objective: we don't need to know anything about how the images were captured, but only look at the image

[swift heart]

The image? Or the data?

[rain void]

What's the difference?

[swift heart]

Well what are you interested in? The performance of your rig or the quality of your final image?

[rain void]

Once again, what is the difference? Surely the performance of your rig is exactly equal to the quality of the images it produces

[swift heart]

Do you process your data into images?

[rain void]

Unless ofc you are specifically interested in those things you can't read from the images, but that's another discussion

[rain void]

I would say my data is the images?

What other data are you getting? The fits header or what are you talking about??

[swift heart]

I wouldn't. I mean, the image represents my data, but I would measure the quality of the image differently from how I would measure the quality of my data

[rain void]

what

lmfao

[tender hollow]

Pro astrophotographers just paint

[brisk copper]

i draw every so often

look at the resolution on that image!!

quite a bit of star bloat tho

[swift heart]

Processing is transforming your data into an image. (Unless you're a linear image enjoyer who loves looking a a black screen with the occasional star, then sure, the identity transform suffices )

[rain void]

Well I'll let you think that, but it doesn't really relate the discussion of the SNR of images we were talking about

[brisk copper]

you can see androm and orion in linear images

[swift heart]

True

[marble hearth]

67 new messages

Kill me

[rain void]

Sure, but surely it's the SNR of the linear image we're discussing

[brisk copper]

if you spend every hour of a year you could probably get some pn's too

[swift heart]

fair enough, I just treat "data" and "image" in a conceptually very different way

[ivory mural]

obJCt

[rain void]

Sure, but how is that relevant to the definition of SNR

[brisk copper]

it says objet actually

[brisk copper]

@rain void whats the snr?

[zealous cloak]

Fray balls my beloved

[rain void]

Well there would be no problem calculating it, just not meaningful to compare with an actual image

[ivory mural]

the loop in the e isnt even big enough to make a hole

that is a fancy c

[brisk copper]

it is :(

[brisk copper]

its a prediction of my eye sensor tho

[brisk copper]

its roght ther

[rain void]

But once again, I would love to see your strict mathematical definition of SNR if you don't agree with mine

It's totally fine if you don't have one, then there's just not much to discuss

[swift heart]

If you're comparing the quality of linear images, sure, that's fine. My question for you is: if you have two datasets taken with the exact same setup with only one difference (one of the cameras has 2x smaller pixels than the other), would you say that one dataset is of better or worse quality than the other?

(or maybe not 2x, you can make that scale value any positive real number)

[brisk copper]

you cant say it betteer or worse tho

ig it could become oversampled

[rain void]

There's a million subjective ways to measure "quality" of an image/dataset. Personally I highly prioritize SNR and fwhm/field radius, as that's just two completely objective ways of measuring "how grainy it looks" and "how sharp it looks", without needing to know any details about the scope. Not saying they're the perfect measurement, but I'd say they're close.

SNR, as opposed to "quality", of my image is not subjective tho. That has a completely objective and mathematically precise definition

And that's not even taking the more artistic parts like composition or nice colors into account, which I ofc also weigh highly

[swift heart]

Would you say that one of the datasets is better or worse than the other?

[rain void]

So you can say anything about how you judge whether an image is good or bad and all I can do is accept that, but when it comes to SNR specifically, that's just not the case

[rain void]

Of which datasets?

[swift heart]

The two taken with the same optics and conditions, and sensors but with different pixel scales

That is all the information you have to make the judgement.

[rain void]

In many cases yes, but that's not the same as saying better SNR

[swift heart]

Exactly!

[rain void]

exactly what?

[swift heart]

It isn't the same thing as saying better SNR.

My point here is: SNR might be an objective measure of your data, but is it measuring what you want it to measure

[rain void]

I would simply say that the dataset with smaller pixels is better (assuming read noise to be irrelevant), because it can be transformed into the other by simply resampling

[rain void]

Yeah that's what I also just explained, so I guess we agree on that

[swift heart]

Now if you're in this camp, then SNR might not be a good measure of your data quality, because it varies with pixel scale.

(obviously I'd take my 3008x3008 dataset over a 1x1 dataset, so pixel scale is relevant to some extent)

[rain void]

Well that's a completely other question that I would love to discuss, but that's not really what I asked for to begin with

[brisk copper]

you use 1,1 pixel sensor like a radioscope

[rain void]

I specifically asked about other definitions of SNR that people use, because I know they must exist when they make claims such as SNR not changing when you resample

[swift heart]

People have less rigorous working definitions of "signal" and "noise" honestly

[rain void]

And that's totally fine, I just want to understand how anyone can reach rigorous conclusions like the one I mentioned

Because "SNR not changing when you resample" is a pretty rigorous statement, and surely isn't just vibes based lol

Or maybe it really is just vibes based idk

Anyways it's getting really late and I have lectures in the morning, but it was nice chatting with you :)

[swift heart]

Tbh I think it is. One of the things I've been surprised by is how few people are interested in this kind of rigor. (That's not a knock on the community, I think it's totally fine if people are just in it for the pretty pictures!)

[swift heart]

Yeah get some rest!

[brisk copper]

pretty picters

[cerulean elm]

I go for Relflector. But the superior reflector imo is the TMA Design

[summer turret]

good knowing you @undone sinew

[delicate lance]

The question shouldn’t be „should we“ but „when should we“

[fossil temple]

This is true

https://tenor.com/view/sacrifice-gif-21845160

[low lance]

What what happened

[summer turret]

hes being sacrificed

[low lance]

Oh lmao

[void jolt]

i mean, come on

[brisk copper]

dont let the hobby win tho

dont quit

[delicate lance]

Astro hates everyone

[lofty jetty]

trial, yeah
when I get pixinsight, yeah I'll buy it too

[somber belfry]

can u bx smth for me then

[lofty jetty]

uhm sure what about #1262558166898446416

Oh and what settings do you want

[somber belfry]

ive already posted the data there

[undone sinew]

❔

[brisk copper]

can u bxt my data too 🥺

[lofty jetty]

have I become the blurx guy :(
ok fine but you are the last person I'm doing it to because I barely have time for my data myself

[brisk copper]

okie

[ivory mural]

All protostars are being sacrificed

[zealous cloak]

https://tenor.com/view/naocara-gif-951861008809332010

[fossil temple]

Rip

[undone sinew]

Sacrificed to who??

[ivory mural]

Rory

[undone sinew]

damn

disgusting mininerd filth

[delicate lance]

watch your tounge as you are beneath us

peasant

[undone sinew]

you wear your pink attire like a fool, acting as if your beginnings weren’t blue either

never insult protostars again lest the meganerds will whoop you

[delicate lance]

i reject my protostar self

for i am pure in my nerd form

[undone sinew]

the tree never sprouts without the seed

[delicate lance]

the seed however would not be able to grow without the tree

[undone sinew]

yet you still poison the seeds to snuff out the protostars in preference to coexisting

implying that you are killing your kind as a result

[ivory mural]

Sounds fun

[brisk copper]

hehe :3

[undone sinew]

SIX SEVEN

HDSKSHJASHHDEBWHDBSJDHRJDJ

HDJDJDNKD

JDJSOWEKDBS

HEISJSJDJSSJSS

[delicate lance]

https://cdn.discordapp.com/attachments/1370449380578033694/1432434755185545337/geto-no-brain.gif

[fossil temple]

https://tenor.com/view/monke-brain-shrinking-gorilla-stupid-gif-21102851

[low lance]

My friend quite 😔

A moment of silence for the fallen

@violet tinsel

[summer turret]

He did?

Lol

[silver holly]

Correct, it much larger than people think.

[low lance]

Smaller the difference is smaller not larger

Each channel isn't narrowband

[silver holly]

I’ve used both. The channel overlap on osc makes mono worthwhile even if it had lower snr. Which it doesn’t.

[low lance]

Oh no mono is still better by all means especially in terms of color accuracy

[silver holly]

If shooting Ha on a colour camera you will have roughly 25% of the signal of a mono camera with the same size pixels.

Same with blue. And red.

Green is 50%

Basically 75% of your signal is wasted on osc if you are after red, Ha or blue signal.

Oiii is approximately 60% waste

The 585 also has hugely better QE for Ha. 60% on the 571 versus 80% on the 585.

Purely QE

Both mono

Having a colour 571 means your QE is kinda similar but your actual signal is 75% lower again.

[low lance]

Absolutely i agree

[delicate lance]

That only holds true in Broadband lol

Shoot narrowband and OSC becomes more colour accurate assuming you use the typical palleting

[low lance]

Yeah thats what i meant when people say color they are referring to broadband

[fossil temple]

It is decided

Who shall be the first Sacrifice

[tribal coral]

I feel like one option is clearly missing

[bronze minnow]

whar

[tribal coral]

you forgot G3

[bronze minnow]

g3 was barely in 2025 but I'll probably put it in the end of year one

[tender hollow]

You will always be a protostar

You hide beneath a pink facade

But I know what you really are

[delicate lance]

the matrix trying to take me down again with ai images i see

[brisk copper]

now now, do you need help making your way back to your end of life care room?

