Is this a good benchmark ? - Nexus S Q&A, Help & Troubleshooting

{
"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"
}

yes its good. custom kernels will increase it (3000+) when set to performance, but as an everyday driver giving good battery life, i consider anything over 2000 to be just fine.

Benchmarks mean nothing. They dont reflect real time performance
Sent from my Transformer TF101 using XDA Premium App

zephiK said:
Benchmarks mean nothing. They dont reflect real time performance
Sent from my Transformer TF101 using XDA Premium App
Click to expand...
Click to collapse
i wouldn't say they mean NOTHING. I mean, i'm getting higher scores on this device and i CAN feel the difference in its performance compared to others that score under 2000. They may not be accurate representations, especially if a file was manipulated to show a faux score, but if you know it hasn't been manipulated you can judge your score appropriately.

itsallsubliminal said:
i wouldn't say they mean NOTHING. I mean, i'm getting higher scores on this device and i CAN feel the difference in its performance compared to others that score under 2000. They may not be accurate representations, especially if a file was manipulated to show a faux score, but if you know it hasn't been manipulated you can judge your score appropriately.
Click to expand...
Click to collapse
So let's get a phone by Motorola featuring Motoblur such as the Atrix 4G and run Quadrant on that. Quadrant means absolutely nothing, many developers have degraded the importance of these benchmark tests.
All you're simply feeling is a placebo effect.

zephiK said:
So let's get a phone by Motorola featuring Motoblur such as the Atrix 4G and run Quadrant on that. Quadrant means absolutely nothing, many developers have degraded the importance of these benchmark tests.
All you're simply feeling is a placebo effect.
Click to expand...
Click to collapse
Just noticed it was u I was talking with in the other topic. I see why u think I was trying to contradict u lol.
I suppose a better benchmark app will need to surface because the current ones are easily manipulated. IMO The benchmarks were more relevant on older phones not as powerful.
Sent from my Nexus S 4G using XDA Premium App

itsallsubliminal said:
Just noticed it was u I was talking with in the other topic. I see why u think I was trying to contradict u lol.
I suppose a better benchmark app will need to surface because the current ones are easily manipulated. IMO The benchmarks were more relevant on older phones not as powerful.
Sent from my Nexus S 4G using XDA Premium App
Click to expand...
Click to collapse
Lol yeah. Other benchmark apps are:
SmartBench 2010, SmartBench 2011 (2010 is more featured towards Single Core CPUs and 2011 towards Dual Core) and Linpack.
The reason why I hate Benchmarks so much is because Samsung-based phones tend to have lower linpacks whereas other manufacturers have 40+ linpack and Samsung phones have <30.
Hell, Quadrant Standard hasn't even updated their list anymore. It's the same thing everytime. HTC Magic is still listed there.

Lol my girlfriend LOVED the white magic. We r finally getting big boys toys
you're right, qualcomms chips are snappier. I like graphics to make a UI more pleasing so I enjoy Samsung and the obundance of phones they've been releasing. I'm glad HTC is starting to implement the Adreno GPU tho. I've been excited for NVIDIA since back when it was the Tegra 650 prototype, and now they are really stepping up the competition. Quadrant Standard doesn't play nice with the tegra 2 and seriously needs to be updated, I agree
zephiK said:
Lol yeah. Other benchmark apps are:
SmartBench 2010, SmartBench 2011 (2010 is more featured towards Single Core CPUs and 2011 towards Dual Core) and Linpack.
The reason why I hate Benchmarks so much is because Samsung-based phones tend to have lower linpacks whereas other manufacturers have 40+ linpack and Samsung phones have <30.
Hell, Quadrant Standard hasn't even updated their list anymore. It's the same thing everytime. HTC Magic is still listed there.
Click to expand...
Click to collapse
Sent from my Nexus S 4G using XDA Premium App

Yeah I think its just a matter of fact that in the old days, the only CPU choice was Qualcomm CPUs. Now theres all types of CPus, NVIDIA Tegra, Qualcomm, Samsung's Hummingbirds and now Intel & AMD are joining in.
With all these CPUs, the benchmarks are only weighing on certain CPUs and optimizations. So some CPUs will get the benefit. I have a TF101 tablet and it scores lower than my Nexus S but the TF101 is running SMOOTH and snappy. It's purely on stock right now, just playing around with Honeycomb but I don't know if I want to root it just yet. Root on 3.X doesn't seem to offer me much but it'd be nice to get my game saves from Angry Birds translated over.
Edit: Just did a benchmark on Quadrant
NS: 2067 (1.0 GHz Single Core - ROM/kernel below)
TF101: 1492 (1.0 GHz Tegra 2- 100% stock)
doesn't prove much though. both devices are VERY snappy and responsive. Now if I try Linpack, it'll tell another side of the story.
NS MFLOPS: 10.963
TF101 MFLOPS: not even gonna try.

