Texture quality and distance to object

#1 If you're using MilkShape, you can export with no mip-mapping.

But it would be nice to know if / where the global setting is in the code.

#2 There's a small textures check. Try changing that? :)

#3 Thank you!

Ben, can you give me some coordinates of that textures check point, please? script? code? class?

#4 Try grepping for smallTextures?

#5 Great!

Thanks, Ben, I've got the idea. The code is quite simple.

Sam, look for "smallTextures" in the texture manager class (TextureManager).

I think I'm going to research it more and, then, publish my code snipit.

#6 Would be interested in seeing that. Glad you got it dealt with!

#7 I've changed it a few months ago, it's really just a hack and is very easy to apply.

I'll see if I can dig it out tonight and post it here if you havent already Alex, if I get some spare time.
No promises though.

#8 Stefan,

any help will be appreciated!

#9 Requested by Nathan Tran, this is the hack for more or less removing the check for smalltextures, which is what is making the texture look so flushed out.

In ShapeBase.cc # 2258 - 2262

F32 projected = dglProjectRadius(dist, axis) / 350;
      if (projected < (1.0 / 16.0))
      {
         TextureManager::setSmallTexturesActive(true);
      }

You can remove this part, but if you have a very far view range, and if you keep alot of items visible at once which are far away, it might kill of some fps for you.

I changed (dist, axis) / 350; to (dist, axis) / 25; and the "issue" went away. The value I changed sets the radius at which SmallTextures are used.

Have fun.

#10 Stefan,

Thanks a lot. It is so nice of you to look into this even though you have migrated your code base to TSE.

Nathan.

#11 Stefan, you are correct, it's the right place. I've prepared 2 code resources to be published soon, they will allow to control the distance from the script. Stay Tuned ;)

#12 @Alexander: Ah, those sound like good candidates for HEAD inclusion.

#13 @Ben: good idea, just need time to make it in proper format, but soon it will be done.

#14 Well, I've included the hacky fix. This is an issue we'll have to revisit at some point. Thank you everyone!