@Flame
Can you shed any light on this thing about different comps giving different results for cos and sin.
Because if that does happen then the reason for lookup tables increases.
Also I dont think you worded "uses the same index all the time" correctly, It uses the same index for all calculations each cycle. (A small point, but when I first read what you had written I couldnt see what you ment)
I think that 360 floats for each of the trig functions, is an arbitary amount of memory, and Im very supprised you used it. (Its less that a 512 By 1 Texture)
I dont doubt that in a real situation the 320%, (Which I quoted as the speed increase from the data given), would be much less. But it would be an increase.
However Jessie, wanted to know which was faster. And the tossup is, an arbitary increase in speed in exchange for an arbitary increase in size/memory. Yet even after jessie showed this increase in speed, Decided not to use it.
My point still remains. If its speed rather than memory that you are worried about, and you have already written the speedier version. Then use it.
Unless, you are saying that it isnt quicker? (As apposed to, it isnt as much quicker as we think)
@Fred
The lookup doesnt provide a result with any lack of precision. (But I know what you mean, which is why I did say for whole angles in my post)
Can you shed any light on this thing about different comps giving different results for cos and sin.
Because if that does happen then the reason for lookup tables increases.
Also I dont think you worded "uses the same index all the time" correctly, It uses the same index for all calculations each cycle. (A small point, but when I first read what you had written I couldnt see what you ment)
I think that 360 floats for each of the trig functions, is an arbitary amount of memory, and Im very supprised you used it. (Its less that a 512 By 1 Texture)
I dont doubt that in a real situation the 320%, (Which I quoted as the speed increase from the data given), would be much less. But it would be an increase.
However Jessie, wanted to know which was faster. And the tossup is, an arbitary increase in speed in exchange for an arbitary increase in size/memory. Yet even after jessie showed this increase in speed, Decided not to use it.
My point still remains. If its speed rather than memory that you are worried about, and you have already written the speedier version. Then use it.
Unless, you are saying that it isnt quicker? (As apposed to, it isnt as much quicker as we think)
@Fred
The lookup doesnt provide a result with any lack of precision. (But I know what you mean, which is why I did say for whole angles in my post)