Related
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
http://www.engadget.com/lg-star/preview/
Definitely an impressive piece of hardware, the Nexus S will have to do for now I guess.
A key question:
Does it have any US 3G (850/AWS/1900) band?
If not, it lost a leg.
http://www.youtube.com/watch?v=cULT5ZeI44s&feature=player_embedded
^ video
Yes, such a beautiful phone, is it SLCD? I guess I can bypass SAMOLED if it is.
I am still waiting to see, the first orion-based phones, as they will be without a doubt the fastest phones. I just hope Samsung up the looks of the phones, maybe not so plasticy, more glass is nice though.
Impressive but not a fan of LG phones. Samsung galaxy 2 will be the talk bigining of next year
Sent from my SGH-T959 using XDA App
irishrally said:
Engadget just put the LG Start dual core tegra2 running on 2.2 through Quadrant and it only scored 2100. Galaxy S phones have scored much higher than this which makes me feel good about Google's choice. I just hope the file system issues are gone in the Nexus S.
If the dual core LG star was going on sale next Thursday, I would still go for the Nexus S, without a doubt.
Click to expand...
Click to collapse
My thoughts exactly
Those high 20k plus quadrant scores are due to lagfix that tricks quadrant I/O score not actual performance increase
Sent from my SGH-T959 using XDA App
also keep in mind that that phone will probably come to the US 5-10 months from now and probably only be on one provider, and it won't be t-mobile, so if your a t-mobile fan, nexus s is still for you.
You can get a better real world estimate of galaxy s series without using ext2 loop. ie all ext4. Which gives 1400 - 1700 depending on setup. I saw where nexus s did quadrant with ext4 plus some other fs and was comparable: 1400–1500. I'm with above poster, can't wait to see Orion vs tegra2.
Sent from my SAMSUNG-SGH-I897 using XDA App
Not really impressive if I see the Neocore benchmark.
LG-Star = 67 fps
Nexus S = 55 fps (Engadget review)
My friends HTC Desire HD = 58 fps
My HTC Desire = 27 fps
Well, I would expect at least 1.5 times compared to single core, more than 80 fps.
The difference right now is only 12 fps average ...
But, it might change if the LG Star is also using Android 2.3, that might be more than 67 fps.
Still, 55 fps is more than enough for games !
Also for those who don't know...galaxy s phones gpu is capped at 56 fps seems to be same with NS..not that you really need more that 60fps but score could be much higher.
Sent from my SGH-T959 using XDA App
gogol said:
Not really impressive if I see the Neocore benchmark.
LG-Star = 67 fps
Nexus S = 55 fps (Engadget review)
My friends HTC Desire HD = 58 fps
My HTC Desire = 27 fps
Well, I would expect at least 1.5 times compared to single core, more than 80 fps.
The difference right now is only 12 fps average ...
But, it might change if the LG Star is also using Android 2.3, that might be more than 67 fps.
Still, 55 fps is more than enough for games !
Click to expand...
Click to collapse
Heh...my MT4 got 56FPS.
Not bad.
gogol said:
Not really impressive if I see the Neocore benchmark.
LG-Star = 67 fps
Nexus S = 55 fps (Engadget review)
My friends HTC Desire HD = 58 fps
My HTC Desire = 27 fps
Well, I would expect at least 1.5 times compared to single core, more than 80 fps.
The difference right now is only 12 fps average ...
But, it might change if the LG Star is also using Android 2.3, that might be more than 67 fps.
Still, 55 fps is more than enough for games !
Click to expand...
Click to collapse
Nexus S is capped to 55.55* fps, just like the other Galaxy S phones. The phone is maxing out the benchmark
The nenamark1 scores on Nexus S are higher than LG star.
Tegra2 ~ Hummingbird. About the same power.
Rawat said:
Nexus S is capped to 55.55* fps, just like the other Galaxy S phones. The phone is maxing out the benchmark
The nenamark1 scores on Nexus S are higher than LG star.
Click to expand...
Click to collapse
Nice try to out do me and talk over me but its 56 not 55.55 lol i just ran 55.8 on neocore
demo23019 said:
Nice try to out do me and talk over me but its 56 not 55.55 lol i just ran 55.8 on neocore
Click to expand...
Click to collapse
Haha it is definitely 56. Quadrant maxes out at 56 as does quake 3 arena.
demo23019 said:
Nice try to out do me and talk over me but its 56 not 55.55 lol i just ran 55.8 on neocore
Click to expand...
