This depends on the texture format. Yes, tga can be compressed (I use GIMP as well). And you could also use DDS. DDS DXT1 as an example can even compress TGA files to about 25 percent of original size. And a game can transfer it in this compressed small size to the GPU. The GPU will decompress it with the S3T algorithms.

But as you can see: All this could be tested when you try to work with real textures instead of untextured place holders. And it can be a waste of time when you have to rework each texture and model just to reduce number of draw calls or to reduce data transfer to the GPU, or memory usage.

Apart from DDS format, all formats like PCX and TGA will stay uncompressed in video memory.
So what will differ probably is the loading time. But hey - you're in development. You will have no idea what the total loading time will be in the end as you still have to add many things.

Important is the performance during play - and for this dummy textures are fine.

As far as I know it not only influences the loading time. It also influences the transfer to the graphics card and thus can have an impact at rendering speed when there are many switches between different textures.

With that in mind it could be in some situations (especially with bigger textures) that DDS also improves the performance. But to be honest, I did no benchmark so far. This is pure theory.

As I develop my levelgeometry in conjunction with the textures, this would be no option for my project.
However, as I´ve stated - each master to his own technique.

I'm not sure if we're still talking about the same thing.

If you're talking about multiple textures vs. single texture you're right.

I was talking about single "empty" texture. Apart from DDS the engine uncompresses any compressed format anyway internally which my result in different loading time for "empty" and "filled" textures. But rendering time should be the same.

If that is true then it is not a very smart design of 3DGS. DDS should not be internally decompressed, it should be sent in compressed state to the graphics card.

What I am talking about is indeed a good testbed for checking:
- how big can textures be
- what format is best
- is it better to have little small vs. a few big textures (containing other textures via atlas)
- is it better to have bigger poly models vs. many with small poly budget
- how many texture memory can be used (with or without lightmaps)
- how much does it help to remove non-visible polygons
- how big is the impact of several shaders
- how big is the impact of animated models
- ...

All this can be seen best with a sample scene containing everything in the aimed quality. And if someone dont has the models and textures, it is easy and really cheap to get them from decent indie-friendly shops

"Apart from DDS"...

I do not know what the engine does with DDS. My guess is it decompresses only when hit information and the like is needed, but does not interrupt the pipeline to the graphics card.

sorry i took so long, i am very busy.
i say HE cant do it. I know the engine CAN but HE cannot...

Soooo, how far are you?

Ah, someone searching for podcast topics.