GXMark can it really be fooled??

[ryncppr]

Ok so since I have been building roms for the last few months, I noticed a few things...all the newest ROMS have a GXMark rating of 1588. Now don't get me wrong all the roms I build have the Pagepool set to Dynamic. Why waste memory if you don't need it right???

Anyways, I noticed if I open all my folders (i.e. Tools, multimedia, communications ect..) at least once and get the pagepool up I get a higher GXMark rating...

For example, right after a soft reset, I get 1528, after I mess with the phone for a few minutes the rating goes up to 1588. After letting it sit a bit longer, I got a rating of 1617????

So does anyone know why this is happening??? If so what is the best Pagepool to have the phone on in order to get an accurate reading?

Just a thought and my finding after a long day of cooking:confused:

[redbandana]

I have read reports of a user touching his screen while GXMark was running and the final results appeared to be much lower (I think it was 650-750). Then reran the test without touching and received a 1588. The user then performed the test yet again with similar results.

Conclusion: I do not base rom performance by running an application. I would rather base performance on my opinion.


I like that you noticed how unreliable GXMark results prove to be.

[omar302]

I have stock WM6.0 and get about 1400. But if I hace a video running in WMP while GXmark is running, the score will go as high as 1588.