Lets talk about benchmarks one last time. Evo 3d > Atrix

[MannyZ28]

When my friend first got his Atrix, I was all over benchmarking this thing. It scored well in Quadrant getting around 2600. The Evo 3d always hovers around 1900 to 2100 as we all know. So the day came that the Atrix got its update to Gingerbread. Seeing that the benchmarks on the Photon for Quadrant were in the Evo 3d territory, I felt frisky and decided to download quadrant again on both me Evo 3d and his Atrix. Low and behold, the Atrix lost. 2023 vs 2145. They performed pretty much identical in Nenamark, and at that time, my friend was tired of benchmarking so I couldn't run any other tests.

Isn't it weird how the Tegra 2 device scored higher with Froyo than with Gingerbread? Seeing the reviews of the Photon made me take a double take after seeing the Quadrant scores.

I wonder how an Evo 3d would score with Froyo? Not that it matters, I just figured GB would be more optimized for 3d graphics, but now it seems Froyo was better at it?

Strange.

 

[Averix]

I think part of it is the very nature of the benchmark apps. They tune them as deep to the metal as they can. When you deal with substantially different CPUs like Tegra vs. Snapdragon vs. whatever else, the benchmark apps may all be handling things a bit differently and skewing scores. Android itself does impact this too. When graphics libraries underlying functionality is changed, it will impact the benchmarks. This may be because the benchmark app is expecting something a bit different or because the abstraction of the libraries is a bit different.
All told, I don't give the benchmarking much more than a general ballpark type credit. That and fanboys of whatever device/platform latch onto the things like a lifeline. It's a fun distraction, but not the end all beat all of truth.

 

[MannyZ28]

I think part of it is the very nature of the benchmark apps. They tune them as deep to the metal as they can. When you deal with substantially different CPUs like Tegra vs. Snapdragon vs. whatever else, the benchmark apps may all be handling things a bit differently and skewing scores. Android itself does impact this too. When graphics libraries underlying functionality is changed, it will impact the benchmarks. This may be because the benchmark app is expecting something a bit different or because the abstraction of the libraries is a bit different.
All told, I don't give the benchmarking much more than a general ballpark type credit. That and fanboys of whatever device/platform latch onto the things like a lifeline. It's a fun distraction, but not the end all beat all of truth.

I just find it interesting how much flak the Evo 3d and the Sensation got from reviewers about the low scores, esp compared to the GS2, but not too much is said about it with the Photon. Then again, I am not watching too many reviews and I am not reading their forums.

 

[CarrieK]

I base it on how the phone performs day to day. Wow... Must have been lit when I posted last... I apologize

 

[milominderbinde]

Don't get hung up on benchmarks. As many have shown, manufacturers know tricks to manipulate the numbers.

Try them out and see what feels best to you. Look at the s[ecs and features in the Phone Comparison.

Regardless, vote in the Readers choice awards: Sprint's best Android phone.

 

[Quis89]

I dont understand why people care about benchmarks so much. I've honestly never ran one on my 3D. It runs just fine...I couldnt care less about those numbers.

 

[Wiley_11]

I dont understand why people care about benchmarks so much. I've honestly never ran one on my 3D. It runs just fine...I couldnt care less about those numbers.

I'm with you. It ain't like we're racing for pink slips or money. Heck you don't even get a trophy.

 

[warpdrive]

I dont understand why people care about benchmarks so much. I've honestly never ran one on my 3D. It runs just fine...I couldnt care less about those numbers.

running benchmarks comes from the days when people like me built their own gaming computers.
we would use benchmarks to show how fast at stock and overclocked the computer was.

now we do the same with phones. but let's face it, benchmarks don't tell the whole story.

in the case with the OP and going from Froyo to GB, I'm not suprised at all that the phone runs drop slower. the new OS will take more room on the RAM to run, and if you are already low on memory, then it can run slower then the lighter Froyo. just wait for ICS and watch all the update problems for phones with only 512 or less of RAM. I'll almost bet it wont be updated using OTA on many phones.

 

[MannyZ28]

running benchmarks comes from the days when people like me built their own gaming computers.
we would use benchmarks to show how fast at stock and overclocked the computer was.

now we do the same with phones. but let's face it, benchmarks don't tell the whole story.

in the case with the OP and going from Froyo to GB, I'm not suprised at all that the phone runs drop slower. the new OS will take more room on the RAM to run, and if you are already low on memory, then it can run slower then the lighter Froyo. just wait for ICS and watch all the update problems for phones with only 512 or less of RAM. I'll almost bet it wont be updated using OTA on many phones.

On my Evo 4g, the benchmarks going from Froyo to GB actually improved quite a bit, which is why I am so surprised that they fell so much on the Atrix, so that doesn't really make any sense. Also, I have high hopes for a good performance gain on my Evo 3d going to ICS since it will finally take advantage of the dual core. Then of course it's still up to the devs to make their apps run properly and take advantage of it.

Again, I am not a benchmark junky, I just observed something which peaked my curiosity. The Evo 4G gained tremendously from GB, including benchmarks, not just overall daily performance. The Atrix gained overall day to day performance, but the benchmarks sank to Evo 3d levels after the update.

 

[warpdrive]

You just described why benchmarks are flawed and don't show what you will get while playing a real game. They mean little when high end titles will optimize for the latest phones. Now give me fraps or a good timedemo of a real game.