Develop Bugs

[Q] Benchmark low - how to push the benchmark higher?

Hello!
So, I have a DHD with the Revolution HD ROM (2.0.8) and the Buzz Sense kernel. I have locked the CPU to min/max 1516MHz Perf. scaling and, upon running the AnTutu System Review 1.2("System Benchmark" under Market) I get a score of 1879.
Yet in the Rankings I see that the #2 device is a DHD with 2.2 @1516MHz with a whopping score of 2774!!!!!
So, how could I reach those heights? The only thing I can think of is the slow SD card that came with the mobile (read/write scores: 34/117), but other than that what could be holding me back?
Cheers
T

I have reached pretty high scores while testing, but then the overall rom smoothness has pretty much sucked. It is normal that synthetic benchmarks give lower results when you use a custom rom, the optimizations that ensure general smoothness and usability cause it. One big reason is CPU scheduling, in custom rom (custom kernel) cpu time is divided for each task in a different way, thus reducing the overall benchmark score.

jkoljo said:
I have reached pretty high scores while testing, but then the overall rom smoothness has pretty much sucked. It is normal that synthetic benchmarks give lower results when you use a custom rom, the optimizations that ensure general smoothness and usability cause it. One big reason is CPU scheduling, in custom rom (custom kernel) cpu time is divided for each task in a different way, thus reducing the overall benchmark score.
Click to expand...
Click to collapse
Aha... but then this means that this or the other ROM do not let a specific app (say, a heavy 3D game) take the phone to its full potential...

Incorrect, it is just the synthetic benchmark that suffers. Games will run fine, faster than stock.

Jkoljo's right.The overall system speed is much better,performance is better,but it sucks a little in benchmarks.If you are so desperate to see high benchmark scores(I was once too! )try kamma's 1.4 kernel.Quadrant 3200.Need I say more?

tolis626 said:
Jkoljo's right.The overall system speed is much better,performance is better,but it sucks a little in benchmarks.If you are so desperate to see high benchmark scores(I was once too! )try kamma's 1.4 kernel.Quadrant 3200.Need I say more?
Click to expand...
Click to collapse
Really? 3200? Wow. Just out of curiosity, what scores are you getting with other kernels?
Well, at the end of the day, it doesn't matter - I haven't seen an app or game that runs slow or choppy or whatever. It's just for the heck of it

krakout said:
Really? 3200? Wow. Just out of curiosity, what scores are you getting with other kernels?
Well, at the end of the day, it doesn't matter - I haven't seen an app or game that runs slow or choppy or whatever. It's just for the heck of it
Click to expand...
Click to collapse
Oh yeah,I know what tou mean!
Anyway,with Lee's and Apache's kernels I get about 2500-2600 in Quadrant.CPU clocked all the way up to 1.5 GHz of course!

High Quadrant Scores like 2500+

Are we ever going to see quadrant scores like the hd2 is getting now ugh!!
A new rom has been imported to the hd2 that has it at 2800 quadrant score evo is better hardware and still at 1100-1500 whyyyy????
Sent From My HTC EVO4G

Lol.
What is the benchmark going to do for you?

garekinokami said:
What is the benchmark going to do for you?
Click to expand...
Click to collapse
Then what does it mean and why does it even exist?

Where can I download this at

If you really want it, you can get CM6.0 here:
http://forum.cyanogenmod.com/files/category/12-htc-evo/
and Snap 7.6 with Turbo here:
http://forum.xda-developers.com/showthread.php?t=733099
The current Snap only works with CM6.0 though which is a few versions behind, so you might as well wait though.

So the kernel is not universally compatible???

poulosjr said:
So the kernel is not universally compatible???
Click to expand...
Click to collapse
he did clearly say that it was only compatible with cm 6.0
no it is not universally compatible
the benchmark doesn't do anything for you anyway

Quadrant scores are a joke and easily manipulated, just look at CM6 with Snap Turbo. 2800 was my highest quadrant on it and speed was identical to all other roms.

Benchmark..Schmenchmark.....

The reason were all here is to improve the performance and functionality of our devices isn't it?? These scores are posted by every reviewer, manufacturers are giving us more powerful processors with each new device. Kernels and overclock tools are a staple in the dev community. How are these benchmarks not important? Sure they may not reflect the overall performance of the device in every case but if everyone viewed benchmark scores as useless data where would we be?? Not here trying to improve on them

DirtyShroomz said:
Quadrant scores are a joke and easily manipulated, just look at CM6 with Snap Turbo. 2800 was my highest quadrant on it and speed was identical to all other roms.
Click to expand...
Click to collapse
This. Even Cyanogen stated that the quadrant scores are fudged. I read his post somewhere, it had something to do with cheating on the I/O portion of the test.
My EVO with MIUI gets around 1750 and performs as good or better than when I was scoring 2600+ with cm6 and snap turbo.

