Ultimate space simulation software

 
Leyo
Space Tourist
Space Tourist
Posts: 34
Joined: 17 May 2021 03:12

Work progress 0.991

18 Jul 2021 02:25

I made a video where you can easily see the difference between the old and the new algorithm. Please watch it in 1080p minimum ; it is better in 1440p.

 
User avatar
SpaceEngineer
Author of SpaceEngine
Author of SpaceEngine
Topic Author
Posts: 1125
Joined: 17 May 2016 22:16
Location: Saint-Petersburg
Contact:

Work progress 0.991

18 Jul 2021 12:12

This is not beacuse new algorithm has lower sharpness limit (well, it has, because values > 1 are not allowed because of graphical glitches). The thing is that galaxy/nebula rendering used its own sharpness pass, but in 1955 I switched it off. Because otherwise skybox/impostor nebulae will be oversharped (sharpness applied twice). I will consider to find another solution.
 
Leyo
Space Tourist
Space Tourist
Posts: 34
Joined: 17 May 2021 03:12

Work progress 0.991

18 Jul 2021 12:48

Thanks a lot for considering this subject.
wouldn't it be easier for you to just let the choice with a tick or a switch in the menu ? 
People affected by black borders could choose CAS while people with higher res could choose the old one. Everyone would be served, and you won't have to spent many time on researching and testing a new solution (a charge you probably weren't expecting).
 
User avatar
Khorrah
Explorer
Explorer
Posts: 170
Joined: 11 Nov 2019 17:50
Location: Earth

Work progress 0.991

18 Jul 2021 18:27

There will be an galaxy visual improvement?
Hello, I LOVE lasagna and Space Engine. ♥

My Specs: Intel Core I5 12600K, Corsair Vengeance Pro SL 32GB, MSI Z690-A WIFI DDR4, NVIDIA RTX 4070 SUPER AORUS MASTER 12GB.
 
A-L-E-X
Galaxy Architect
Galaxy Architect
Posts: 3498
Joined: 06 Mar 2017 20:19

Work progress 0.991

18 Jul 2021 23:33

Thanks a lot for considering this subject.
wouldn't it be easier for you to just let the choice with a tick or a switch in the menu ? 
People affected by black borders could choose CAS while people with higher res could choose the old one. Everyone would be served, and you won't have to spent many time on researching and testing a new solution (a charge you probably weren't expecting).
where are there black borders?  My screen resolution is 1280x1024 but my videos are in 720P using OBS, I find it useful because cutting off the bottom of the screen cuts off the console displays, so it's mostly about scenery and not the control panel.
 
Leyo
Space Tourist
Space Tourist
Posts: 34
Joined: 17 May 2021 03:12

Work progress 0.991

19 Jul 2021 10:32

AS far as i understood, Vladimir set up a new algorithm mostly because  some people suffers from seeing black lines on borders of some objects. On my side i'v never seen them too.

@Khorrah
This is just a new algortihm for general sharpness, not especially the galaxies. I asked Vladimir if he could keep the old one available because the new one can has unwanted effects on some config/resolution.
 
A-L-E-X
Galaxy Architect
Galaxy Architect
Posts: 3498
Joined: 06 Mar 2017 20:19

Work progress 0.991

19 Jul 2021 10:45

AS far as i understood, Vladimir set up a new algorithm mostly because  some people suffers from seeing black lines on borders of some objects. On my side i'v never seen them too.

@Khorrah
This is just a new algortihm for general sharpness, not especially the galaxies. I asked Vladimir if he could keep the old one available because the new one can has unwanted effects on some config/resolution.
Oh thanks, I thought it was black borders on the edges of the screen when 4:3 video is displayed in 16:9 format lol.  I've never seen it so never knew this problem even existed lol.
 
User avatar
SpaceEngineer
Author of SpaceEngine
Author of SpaceEngine
Topic Author
Posts: 1125
Joined: 17 May 2016 22:16
Location: Saint-Petersburg
Contact:

Work progress 0.991

19 Jul 2021 15:38

Just look at stars in front of Milky Way with sharpness 1 or more, and you will see (in older versions).
 
User avatar
longname
Pioneer
Pioneer
Posts: 381
Joined: 13 Apr 2017 07:20
Location: ∞/The Multiverse/The Universe/Local Cluster/The Milky Way/Orion Arm/Sol System/Earth-Moon/Earth/UK

Work progress 0.991

19 Jul 2021 16:37

Just look at stars in front of Milky Way with sharpness 1 or more, and you will see (in older versions).
I don't see the problem
[dah<500,26>dah<180,14>dah<180,21>dah<500,19>dah<180,26>dah<500,21>]
 
User avatar
SpaceEngineer
Author of SpaceEngine
Author of SpaceEngine
Topic Author
Posts: 1125
Joined: 17 May 2016 22:16
Location: Saint-Petersburg
Contact:

Work progress 0.991

19 Jul 2021 16:40

This is just a new algortihm for general sharpness, not especially the galaxies. I asked Vladimir if he could keep the old one available because the new one can has unwanted effects on some config/resolution.
Galaxies/nebulae have theor own sharpness pass, but it was disabled, because the global (scene) sharpness did the job even better. And having double sharpness makes them over-sharpen. Now I added adaptive sharpness pass for galaxies/nebulae, which is reversed to the global sharpness slider value. You can disable this "auto sharp" in the debug window, if you with to have oversharpen nebulae (SE will remember your choise in the config).