GeNeTiiCs said:
Click to expand...
Click to collapse
did you mode it?
my benchmark was 1423, but i did not mod my phone, thought i ask perhaps something wrong with my phone

nixguy said:
did you mode it?
my benchmark was 1423, but i did not mod my phone, thought i ask perhaps something wrong with my phone
Click to expand...
Click to collapse
It's because kernels add "ext4 optimization" onto it. It is a real optimization. Feature of most (if not all) kernels.

zephiK said:
Yeah I think its just a matter of fact that in the old days, the only CPU choice was Qualcomm CPUs. Now theres all types of CPus, NVIDIA Tegra, Qualcomm, Samsung's Hummingbirds and now Intel & AMD are joining in.
With all these CPUs, the benchmarks are only weighing on certain CPUs and optimizations. So some CPUs will get the benefit. I have a TF101 tablet and it scores lower than my Nexus S but the TF101 is running SMOOTH and snappy. It's purely on stock right now, just playing around with Honeycomb but I don't know if I want to root it just yet. Root on 3.X doesn't seem to offer me much but it'd be nice to get my game saves from Angry Birds translated over.
Edit: Just did a benchmark on Quadrant
NS: 2067 (1.0 GHz Single Core - ROM/kernel below)
TF101: 1492 (1.0 GHz Tegra 2- 100% stock)
doesn't prove much though. both devices are VERY snappy and responsive. Now if I try Linpack, it'll tell another side of the story.
NS MFLOPS: 10.963
TF101 MFLOPS: not even gonna try.
Click to expand...
Click to collapse
I had the viewsonic GTablet (underrated for its price for sure) and like you said, it would run buttery smooth but scored poorly. Graphics were almost 70fps with neocore tho, that was exciting
i haven't messed with linpack since getting a samsung device 6 months ago, mainly because i know the scores are junk compared to my old Evo and would be pointless to compare.
But for some reason Linpack is not showing up in the market for me now. odd

zephiK said:
It's because kernels add "ext4 optimization" onto it. It is a real optimization. Feature of most (if not all) kernels.
Click to expand...
Click to collapse
no its not its cheating
some of you guys a are ridicules when its comes to this stupid benchmark
not to mention he give no indication on his CPU speed.....his score is all green we cant see **** per hardware score

The I/O "optimization" is like cache...same with a hard drive running on a dedication high end controller card with cache memory on the controller. If you enable cache in the controllers bios it will give out inaccurate scores when benchmarking the hard-drive read/write speeds it will double even tripple score.
Its the same **** with what you guys are doing with I/O optimization.
Someone please chime in if you know what im talking about ...another PC hardware guru benchmarking enthusiast.
Benchmarking.... you want to compare your HARDWARE with others you're suppose to run software standards with absolutely no software tweaks only hardware tweaks when benchmarking

itsallsubliminal said:
I had the viewsonic GTablet (underrated for its price for sure) and like you said, it would run buttery smooth but scored poorly. Graphics were almost 70fps with neocore tho, that was exciting
i haven't messed with linpack since getting a samsung device 6 months ago, mainly because i know the scores are junk compared to my old Evo and would be pointless to compare.
But for some reason Linpack is not showing up in the market for me now. odd
Click to expand...
Click to collapse
Yeah certain apps don't appear on the market when using a tablet.
@demo: I never said its cheating. There's a cheating script added in some kernels, but the ext4 optimization found by supercurio implemented by many devs is INDEED legit. Since everybody uses Quadrant Standard, you cant tell if its cheating or not because the Quadrant Advanced goes more in depth listing it by each factor. But even then, the quadrant benchmark is still useless for the reasons I listed.
And I honestly don't know the purpose of you're post if you're trying to attempt to contradict my post or not. My saying is that I find benchmarks to be pointless and everybody makes it seem like it's a bigger deal than what it really is.

zephiK said:
Yeah certain apps don't appear on the market when using a tablet.
Click to expand...
Click to collapse
I'm talking about my NS4G tho

zephiK said:
Yeah certain apps don't appear on the market when using a tablet.
@demo: I never said its cheating. There's a cheating script added in some kernels, but the ext4 optimization found by supercurio implemented by many devs is INDEED legit. Since everybody uses Quadrant Standard, you cant tell if its cheating or not because the Quadrant Advanced goes more in depth listing it by each factor. But even then, the quadrant benchmark is still useless for the reasons I listed.
And I honestly don't know the purpose of you're post if you're trying to attempt to contradict my post or not. My saying is that I find benchmarks to be pointless and everybody makes it seem like it's a bigger deal than what it really is.
Click to expand...
Click to collapse
no my whole posts werer not aimed at you just the cheating part
demo23019 said:
no its not its cheating
Click to expand...
Click to collapse

demo23019 said:
no my whole posts werer not aimed at you just the cheating part
Click to expand...
Click to collapse
Ahhh.. yeah, well I think it was Cyanogen that started it making his over 3000 to degrade the importance of benchmarks.