Click to collapse
Not trying to talk over you. It is capped to 55.555*. There's a parameter that can be changed when compiling the kenrel, and it'll be capped at 65fps. Neocore scores 65fps with a kernel that's compiled with the parameter
Anderdroid said:
Haha it is definitely 56. Quadrant maxes out at 56 as does quake 3 arena.
Click to expand...
Click to collapse
55.5 rounded up is 56. And 55.55 rounded up is 55.6, which is what my Galaxy S has scored since they day I got it.
EDIT: you can check build.prop, and it mentions (a rounded down) 55fps: windowsmgr.max_events_per_sec=55
this phne may be fast but it's ugly and it also has a skined version of android
The Nexus S does
both running 2.2 (epic on dk28)
epic gets 991-1000 and the evo just scored 1241... wtf?!
the evo isn't EXT4 is it?
the epic is currently RFS... I'm having problems going to EXT4
The Quadrant was made specifically for the Snapdragon process; which the Evo uses if i'm not mistaken.
Besides; they don't prove real world performance. I've gotten my hands on numerous evo's and all seemed to be 'laggy' they arn't nearly as responsive as my Epic.
Quadrant scores do not accurately represent Snapdragon vs the Galaxy S Hummingbird (or any other Cortex-A8 like those in Verizon's Droid lineup). Certain functions such as the Virtual Floating Point extension in Snapdragon allow for artificially inflated Linpack scores which do not represent real-world performance. Additionally, the RFS file system that Samsung uses on the Galaxy S phones is not well-understood by Quadrant, resulting in drastically lower scores.
For example, when we switch to EXT4 from RFS, our FroYo Quadrant scores can jump from 1100 to the 1600 range (which kicks the pants off that EVO). The actual performance increase however is hardly perceptible.
In short, Quadrant sucks, and Linpack is susceptible as well. For ****s and giggles, a little while back I made a minor modification to my Epic and produced a Quadrant score of 2597, which is currently listed as the 10th highest score on SmartphoneBenchmarks.com, and is the highest stock clock Quadrant score recorded. I accomplished this score utilizing a simple hack devised by a few developers here on XDA that fools the Quadrant application by utilizing a ramdrive for the I/O test. I posted about it on their forums and was advised by an administrator that they are aware of the problems with Quadrant and are releasing their own benchmark tool.
Android runs on top of a virtual machine so really what you're testing is virtual machine performance, and currently that virtual machine is tweaked for Snapdragon. With the Nexus S now a Google flagship phone, we'll likely see the VM better optimized for Hummingbird in the near future, and in fact, Gingerbread had a few more JIT enhancements as well.
Hummingbird is a slightly better performing chip MHz for MHz than the first-generation Snapdragons, if by a small amount. However it is significantly more power-efficient. Second-gen Snapdragons do draw even in terms of efficiency and performance, but both are going to be blown away by the Cortex-A9 Tegra phones we'll be seeing this spring.
EDIT - So I just noticed that SmartphoneBenchmarks.com has released their own benchmark tool recently, Smartbench 2010. I just ran it on my phone and scored 1178 on the Productivity Index (CPU) and 2610 on the Games Index (GPU). The highest-scoring competitor, the HTC G2, scores 1045 and 1396 respectfully. DRockstar on IRC ran the benchmark on his Epic that has RFS, and scored 1133 and 2521. So, this benchmark tool actually performs fine on RFS. Amazing! Grab it off the Android Market!
Electrofreak said:
Quadrant scores do not accurately represent Snapdragon vs the Galaxy S Hummingbird (or any other Cortex-A8 like those in Verizon's Droid lineup). Certain functions such as the Virtual Floating Point extension in Snapdragon allow for artificially inflated Linpack scores which do not represent real-world performance. Additionally, the RFS file system that Samsung uses on the Galaxy S phones is not well-understood by Quadrant, resulting in drastically lower scores.
For example, when we switch to EXT4 from RFS, our FroYo Quadrant scores can jump from 1100 to the 1600 range (which kicks the pants off that EVO). The actual performance increase however is hardly perceptible.
In short, Quadrant sucks, and Linpack is susceptible as well. For ****s and giggles, a little while back I made a minor modification to my Epic and produced a Quadrant score of 2597, which is currently listed as the 10th highest score on SmartphoneBenchmarks.com, and is the highest stock clock Quadrant score recorded. I accomplished this score utilizing a simple hack devised by a few developers here on XDA that fools the Quadrant application by utilizing a ramdrive for the I/O test. I posted about it on their forums and was advised by an administrator that they are aware of the problems with Quadrant and are releasing their own benchmark tool.
