Celestia Forums

Hey all, I have another (possibly stupid) question for you...

Is there any way to disable specular reflections within Celestia for
ALL objects? I have tried searching for this answer with no luck, and
I'm afraid that it's simply my PC and/or graphics card which are giving
me problems here, but I would simply like to GET RID of specular
reflections altogether. How can I do this?

Any help appreciated. Even when I remove specular textures, and/or
increase the specular reflection value to 100, I still have unwanted
bright circles of light distorting the images of such places like Io, or
the Earth.

Thanks, Brain-Dead

One other note here...

Must just be my HP implementation of OpenGL 2.0. If I use any
OTHER render path, I have no problems.

Just FYI.

Thanks, Brain-Dead

Bob,


Specularity is not visible in Basic or Multitexture.
It is supposed to work in OpenGL Vertex Program and in OpenGL 2.0.

Are you saying it's working in the OpenGL Vertex Program path?

Bob,

how about commenting out

SpecularPower xx.y

in solarsys.ssc? I really don't understand your problem, but that doesn't matter, really

F.

selden wrote:Bob,


Specularity is not visible in Basic or Multitexture.
It is supposed to work in OpenGL Vertex Program and in OpenGL 2.0.

Are you saying it's working in the OpenGL Vertex Program path?

Selden,

I'm having what I'm calling "Specularity Problems" with objects ONLY
in the OpenGL 2.0 render path. Since this is the path for which Celestia
is designed, then I am in trouble when trying to use it. All of my planets
reflect a VERY bright glow, which I simply cannot get rid of no matter
what I do in render path OpenGL 2.0.

If I switch to any of the other render paths, then this "brightness"
goes away, and I can view the planets/moons in a satisfactory manner.

Probably just my machine again. Sigh...

t00fri wrote:Bob,

how about commenting out

SpecularPower xx.y

in solarsys.ssc? I really don't understand your problem, but that doesn't matter, really

F.

Yes, I already did that Fridger. Thanks though.
That's why I asked this question. I wondered if Celestia was somehow
causing specularity via some other means that I was unaware of.
Apparently, NOT...

Thanks anyway, Bob

you should show us a screenshot here so we know how your problem looks like

Ask, and ye shall receive...

First image with absolutely NO specularity images involved or defined
within Celestia, and under OpenGL 2.0 path:



Second image under same circumstances but rendered via OpenGL with vertex path:



Comments?

Bob,

are you sure that you don't have some Io add-on installed in the extras folder where some specularity commands may be located, overiding the default settings?

F.

t00fri wrote:Bob,

are you sure that you don't have some Io add-on installed in the extras folder where some specularity commands may be located, overiding the default settings?

F.

You know, I just KNEW that you were going to ask that after I selected
Io as my example, but this occurs also on Jupiter, Saturn, and any number
of OTHER planetary objects having absolutely NO specularity commands
enabled or textured. And to answer your question, there are NO
specularity-related objects, textures, or commands enabled in
these views.

Thanks, Brain-Dead

Just so you BELIEVE me here:

Venus via render path OpenGL 2.0:



Venus via OpenGL with Vetex render path:



Thanks, Brain-Dead

Hey Bob,
You may have tried this already, but in case you didn't here it goes.
Make sure your ambient light is off. I see a slight increase in brightness when my ambient light is maxed and I switch into OGL2 . With ambient light off, I see no changes no matter what render path I use.

If that doesn't work, try installing Celestia 1.5.0 again under a new name, to a different place.
When the installer script asks for the directory to install to type -spectest. When the script askks for startmenu name, same thing. Rename it. So it will install in C:\Program Files\Celestia-spectest. When the script is almost finished and asks to associate cel,etc. , uncheck so it doesn't. This way you'll have an installation right out of the box with no changes or addons. And it should have no ties to any other install.
cartrite

cartrite wrote:Hey Bob,
You may have tried this already, but in case you didn't here it goes.
Make sure your ambient light is off. I see a slight increase in brightness when my ambient light is maxed and I switch into OGL2 . With ambient light off, I see no changes no matter what render path I use.

I have played around with the ambient light settings to see if it makes
any difference, and it seems to, but not in the way I expected. I'll keep
playing with it, but if this is the problem, then IT needs to be repaired
does it not?

As to your other suggestions concerning installation, I have already
re-installed Celestia in many different configurations, and under the
circumstances you have described also. This problem is still apparent
no matter HOW I install the thing.

At any rate, thanks very much for the suggestions. I'll play
around with it some more.

Take care, Bob

BobHegwood wrote:

cartrite wrote:Hey Bob,
You may have tried this already, but in case you didn't here it goes.
Make sure your ambient light is off. I see a slight increase in brightness when my ambient light is maxed and I switch into OGL2 . With ambient light off, I see no changes no matter what render path I use.

I have played around with the ambient light settings to see if it makes
any difference, and it seems to, but not in the way I expected. I'll keep
playing with it, but if this is the problem, then IT needs to be repaired
does it not?

Bob,

Cartrite is correct: what you're seeing is a difference in the way that ambient light is handled in the OpenGL 2.0 render path. It has nothing to do with specular reflections. If you turn ambient light down to zero (which is the only realistic mode), Venus will look nearly identical in the different modes.

As for whether this is a bug that needs to be fixed, I'd say no. The differences have to do with clamping of color values to the range [0, 1]. In the OpenGL 2.0 path, clamping occurs later in the graphics pipeline (when actually writing the pixel values) instead of earlier (the vertex colors.)

--Chris

chris wrote:Bob,

Cartrite is correct: what you're seeing is a difference in the way that ambient light is handled in the OpenGL 2.0 render path. It has nothing to do with specular reflections. If you turn ambient light down to zero (which is the only realistic mode), Venus will look nearly identical in the different modes.

As for whether this is a bug that needs to be fixed, I'd say no. The differences have to do with clamping of color values to the range [0, 1]. In the OpenGL 2.0 path, clamping occurs later in the graphics pipeline (when actually writing the pixel values) instead of earlier (the vertex colors.)

--Chris

Okay the, I really appreciate the responses here. It does make
me wonder what the ambient light values are used for though. Is
it fair to say that they are applicable only to spacecraft? If I have
to remove these for any planetary objects, then I don't understand
why they're there in the first place. That's okay though, I'll fix this
up quickly now. (I hope.)

Thanks again, Brain-Dead

Bob,

Ambient light lets you "see in the dark". When you're looking at an object which is on the night side of a planet, you need to adjust the ambient light in order to be able to see the object. The feature could be improved in various ways to better distinguish among similarly colored surfaces, but it's adequate. I've been using it quite a bit recently while working on the control desks for the Hale Telescope Addon. (It'd help me a lot if Celestia had controllable local light sources in addition to Stars; I've submitted a feature request.)

selden wrote:It'd help me a lot if Celestia had controllable local light sources in addition to Stars; I've submitted a feature request.

Good for you Selden. I was going to request the same or similar. Would you let me know by PM (or post here) the URL of your feature request.

Cheers
CC

Feature requests are at
https://sourceforge.net/tracker/?group_ ... tid=371302

It's # 1871939 "local illumination sources"
at
https://sourceforge.net/tracker/index.p ... tid=371302