[swift heart]

Mono is worse than OSC for color accuracy if you're using standard RGB filters with no overlap. It will never be able to accurately reproduce spectral colors

[low lance]

How so

[swift heart]

If there's no overlap between the filters, a pure spectral source can only ever pass through one of the filters. So spectral colors end up pure red, green, or blue

Star colors aren't affected by this (as are any broadband sources)

[low lance]

Ohh i didn't know that

So it's normal to look thru an astro rgb filter and see some green thru the red filter

[swift heart]

Your eye's cone cells overlap in a way that makes it impossible to just activate the red cones with visible light

[low lance]

No i mean look at an objet emitting green light and see the green from it

[swift heart]

I don't think you'd see pure green, but definitely some yellow

If you're looking through most astronomical R filters

[low lance]

For example when i look at a green traffic light thru the red filter it still looks green albiet a bit dimmer

[delicate lance]

yea

and no

[undone sinew]

nir wins always

[swift heart]

I was thinking I could probably get all the NIR data I need on NGC 2403 tonight

[low lance]

Do NIR on a molecular cloud

Get all the juicy protostars forming

[cerulean elm]

NIR on NGC-1333

[swift heart]

I have a list of targets I need data on...either way looks like I'll be following the poll since I can't exactly shoot long subs with the gusts tonight

[low lance]

Yes

We need more people doing nir it's like seriously underrated

[cerulean elm]

#1326833769495269376 message
I'd call this serious

[low lance]

Idk kinda looks cooked maybe try keeping ha red

Basically just switch the spots between the Ha and NIR 720

[cerulean elm]

Why?

[low lance]

Idk right now it looks like your data got corrupted like if i didn't know ha was blue i would have 100% thaught the data got corrupt

[cerulean elm]

Euclid Ha is blue. The channels are ordered in their spectral order. The longest wavelength being the red and the shortes being blue

[low lance]

Yeah but like the nebulas just look like blue blobs

Or maybe its the color balance

Idk for this type of stuff you kinda just gotta mix around until something looks good because there is not set way to do it

[brisk copper]

guess what im doin :3

[limpid geode]

Holy shit

[cerulean elm]

Well... this in the first place is an extremely unorthodox palette. So even this is more than what anyone else achieved around here😂

[low lance]

Thats true

But that doesn't mean there isn't room for improvement

[brisk copper]

is it a dual color pallete ?

[low lance]

Nope

Triple

[cerulean elm]

Trichromatic

[brisk copper]

ah

[low lance]

Ha is blue 640 is green and 720 is red

[brisk copper]

ah

[low lance]

Which means those random blue blobs are actually nebulas

[brisk copper]

ye

[cerulean elm]

The reason why the image is greenish is because I had 4.5h of IR 685 and I think sub one h of IR720

[brisk copper]

i wanna at some point do subtraction on ir so i can get ir exclusive data

[brisk copper]

[low lance]

Under stretch the green then or over strech the red

Or both

[brisk copper]

or remove ir 720 and do a duochrome imag

[cerulean elm]

This is the continuum subtracted Ha.

[brisk copper]

orange blue / orange cyan palette ftw

[brisk copper]

u should continuum subtract the ir

[low lance]

Oh damn didnt know there was detail in the Ha

[low lance]

Yes

[cerulean elm]

You mean like this?

[brisk copper]

how do you do it?

i wanna do ir continuum subtraction

[cerulean elm]

The problem is, subtract it from what? I had to subtract it from L

[brisk copper]

(without pix)

[brisk copper]

i have luminance

but i used ir cut

[cerulean elm]

Same

[brisk copper]

huh, ig ill have to try it

[low lance]

Wait a minute is that a negative or normal

[cerulean elm]

Normal

[low lance]

Huh i did expect it to look that different from visible

[obsidian plover]

Haven't done astro in 2-3 weeks

Just all rain

[urban quiver]

October was good for me. Weather got bad only after my mini pc died

[swift heart]

October has been pretty bad in Wisconsin

But the few good nights have been really nice

[sharp berry]

same

my mount decided to have problems in b1

[obsidian plover]

Your kidding 😭