On Aug 16, 2011, at 11:37 PM, David Brownlee wrote: > On 16 August 2011 21:19, Lars Heidieker <l...@heidieker.de> wrote: >> Real world benchmark is almost the same. >> a release build eg >> >> real user system >> current: 2735 8422 1390 >> vmem-heap: 2716 8419 1383 >> >> variations between runs are easily larger then this. > > Would the benchmark be expected to show any difference in a high > memory pressure situation?
I don't expect much of a difference except for the fact that there should be no failed allocations with the changes, so performance wise probably a slight edge for the vmem solution...