Develop Bugs

[Benchmark] 3D/OpenGL Performance Across ROMs [28/07/10]

Hi.
I'm have now rewritten the Spreadsheet and will shortly be integrating my CPU benchmarks and GPU benchmarks into the same sheet (not really that interesting or anything but it's there if you want to see exactly how things are worked out) but it explains why the progress has been slow recently. I have now taken this thread over from Arkymedes - and a HUGE thanks to him for creating the spreadsheet and doing most of the benchmarks which are still being referenced in them. He has done a brilliant job!
This page will continue to keep the tests strictly to GPU performance. CPU performance can be viewed here.
(like Arkymedes) I will take in consideration the clock used in each ROM and try to use the maximum allowed by the kernel using SetCPU. HOWEVER, all rom's will be tested at 691 MHz for consistency and any other frequencies will be detailed.
About JIT, unfortunately, all Arkmedes tests show that it doesn't affect 3D performance YET, since all this apps are not properly optimized and didn't take JIT into account. However, this may change with Froyo builds so keep an eye out.
All the ROM's clocked @ 691MHz "Performance" under SetCPU, unless specified otherwise. All tests are conducted 5 times and an average is taken. The tests are carried out in Aeroplane mode with no widgets running on the desktop. No apps are installed aside from the benchmark apps and Apps2SD is used (if enabled by the ROM). The latest version of the benchmark software is always used (they usually only add support for new phones in updates rather than changing benchmark parameters).
Benchmark apps:
Neocore - Qualcomm's own benchmark tool for 3D performance. Measured in FPS
Benchmark - Softweg's test suite for benchmarking. Only the GPU test done here to account for 2D performance. Measured by an internal score.
An3DBench - Tests based on jPCT, a free 3D engine for Java. 7 tests showing the capabilities of the engine. Measured by an internal score.
Note that all the tests mean: The higher the score, the better.
Anyway, here it goes
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}

A little suggestion: you should specify wether whe rom is sense or vanilla, and also the CPU clock used for the tests.

Can't wait!

that's a long couple of hours.......

Are you going to do VillainROM 10.2 with Sense + JIT enabled? If not I can try and do a benchmark.

Thread updated

Please Note:
I believe in the EzHero 2.7 stats, that jit is not enabled, with jit EzHero 2.7 is equal or greater than EzHero 2.0

maybe link each of the roms in the first thread for people to easily find...being quite new here i cant fins the Androbin-cr2-5.0.7...what ever that one is...

Your efforts are highly appreciated, good job.
But there is a problem: you didn't say which ROM is JIT enabled and which is not, because as you see CPU benchmarks have better results on JIT enabled ROMs, even though, as you can see, 3D is not affected.

My bench marks
My benchmarks with EzHero 2.7+ JIT:
Linpack- 4.886
Neocore w/ sound- 30.2FPS
Neocore w/o sound- 33.2FPS
Quadrant Standard- 280
Benchmark Pi- 7222 milliseconds
Very good job on the post, well done keeping us informed!

Hello!
What I'm trying to do here is to benchmark any version that I can find in these forums WITHOUT any further optmization. The benchmarkings are done OUT OF THE BOX
So, if I have to MANUALLY enable JIT or any OC AFTER the flashing, it will NOT show in these graphics!
I didn't do that because it is just impossible for me to keep track of every single optmization that can be done to every single one of this ROM's. I just don't have the time and resources for that.
So again, all this tests are done straight after the flashing of the downloaded ROM (logically, after some reboots and bla bla bla).
I really hope that you guys understand that. It's not a way to show which ROM is better, since there are MANY more factors that dictate that, like BT calls, GPS and so on. It's just a way to measure mostly 3D performance out of the box with the ROM's you provide.

sheridan2000 said:
maybe link each of the roms in the first thread for people to easily find...being quite new here i cant fins the Androbin-cr2-5.0.7...what ever that one is...
Click to expand...
Click to collapse
That's a good idea, but I guess the WIKI sticked in the first page should have done the job. It will be redundant but I can try to keep track of them.

