Vc branch ready for testing

[Arjen Hiemstra]

On Tuesday 11 September 2012 17:35:55 JL VT wrote:
> 
> Before purchasing my RAM I looked at a long list of benchmarks on Tom's
> hardware done on different types of software, and the only time the effect
> of faster RAM was really noticeable was on artificial benchmarks.
> Everywhere else, like games, encoding, video editing, it didn't go past 7%
> speedup, even going from 1333 Mhz to 2100 Mhz, or single to dual, or dual
> to quad channel. That's why I think it's not a memory problem. And if it
> was, maybe we can program things differently; after all, if other graphical
> software isn't affected by memory speed, why are we?.

Modern games are heavily GPU and Disk I/O bound, since they need to constantly 
load textures from disk and use a gazillion of shaders. Encoding is heavily 
Disk I/O bound since each frame needs to be written to disk. Video Editing is 
also Disk I/O bound as most videos are too large to be contained completely in 
memory. Neither of these workloads really compares to what Krita does. Krita 
has all data of the image in memory and performs most of the calculations on 
the cpu, meaning that memory bandwith is most likely one of the biggest 
bottlenecks here. (Fun fact: Games are actually often memory bound, but 
graphics memory bound rather than system memory.)

- Arjen