Our phones cannot display more then ~56 fps due to vsync
Test settings: JM2, samset 1.8, mimocans ext4 fix + dalvik cache sent to nand, 8gb class 6 sd, all benchmarks were only run once.
Neocore:
200mhz : 36.8 fps
400mhz : 56 fps
800mhz : 56 fps
1000mhz : 56 fps
Yes, with our cpu reduced by 80% we still beat the nexus one and other froyo omapp/snapdragon cpu's in neocore.
Quadrant: (world & moon, the others are more difficult to test)
200mhz : 40fps : 410 Quadrant
400mhz : 56fps : 789 Quadrant
800mhz : 56fps : 1532 Quadrant
1000mhz: 56fps: 1762 Quadrant
Fps2d is 56 fps at all cpu settings.
-IMPORTANT-
Due to the nature of vsync we have no idea what fps we could be getting, Generally when you scale down cpu performance the bandwidth available for the gpu decreases dramatically as the processor cannot handle the new frames to keep up with the gpu. What this is showing us is that in neocore at 1000mhz we only require 20% of the cpu to get 36.8 fps and somewhere around 300mhz to get to the 56fps vsync cap. For all i know the gpu might only be able to output 70 fps without vsync at 1,000mhz, the question is by how much does reducing the cpu speed effect the available bandwidth to the gpu
This shows us:
1. Our gpu is largely held back by the frame-limit, the cpu also has an impact on the gpu performance.
2. We need better benchmarks to test only 3d performance, many of the 3d tests in quadrant are either heavy on the cpu, or are 2d tests. And neocore is simply not taxing enough on our gpu.
3. Need to find a way to remove vsync
4. Its a powerful beast! looking forward to froyo
Look foward to GLbenchmark 2.0
Even currently in 1.1 it's showing some impressive results, even though it's not supported very well on latest-gen gpu's (SGX530/535/540 mostly)
*off topic*
Would you please link / pm me to the Sd card you are using?
Its a Samsung 8gb class 6
would underclocking the cpu to 600mhz for example help increase battery performance, or it it mostly the screen that drains the battery?
its mostly the screen
But really really great research there !
Its nice to know.. Hehe..
The next thing i would like too see is quadrant tests where the SGS has undergone the Lagfix and then with different SD cards
Related
One of my coworkers has a tmobile vibrant with some lag fix according to him.. he did a quadrant benchmark right in front of me and it was showing 2500 plus everytime.. Im very curious as to what is making his phone so fast. And can it be dont to ours. Hes not running a custom rom or overclocking. Im only getting 1030 with mine clocked at 1.2ghz. Any Ideas? I couldnt get into too much details with him yesterday and I dont know whens the next time ill see him..
If you were to look at a test break down you would see generally all the scores are identical or the epic a little ahead except in the read/write area. The scores from their read/write are just inflating their overall score. It's a issue with quadrant and how it handles its overall score. Basically it just makes the system easy to abuse/cheat. So I wouldn't worry much about the difference in your score and his.
Sent from my Samsung Epic
The reason other Galaxy S phones score high in quadrant is because of the lag fix they use. The lag fix mounts a different file system on the phone with DRAMATICALLY increases read-write times. That portion of the quadrant benchmark gets inflated beyond reason. Using this game technique, Cyanogen was able to score more than 3000 on a snapdragon phone.
All of the Galaxy S phones have the same processor. Also, quadrant is a terrible benchmark. It's the most over-quoted and abused benchmark for android phones
Ahh ok.. thats good to know.. so what would be a better benchmark to use? Linpack?
jok3sta said:
Ahh ok.. thats good to know.. so what would be a better benchmark to use? Linpack?
Click to expand...
Click to collapse
Linpack is good for measuring raw CPU processing power... but only on devices running the same version of android. Phones with 2.2 will score insanely high due to the JIT compiler. For example, a snapdragon phone with Froyo can score ~40 Mflops. A snapdragon phone with eclair scores around 7 Mflops. Does Froyo make the phone run 5-6X faster? Hell no. In some cases, the difference is almost unnoticeable to the human eye.