Man sad day, all you have to do in EzHero to enable jit is go to the terminal app and type jit-on.
Still, job well done! Kudos!

xenoside said:
Man sad day, all you have to do in EzHero to enable jit is go to the terminal app and type jit-on.
Still, job well done! Kudos!
Click to expand...
Click to collapse
I will add EZHero 2.7 with JIT enabled after Italy x Paraguay

Arkymedes said:
I will add EZHero 2.7 with JIT enabled after Italy x Paraguay
Click to expand...
Click to collapse
World Cup always comes first my friend lol

its raining pretty hard in the game...
anyways, so 2.0 is JIT enabled and 2.7 is not in your graph? because huge difference in linpack scores with the older having a better score.
you should also mention Benchmark Pi - lower the score, the better.

-2y2k- said:
its raining pretty hard in the game...
anyways, so 2.0 is JIT enabled and 2.7 is not in your graph? because huge difference in linpack scores with the older having a better score.
you should also mention Benchmark Pi - lower the score, the better.
Click to expand...
Click to collapse
Yeah, 2.0 is jit enabled, but now 2.7 is optional due to complaints and split decisions. I posted the benchmarks with jit on the last post of the front page

im a Sense guy but its been awhile since I tried a non-sense one, your ROM is rrreally tempting right now lol

-2y2k- said:
im a Sense guy but its been awhile since I tried a non-sense one, your ROM is rrreally tempting right now lol
Click to expand...
Click to collapse
Well vanilla Eclair and EzHero are both great roms, heck even Androbin is tied with those two.

In the Quadrant Graph, androbin doesn't have his bar.

[Q] 30 FPS cap????

Does anyone have a work around for the 30 fps cap the Incredible has on the GPU?

What 30fps cap?
More like, it just doesn't have enough balls to hit 30+ fps, most the time

hahaha true true. I've read that HTC has capped the EVO and the Incredible at 30fps on the GPU.

No, just the Evo. Something to do with HDMI.

I would expect a higher quadrant score with the 1ghz processor the Droid X almost puts up the same numbers and it's running 2.1.

The GPU is making up for it. Quadrant is very GPU oriented.
That and the OMAP is faster. So its a lose-lose situation for us. The GPU in the X walks all over ours.
Kind of like integrated graphics vs a discrete graphics card in a computer.

Makes sense...too bad though I'm the type of guy that always has to have the latest and greatest don't get me wrong the incredible is a great phone can't beat the form factor but I wish it had a little more power under the hood in the area of graphics processing.

HeyItsLou said:
Makes sense...too bad though I'm the type of guy that always has to have the latest and greatest don't get me wrong the incredible is a great phone can't beat the form factor but I wish it had a little more power under the hood in the area of graphics processing.
Click to expand...
Click to collapse
I agree 100%.

adrynalyne said:
The GPU is making up for it. Quadrant is very GPU oriented.
That and the OMAP is faster. So its a lose-lose situation for us. The GPU in the X walks all over ours.
Kind of like integrated graphics vs a discrete graphics card in a computer.
Click to expand...
Click to collapse
It's pretty interesting actually. I thought the same thing, until I saw quadrant run on a droid X. It does get higher fps in most of the 3d tests, (the rotating planet/moon is really choppy and in the low teens of fps)
Anyway, because I had nothing better to spend my money on, I bought the advanced version of quadrant.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
The GPU does score a bit higher in 3d, though for whatever reason the I/O (read/write test) scores crazy high compared to anything else. That's where the most
of the score is coming from...
but on topic, there is no framerate cap. Download fps2d fro the android market and see for yourself.

It's interesting that you say that. I've been doing a lot of development work lately on the DInc and when transferring files I have been getting blazing speeds... much higher than on any other solid state/card reading device I own (why couldn't Nvidia stick a sexy little GPU in there... *sigh* just have to wait a couple years).

I think the lack of 3D performance on the Incredible is due to the drivers HTC includes for the Adreno GPU. I've read that the drivers they include aren't optimized or tweaked at all.
I remember there being a thread a while back about porting drivers from the Samsung devices using it since they were getting much better performance out of the same GPU. Anyone remember that or have any new info on that development?

