Tim,
While i respect you for being on the forum more than me, i have to strongly disagree whith what you have said - I didnt say it would do it 4 x faster.
I said
"this basically means that when you send a request to the CPU it splits it up into 4 and can get the job done 1/4 of the time faster than a single core...."
You are correct that SLI is a brand, i didnt say it was a standard. Mearly pointing out that having two GPU's that are SLI interfaced and that are NON AGP (Therefore PCIe) will work a lot faster.
But you say "
Most software does its computation using the CPU rather than the graphics processor so the "power" of the graphics card is irrelevant" i feel is wrong.
Try this, remove your PCIe graphics card and pop one in that is 128meg £20 graphics card, lets see if it shows and plays back the rendered HD footage just as good as 2 x PCIe graphics cards in SLI with 512meg or more on each, that are hyperthreaded and clocked. I never once said that the power of the GPU was the thing to go for the most, but to show the rendered footage on an external TV so you have a good representation of what you will actually see, you will need something that has a bit of power with either DVI out or S Vid out.
You also said -
"A basic graphics card with 256MB RAM will be more than capable" Funny, you havent mentioned DVI out, S Vid out and the like's.
If this chap goes out and buys a cheap £20 graphics card with 256 meg on, he may not even know that he needs DVI out for a good monitor, or that S Vid out is a must if you want to pop the image onto a TV.....
Yes a 256meg Graphics card will do, but i was letting him know that the graphics card will need to be a little better, simply because getting the rendered HD footage onto an external TV can not be done unless you have a card with DVI out or S Vid out. I didnt mention them, because a good card with 512 or more on it should have them on.
Sorry Tim but you made me out to be wrong on all accounts, thats just not correct.
Snippits from a small problems page on CPU's.
The amount of performance gained by the use of a multicore processor depends on the problem being solved and the algorithms used, as well as their implementation in software (Amdahl's law). For so-called "embarrassingly parallel" problems, a dual-core processor with two cores at 2GHz may perform very nearly as fast as a single core of 4GHz. Other problems though may not yield so much speedup. This all assumes however that the software has been designed to take advantage of available parallelism. If it hasn't, there will not be any speedup at all. However, the processor will multitask better since it can run two programs at once, one on each core.
With the above in mind, running windows and other essential tasks will take up a small amount of power from the CPU, if you have a quad core, then this can be split into 4 and the larger requests can be done quicker, while the basic task of running windows can be dealt with by one of the other cores...
Have a look at the attached gif, it will explain a lot.
Alienware: Understanding Processor Performance! for more information.
Mitch