Leedroid but what kernel

Hi all,
I've got Leedroid and would like to know how the hell people get the 3000+ scores on quadrant ?
I've used most of the roms with a lot of the kernels but the most i get is around 2500 - 2600 , i've heard that the kernel by Kamma is a good one but i'm at a loss as to which one i should use.
So if anyone could point out which one i could use for .
1. Stability.
2. Battery saving.
3. Benchmarking (without smoking my device of course)
Any answers would be appreciated.
Thanks.

dladz said:
Hi all,
I've got Leedroid and would like to know how the hell people get the 3000+ scores on quadrant ?
I've used most of the roms with a lot of the kernels but the most i get is around 2500 - 2600 , i've heard that the kernel by Kamma is a good one but i'm at a loss as to which one i should use.
So if anyone could point out which one i could use for .
1. Stability.
2. Battery saving.
3. Benchmarking (without smoking my device of course)
Any answers would be appreciated.
Thanks.
Click to expand...
Click to collapse
the first thing you need to understand is quadrant is ****, it is not an acurate indication of the speed of your device, a few devs have made kernels that trick quadrant into giving it massive scores, but in reality there kernels are not the best that being said, you wont find a better kernel for stability or speed than leedroids one, it is flawless you can download it here
http://www.multiupload.com/U2ECLTRT7V

AndroHero said:
that being said, you wont find a better kernel for stability or speed than leedroids one, it is flawless you can download it here
Click to expand...
Click to collapse
Have you noticed much difference between 2.0.5 and 2.0.6 kernels? The LeeDroid thread is getting very lengthy to read through these days to catch up on new updates lol!

RobSimmo said:
Have you noticed much difference between 2.0.5 and 2.0.6 kernels? The LeeDroid thread is getting very lengthy to read through these days to catch up on new updates lol!
Click to expand...
Click to collapse
Yeah the new one is based on the latest kernel source, with the 1.75 libs,
Sent from my Desire HD using Tapatalk

Tbh I use leedroids newest one an its really good, quadrant gets 2500 +
The device is smooth as silk, bit then again it always was.
I didn't know that quadrant was pants, I thought it was a good benchmarking tool.

Confirm that the newest kernel of Lee is the best!but in a few days he prepare a new awesome BFS kernel...

Mmm badly ****ed scheduler. My favourite.
If you want good quadrant, buy the paid app which gives higher scores than the free one. Do the I/O hack which core and lee droids use. Use Leedroid kernel at max freq, performance governor, and finally run it in airplane mode.
Follow all those steps and you will get around the 3000 score.
You should state if you use quadrant standard or advance when posting results as they do differ. Otherwise it's like comparing oranges to mandarins. They are both orange but one is always bigger.
Sent from my super slick Android device.

dr.m0x said:
Mmm badly ****ed scheduler. My favourite.
If you want good quadrant, buy the paid app which gives higher scores than the free one. Do the I/O hack which core and lee droids use. Use Leedroid kernel at max freq, performance governor, and finally run it in airplane mode.
Follow all those steps and you will get around the 3000 score.
You should state if you use quadrant standard or advance when posting results as they do differ. Otherwise it's like comparing oranges to mandarins. They are both orange but one is always bigger.
Sent from my super slick Android device.
Click to expand...
Click to collapse
Yea it's just the standard version of quadrant, I thought that it would do the job.
Also i don't want to trick it or actually buy an app just so it can give me higher scores, that's not actually helping at all, i'll try the airplane mode to see if that helps, my phone is rapid anyway so it's only for my own satisfaction and to see how high i could get the score to go to.

Quadrant: Worse than you thought

As we all know quadrant is no reliable measure for speed. At least I knew this for a while now and it was repeated and quoted many times.
This article tells anybody with a functioning brain (that is used of course) that quadrant means pretty much nothing.
I can't help to run it from time to time anyway
So I sat on the to... in my room in front of my computer with my phone. I9000 with supersonic ROM and the remount script from adrenaline shot 7. I sat there and said to myself "how hight can you score in quadrant LOL"
I started quadrant up and ran the benchmark: 2309
Then I opened the task manager-> Exit all & Clear memory
Then via long press homebutton back to quadrant to run the benchmark again score: 2453
But since I am a programmer and can imagine all kinds of optimizations and caching I pressed the back button and just ran it again just after it finished
Score: 2675
How the hell could anyone call that a benchmark?^^
just to be sure could anyone confirm that behavior? And does anyone know of a mor reliable alternative? I'd like to collect that knowledge in this thread.
TL;DR: quadrant sucks, you know anything better or want to flame away: do it here
Those are not the actual numbers from my first experiment, I repeated the scenario just now and took the numbers from those runs.
Additional runs scored 2775, 2907 and 2820, that's just silly

I think this behaviour is well known and has to do with JIT optimizations or something like that

allotrios said:
I think this behaviour is well known and has to do with JIT optimizations or something like that
Click to expand...
Click to collapse
The reason is irrelevant. The fact it doesn't provide a reliable benchmark is.

no benchmark is precise if you don't use it as intended. Quadrant produces a reliable comparative benchmark when used as designed: run it five times, remove the lowest and highest scores and average the remaining 3 -- that is your benchmark. You may not like it, but that is how it is designed to be used.
Now if you want to be pedantic, you could reasonably test again, by running quadrant 5 times, removing the outliers and average your 3 remaining scores. Repeat 10 times and then tell me how your average scores do or do not vary: they will in fact be within a narrow range, your actual benchmark.
Alternatively, tell us which benchmark produces the same score each run, as that appears to be the sum total of your objection to quadrant.
There are other benchmarks, such as Caffiene Mark, AnTuTu and NenaMark, but they are all apps just as Quadrant is and all require several runs and averaging to produce a comparable benchmark.
Moreover, the primary use of any benchmark is to compare firmware (kernel and rom) builds on the same phone to see relative performance gain and drop.

