Posted Friday, 23 November, 2007 - 22:39 by nythrix
(I've always wondered how it is possible to use quaternions for translations). I'll be doing some reading.
That's what drove me to them in the first place.
Check wikipedia for dual numbers. Especially the linear approach to "e" is brilliant. I can't visualize the rest without painful headaches or additional dimensions to expand myself to :)
Posted Saturday, 24 November, 2007 - 13:12 by Inertia
This brings up an interesting question: What's the real scope of the mathlib?
If it's goal is to provide a base layer to build your own Animation, simple Frustum-culling or Particle Systems ontop, i'd say it's almost there. Should it become a full replacement for Math.Net or Collision/Physics SDKs? Or some shade of grey inbetween?
(If this isn't 100% clear yet it might be wise to start a new topic to discuss this)
[double precision/dual quaternion]
OpenGL 3 will not support double precision floating point, so that would be largely pointless. Although a future OpenAL version could put them to use, I seriously doubt anyone can notice the precision difference (which is probably why there are no states using double precision right now).
Regarding the dual Quaternions I don't really see a reason to add them, hardware understands matrices natively which can also contain scaling and perspective projection.
Regarding double precision floats, Ati 3xxx cards already support these. While OpenGL 3 won't require support, Mt. Evans probably will (and that's said to come only a few months after GL3) - adding support won't be bad, but it's not high priority either.
Comments
(I've always wondered how it
(I've always wondered how it is possible to use quaternions for translations). I'll be doing some reading.
That's what drove me to them in the first place.
Check wikipedia for dual numbers. Especially the linear approach to "e" is brilliant. I can't visualize the rest without painful headaches or additional dimensions to expand myself to :)
http://en.wikipedia.org/wiki/Dual_numbers
This brings up an
This brings up an interesting question: What's the real scope of the mathlib?
If it's goal is to provide a base layer to build your own Animation, simple Frustum-culling or Particle Systems ontop, i'd say it's almost there. Should it become a full replacement for Math.Net or Collision/Physics SDKs? Or some shade of grey inbetween?
(If this isn't 100% clear yet it might be wise to start a new topic to discuss this)
[double precision/dual quaternion]
OpenGL 3 will not support double precision floating point, so that would be largely pointless. Although a future OpenAL version could put them to use, I seriously doubt anyone can notice the precision difference (which is probably why there are no states using double precision right now).
Regarding the dual Quaternions I don't really see a reason to add them, hardware understands matrices natively which can also contain scaling and perspective projection.
Let's continue the scope
Let's continue the scope discussion at http://opentk.sourceforge.net/home/node/98, this topic is getting too long.
Regarding double precision floats, Ati 3xxx cards already support these. While OpenGL 3 won't require support, Mt. Evans probably will (and that's said to come only a few months after GL3) - adding support won't be bad, but it's not high priority either.