Why would you want to upgrade a perfectly good 285 for a 570? You will be hard pressed to notice any difference in performance, at most is will be marginal for rendering and have no effect on encoding. I think that the $ 350 or there about can be better spend on adding memory and disks, improve cooling or PSU and overclocking.the current setup.
There are some issues--actually, more like side effects--with the linear color processing used my hardware MPE. These side effects show themselves in a number of different ways, but one of the most telling are the harsh/chunky edges created by blurs, shadows or glows on clips with alpha channels. It's not an issue when blurring regular video assets.
Also, it's not a result of a particular GPU or driver, but rather, part of hardware MPE/CUDA itself. It's also not dictated by an "official" card or a "hacked" card--it affects all GPUs of all levels. So, you would not likely see any change by going from a GTX285 to a GTX570; if you don't see it on the former, you won't see it on the latter. But as Harm says, there is little motivation at this point to go from the 285 to the 570.
>I'm about to upgrade my GTX 285 to either the GTX 570 or GTX 470.
I realize I will have to apply the "hack" to make either of those work properly.
The GTX 470 is on the list of cards that provide the CUDA acceleration features.
The official and up-to-date list of the cards that provide the CUDA processing features is here:
Some of the cards on that list are only enabled if you have the recent updates.
Harm, because my GTX 285 is getting old, I'm noticing some small problems like improper detection of displays, and I'm seeing all top performers in the benchmark using something better than a 285. It's worth a shot, funds aren't a problem.
I was being unclear, I meant to say that I'm aware that I'd have to apply the hack to the 570, if I choose that one. Thanks
Exactly what I was looking for, thank you Colin.