Android runs on top of a virtual machine so really what you're testing is virtual machine performance, and currently that virtual machine is tweaked for Snapdragon. With the Nexus S now a Google flagship phone, we'll likely see the VM better optimized for Hummingbird in the near future, and in fact, Gingerbread had a few more JIT enhancements as well.
Hummingbird is a slightly better performing chip MHz for MHz than the first-generation Snapdragons, if by a small amount. However it is significantly more power-efficient. Second-gen Snapdragons do draw even in terms of efficiency and performance, but both are going to be blown away by the Cortex-A9 Tegra phones we'll be seeing this spring.
EDIT - So I just noticed that SmartphoneBenchmarks.com has released their own benchmark tool recently, Smartbench 2010. I just ran it on my phone and scored 1178 on the Productivity Index (CPU) and 2610 on the Games Index (GPU). The highest-scoring competitor, the HTC G2, scores 1045 and 1396 respectfully. DRockstar on IRC ran the benchmark on his Epic that has RFS, and scored 1133 and 2521. So, this benchmark tool actually performs fine on RFS. Amazing! Grab it off the Android Market!
Click to expand...
Click to collapse
Electrofreak,
I just wanted to thank you (I pushed the button, too!) for this post. I'm trying to decide between three phones for my Sprint upgrade next month. My three candidates are the Epic, Evo, & new Evo Shift.
I was not aware of everything you stated, so it helped me look at the Epic in a different light.
Again, thanks.
tps70 said:
Electrofreak,
I just wanted to thank you (I pushed the button, too!) for this post. I'm trying to decide between three phones for my Sprint upgrade next month. My three candidates are the Epic, Evo, & new Evo Shift.
I was not aware of everything you stated, so it helped me look at the Epic in a different light.
Again, thanks.
Click to expand...
Click to collapse
No probs, and if you're interested in still more info, you're welcome to read an article I wrote comparing the hardware in multiple smartphones back in April (though the focus was on the EVO 4G and the Samsung Galaxy S I9000). The article is starting to get a little outdated, (neither the EVO nor the Galaxy S line had been released at that point yet) and it also doesn't cover some other details I've unearthed since then (my blog in my signature is where you'll find that) but most of it is still relevant.
http://alienbabeltech.com/main/?p=17125
Edit - I think I have an addiction to parenthesis (which I'm ashamed to admit)
Electrofreak said:
Quadrant scores do not accurately represent Snapdragon vs the Galaxy S Hummingbird (or any other Cortex-A8 like those in Verizon's Droid lineup). Certain functions such as the Virtual Floating Point extension in Snapdragon allow for artificially inflated Linpack scores which do not represent real-world performance. Additionally, the RFS file system that Samsung uses on the Galaxy S phones is not well-understood by Quadrant, resulting in drastically lower scores.
For example, when we switch to EXT4 from RFS, our FroYo Quadrant scores can jump from 1100 to the 1600 range (which kicks the pants off that EVO). The actual performance increase however is hardly perceptible.
In short, Quadrant sucks, and Linpack is susceptible as well. For ****s and giggles, a little while back I made a minor modification to my Epic and produced a Quadrant score of 2597, which is currently listed as the 10th highest score on SmartphoneBenchmarks.com, and is the highest stock clock Quadrant score recorded. I accomplished this score utilizing a simple hack devised by a few developers here on XDA that fools the Quadrant application by utilizing a ramdrive for the I/O test. I posted about it on their forums and was advised by an administrator that they are aware of the problems with Quadrant and are releasing their own benchmark tool.
Android runs on top of a virtual machine so really what you're testing is virtual machine performance, and currently that virtual machine is tweaked for Snapdragon. With the Nexus S now a Google flagship phone, we'll likely see the VM better optimized for Hummingbird in the near future, and in fact, Gingerbread had a few more JIT enhancements as well.
Hummingbird is a slightly better performing chip MHz for MHz than the first-generation Snapdragons, if by a small amount. However it is significantly more power-efficient. Second-gen Snapdragons do draw even in terms of efficiency and performance, but both are going to be blown away by the Cortex-A9 Tegra phones we'll be seeing this spring.
EDIT - So I just noticed that SmartphoneBenchmarks.com has released their own benchmark tool recently, Smartbench 2010. I just ran it on my phone and scored 1178 on the Productivity Index (CPU) and 2610 on the Games Index (GPU). The highest-scoring competitor, the HTC G2, scores 1045 and 1396 respectfully. DRockstar on IRC ran the benchmark on his Epic that has RFS, and scored 1133 and 2521. So, this benchmark tool actually performs fine on RFS. Amazing! Grab it off the Android Market!
