Re: [Freetype] LoadGlyph Time

[James H. Cloos Jr.]

>> Would cpu-specific optimizations using [SIMD] instructions help
>> speed things up?

David> This wouldn't add much I believe. Since it's not possible to
David> automatically parallelize TrueType bytecode.
 
>> How about a jit for the bytecodes?

David> You won't win much in terms of performance, and you're going to
David> explode the memory requirements of the engine (and the code's
David> complexity).
 
I didn't think there would be much of a gain, but dialog is always
better than thinking to oneself....

>> Or do patent issues start popping up with such optimiaztions?
>> 
David> Nope, the patents protect the _result_ of computations, not the
David> way they're done. Otherwise, we would have side-stepped them
David> for long by now :-)

I should have said *more* patent issues; but it looks like they didn't
need to bother with anything beyond what they did go for.  

-JimC