I'm very disappointed with the Slackdragon chip. I hate the slow downs during game play. Asphalt consistently crashes on this phone.
The guys using the more than 3yr old HTC Vogue just got a new kernel and are enjoying over 30fps on Neocore. I have never reached 26fps. Granted they have DZO who's a brilliant programer but there is no reason why a phone that's years old and is not even an Android phone (at least not native) should be anywhere close to our phone. That phone came with a 400mhz chip and no drivers for the GPU.
I love the form factor and screen on this phone but I doubt any amount of overclocking will ever solve its problems. My next phone will definitely not have Qualcomm chip.
Sent from my ADR6300 using XDA App

I enjoy the phone for what it is.
Otherwise I would have returned it. If someone bought this phone thinking it was a gaming powerhouse...LOL?
Buyer beware, research first.
I will take a 1ghz CPU over a 550mhz CPU, even if it plays games better. I need it to be fast as a phone, not crappy slow except for games.
When Android starts being GPU driven, then I will pay more attention to faster gpus.
You can have the fastest GPU of all, and still run like crap. Ask Samsung Galaxy S owners who are suffering lag and slowness due to apps installing only to sd internal storage But hey, they run games fast!

Dude I didn't buy the phone to replace my xbox but it'd be nice if games didn't crash on my phone. On paper the snapdragon seems great but in all actuality the first Droid isn't far behind at all at half the speed and it out performs this phone when overclocked to 1.2ghz.
I still really like this phone especially now with Froyo but that doesn't mean I can't wish the snapdragon chip wasn't a slacker. I did my research and unless I went with the iphone (which I'd never get) I was not get better 3d performance...on paper. Even video slows down on this phone sometimes. That just doesn't seem right
Sent from my ADR6300 using XDA App

Take Froyo on both phones.
Run Linpack with Moto at 1.2ghz, and Dinc at 1ghz.
Chuckle at how much higher score the Dinc gets.
The game performance is about the same (with oc'ed Moto Droid taking the lead), but the rest is not even in the same ballpark as the Dinc.

It's likely capped at 30 fps to sync with display refresh rate (otherwise known as v sync ) and yes with modifications to the drivers it will be possible to increase this. But as for now I agree with adrenaline and enjoy the phone for what it is.
Don't get me wrong I'm always game for more performance but even at stock clock speeds the incredible is plenty fast. Faster gpu would be at the bottom of my list for tweaks to be made. Even so I'm sure someone will be working on it soon enough.
Sent from my ADR6300 using Tapatalk

Odd CPU scaling behaviour

I've been playing around with TegraStats for a while now and found out, that the CPU's up- and downscaling acts kinda weird from time to time.
For example, running CPU intensive tasks like Antutu benchmark will fire up all four cores, but maximum clock was always 1200 MHz instead of 1400 MHz. Seems kinda "capped" through kernel parameters (correct me if I'm wrong).
So I've tried Shadowgun after benchmarking. In this case, the CPU scaled down to around 360 MHz from time to time, resulting in heavy stuttering during gameplay in huge environments. That's what really bothered me, 'cause same happened with Riptide GP from time to time, just random encounters.
So I installed good old ROM Toolbox (very nice tool by the way) and enforced Performance governor.
Result of this was that cores 0 and 1 scaled very nicely during gameplay (Shadowgun), but cores 2 and 3 were shut down permanently. Really weird stuff, but the game ran much, much better with just those 2 very stable scaling cores!
My best guess is that HTC really messed up the default governor, I hope they'll release kernel sources soon, so our kernel devs can take a look at this.

more2come said:
My best guess is that HTC really messed up the default governor, I hope they'll release kernel sources soon, so our kernel devs can take a look at this.
Click to expand...
Click to collapse
Nvidia's low-level code and drivers are proprietary and have been since Teg2. So there's only so much the devs (and HTC) can do.
What you're describing is exactly what happened to the Asus Prime when the first few updates were released. Clock speeds were locked (that hadn't been) and the way the cores worked together was modified and in the process both performance and battery life were lost. It appears that in real world use the way the processors step is very different than Nvidia's marketing material.
I'd be willing to bet you that you'll observe different results using different updates. Code and drivers are going to be modified to fix the graphics issues and improve battery life and those changes are going to alter the way the processor taps and manages the different cores.
Does your phone flicker? It would be interesting to see the results of TegraStats on flickering vs. non-flickering phones. It's a shame you have to be rooted to use it because that limits the number of people who can post results.
Is there some sort of test you can come up with that people can use to compare different phones to see if the results are different? I'm not rooted or I'd help.

