I had two things I wanted todo today;
- Complete a request from Coder-R to allow the origin to be changed on images with regards to the flipping
- Add the ability to decompress DDS images
These are infact practically done and I even caught a silly bug in the process [smile]
Firstly, the origin business. You can now request that GTL flips/mirrors the texture so that the origin is in ANY of the four corners of the image, which means things might well make more sense of the OGL coders out there [grin]
- This only applies to none DDS images atm. I need to investigate how tricky the flipping and mirroring would be to allow for the same trick to be pulled on those images
- Its a state thing, so when you set it it applies from that moment on.
- Once loaded images can't be flipped. This functionality could be added if someone wants it [smile]
GTL::SetOrigin() is your new function of choice for this operation [grin].
After Deranged's comment about Ogre looking for an image loader I shuffled over to the Ogre site and had a word with one of the high up guys. One of the issues he raised was the lack of ability to decompress a DDS image, as such the 'Decompress' function has been born.
It's not 100% completed as yet, none mipmap'd DXT1 images will decode fine, the mipmaped one has come out a bit... odd... I suspect some maths is out somewhere onmy part [grin]
But you can now decompress a texture. However, with the lack of flip ability it'll stay with its origin at the top right... this might be a good reason to add a function to change things to the new origin, anyone got any thoughts?
Tomorrow... well, today... should hopefully see the decompress function fixed up. I also need to expand the image formats a bit; 16bbp support is lacking in a few image formats, plus single and dual channel formats might well be handy to add to support floating point textures and decompressed 3Dc images. Infact, 3Dc needs to be split to allow for ATI1N and ATI2N modes.
So, a bit of work ahead, but things are going well and as monday isn't a real day I'll be spending a few hours on it. After tomorrow/today I need to focus on some assignment work a bit... which means I'll probably end up doing some more tuesday evening and wednesday... its Java programming week, no bother [grin]
Hopefully, a new release before the weekend!