@dougbinks Yeah, that doesn't help here. I'm talking about the case where two textures _abut_, so they need to be sampled _across_.
-
-
Replying to @cmuratori
@cmuratori Could semi-solve that if you generate mipmaps yourself and all textures tile with themselves I'd have thought.1 reply 0 retweets 0 likes -
Replying to @dougbinks
@dougbinks Did you read the pastebin I posted? It explains why that doesn't work.1 reply 0 retweets 0 likes -
Replying to @cmuratori
@cmuratori I read it. Doesn't appear to consider the case I mention.1 reply 0 retweets 0 likes -
Replying to @dougbinks
@dougbinks I guess I don't understand the case then? The goal is just "break up one image into smaller textures, then draw it."2 replies 0 retweets 0 likes -
Replying to @cmuratori
@cmuratori You have 2 cases mentioned. 1 is that, the other atlases. Atlas filtering usually solved by texture arrays.1 reply 0 retweets 0 likes -
Replying to @dougbinks
@dougbinks You're thinking of only one type of atlas, the kind where each is applied to a different piece of geometry.1 reply 0 retweets 0 likes -
Replying to @cmuratori
@dougbinks Texture atlasing in general does not have that limitation - you may have an atlas of maps for a single object (eg., a globe).1 reply 0 retweets 0 likes -
Replying to @cmuratori
@dougbinks (so that you can page in textures and use textures that are larger than the largest supported texture size on the hardware)1 reply 0 retweets 0 likes -
Replying to @cmuratori
@cmuratori Yes megatexutre/virtual texture style. Hence I mentioned 'if all textures tile with themselves'. Array only solves one situation.1 reply 0 retweets 0 likes
@dougbinks Ah, yes - no problems there. If your charts are all wrapping or clamping, arrays are perfect.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.