Here is a rundown of what I believe to be the pros and cons of various benchmarks:
Linpack
Pros:
- Good for measuring CPU processing power on the same version of Android
- Great tool for measuring the performance gain from overclocking
Cons
- Scores are boosted unreasonably by Froyo's JIT compiler on snapdragon phones
Quadrant
Pros:
- Great tool for measuring the performance gain from overclocking
- Decent tool for measuring 3D graphics performance (just pay attention to FPS, not the end result)
- Decent tool for measuring 2D graphics performance (again, look at FPS)
- The paid version ("Quadrant Pro" I believe) shows which parts of the benchmark contributed to the score. Easier to spot the inflated CPU or I/O inflation
Cons:
- I/O portion isn't valued as much as others, but can boost scores beyond reason via exploits, hacks, fixes, etc.
- CPU portion is inflated on phones running 2.2. A Nexus One is not faster than any Galaxy S, Droid X, Droid 2, etc.
Neocore
Pros:
- Good tool for measuring graphics processing power
Cons:
- Graphics are not intense enough to push the power of very fast GPU's. Some phones will hit their FPS limit
- Only measures graphics processing power.
Nenamark1
Pros:
- Great tool for measuring graphics processing power
- Effects are advanced enough to show the performance of faster GPUs in relation to phones with lesser GPUs.
Cons:
- Only measures graphics processing power.
Sweet thanks for all the info man..
Agreed, this is great info thanks. I think the quadrant score is the most quoted becuase it provides a very easy to read graph built in with it for instant comparing/gratification. I guess I am gonna start going by linpack and nenamark1.
hydralisk said:
Linpack is good for measuring raw CPU processing power... but only on devices running the same version of android. Phones with 2.2 will score insanely high due to the JIT compiler. For example, a snapdragon phone with Froyo can score ~40 Mflops. A snapdragon phone with eclair scores around 7 Mflops. Does Froyo make the phone run 5-6X faster? Hell no. In some cases, the difference is almost unnoticeable to the human eye.
Click to expand...
Click to collapse
Linpack is ok for when your using same CPU comparison, different CPU's can cause issues...The reason why snapdragon gets scores of 5-6x is for some reason the snapdragon utilizes the VFP rather then using raw processing power..aka snapdragon cheats on the Linpack.
In reality our I/O scores should be a lot higher then it is as even in the Epic some of samsung's crappy file system still exists. But not as high as the lagfixed Vibrant of course.
Quadrant Pro is probably best indicator out of them all(The non-pro version is pretty much useless unless your comparing the same phone)...the con of having 2.2 show is higher is expected as it is a measure of efficiency of JIT in comparison to the current. The OS always played a role in Benchmarks so it is expected.
it can be faked by using a different partition to test on. IIRC the data partition making the speeds much faster than they should be so be careful when accepting those high scores
rjmjr69 said:
it can be faked by using a different partition to test on. IIRC the data partition making the speeds much faster than they should be so be careful when accepting those high scores
Click to expand...
Click to collapse
It is not exactly faking it..as you are increasing performance..thing is you cannot see at what it performs well at unless you see the individual scores from the Pro version....
whilst running Quadrant, there were a few issues that occured to me, The TF is supposed to have a superior GPU so why has Asus capped the frames per second in the graphics benchmark my Xperia X8 which is far less superior than the TF can reach around 50 FPS on the moon orbiting sun graphics, while the TF can only do about 30-35, i was just wondering if this should be an issue
Resolution
rilot said:
Resolution
Click to expand...
Click to collapse
+1
very straightforward answer man.
Nenamark1 => 35fps
Nenamark2 => 18fps
Quadrant => 2528 avg
Smartbench 2011 = 3232
Running Roach's Prime 1.4 (HC 3.1)
Terrible FPS scores in comparison to other inferior devices. The Smartbench scores show it's pretty close to other 1.0ghz HC Tegra2 devices.... but still not great....
I installed CM7 with DSC Phoenix test2,overclock to 1344MHz.
when I bench with Quadrant Advanced ver2.1, my score is >1300 to 1400.
It's lower than my old milestone 1 (with CPU 600 MHz Cortex-A8 and GPU PowerVR SGX530, overclock to 1,1Ghz). quadrant score of milestone1 is > 1700. I dont know why?
And this morning I tested with HTC Incredible S of my friend (CPU 1 GHz Scorpion
and GPU Adreno 205, ICS rom without overclock). Score is > 1900.
