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
SpaceEngineerDate: Saturday, 21.11.2015, 23:19 | Message # 826
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
Quote DeathStar ()
If so is true, I have an idea on how to represent the Cosmic microwave background, but I don't know how feasible it would be.

Yes, this could work. Noise texture is not really needed, SMBR spectrum is uniform black body radiation with temperature fluctuations of ~10-5. It's impossible to see them even on highly blueshifted image. CMBR will look just like uniform glow of the sky.





 
DeathStarDate: Saturday, 21.11.2015, 23:50 | Message # 827
Pioneer
Group: Users
Croatia
Messages: 515
Status: Offline
Quote SpaceEngineer ()
Yes, this could work. Noise texture is not really needed, SMBR spectrum is uniform black body radiation with temperature fluctuations of ~10-5. It's impossible to see them even on highly blueshifted image. CMBR will look just like uniform glow of the sky.


Great to hear! All those pictures of the CMBR have led me to the false conclusion that the low and high temperature areas have a sharp contrast, hence my suggestion for the (as you pointed out, unnecessary) noise. Even without the noise, though, the CMBR would be nice to have, with space in front of you starting to glow as you travel at intergalactic speeds(or even lower speeds).
 
InariusDate: Sunday, 22.11.2015, 23:24 | Message # 828
Explorer
Group: Local Moderators
France
Messages: 237
Status: Offline
That's really great work ! Such details ! Wonderful !
 
parameciumkidDate: Monday, 23.11.2015, 04:54 | Message # 829
Explorer
Group: Users
United States
Messages: 277
Status: Offline
I love the fact that you're doing doppler effect redshift for the hyperdrive, but allow me to rehash my earlier statement about bright colors.
I DO know that severe doppler effects would cause ultraviolet or infrared light to become visible IRL and that Space Engine does not and is not planned to incorporate these. As a solution, though, rather than simply truncate the red/blue shift and get left with highly saturated blue or red backgrounds, I suggest scaling down or muting the colors so that even if not as severe as would be realistic, it also doesn't look unrealistically oversaturated.
Pursuant to this, I'd like to point out that blueshifting would also drive things toward white and redshifting toward black. I already observed that this is implemented somewhat in the above screenshots, so my suggestion is perhaps trying a little more brightness shift and less color shift.
I could maybe post a picture of what I'm imagining if it helps xP (See below)

Oh, and while I understand an exact warp effect isn't feasible, are you able to implement any degree of reduced field of view a la the black hole upgrade (e.g. the shot looking out from near the event horizon)? From what I've read about relativistic effects, at ludicrous speed the sky in front is reduced to a blinding blue-white point while stars in back fade from red straight to black.

EDIT: Here, this is sort of what I meant:

Rather than shift the colors uniformly toward blue, do some sort of additive color temperature shift so that things get super bright while the bloom and shading shift toward bluish and purplish. If you can, of course. Don't do it if you don't want to; I'm just throwing in my aesthetic opinions after all ;)
Oh, and ignore the fact that the ship(s) are getting shifted too. That's just a random screenshot I had. It more or less represents the camera whooshing toward them rather than the ships themselves moving. And of course it leaves out the part I mentioned about the field of view changing because I couldn't be arsed to write my own shader just for an illustration xD





Intel HD Graphics 4000 ;P

Edited by parameciumkid - Monday, 23.11.2015, 05:17
 
SpaceEngineerDate: Monday, 23.11.2015, 06:31 | Message # 830
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
Quote parameciumkid ()
As a solution, though, rather than simply truncate the red/blue shift and get left with highly saturated blue or red backgrounds, I suggest scaling down or muting the colors so that even if not as severe as would be realistic

There is no "simple truncate", but simulation of hr actual red/blue shift based on some assumption about spectrum. On these screenshots, each pixel is assumed to have sun-like spectrum, so missing data from IR/UV is simple constants in the code. In future updates I plan to make better approach by analyzing of the pixel color to get idea about it's full spectrum (something like conversion RGB -> Planck temperature -> shift -> RGB).

Quote parameciumkid ()
it also doesn't look unrealistically oversaturated.

Lol did you saw doppler/gravity shift with your eyes, to know which is "realistic"?

Quote parameciumkid ()
so my suggestion is perhaps trying a little more brightness shift and less color shift.

Brightness change is already based on right formulas, so don't need any tweaks. Aesthetics have nothing to do with real nature.

Quote parameciumkid ()
Oh, and while I understand an exact warp effect isn't feasible, are you able to implement any degree of reduced field of view a la the black hole upgrade (e.g. the shot looking out from near the event horizon)? From what I've read about relativistic effects, at ludicrous speed the sky in front is reduced to a blinding blue-white point while stars in back fade from red straight to black.

Ships in SE are moving using warp drive, not subluminal flight. Wiew from inside the warp bubble also shows effect of field of view shrinking, but without raytracing in real Alcubierre metrics it's hard to implement.

Also, ship movement required updating of the skybox texture each frame. This is heavy operation currently, on my system FPS drops from 300 to 3 in interstellar flight. So I probably will be forced to change back to old method of recontruction of the missing frame data (mirror repeat of the frame texture instead of using skybox in the offscreen areas of the warped image).





 
n0b0dyDate: Monday, 23.11.2015, 19:23 | Message # 831
Explorer
Group: Users
Pirate
Messages: 297
Status: Offline
SpaceEngineer, can you have a look at this please. Thank You smile .
 
SalvoDate: Monday, 23.11.2015, 20:10 | Message # 832
Star Engineer
Group: Local Moderators
Italy
Messages: 1400
Status: Offline
Quote SpaceEngineer ()
on my system FPS drops from 300 to 3 in interstellar flight.


I'm afraid of what could happen on mine...





The universe is not required to be in perfect harmony with human ambition.

CPU: Intel Core i7 4770 GPU: ASUS Radeon R9 270 RAM: 8 GBs

(still don't know why everyone is doing this...)
 
EliamDate: Monday, 23.11.2015, 20:13 | Message # 833
Observer
Group: Users
France
Messages: 11
Status: Offline
Amazing.

With the blueshift/redshift it would be fun to create a mod with relativistic effects on time. Speeds would be limited under c but due to time dilation effects the travel between two different points would only take few seconds in camera proper time with an enormous acceleration.

For example starting at Earth in 2014 and choosing to go to Adromeda would take a little more than 10.35 seconds in camera proper time with a 200 million g acceleration + deceleration phase. The arrival year would then be around 2,502,014.
 
Tac1017Date: Monday, 23.11.2015, 20:23 | Message # 834
Explorer
Group: Users
United States
Messages: 167
Status: Offline
Would there be a way to keep a procedural set universal for future releases so people can be looking from the same set of planets regardless of version, and so "Terraphase", or where people need to find another planet that satisfies them? I would HATE to do that again, being I found a Pinkworld Terra that is just off the small lobe of Barnard's loop.

(Just for those who are wondering, I put scans in ''My latest finding that's earth-like"





The Terra Hunter of the Milky Way!

(By the way, I was born in 2001, NOT 1972 XD)


Edited by Tac1017 - Monday, 23.11.2015, 20:26
 
OMGspace_engineDate: Monday, 23.11.2015, 23:17 | Message # 835
Observer
Group: Users
Brazil
Messages: 19
Status: Offline
About an older post "reversed floating-point depth buffer", it will be available in 9.7.4 ???
 
anonymousgamerDate: Monday, 23.11.2015, 23:26 | Message # 836
World Builder
Group: Global Moderators
United States
Messages: 1011
Status: Offline
Quote OMGspace_engine ()
About an older post "reversed floating-point depth buffer", it will be available in 9.7.4 ???


It was already implemented in patch 6.





Desktop: FX-8350 4.0 GHz, 8 GB DDR3 RAM, EVGA GeForce GTX 1080 FTW 8 GB, 2 TB HDD, 24 inch 1920x1080 screen
Laptop: Core i5 480M 2.66 GHz (turbo 2.93), 8 GB DDR3 RAM, AMD Radeon HD 6550m 1 GB, 640 GB HDD, 17.3 inch 1600x900 screen
 
Destructor1701Date: Tuesday, 24.11.2015, 04:22 | Message # 837
Pioneer
Group: Users
Ireland
Messages: 533
Status: Offline
As usual, fantastic work, Space Engineer. I had a thought...

On the topic of red/blue shift:

Quote
On these screenshots, each pixel is assumed to have sun-like spectrum, so missing data from IR/UV is simple constants in the code.


So, as I understand it, blue-shift is so-called not because everything turns blue, but because everything shifts frequency towards the blue end of the spectrum - to higher frequencies.

At low relativistic velocities, visible colours will subtly start to re-assign to higher-energy visible colours. This ought to mean that red tends to yellow and green, green tends to blue, and so on. As the field-strength increases, the universe seen through the forward lobe would shift kaleidoscopically through the spectrum, and lower, sub-infrared radiations would start to slip into the frequency bands of visible colours and follow them in the kaleidoscope.

So, I'm not sure I understood your post fully - is that what you're saying you're planning?

The visible light portion of that should be easy enough - just shift the colour values of the pixel by some quantity commensurate with the field strength through which that pixel is seen. After that - outside the visible spectrum - your Sun Spectrum model would work... or you could cheat and wrap back around the colour values - it's not like we'd know the difference :p

The "Hue/Saturation" tool in Photoshop illustrates the effect (though not along the scientific spectrum, it would seem):





Combined with the g-lensing and the varying warp field strength gradient, this could end up looking '60s tie-die-T-shirt trippy

A bit like the MIT indie game "A Slower Speed Of Light":








Edited by Destructor1701 - Tuesday, 24.11.2015, 04:22
 
OMGspace_engineDate: Tuesday, 24.11.2015, 11:41 | Message # 838
Observer
Group: Users
Brazil
Messages: 19
Status: Offline
Quote anonymousgamer ()
It was already implemented in patch 6.

Where I can download it?
 
CanleskisDate: Tuesday, 24.11.2015, 14:07 | Message # 839
Space Pilot
Group: Users
France
Messages: 117
Status: Offline
Quote OMGspace_engine ()
Where I can download it?


Here http://en.spaceengine.org/forum/21-3022-1
 
OMGspace_engineDate: Wednesday, 25.11.2015, 14:37 | Message # 840
Observer
Group: Users
Brazil
Messages: 19
Status: Offline
I'm having an issue with the HUD, it looks like all images was changed. The mouse icon was changed to an random icon too:
http://i.imgur.com/rGvLnH2.jpg
 
Forum » SpaceEngine » Archive » Work progress and public beta test - 0.9.7.4
Search: