[citation][nom]acku[/nom]I've tried passmark, but I'm more hesitant to adopt their tests. Putting aside whatever you feel about their desktop software, which is a whole different discussion.GLBenchmark is the mobile industry standard for graphics performance. It's what the companies use internally to evaluate performance during an R&D phase. That said, we've had a pretty good line of communication with Kishonti (and GeekBench), so we have a good understanding of what the programming looks like on each operating system. As a result, it's easy for us to use those titles in our reviews in a cross platform comparison. Right now, I can't say the same for Passmark. However, I've been playing the benchmark you mentioned here and there. Since it originally came out, I've noticed some shifting (post updates), so I'm still on the fence on how we might use it.Cheers,Andrew Ku[/citation]
Hmmm i see....yeah i think i sort of know what you mean...they changed the way they calculated CPU encryption bandwidth...my ipod shot up from ~5 MBps to ~22 MBps...and yeah it seems to output fluctuating results, even if they are only small fluctuations.
I was originally excited about passmark because it was available on Android, iOS and the PC, so i thought it was a nice way to see how mobile devices are coming along...but i guess we have geekbench for that too.
But i agree with you, better to use something you trust. What about Vellamo and AnTuTu (plus their battery benchmark)? Have you tried them? Their battery benchmark seems to be a very worst-case method of testing, just loads up the CPU at 100% till the battery hits 20% (or 15%, can't remember)? But then again they're not suitable for cross-platform evaluation...
Hey i just had a brainwave: Why doesn't Tom's develop it's own benchmark! That would be awesome! Or you could hold a contest or something, people could submit stuff.