Is my streak weak? I cant believe streak5 is slower than milestone1 and HTC Incredible S.
how about you?
I think many on here will agree with me when I say forget about the benchmark scores. They largely mean nothing as it relates to real world performance and you can drive yourself crazy trying to tweak your phone to death to increase the numbers. With that being said, the performance hit is likely in the file i/o section.
As normal user and not a programmer, I dont understand about the file i/o section. But thanks you, Lordmorphous. So I'll forget foolish scores, what a stupid app =))
File I/O = File input and output, which in simpler language is the act of reading and writing to the MicroSD cards.
Quadrant and other such benchmark tools are fun to run, but they don't mean much. To properly test a device requires a benchmark program like Futuremark's PCMark, which includes real-world applications and simulates real-world usage. There is no such benchmark app for Android however.
though, if you guys notice the benchmark for the quadrant advance 2.x, the i/o score for our DS5 does a bit low compared to others but... well you know, it just numbers... nothing more.. nothing less...
however, cpu power does have the impact when overclocked, for let say using default cpu speed of 1ghz, software decoding is a bit slow compared to overclocked 1.2ghz...
just my 2 cents
So i've seen in internet that people who uses trinity kernel scores over 7k (in quadrant bench) @ 1.64ghz.And im with 1.7ghz and 700mhz gpu still get arround 6696~ [Got nexus 7 32gb 3g model]...So what's the problem?How can i increase scores?And improve overall performance?
Btw is there any software or something that could fix blown speakers ?
FatalaS said:
So i've seen in internet that people who uses trinity kernel scores over 7k (in quadrant bench) @ 1.64ghz.And im with 1.7ghz and 700mhz gpu still get arround 6696~ [Got nexus 7 32gb 3g model]...So what's the problem?How can i increase scores?And improve overall performance?
Btw is there any software or something that could fix blown speakers ?
Click to expand...
Click to collapse
the gpu speed doesnt matter as quadrant doesnt test gpu. it looks at our fps, and our fps will be 60fps no matter how high your gpu is oc'd. lower your gpu speed as it increases how hot the device gets. are you using the trinity kernel toolbox app btw? at 1.7ghz i get around 6900-7200. make sure that you disable tegra hotplug and enable all four cores, for that extra bonus in your scores. and that you bench at 1700/1700(high/low)(dont let the device scale). also, make sure that your device isnt getting thermally throttled. if it is, it automatically reduces your cpu speed no matter what you set it to.
simms22 said:
the gpu speed doesnt matter as quadrant doesnt test gpu. it looks at our fps, and our fps will be 60fps no matter how high your gpu is oc'd. lower your gpu speed as it increases how hot the device gets. are you using the trinity kernel toolbox app btw? at 1.7ghz i get around 6900-7200. make sure that you disable tegra hotplug and enable all four cores, for that extra bonus in your scores. and that you bench at 1700/1700(high/low)(dont let the device scale). also, make sure that your device isnt getting thermally throttled. if it is, it automatically reduces your cpu speed no matter what you set it to.
Click to expand...
Click to collapse
Yes i use trinity kernel toolbox.I always disable tegra hotplug,using fsyn for faster all options to fastest .But still max score was 6600~...
FatalaS said:
Yes i use trinity kernel toolbox.I always disable tegra hotplug,using fsyn for faster all options to fastest .But still max score was 6600~...
Click to expand...
Click to collapse
Don't worry about the score. It doesn't matter as long as your device is running nice and smooth, right?
I need answers from hardware specialists rather than software specialists.
I am trying to overclock my Galaxy S4 9505 GPU (Adreno 320) as much as possible. All I know that I need to keep an eye on temperature and NOT to increase it over 90C (to be safe).
But I noticed that I have the only one temperature sensor in galaxy S4..? is it only for CPU, or both - GPU and CPU?
I see it masivelly increases when my CPU is running at frequency of 2160mhz, but when I overclocked my GPU from 450mhz to 545mhz I didn't notice any bigger temperature increasing. Only a better score in benchmarks
My BIG comment in other overclocking for gaming thread with all information what I was trying to do etc : http://forum.xda-developers.com/showpost.php?p=53975711&postcount=27