Quote:
Originally Posted by
solipsism 
Anand updated his reason speed tests to show the speed ups with v3.0 over v2.x. I don't think it was much but their is a slight benefit overall.
There was an interesting anomaly in his statements about the tests.
He states that a good part of the 3Gs speed-up over the 3G is understandable because " this is a new microarchitecture".
He then mentions that the performance of the 3Gs is about that of the Pre, just 11% faster.
But that all SCREWY!
Why? Because that 11% is just for downloading web pages!
However when he gets to that last chart, we can see something much more interesting, which is the speed of opening some apps.
First of all, in the last chart, you notice that for some reason, Palm's name got reversed into white. At first, I couldn't read that it said palm, and wondered what the numbers represented.
Then I got closer and could read it.
Please make sure your window is wide enough to display the columns properly. I don't have the energy to do this as an image.
Application Launch Time in Seconds...Web Browser.......Dialer.......Google Maps......Camera....Email
Apple iPhone 3GS.................................0.7s.......... .........0.7 s............2.7 s.................2.8 s......0.8 s
Apple iPhone 3G...................................0.8s......... ..........1.2s.............3.3 s.................3.9 s......1.2 s
Palm Pre............................................... 3.0 s..................1.5 s............8.6 s.................4.4 s......3.3 s
T-Mobile G.........................................15.4 s..................2.0 s............4.4 s.................4.9 s......2.0 s
iPhone 3GS vs. 3G Perf. Advant...........13%.....................42%...... ........18%..................28%........33%
But what about the performance advantage over the Pre and "G"? It's far better!
And the Pre, at least, shares the same microarchitecture, as he brought up earlier. It may even have a slightly better graphics chip, possibly the 530 vs the 520 for the 3Gs.
So, what gives? Apparently something else is going on here. Is it possibly the crappy way programs are implemented on the Pre?
How about that Android "G"? That really pretty dreadful.
We're talking about anywhere between 200% and 600%
Why didn't he comment on this much more important data that he produced? That odd for him.