Jump to content
  • Advertisement
Sign in to follow this  
greeniekin

Generating non linked neighbor uv's

This topic is 2300 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I was fiddling with making an editor. I know it's not smart to reinvent the wheel, but it's a hobby.
I was thinking about uv's. A lot of editors you mark seams and it generates those uv's most of us have seen. With a fair amount of wasted space and some editing required to choose the level of detail you want.

What i was wondering is it possible to generate uv's of rectangles and have it use 100% of the texture.
Generating this would be rather simple as-well, even when giving more detail to places like a human face. The uv's may not align on the texture with there neighbor face though. Also trying to edit in Photoshop would be impossible. Though with 3d editor paint tools it could compensate, and removes knowing about the uv's from the user.

The thing I wonder about is what problems could arise from uv not having the same neighbor face?
Would there be problems with getting pixels from outside the allocated area? Would mipmapping cause big problems? How could these be avoided?

I know I'll get a lot of information on why the other way is better. I know these already, so just humor me and tell me of potential pitfalls and ideas.

I appreciate any help you can give me.

Share this post


Link to post
Share on other sites
Advertisement

he thing I wonder about is what problems could arise from uv not having the same neighbor face?
Would there be problems with getting pixels from outside the allocated area? Would mipmapping cause big problems? How could these be avoided?

Mipmaps could be a real problem. Many pre-calculated lightmap techniques use this kind of unwrapping.

Share this post


Link to post
Share on other sites
Ignoring the problems with mipmaps bleeding as you get in lightmaps.

Consider that the mip chosen when it samples the texture is based on a calculation of the pixel density on the screen. So if you try to have higher density in one area it can actually choose a lower mip than you would want.

Also if you split a model into rectangles you have to store more data and potentially you have a negative effect on your vertex cache

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!