Click to expand...
Click to collapse
That was great!
I scored 1147/2704 but i'm rooted/rommed.
the evo scored 700/910
razorseal said:
That was great!
I scored 1147/2704 but i'm rooted/rommed.
the evo scored 700/910
Click to expand...
Click to collapse
That EVO on 2.2? I would have expected it to score around 1000 at least. I wonder how it would score on EXT4 running CM6...
Smartphone benchmarks is a great benchmark.
Sent from my SPH-D700 using XDA App
Electrofreak said:
That EVO on 2.2? I would have expected it to score around 1000 at least. I wonder how it would score on EXT4 running CM6...
Click to expand...
Click to collapse
yup, it's a stock evo running whatever sprint updated for it
Scored 1257/2751 CM6 EXT4
Sent from my CM6 EXT4 Epic
1255p 2945g,Im running my ROM,how could I be faster then CM6? maybe not the best benchmark.
Sent from my SPH-D700 using XDA App
My Epic w/ Cmod's latest gave me 1115 and 2600
My wife has an Evo with CMods latest as well but only got 1089 and 1050, why so low on the second one?
My Epic scored 606\1901 in smartbench 2010. Weird... much lower productivity score than other people, but really high gaming score.
My Epic is stock.
EDIT: I ran it a few more times and watched it carefully.
603/1808
618/1954
633/1941
Seems I/O is pretty slow...
I'm just wondering why it matters? It's not like Android has a robust collection of high performance games.
razorseal said:
both running 2.2 (epic on dk28)
epic gets 991-1000 and the evo just scored 1241... wtf?!
the evo isn't EXT4 is it?
the epic is currently RFS... I'm having problems going to EXT4
Click to expand...
Click to collapse
Probably because she keeps her Evo in her bra and you keep your Epic in your pocket. It's a fact (check quadrant scores) that smartphones prefer boobs to guys hips 9 out of 10 days of the week. So obviously her Evo is happier and therefor performs better than yours. Do your Epic a favor and give him some booby time and watch those Quadrant scores rise!
+1,agreed and its been proven time and time again...
jirafabo said:
Probably because she keeps her Evo in her bra and you keep your Epic in your pocket. It's a fact (check quadrant scores) that smartphones prefer boobs to guys hips 9 out of 10 days of the week. So obviously her Evo is happier and therefor performs better than yours. Do your Epic a favor and give him some booby time and watch those Quadrant scores rise!
Click to expand...
Click to collapse
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
Hi all,
I can't find any benchmark about Nexus 10 on GeekBench. I need to compare it with Apple Ipad 4 CPU (already listed on geekbench website).
Someone could help me?
Where can I find a benchmark?
crashlab said:
Hi all,
I can't find any benchmark about Nexus 10 on GeekBench. I need to compare it with Apple Ipad 4 CPU (already listed on geekbench website).
Someone could help me?
Where can I find a benchmark?
Click to expand...
Click to collapse
It will smash the iPad 4. But no-one has had a chance to bench it yet.
Dual Exynos 5250 ARM-Cortex A15 CPU vs @1.7GHz
Dual Apple A6 ARM-Cortex A9 (with some architectural enhancements from A15) @ 1.3GHz
Mali-604 GPU which is 4-5x faster than previous generations Mali-400 GPU
PowerVR SGX543MP4 GPU which is great compared to Tegra and OMAP, but not compared to Mali-604
RAM is double on Nexus 10.
Geekbench on N10 is 2315 as per ZDNET. On N4, it's 1961.
I think they should improve further as Google clearly haven't optimized the drivers properly. A lot of work needs to be done from google to to optimize the devices. Expect to improve further in future.
Just got 2657 on geek bench on my nexus 10. It's rooted with a custom rom but not overclocked
I got only 1590 with geekbench on my CM11+KTmanta Nexus 10..
so how can you get 2700..?
rhyxos said:
I got only 1590 with geekbench on my CM11+KTmanta Nexus 10..
so how can you get 2700..?
Click to expand...
Click to collapse
Hi
Geekbench 3 scores shouldn't be compared with Geekbench 2 scores. AFAIK your score is related to GB3 while theirs (that are older) are related to GB2 .
~Lord
XxLordxX said:
Hi
Geekbench 3 scores shouldn't be compared with Geekbench 2 scores. AFAIK your score is related to GB3 while theirs (that are older) are related to GB2 .
~Lord
Click to expand...
Click to collapse
I thought my device had some power issues with that score (1500)