That's an interesting benchmark, but I am more interested in a cold start test on a much smaller file (think 2MP, not 1200MP), because that's my usual use case.
I agree. That is another problem with the benchmark- it gives the gold medal to ACDsee, which in my experience has the longest cold start of the three.