ENG New site

Advanced search

[ New messages · Forum rules · Members ]
Forum » SpaceEngine » Archive » Work progress and public beta test - 0.9.7.4
Work progress and public beta test - 0.9.7.4
RedRuinDate: Sunday, 24.01.2016, 01:06 | Message # 1291
Space Tourist
Group: Users
United States
Messages: 34
Status: Offline
metallicgod, I usually fix those kinds of errors by upgrading my drivers.
 
LookAtDatDakkaDate: Monday, 25.01.2016, 02:49 | Message # 1292
Explorer
Group: Users
United States
Messages: 156
Status: Offline
Quote RedRuin ()
metallicgod, I usually fix those kinds of errors by upgrading my drivers.


Agreed! smile





NVIDIA 960 GTX 2048MB
 
The_BlazerDate: Monday, 25.01.2016, 17:15 | Message # 1293
Space Tourist
Group: Users
Italy
Messages: 34
Status: Offline
In general, this patch seems to be extremely crash-prone to me. I can play for maybe a few minutes before it crashes. My setup is as follows: GTX 770 (2gb VRAM), i5 4570, latest nVidia drivers and I also disabled Shadowplay to see if it was causing problems. My settings are the same as with 0.973 (which almost never crashes), LOD is set to maximum (just like in 0.973). Crashing occurs especially when exploring planets with terrain, I know that max terrain LOD is hard to render but it really shouldn't be downright crashing on a setup with a GTX 770.

Example of crash-causing planet: RS 8409-2267-6-185835-90 5.1 It's worth noting that crashes here are much more frequent, but they still happen way too often even elsewhere.

EDIT: I also noticed that many atmospheres seem to have an extremely sharp cutoff when you are a certain height above the surface. If you go higher the atmosphere will be below you and render properly, if you enter it, it will also cover the sky and render properly. If you are near the "edge", this happens.






Born too early to explore space... But just in time for Space Engine.

PC Specs:
GTX 770 (2GB VRAM) - i5 4570 - 8GB RAM


Edited by The_Blazer - Monday, 25.01.2016, 17:35
 
HarbingerDawnDate: Monday, 25.01.2016, 17:34 | Message # 1294
Cosmic Curator
Group: Administrators
United States
Messages: 8717
Status: Offline
Quote The_Blazer ()
I know that max terrain LOD is hard to render but it really shouldn't be downright crashing on a setup with a GTX 770.

You're doing something that fills up your entire (and very modest) VRAM in seconds, and SE has always had memory issues. It's amazing you didn't have more crashes before.

That said, the current 0.974 release is called beta for a reason. It's not finished, and of course will have bugs and stability issues.





All forum users, please read this!
My SE mods and addons
Phenom II X6 1090T 3.2 GHz, 16 GB DDR3 RAM, GTX 970 3584 MB VRAM
 
CanapinDate: Monday, 25.01.2016, 19:51 | Message # 1295
Astronaut
Group: Users
France
Messages: 52
Status: Offline
Quote The_Blazer ()
EDIT: I also noticed that many atmospheres seem to have an extremely sharp cutoff when you are a certain height above the surface.

Yup : http://en.spaceengine.org/forum/4-3295-1
 
HarbingerDawnDate: Monday, 25.01.2016, 19:54 | Message # 1296
Cosmic Curator
Group: Administrators
United States
Messages: 8717
Status: Offline
Yes, that atmosphere issue exists for planets which have extremely compact atmospheres.




All forum users, please read this!
My SE mods and addons
Phenom II X6 1090T 3.2 GHz, 16 GB DDR3 RAM, GTX 970 3584 MB VRAM
 
SpaceEngineerDate: Monday, 25.01.2016, 21:57 | Message # 1297
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
Updates what you guys are wanted really long time: chemical composition of atmospheres. Current implementation is far from complete and may give unrealistic results.

Good old CO2 atmosphere:


This is sauna planet - looked like a terra with boiled out oceans. Not a far future of Earth:


Terra with life - atmosphere seems suitable for humans... Well, no, partial pressure of O2 is just 0.089 atm (human needs at least 0.16):


This planet could be titan if it were a bit colder:


And of course gas giant:


This ice giant seems lost all it's H2 and He... This is a "chtonian" planet. It's radius must few times smaller than showed here, but this is not implemented yet.

Attachments: 6080864.jpg(329.0 Kb) · 6631886.jpg(297.6 Kb) · 6133489.jpg(296.5 Kb) · 4917574.jpg(382.9 Kb) · 3232167.jpg(289.8 Kb) · 5504579.jpg(274.6 Kb)





 
SpaceEngineerDate: Monday, 25.01.2016, 22:03 | Message # 1298
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
You may also notice what planets have new palettes. Well, no, it's not a mew palettes, but a new feature: recolouring of the atmospheric scattering. I added ability to set custom hue and saturation for atmosphere of any planet. SE uses this to randomize colors of the standard atmo models, to make new ones (pink, gree, yelow etc), and of course this settings are now available in the scripts and editor:







Performance hit is beyond FPS counter accuracy, but possibilities is great. No need to create many new precumputed atmopshere models, which uses a lot of video memory: just take one and change it's color on-the-fly. This is not scientifically accurate of course, but if you are not going to make crazy acid colours, it could look believable.

Edit: and there is a new feature introduced by HarbingerDawn: option to switch diameter display to radius. It will be in the Player menu.

Attachments: 3000040.jpg(364.4 Kb) · 2118575.jpg(358.4 Kb) · 0924920.jpg(356.8 Kb)





 
InariusDate: Monday, 25.01.2016, 22:04 | Message # 1299
Explorer
Group: Local Moderators
France
Messages: 237
Status: Offline
@Spaceengine : That's good news !I was expecting this for a long time !
Great !
 
AerospacefagDate: Monday, 25.01.2016, 22:10 | Message # 1300
Pioneer
Group: Users
Russian Federation
Messages: 401
Status: Offline
Quote SpaceEngineer ()
Performance hit is beyond FPS counter accuracy, but possibilities is great. No need to create many new precumputed atmopshere models, which uses a lot of video memory: just take one and change it's color on-the-fly.

But, I never actually understood, what is the actual difference between different atmospheric models, other then in colour?
 
Thomas988Date: Monday, 25.01.2016, 22:30 | Message # 1301
Space Pilot
Group: Users
United States
Messages: 125
Status: Offline
*reads update*

*dies*

God's work, SpaceEngineer! You have my applause! biggrin





All you need in life are space games and typhlosions.
 
HarbingerDawnDate: Monday, 25.01.2016, 22:32 | Message # 1302
Cosmic Curator
Group: Administrators
United States
Messages: 8717
Status: Offline
Quote Aerospacefag ()
I never actually understood, what is the actual difference between different atmospheric models, other then in colour?

There's Rayleigh scattering (all-sky scattering of light by molecules of air), Mie scattering (back-scattering of light by dust and aerosols), and extinction (dimming of certain wavelengths of light as they pass through the atmosphere). All of these things, and variables associated with them, can be adjusted independently for each atmosphere model. For example, the atmosphere models of Earth and Neptune are identical except for differing Mie extinction values, and Mie scattering is responsible for the reddish color of the atmosphere of Mars. So colorizing the atmospheres cannot replace having different atmosphere models, but it does mean you can have a large variety using only a small number of models.





All forum users, please read this!
My SE mods and addons
Phenom II X6 1090T 3.2 GHz, 16 GB DDR3 RAM, GTX 970 3584 MB VRAM
 
AerospacefagDate: Monday, 25.01.2016, 22:48 | Message # 1303
Pioneer
Group: Users
Russian Federation
Messages: 401
Status: Offline
Quote HarbingerDawn ()
There's Rayleigh scattering (all-sky scattering of light by molecules of air), Mie scattering (back-scattering of light by dust and aerosols), and extinction (dimming of certain wavelengths of light as they pass through the atmosphere).

So these effects are integral part of the atmosphere model, and cannot be adjusted separately?

I hope that freeing some video memory in that way will not only improve performance in the long run, but will also allow to add more different effects.
 
CanapinDate: Monday, 25.01.2016, 23:16 | Message # 1304
Astronaut
Group: Users
France
Messages: 52
Status: Offline
Okay so now that we have the chemical composition of the atmospheres, I guess we can imagine the shape of crystals we can found in them, import and use them in Halosim and use photoshop to simulate realistic ice halo displays on unknown planets.
My dream, but I lack knowledge. And motivation, too, seems to be a hard work.


Edited by Canapin - Monday, 25.01.2016, 23:21
 
VilfateDate: Tuesday, 26.01.2016, 00:18 | Message # 1305
Astronaut
Group: Users
China
Messages: 49
Status: Offline
Quote SpaceEngineer ()
This is sauna planet - looked like a terra with boiled out oceans. Not a far future of Earth:


Finding significant amounts of water in a desert's atmosphere looks extremely impressive to me.

The planet is spinning considerably fast by the way...

=================Some more questions about compositions================
Is the data somewhat linked to other characteristics? Like Cl2 and CH2Cl2 etc. for a Greenworld, or mandatory O2 for all organic multicellular planets.

Will the data concern type and age of host stars? For example, a hot desert near an O8 V and one orbiting a K2 III may behave totally differently in terms of atmpspheric composition (like due to various strengths of solar winds)







Edited by Vilfate - Tuesday, 26.01.2016, 00:18
 
Forum » SpaceEngine » Archive » Work progress and public beta test - 0.9.7.4
Search: