Android 2.2 - as fast as you wished!

RinTinTigger

Premium Member
Premium Member
Joined
Nov 4, 2009
Messages
2,847
Reaction score
2
Location
Germany, Near Frankfurt/Main
Website
www.droidforums.net
android22-fast.jpg

AndroidPolice now got a Nexus One with Froyo (Android 2.2) for testing and one thing they did, was Benchmarking the System, to show how fast it is. Expectations are pretty high, as FroYo is expected to come with JIT (Just in Time Compilation).

Now, they used the app Linpack, which measures all Benchmarks in FLOPS (floating point operations per secound).

the nexus one has been tested with 6.5 MegaFLOPS, which is very fast, compared to a Hero with 2 MFLOPS. but now, how is FroYo doin on the Nexus?

PWNAGE! It shot 37.5 MFLOPS!! this is 5 times faster as 2.1 is....just one more reason to sincerely wait for android 2.2!!

Tigger
via
 

Hugh Jass

Senior Member
Joined
Feb 7, 2010
Messages
1,659
Reaction score
121
I'm dying to know how this will translate to the Droid. Hopefully we get similar results with the JIT working good.
 

Darkseider

Senior Member
Joined
Mar 12, 2010
Messages
1,862
Reaction score
0
Just ran the linpack on BB v1.0 running @ 800Mhz and I score a consistent 6.4xx MFlops. So it seems that the Droid at 800Mhz is roughly the equivalent to the Nexus at 1Ghz when it comes to floating point ops. Now the question is whether or not JIT translates to real world performance or just some increase on a synthetic benchmark.
 

Hugh Jass

Senior Member
Joined
Feb 7, 2010
Messages
1,659
Reaction score
121
I get 9.915mflops in linpack with BB1.0 at 1.2ghz lv. The JIT is doing something, that's for sure be it synthetic or not.
 

Darkseider

Senior Member
Joined
Mar 12, 2010
Messages
1,862
Reaction score
0
I get 9.915mflops in linpack with BB1.0 at 1.2ghz lv. The JIT is doing something, that's for sure be it synthetic or not.

Umm... from what I see here LinPack scales nearly linearly with clockspeed.

800 Mhz Droid = 6.4
1.2 Ghz Droid = 9.9

6.4 / 8 = .8
.8 * 12 = 9.6

So now the question is what JIT does to increase those numbers and is it an artificial result solely based on the benchmark? In order to see a real impact one would need to run some type of process intensive task, say converting a file to an .mp3, from the Droid console on a rooted device. Otherwise I would take this with a grain of salt.
 

Zarniwoop

New Member
Joined
Nov 6, 2009
Messages
21
Reaction score
0
Benchmarks, shmenchmarks.

I don't give two craps about benchmarks...

I want the snappy response the Incredible seems to have out of the box, and the battery life of a Saharan Camel. :)
 

gatewayad

Member
Joined
Nov 6, 2009
Messages
521
Reaction score
0
Location
St. Louis, MO
I wonder how fast it is on 3g vs. the wifi they are running, haven'tweI seen high 25's on the droid on wifi
 

Hugh Jass

Senior Member
Joined
Feb 7, 2010
Messages
1,659
Reaction score
121
The Droid News section has no edit, for whatever reason.
 

gatewayad

Member
Joined
Nov 6, 2009
Messages
521
Reaction score
0
Location
St. Louis, MO
Sorry what I'm saying is isn't there a big speed difference doing a benchmark test on WiFi vs doing it on 3G
 
Top