Page 2 of 6

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 2:07 pm
by cim
Kaks wrote:
or you could put a low density, physically weak real oolite entity in exactly the same place, with a transparent shader as texture. All you'd need then is a script that recreates it every time it's destroyed... ;)
For a typical planet the ring is several kilometres wide and a couple of hundred in circumference, so keeping collidable entities in place just in case is going to get expensive on the entity count. Dropping some invisible flat sheets into the parts of the ring near real ships with the frame callback would deal with the NPC AI problem, though.

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 2:45 pm
by JazHaz
cim wrote:
For a typical planet the ring is several kilometres wide and a couple of hundred in circumference, so keeping collidable entities in place just in case is going to get expensive on the entity count.
I think just leave it as is, after all I believe the Cassini probe to Saturn flew through the rings without damage.

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 3:03 pm
by Selezen
Hey, if the USS Voyager can do it in the title sequence EVERY WEEK without damage, then why not Oolite ships?? They're generally a lot smaller... ;-)

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 6:18 pm
by Agis Silverfish
That's fantastic! Can you mine the rings? I am always looking for that extra credit. Sorry, that was silly. :mrgreen:
But it's going to be great to ambush pirates from above or below the rings. :D

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 6:35 pm
by Wildeblood
cim wrote:
Image
What I'd like to know is how you have a textured planet. Over here, planet textures are still all scrambled in trunk.

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 6:37 pm
by JazHaz
Wildeblood wrote:
What I'd like to know is how you have a textured planet. Over here, planet textures are still all scrambled in trunk.
In my screenshots, I used Povray Planets OXP. Dunno what cim used though.

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 7:17 pm
by Cody
Wildeblood wrote:
What I'd like to know is how you have a textured planet.
I presume that's using submersible's planet branch of trunk (or whatever it's called)... but I may be wrong, of course.

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 7:24 pm
by Tricky
El Viejo wrote:
Wildeblood wrote:
What I'd like to know is how you have a textured planet.
I presume that's using submersible's planet branch of trunk (or whatever it's called)... but I may be wrong, of course.
Shady Planets branch. That gives proper shading to the planets in the dev program. That hasn't been updated for a while (unless someone has done it lately). cim may have back-ported the new visual effects code into it.

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 7:51 pm
by Thargoid
I guess maybe my little netbook isn't up to the job (even though it happily runs full shaders):

Image

That or the definition of a ring has changed somewhat ;)

Log reports an error:

Code: Select all

  20:45:21.916 [shader.compile.failure]: ***** ERROR: GLSL fragment shader compilation failed for vedemo-rings.fragment:
>>>>> GLSL log:
Fragment shader failed to compile with the following errors:
ERROR: 2:74: error(#202) No matching overloaded function found dot
ERROR: error(#273) 1 compilation errors.  No code generated


20:45:21.916 [shader.load.fullModeFailed]: ----- WARNING: Could not build shader vedemo-rings.vertex/vedemo-rings.fragment in full complexity mode, trying simple mode.
  20:45:21.939 [shader.compile.failure]: ***** ERROR: GLSL fragment shader compilation failed for vedemo-rings.fragment:
>>>>> GLSL log:
Fragment shader failed to compile with the following errors:
ERROR: 2:74: error(#202) No matching overloaded function found dot
ERROR: error(#273) 1 compilation errors.  No code generated


20:45:21.939 [shader.load.failed]: ***** ERROR: Could not build shader vedemo-rings.vertex/vedemo-rings.fragment.
20:46:53.036 [gameController.exitApp]: .GNUstepDefaults synchronized.
Looks like the same as A_C reported, will try his fix.

Editted to add - OK, that fixed it :) This is under AMD Radeon HD6310 Graphics by the way, so A_C isn't the only one whose drivers need babysitting.

Image[/color]

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 9:19 pm
by Eric Walch
Thargoid wrote:
This is under AMD Radeon HD6310 Graphics by the way, so A_C isn't the only one whose drivers need babysitting.
On my Renderer: "NVIDIA GeForce 9400 OpenGL Engine" I got the same graphical bug and with the same solution. So, more cards are a bit more precise.

I do see a hairline of wrong bits running through the rings. I do like that the planet is able to shade the rings. :wink: Maybe add a few small asteroids as solid entity in the rings so the player does hit something sometime. :D

Re: [Proof of concept] Planetary Rings OXP

Posted: Thu Sep 20, 2012 9:50 pm
by Commander McLane
JazHaz wrote:
In my screenshots, I used Povray Planets OXP.
Then Povray Planets has a problem. There is a very noticeable seam in the texture in your first screenshot.

Re: [Proof of concept] Planetary Rings OXP

Posted: Fri Sep 21, 2012 1:23 am
by JazHaz
Commander McLane wrote:
JazHaz wrote:
In my screenshots, I used Povray Planets OXP.
Then Povray Planets has a problem. There is a very noticeable seam in the texture in your first screenshot.
No. Wildeblood asked about textured planets not working in trunk, and my screenshots (and Cim's) prove that is not the case. Yes there is a seam in the texture, but that was not being discussed, and in fact that whole question of trunk's handling of textured planets is off topic. That should be discussed elsewhere.

Re: [Proof of concept] Planetary Rings OXP

Posted: Fri Sep 21, 2012 3:49 am
by JazHaz
JazHaz wrote:
cim wrote:
For a typical planet the ring is several kilometres wide and a couple of hundred in circumference, so keeping collidable entities in place just in case is going to get expensive on the entity count.
I think just leave it as is, after all I believe the Cassini probe to Saturn flew through the rings without damage.
Plus there might be a problem if the planet's station is spawned in the ring, and gets damaged.

Re: [Proof of concept] Planetary Rings OXP

Posted: Fri Sep 21, 2012 5:39 am
by Wildeblood
JazHaz wrote:
Commander McLane wrote:
JazHaz wrote:
In my screenshots, I used Povray Planets OXP.
Then Povray Planets has a problem. There is a very noticeable seam in the texture in your first screenshot.
No. Wildeblood asked about textured planets not working in trunk, and my screenshots (and Cim's) prove that is not the case. Yes there is a seam in the texture...
Actually, your screen shots prove that it is so. That texture is completely scrambled, and is exactly as it would appear to me. It looks nothing like it should.

Woe is me! I get the square ring, like Thargoid, and the scrambled planet texture, like JazHaz. This is Lave:
Image
JazHaz wrote:
...but that was not being discussed, and in fact that whole question of trunk's handling of textured planets is off topic. That should be discussed elsewhere.
Pfft!

Image

Re: [Proof of concept] Planetary Rings OXP

Posted: Fri Sep 21, 2012 6:51 am
by cim
Okay - shader-fixed OXP uploaded and first post updated.

As regards textures in trunk, the answer is simple: I compiled my personal copy with -DNEW_PLANETS=0, which returns planet rendering to 1.76 mode. Since the current trunk new planet code will eventually be replaced by submersible's shader branch, I figure I might as well enjoy the existing textures while I fly about and test other things.

Jazhaz and Wildeblood's screenshots show the interesting bug in the current trunk planets code where a cube map is treated as if it were a latlong map, rescaled, and applied. So every planet looks like a ball of stripes with a seam. Admittedly, this makes it a lot easier to find a planet which looks like a gas giant, so maybe it's not all bad.