I never experienced any flickering so far, seems to be just fine!
Well, easiest part to reproduce is test 1, I guess. Your phone needs to be rooted, you need TegraStats from the Play Store and Antutu Benchmark.
During CPU benchmarking, all four cores are at 100% load, but they're stuck at 1.2 GHz.
Another thing that bothers me is the browser performance with web pages that include alot of different content, like the desktop page of engadget.com .
The One X lags alot during pinch to zoom, I think it's because of text reflow and constant repositioning of pictures and flash content. But with performance governor, the browser runs like hell!
I'll keep playing around with some settings, will post them here if something useful comes out.

What sort of Quadrant/ Antutu scores are you getting with performance governer?

When I test my phone they stay at 1.4 but do go to 1.5 in games but not in benchmarks, maybe the CPU has a turbo boost of some sort.
Sent from my HTC One X using xda premium

Cpu also spikes when you scroll between screens. Largely due to the carousel effect on Sense. Spikes too much imo. They need to optimize Sense for the Tegra 3.
Sent from my HTC One X using xda premium

I read that 1.5Ghz was the fastest single core speed and for multi cores the fastest is 1.4Ghz. So a benchmark using all cores will hit 1.4Ghz but a game running on a single core will hit 1.5Ghz...
"Up to Single Core 1.5 GHz /Quad-Core 1.4 GHz"
Has anyone been able to test this with the One X yet? I know it works with my Transformer Prime
"3D Stereo Gaming - This capability leverages NVIDIA’s award-winning NVIDIA 3D Vision® technology to automatically, and in real-time, convert OpenGL based games and apps into Stereo 3D."

Tried setting up Performance and my system became somewhat unstable.
Running ARHD 5.0 @ Firmware 1.29
Browser didnt rly improve in performance, too. Compared the desktop site of engadget with interactive and performance. Both performed pretty poor here.
Plus: With performance I get 1400Mhz constantly, and my phone heated up pretty fast pretty bad. In regular homescreen use I easily reached 40 degrees whereas with interactive I usually reach tops of 32-33 degrees....

TremF said:
I read that 1.5Ghz was the fastest single core speed and for multi cores the fastest is 1.4Ghz. So a benchmark using all cores will hit 1.4Ghz but a game running on a single core will hit 1.5Ghz...
"Up to Single Core 1.5 GHz /Quad-Core 1.4 GHz"
Has anyone been able to test this with the One X yet? I know it works with my Transformer Prime
"3D Stereo Gaming - This capability leverages NVIDIA’s award-winning NVIDIA 3D Vision® technology to automatically, and in real-time, convert OpenGL based games and apps into Stereo 3D."
Click to expand...
Click to collapse
just tested for you
1) all 4 cores on during benchmark
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
2) 2 cores disabled (dual core)
as you can see with 4 cores the max was 1.2GHZ where as on only 2 cores the max was 1.4GHZ
even with setting the max cores to 1 it still used 2 cores so i concluded it wouldn't be a fair test.

I wonder if HTC have messed with the CPU working somehow? Nvidia def state "Up to Single Core 1.5 GHz /Quad-Core 1.4 GHz" here.
and the test I was asking about was the 3D Stereo Gaming
I like how with my Prime there's 3 diff settings. Powersave - 1.0Ghz, Standard 1.2Ghz and Full speed (not the actual names but I don't have my Prime in front of me) 1.4Ghz so you can set it depending on what you are doing. Plus with root and overclock upto 1.6Ghz or 1.7/8Ghz with kernal change/custom rom.

