ENG New site

Advanced search
[ New messages · Forum rules · Members ]
Work progress - 0.9.7.3
second-ichDate: Sunday, 29.03.2015, 13:07 | Message # 151
Space Pilot
Group: Users
Germany
Messages: 87
Status: Offline
Quote SpaceEngineer ()
main of them is impossibility to have various fonts.

My interface skin contains only one font (Calibri). Or do you mean you can't use anything else than the current one?


Edited by second-ich - Sunday, 29.03.2015, 13:38
 
SpaceEngineerDate: Sunday, 29.03.2015, 14:50 | Message # 152
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
Quote second-ich ()
My interface skin contains only one font (Calibri). Or do you mean you can't use anything else than the current one?

it contains text of various size and italic/non-italic.





 
FastFourierTransformDate: Tuesday, 31.03.2015, 16:07 | Message # 153
Pioneer
Group: Local Moderators
Spain
Messages: 542
Status: Offline
Quote SpaceEngineer
You are asking to model rivers. You must know this is impossible task for pure procedural fractal generation. Only erosion post-process may model something river-looking (for steep slopes like volcano it actually must give nice results). Wait for SE 1.0 for this feature.


If it helps:
I have found a paper that explains how to perform proceudural generation of terrains accounting for hydrological processes

By the way, I've encountered a person that is trying to simulate the erosion caused by hydrological processes post-process and is quite awesome how well the terrain gets after the erosion. It could be added for polar ice caps on planets or glaciers (that are easily eroded) or for terrains on planets with corrosive atmospheres and rain.

 
jchomaarDate: Wednesday, 01.04.2015, 22:54 | Message # 154
Observer
Group: Newbies
Pirate
Messages: 1
Status: Offline
Quote FastFourierTransform ()
Quote SpaceEngineer
You are asking to model rivers. You must know this is impossible task for pure procedural fractal generation. Only erosion post-process may model something river-looking (for steep slopes like volcano it actually must give nice results). Wait for SE 1.0 for this feature.

If it helps:
I have found a paper that explains how to perform proceudural generation of terrains accounting for hydrological processes

By the way, I've encountered a person that is trying to simulate the erosion caused by hydrological processes post-process and is quite awesome how well the terrain gets after the erosion. It could be added for polar ice caps on planets or glaciers (that are easily eroded) or for terrains on planets with corrosive atmospheres and rain.



That video reminds me of this Chrome Experiment, called Craftscape.
 
fbDate: Friday, 10.04.2015, 12:48 | Message # 155
Observer
Group: Newbies
Pirate
Messages: 4
Status: Offline
Hi all,

I'm dedicated to SE 3D interaction, using twin projector setup and a big silver screen (polarized set, simultaneous L+R projection).

3D setup parameter "stereobase" is static, so no way to find a value good for both planet and galaxy type objects.

Any chance to make it automatically variable when interacting with objects of various size ?
 
SpaceEngineerDate: Friday, 10.04.2015, 20:04 | Message # 156
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
Quote fb ()
Any chance to make it automatically variable when interacting with objects of various size ?

No. This has been discussed earlier. Automatic change of the stereobase will look strange when you trying to zoom between planets or galaxies. And this is not easy to decide the needed value of the stereobase in the code.





 
SpaceEngineerDate: Friday, 10.04.2015, 20:13 | Message # 157
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
A new method of fixing texture tiling on terrain. A set of screenshots for comparison:

1) No fix - tiling is clearly visible.



2) The method used in 0.972 and in older versions. The detail textures are blended with themselves at a larger scale. This hides the repeating pattern, but blurs the surface and look strange sometimes (you have a large rocks blended with a small ones).



3) The new method: detail texture is split into random cells and a random offset to texture coordinates is added to each cell. Transitions between cells are smooth (blended). Contrast is returned back (better than in the method 2), but repeating pattern is still visible in some places.



4) Added random rotation of texture coordinates in each cell. Repeating pattern is completely gone!



The tiling fix method, as long as tiling blend method, can be chosen in the code in the tg_common.glsl shader. Tests on different hardware does not show any impact on the planet generation performance.

Attachments: 8934019.jpg (421.8 Kb) · 7679959.jpg (311.7 Kb) · 2429406.jpg (367.8 Kb) · 7121454.jpg (368.7 Kb)





 
SalvoDate: Friday, 10.04.2015, 20:39 | Message # 158
Star Engineer
Group: Local Moderators
Italy
Messages: 1400
Status: Offline
The "new method" was the one that you implemented in 0.9.7.1 or... a new one again? smile

Also, I like so much the 4th one! It would be nice to see its job on some other kind of surfaces or distances.





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...)
 
PsauceDate: Friday, 10.04.2015, 23:57 | Message # 159
Space Tourist
Group: Users
United States
Messages: 25
Status: Offline
Quote SpaceEngineer ()
Repeating pattern is completely gone!


Clever. Looks great!
 
FireintheholeDate: Saturday, 11.04.2015, 08:59 | Message # 160
Pioneer
Group: Translators
Sweden
Messages: 356
Status: Offline
Looks outstanding! smile




Love SpaceEngine!
 
the_nerervarineDate: Saturday, 11.04.2015, 13:30 | Message # 161
Space Pilot
Group: Users
United States
Messages: 106
Status: Offline
Looks like you're out doing yourself again SpaceEngineer. Fantastic work and I can't wait for the next version.




AMD FX-8320 8 core @4.0ghz , AMD R9 290 4GB, 8GB DDR3 @1866mhz, 1TB Hard Drive
 
SpaceEngineerDate: Sunday, 12.04.2015, 15:00 | Message # 162
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
Implemented support of the Oculus Rift DK2 positional tracknig and the "Direct mode" (SDK-driven rendering). All became very smooth! Direct mode is highly recommended for now!




 
fbDate: Sunday, 12.04.2015, 15:28 | Message # 163
Observer
Group: Newbies
Pirate
Messages: 4
Status: Offline
Quote SpaceEngineer ()
Automatic change of the stereobase will look strange when you trying to zoom between planets or galaxies. And this is not easy to decide the needed value of the stereobase in the code.


Zoom could (and should) be ignored, as it's known that telescope does not provide 3D view of distant objects.
The active stereobase could be a selectable 'percentage' of distance between viewer and currently selected object.

I bet also O'Rifters would benefit from this !


Edited by fb - Sunday, 12.04.2015, 15:31
 
Destructor1701Date: Wednesday, 15.04.2015, 11:38 | Message # 164
Pioneer
Group: Users
Ireland
Messages: 533
Status: Offline
Quote fb ()
The active stereobase could be a selectable 'percentage' of distance between viewer and currently selected object.


That sounds like a reasonable approach, though it could get very weird when you approach a planet and it seems to grow before you, or when you change selection and the universe crunks to a different scale suddenly.

Still, with some finessing - perhaps a transition of a few seconds between scales, and a cut-off at human scale - I think that would be neat.

I don't think the static human option should be removed, though - sometimes you want to just feel the distance.

Quote fb ()
Zoom could (and should) be ignored


I don't think Space Engineer meant "zoom" in the optical sense - just as a synonym for "travel".







Edited by Destructor1701 - Wednesday, 15.04.2015, 11:40
 
Search: