On 11 Aug 2006 at 16:18, Darcy James Argue wrote:

> On 10 Aug 2006, at 10:27 PM, David W. Fenton wrote:
> 
> > On 10 Aug 2006 at 12:45, Darcy James Argue wrote:
> >
> >> Because -- as I've said before a whole bunch of times -- 2D
> >> graphics performance is CPU-bound, not GPU-bound.
> >
> > Um, I think there's an interesting intersection here with Hiro's
> > posts, in that he says that on the Mac, 2D rendering is in the CPU,
> > not in software or the graphics subsystem, whereas 2D rendering on
> > Windows has to be elsewhere (if it's handled in software, it can be
> > handed off to a graphics subsystem).
> >
> > This would be a difference between Windows and Mac that could make
> > your position correct for the Mac and incorrect for Windows.

> You've actually got it backwards . . .

Eh? I don't understand.

> . . . -- OS X has progressively been 
> offloading more and more drawing tasks to the graphics card. It 
began 
> with the introduction of Quartz Extreme, and continued with Tiger's 

> Core Image. (Of course, if the Mac's graphics card isn't powerful 
> enough to support Quartz Extreme or Core Image, those features are 
> disabled.)

That would imply to *me* that you've been wrong all along with your 
CPU-bound argument, unless it's only Quartz's 3D and transparency 
elements that are handed off.

Your description of anything graphics-related being wholly CPU-bound 
sounds wrong to me, from the way I understand the Windows graphics 
subsystem to be interacting with graphics hardware.

I was only describing the Mac side of things as the implications of 
what you've been saying about graphics cards, instructing Tyler at 
length on how his own computers are working (despite the evidence of 
his own in-person hands-on experience).

For what it's worth, I think Tyler has plenty of reason to become 
testy with you -- you seem to be arguing in a manner that ignores any 
point he makes that goes to the heart of the dispute, while insisting 
that certain things are true in the face of his providing evidence 
that clearly shows that they are not.

-- 
David W. Fenton                    http://dfenton.com
David Fenton Associates       http://dfenton.com/DFA/

_______________________________________________
Finale mailing list
[email protected]
http://lists.shsu.edu/mailman/listinfo/finale

Reply via email to