View Full Version : High-end Android Benchmarks
Skele Drew
November 10, 2010, 09:07 AM
If you think that all 1GHz Android smartphones are equally quick, think again. ZDNet Germany’s Christoph H Hochstätter shows that the performance of CPUs, RAM, flash and GPUs varies widely.
The basic specification of all high-end smartphones is pretty standard: a 1GHz ARM processor; a 3.5in. to 4in. display with a resolution of about 480 by 800 pixels; 3D graphics acceleration; and around 512MB of RAM.
What’s not evident from the smartphone manufacturers’ specifications is the fact that there are significant performance differences. Fortunately, these are clearly measurable.
Benchmark showdown: High-end Android smartphones (http://www.zdnet.com/blog/btl/benchmark-showdown-high-end-android-smartphones/39206)
nigelt
November 10, 2010, 05:01 PM
Nice article. I don't have Quadrant Pro, but my Milestone running 2.1 scored 544 in Quadrant Standard (slightly higher than the Nexus One running 2.1)
Skele Drew
November 10, 2010, 07:48 PM
One thing I noticed is that custom roms outperformed stock roms in pretty much EVERY case. Yet another reason for loading custom roms on your device.
ram130
November 10, 2010, 11:51 PM
Nice article. I don't have Quadrant Pro, but my Milestone running 2.1 scored 544 in Quadrant Standard (slightly higher than the Nexus One running 2.1)
my vibrant is around 2000 or more..
Neoprimal
November 11, 2010, 12:04 AM
@Skele Drew.
ROMS aren't truthfully/realistically represented by Quadrant and Linpack because they're synthetic and aren't weighted to/for the changes and tweaks performed in the guts of the software in these ROMs. This isn't to say they don't actually perform a bit better than the stock ROM in many cases, it simply means we can essentially throw the scores we see out the window in terms of comparing how fast device A is vs. device B by looking at these numbers.
For Eg: take a look at this list. (http://smartphonebenchmarks.com/)
The score for the Captivate has been ripped to shreds by the lag fix, which essentially creates a virtual file system for the OS to work with rather than Samsungs redonkulous file system Quadrant however, hasn't been adjusted to weigh and test this properly so the Captivate gets a result that is false. In reality if you were to stick a number on the phone relative to other phones you've seen tested by Quadrant, I'd say I'd give it a 1200-1300.
The score for the stock G2 while also pretty grand is real, because that's an actual hardware overclock and Quadrant is assigning a properly weighted value.
In essence, if you were to run the phones side by side, the G2 on that list would flick the lag-fixed Captivate away with it's pinky.
In fact, I've run my MT4G against a Ryan's Lag Fixed (supposed to be the best lag fix currently available) Vibrant and my MT4G went nom nom nom. The owner quickly sold it and got himself a MT4G as well, lol.
Where ROMs may enhance the experience of using the phone by removing 'bloatware', being 'pretty' at default and allowing users to install whatever they want - they by no means offer as 'potent' a jump in performance as some people think they do or as some may lead you to believe.
LeBoss
November 11, 2010, 03:53 AM
Nice article, so the each device is good its own way.
Vibrant outplays with stock 2.1
Desire blast away with Froyo
Droid for it internal memory
Combine these three phones and you got yourself a well machine Android phone
nigelt
November 11, 2010, 08:04 PM
my vibrant is around 2000 or more..
Using Quadrant standard or pro? And are you using a stock rom? The benchmark for a stock Milestone is like 275, so 544 is pretty good for me since I'm using a stock rom
ram130
November 14, 2010, 05:24 PM
Using Quadrant standard or pro? And are you using a stock rom? The benchmark for a stock Milestone is like 275, so 544 is pretty good for me since I'm using a stock rom
yes with a custom rom and lag fix...but I've since moved on to better worlds.
Powered by vBulletin® Version 4.2.0 Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.