Screenshots for comparison - open in the new tab.

All sharpness is off.
scr00864.jpg
Galaxy/nebula sharpness is at 1 ("Volum sharp"), global at 0 ("Global sharp"). Stars are not sharpen, looks the same as in previous screenshot. Nebulae are great now.
scr00865.jpg
Galaxy/nebula sharpness is at 0 ("Volum sharp"), global at 1 ("Global sharp"). Stars are sharpen, which is acually unwanted. Big nebula looks almost the same. But look at the pink sprite nebula near the left edge, and at the tiny nebula near the bottom. They look much better now! Global (CAS) sharpness works better than a special nebula sharpness pass.
scr00866.jpg
Both sharpness is at 1. This is too much IMHO. Note oversharpen black blobs in the big nebula.
scr00867.jpg
So the "Volum auto sharp" checkbox enables adaptation, so that "Volum sharp" is locked and its value = 1 - Global sharp value. The same system is used for 3D rings ("Solid sharp"), because they look especially horrible then sharpness applied twice.
 
User avatar
longname
Pioneer
Pioneer
Posts: 381
Joined: 13 Apr 2017 07:20
Location: ∞/The Multiverse/The Universe/Local Cluster/The Milky Way/Orion Arm/Sol System/Earth-Moon/Earth/UK

Work progress 0.991

19 Jul 2021 16:51

Galaxy/nebula sharpness is at 1 ("Volum sharp"), global at 0 ("Global sharp"). Stars are not sharpen, looks the same as in previous screenshot. Nebulae are great now.
Personally I preferred the 'oversharpened' nebula. I don't like the blurrier, just generally not as good nebula you've not provided us with the option to not use.
[dah<500,26>dah<180,14>dah<180,21>dah<500,19>dah<180,26>dah<500,21>]
 
User avatar
HarbingerDawn
SE Team Member
SE Team Member
Posts: 694
Joined: 22 Aug 2016 15:11
Location: CT, USA
Contact:

Work progress 0.991

19 Jul 2021 17:28

This is the ideal sharpening experience. You may not like it, but this is what peak visuals looks like.
Image
[hr] Jokes aside, those of you who were using very high sharpening values, did you have FXAA enabled? If so, you should disable it.
Ryzen 7 3700X, 64 GB DDR4-3200 RAM, RTX 3090 24 GB VRAM
Posts on old forum: 8717
 
Leyo
Space Tourist
Space Tourist
Posts: 34
Joined: 17 May 2021 03:12

Work progress 0.991

20 Jul 2021 11:25

@Space Engineer
 I now understand what you were working on. I've always seen those black dots/lines on nebulae, but it always seemed obvious to me that it was caused by the nebulae model/texture itself.

Because we can see it even more when we get close to it (screenshot taken with the new sharpness). And because many others Nebulaes don't have this issue.
scr00079.jpg
FXAA disabled definitly renders better visuals with new sharpness algorithm... but is it normal ? i don't think so.
With this trick and those given by Space Engineer with the debug toolbox, we can approach what we had before, which is a good news.
But we still have lost the clear vision of depth that comes along with the old algorithm, which also sharpened the stars. 

I maintain we lost a bit in global visual quality, but since it has a large (full ?) part of subjectivity and the release has been made, i just hope the old algorithm will be available again in futures update.
 
Marud
Observer
Observer
Posts: 7
Joined: 21 Jul 2021 09:27

Work progress 0.991

21 Jul 2021 09:35

I registered specially to ask one question. After the release of rings, what is the new big goal? Clouds? Water? Please don't say this is ships' fuel or their temperature stuff, nobody wants that, it's no use. I'd wish that the new big update was improving real planets. So painful to see Earth and Mars that flat and boring. Isn't it the time to reimagine them? I mean add detailed landscapes above the existing map.
 
User avatar
longname
Pioneer
Pioneer
Posts: 381
Joined: 13 Apr 2017 07:20
Location: ∞/The Multiverse/The Universe/Local Cluster/The Milky Way/Orion Arm/Sol System/Earth-Moon/Earth/UK

Work progress 0.991

21 Jul 2021 11:28

 After the release of rings, what is the new big goal? Clouds? Water? Please don't say this is ships' fuel or their temperature stuff, nobody wants that, it's no use.
As someone who enjoys the flight simulator mode, it's only no use to you. You don't choose what everyone else wants. To see ships worked inbetween graphical overhauls would give us something to do in the meanwhile.
I'd wish that the new big update was improving real planets. So painful to see Earth and Mars that flat and boring. Isn't it the time to reimagine them? I mean add detailed landscapes above the existing map.
The main problem as I see it is generating procedural details on top of mapped planets. If SpaceEngineer got it to work, then you might end up with unrealistic coastlines. At least SpaceEngineer has a more realistic map for Earth cooking up with better, more accurate visuals, although it still won't add small details.
[dah<500,26>dah<180,14>dah<180,21>dah<500,19>dah<180,26>dah<500,21>]

Who is online

Users browsing this forum: No registered users and 0 guests