TremF said:
I wonder if HTC have messed with the CPU working somehow? Nvidia def state "Up to Single Core 1.5 GHz /Quad-Core 1.4 GHz" here.
and the test I was asking about was the 3D Stereo Gaming
Click to expand...
Click to collapse
still going to count the tests as valid lol, on 4 cores my phone only got to 1.2 GHz where you posted it should be 1.4 which was only the case on 2 cores.

This is going to work the same way it works for "Regular" computer CPUs.
Core i7 scales frequencies (and does turboboost) depending on the overral load. You get the highest boost with just one core, a little lower with two, and very little with all of them (this depends on package heat dissipation and is usually set by BIOS in laptops - not sure about desktops). For computers, there is software called Throttlestop which allows you to tune some of these parameters, not sure what can be done on Tegra3.

treebill said:
still going to count the tests as valid lol, on 4 cores my phone only got to 1.2 GHz where you posted it should be 1.4 which was only the case on 2 cores.
Click to expand...
Click to collapse
Aye. Can't argue with proper test results. I think for games I'll stick to my Prime with 1.6Ghz max, less heat and more battery

Overclock - 1802mhz CPU, 512mhz GPU

I ran Quadrant Pro at 1802mhz on the CPU and 512mhz on the GPU (Franco Kernel & Jelly Belly ROM)
This was the result.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}

Did your phone explode? Because I don't see a picture.

And you posted this here..... Why?

Slemhelge said:
I ran Quadrant Pro at 1802mhz on the CPU and 512mhz on the GPU (Franco Kernel & Jelly Belly ROM)
This was the result.
Click to expand...
Click to collapse
thats lousy! id be embarrassed to show that off to people. i suggest a kernel change. anyways, that post does not belong here in development, nor anywhere near development heres trinity at 1728mhz..

simms22 said:
thats lousy! id be embarrassed to show that off to people. i suggest a kernel change heres trinity at 1728mhz..
Click to expand...
Click to collapse
LMAO! Damn you simms

can't beat simms at his own game =P

First off, that's embarrassingly low. I've gotten 4,000+ at 1.6GHz. Second why the hell did you post that in this section? I thought you needed 10 posts to post in a dev forum. Mods please close and OP read the damn rules.

irizwan said:
can't beat simms at his own game =P
Click to expand...
Click to collapse
lol, unfortunately not true though. there are a few that know how.
i just think that anybody running at 1802mhz should know how to propery use it or not be allowed to run that high of an oc. i mean, the op isnt getting any benefit from it unless you call a higher rate of battery loss a benefit. learn how to benefit from a high oc like this, or dont use it. its pointless to show off that youre using a high oc if a slightly oc'd(1344mhz kernel) can run laps around your phone. dont get me wrong, i know that franco kernel is capable, but this does it no justice.

simms22 said:
thats lousy! id be embarrassed to show that off to people. i suggest a kernel change. anyways, that post does not belong here in development, nor anywhere near development heres trinity at 1728mhz..
Click to expand...
Click to collapse
trinity FTW!

Then someone do tell me a good kernel and how to do a proper OC thats stable, efficient etc etc.. Im **** at OC'ing.. i just put franco kernel 512mhz and was allowed the option to go 1802mhz, which ive never seen before..

What development are you doing here?
None? Then why is it posted in development?!!
Thread moved
FNM

Slemhelge said:
Then someone do tell me a good kernel and how to do a proper OC thats stable, efficient etc etc.. Im **** at OC'ing.. i just put franco kernel 512mhz and was allowed the option to go 1802mhz, which ive never seen before..
Click to expand...
Click to collapse
Franco Kernel is a great kernel, yo must be doing wrong. I have achieved that score with franco's at 1.3 OC and Stock JB ROM...

Slemhelge said:
Then someone do tell me a good kernel and how to do a proper OC thats stable, efficient etc etc.. Im **** at OC'ing.. i just put franco kernel 512mhz and was allowed the option to go 1802mhz, which ive never seen before..
Click to expand...
Click to collapse
try them all, or a few different kernels at least. every phone is different, and will react differently to different kernels. also, just because you can oc to 1802, doesnt mean that your phone will properly handle it. at this point, it looks like you arent getting any of the gain from this high oc(some loss actually), but are going through your battery like you are. first off, at that high oc, turn off thermal throttling if you are going to benchmark. play with adjusting voltages until you find your sweet spot at that oc. try ocing less too, your phone might not like 1802, or 1700, or 1600.. not many phones do. also, when you benchmark, benchmark with your hiest cpu speed/highest cpu speed(1802/1802) so that you dont scale and actually bench 1802mhz.

simms22 said:
turn off thermal throttling if you are going to benchmark.
Click to expand...
Click to collapse
This is the key point that I believe the OP forgot to do.

IAmAN00bie said:
This is the key point that I believe the OP forgot to do.
Click to expand...
Click to collapse
very likely youre right.

Not even worth overclocking honestly.
Only for benchmark scores, your phone most likely doesn't run that much faster.

I Am Marino said:
Not even worth overclocking honestly.
Only for benchmark scores, your phone most likely doesn't run that much faster.
Click to expand...
Click to collapse
funny, only people that dont overclock say that, i guess they really do believe it for themselves. but in reality, there is a difference. if its a slight overclock, its not that different. when its a big overclock, the difference can be dramatic.

simms22 said:
funny, only people that dont overclock say that, i guess they really do believe it for themselves. but in reality, there is a difference. if its a slight overclock, its not that different. when its a big overclock, the difference can be dramatic.
Click to expand...
Click to collapse
I have overclocked before, I've done it before on the GNex as well.
I mean it was a huge increase of ability to a single core phone and early dual cores but for a phone already seemingly perfect on the software side, the phone already screams as stock clock speeds.
Besides, is it worth the probable loss of battery life? A marginally higher benchmark score? Things opening 1 or 2ms faster?
This is all just my opinion, I have no hard facts behind all this, but at least in my own personal testing and experimenting, there was no obvious benefit.
Provided you might be right, I've never OC'ed past 1.5, but beyond that, what's the effect on the health of the hardware, and like I said, battery life?
I could be totally wrong and I'm not against admitting being wrong.
Just personal preference I guess.

I Am Marino said:
I have overclocked before, I've done it before on the GNex as well.
I mean it was a huge increase of ability to a single core phone and early dual cores but for a phone already seemingly perfect on the software side, the phone already screams as stock clock speeds.
Besides, is it worth the probable loss of battery life? A marginally higher benchmark score? Things opening 1 or 2ms faster?
This is all just my opinion, I have no hard facts behind all this, but at least in my own personal testing and experimenting, there was no obvious benefit.
Provided you might be right, I've never OC'ed past 1.5, but beyond that, what's the effect on the health of the hardware, and like I said, battery life?
I could be totally wrong and I'm not against admitting being wrong.
Just personal preference I guess.
Click to expand...
Click to collapse
agreed. i myself use stock frequencies on franco kernel i've used almost all kernels and roms possible but now i've stuck with aokp n franco smooth even when under-clocked to 1 Ghz. i find no point of oc'ing other than just benchmarking usually get 3.4-3.3K on quadrant with normal freq.

simms22 said:
funny, only people that dont overclock say that, i guess they really do believe it for themselves. but in reality, there is a difference. if its a slight overclock, its not that different. when its a big overclock, the difference can be dramatic.
Click to expand...
Click to collapse
Your phone spends most of its life in the 350/700mhz range.
Signed,
An Overclocker

How on earth is this possible?

I have an XPERIA Arc LT15i. Single-core 1-GHz processor, 512 MB of RAM (much less available to the user, though). I've tried different ROMs, and the highest that I'd ever got on the Antutu benchmark was somewhere in the 5000s. Usually, though, it hovers between 4400 and 4800-ish. I tried a CM10-based MIUI ROM a while back, however, and the Antutu scores were completely off the charts.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
I'm wondering how it's possible for a 1-GHz single-core processor to match a much superior (dual-core) processor (the Optimus 2X) in terms of performance. Or are these benchmarks, which the myriads of review sites out there use to make comparisons between phones, actually rubbish? I'd love to have some clarifications/opinions.