A benchmark is supposed to give way of comparing the capabilities of a given device. This means that a device with a high average score implies a better device than a lower score.
But the Quadrant score does nothing of this sort! In a competition with a friend I achieved an average Quadrant score of about 4300, with a peak of 4462. According to Quadrant my device is a lot better than the OP! Which is just not true.
Quadrant is unreliable as a benchmark, no matter how it is "designed to be used".
Sent from my GT-I9000 using XDA App

whaave said:
But the Quadrant score does nothing of this sort! In a competition with a friend I achieved an average Quadrant score of about 4300, with a peak of 4462. According to Quadrant my device is a lot better than the OP! Which is just not true.
Click to expand...
Click to collapse
You're doing it wrong.
lgsshedden said:
Moreover, the primary use of any benchmark is to compare firmware (kernel and rom) builds on the same phone to see relative performance gain and drop.
Click to expand...
Click to collapse

Quadrant scores are useless. I've used custom roms with scores of 2500+ but they aren't as smooth as stock roms, which only have scores of 1600-1800.

Antutu is indeed quite reliable imho. My results never fluctuate more than +-5% on the same config. That's an acceptable range, considering I don't set cpu governor to performance before running my tests.
Sent from my GT-I9000 using XDA App

upichie said:
You're doing it wrong.
Click to expand...
Click to collapse
w00t?
Quadrant does not reflect performance, and therefore can not be used as a comparison parameter.

It can't be much worse than I thought.
My phone with 2.1 and 'lag fix' scored 2200 and lagged so bad I wanted to throw it against a wall multiple times a day.
With stock 2.3 quadrant can be ~1000 but the phone runs much smoother.
Other than the obvious file systems I/O 'cheats' that resulted in the above, there is also the frame rate cap that makes the GPU tests useless as well.

if your trying to measure height with a scale , u wont get your answer .
The only benchmark tool that ever reflected how the phone felt in my hands , in real life usage is linpack .
changing OC / kernel is mainly the only thing that will affect linpack if your trying to use it to compare roms ill efer you to my first statement .
In order to have a good feel of a rom / set up on the phone , use some apps that will use lots of ressources , for example TW4 launcher , go in there scroll a lot open gallery (if you have many pics) scroll thru them and repeat ... Any benchmark tools will basically tell you the 'ability of your device ' ( comparing 2 different models like an inspire and an sgs2 for example will be accurate )

ZioGTS said:
Antutu is indeed quite reliable imho. My results never fluctuate more than +-5% on the same config. That's an acceptable range, considering I don't set cpu governor to performance before running my tests.
Click to expand...
Click to collapse
I recently tried Passmark Mobile. Still a beta version, but I like it. Test results reflect real performance improvement and degradation pretty closely, particularly for what concerns I/O and memory speed.

benchmark confusion

Hi!
i just got my dhd, and i am a bit confused about the bench scores.
firstly, i'm tunning rcmix3d runny 3.3, with defoult kernel. what confuses me is that i get a normal result in quadrant (around 1900), i only get around 9-ish in linpack, where i should get around 40 if i understand correctly. phone feels quite smooth...
could somebody please explane what i dont get here

t0mas_ said:
Hi!
i just got my dhd, and i am a bit confused about the bench scores.
firstly, i'm tunning rcmix3d runny 3.3, with defoult kernel. what confuses me is that i get a normal result in quadrant (around 1900), i only get around 9-ish in linpack, where i should get around 40 if i understand correctly. phone feels quite smooth...
could somebody please explane what i dont get here
Click to expand...
Click to collapse
Even I am getting the same results.
Changing cpu speed using setCPU has no effect on Linpack.
Maybe its something to do with Linpack i suppose.

Most benchmarks have absolutely nothing to do with phone performance. Especially quadrant and Linpack.
Linpack tests CPU Speed, which is pretty pointless since there are few tasks that actually require CPU (everything should still feel smooth @ 768 MHz).
And quadrant (Especially standard) gets quite alot of its points from IO on the Desire HD, which is actually pretty pointless. The 3D Performance is around the same on every ROM.
I would suggest getting a 3D benchmark like Nenamark/etc if you actually want to test performance, since 3D gaming is the only thing that would actually require it.

thanks for info, i figured that everything works quite fast so why bother with some benchmarks
but since i like to understend stuff, i feel better now