On arc s overclocked to 1.6 running CM10 and a Fusion kernel I'm getting 6250 without rebooting before benchmark and with a few processes running in a background.
edit
I'll try with stock 1.4 and report results.
edit
6148 on 1.4ghz. Also keep in mind that anyone can install a GB rom with kernel that supports overclocking up to 2.0ghz to get even bigger score. Arc s is still a pretty capable device.
edit
Oh you were talking about arc not arc s xD But my point is still true because there is only 400mhz of difference between these phones.

NEVER use Antutu or any other benchmark tools to determine the performance. Benchmark score can easily get cheated by overclocking, increasing memory read ahead or using different IO scheduler
____________________________________________________________________________________________________________
if I helped press thanks :good::good::good:

NEVER use Antutu or any other benchmark tools to determine the performance. Benchmark score can easily get cheated by overclocking, increasing memory read ahead or using different IO scheduler
Click to expand...
Click to collapse
But its not cheating... I agree that benchmarks are not really that accurate and every score should be taken with a grain of salt but overclocking is not cheating. Cheating would be making a modifications to antutu apk for bigger scores.

Thank you for your opinions. However, I've never overclocked the processor on my phone to this date. The mere thought gives me the jitters. What if it burns out? Going by those scores, maybe the ROM overclocked the processor itself without my knowing. Have any of you tried overclocking, though? And, have you noticed any difference in performance or stability issues? Maybe the battery would drain out quicker, too.

MarcSinger said:
Thank you for your opinions. However, I've never overclocked the processor on my phone to this date. The mere thought gives me the jitters. What if it burns out? Going by those scores, maybe the ROM overclocked the processor itself without my knowing. Have any of you tried overclocking, though? And, have you noticed any difference in performance or stability issues? Maybe the battery would drain out quicker, too.
Click to expand...
Click to collapse
If you have read my previous post you would know everything that you asked. On CM10 I'm getting more than 6000 points in antutu without any overclocking (arc s CPU's frequency is 1.4ghz) : D Its not that scores on CM10 are high its just that scores on stock firmware are low. Sony's software is bloated and ****ty in overall.
Overclocking is safe if you are doing it with brain. If you set Arc CPU (1ghz) to 2ghz it will fry for sure but setting it to 1.2 should be perfectly safe. Overclocked phone drains battery faster but its pretty obvious everything is stable and a little smoother.

pypcier said:
If you have read my previous post you would know everything that you asked. On CM10 I'm getting more than 6000 points in antutu without any overclocking (arc s CPU's frequency is 1.4ghz) : D Its not that scores on CM10 are high its just that scores on stock firmware are low. Sony's software is bloated and ****ty in overall.
Overclocking is safe if you are doing it with brain. If you set Arc CPU (1ghz) to 2ghz it will fry for sure but setting it to 1.2 should be perfectly safe. Overclocked phone drains battery faster but its pretty obvious everything is stable and a little smoother.
Click to expand...
Click to collapse
No, I would NOT have known everything that I'd asked. My question was whether there were any stability or battery-drain issues with overclocking, but none of those queries were addressed on your previous post.
But thanks for the clarifications on the scores. Yes, stock ICS is absolute rubbish, that's why I've moved on to CM10/Jelly Bean. It doesn't have the "customizations" you can get with stock ICS, but it's smooth, responsive, fluid, and that's what matters to me.
By the way, I do believe 400 MHz SHOULD make quite a bit of a difference. A 600-MHz processor and a 1-GHz processor wouldn't be giving a similar performance. Correct me if I'm wrong.

Overclocking your Arc to 1.4GHz should be a clearly visible upgrade but I'm not sure if this would be healthy for your phone. It's almost 50% more in performance but under heavy load this could cause some high temperatures and degradation of your phone's battery and cpu life. My Arc s is clocked at 1.6 all the time and it doesn't even gets too hot while playing games so 200MHz more seems right and safe to me.
edit
Values described in this post as "safe" or "best" may not be the same for you so google everything before making any changes to your phone.