people want true hardware scores you're not suppose to optimize anything in software its not fair and just makes everything confusing...having one standard and sticking to it will yield better results for comparing with other devices.
unfortunately this isn't the case..devs are implementing ext4 tweaks with kernels
we need a true benchmarking program that only test hardware

itsallsubliminal said:
I'm talking about my NS4G tho
Click to expand...
Click to collapse
Ah.. try market enabler then. Linpack appears on my NS
Or try installing it from http://market.android.com/

Related

Quadrant Results

Hey what are your quadrant results? Stock Rom, Stock Kernel I was doing 1600-1800.....I just flashed the Adrynalyne's Overclocked 1.4ghz (b) kernel..and now I am a solid 2200ish with all governers. I am still running the stock rom with all verizon bloat.
Is 2200 a good score for a Stock Rom w/ Adrynalyne's kernel o.c to 1.4?
and what are your scores with your custom roms/kernel's?
BAMF Remix 1.4 at stock=1436
at 1.4Ghz= 2225
at 1.4Ghz with quadrant cheat on=2853
2.5k.
1.4 ghz, and no cheats, and bmaf 1.4.
Sent from my ADR6400L using XDA App
I'm using the Drod Kernel v1.5 and at 1.8Ghz (max) I'm getting around 2800 quadrant scores (no cheat).
I run the kernel at 1.5 normally (with great battery life I might ad) and get around 2400 (again, no cheat).
monkey1911 said:
BAMF Remix 1.4 at stock=1436
at 1.4Ghz= 2225
at 1.4Ghz with quadrant cheat on=2853
Click to expand...
Click to collapse
What's the point of the quadrant cheat except for meaningless bragging rights?
One of the blogs posted Quadrant scores on one of the new dual core phones (might have been Sensation) yesterday. Damn thing got 3300! Hard to believe on a stock anything.
Everyone uses Quadrant, but it's flawed and easily cheated; in one of the other phone sections I read here people are pretty passionate about not using it.
My TBolt 100% stock (rooted) only gets 1666 (evil number) on Quadrant, vs 2100 or more on my overclocked Galaxy S, but in daily use, the TBolt seems a lot snappier, even with sense. That alone tells me that Quadrant doesn't really provide real-world useful information.
distortedloop said:
What's the point of the quadrant cheat except for meaningless bragging rights?
One of the blogs posted Quadrant scores on one of the new dual core phones (might have been Sensation) yesterday. Damn thing got 3300! Hard to believe on a stock anything.
Everyone uses Quadrant, but it's flawed and easily cheated; in one of the other phone sections I read here people are pretty passionate about not using it.
My TBolt 100% stock (rooted) only gets 1666 (evil number) on Quadrant, vs 2100 or more on my overclocked Galaxy S, but in daily use, the TBolt seems a lot snappier, even with sense. That alone tells me that Quadrant doesn't really provide real-world useful information.
Click to expand...
Click to collapse
That's why i listed both scores, if i wanted to be a turd i could have just claimed the high and called it a day.
Sent from Thor's hammer on a bolt of lightning.
distortedloop said:
What's the point of the quadrant cheat except for meaningless bragging rights?
One of the blogs posted Quadrant scores on one of the new dual core phones (might have been Sensation) yesterday. Damn thing got 3300! Hard to believe on a stock anything.
Everyone uses Quadrant, but it's flawed and easily cheated; in one of the other phone sections I read here people are pretty passionate about not using it.
My TBolt 100% stock (rooted) only gets 1666 (evil number) on Quadrant, vs 2100 or more on my overclocked Galaxy S, but in daily use, the TBolt seems a lot snappier, even with sense. That alone tells me that Quadrant doesn't really provide real-world useful information.
Click to expand...
Click to collapse
Quadrant is not doubt flawed, however since it been around so long, it provides a decent comparison across different phones/platforms/roms etc. I agree that at the end of the day what matters is how the phone feels.
On my tbolt OC'd to 1.92ghz (no cheat) using Imosyeon kernel I get 3123 (no cheat). On my Nexus S OC'd to 1.4ghz (no cheat) I get 3950.
That being said, my Nexus S definitely feels faster to me. Maybe it's because it's on Gingerbread while tbolt is on 2.2. Don't really know but in this case scores match up with perception.
Just curious, but how does your Nexus S score on linpack?
jlevy73 said:
Quadrant is not doubt flawed, however since it been around so long, it provides a decent comparison across different phones/platforms/roms etc. I agree that at the end of the day what matters is how the phone feels.
On my tbolt OC'd to 1.92ghz (no cheat) using Imosyeon kernel I get 3123 (no cheat). On my Nexus S OC'd to 1.4ghz (no cheat) I get 3950.
That being said, my Nexus S definitely feels faster to me. Maybe it's because it's on Gingerbread while tbolt is on 2.2. Don't really know but in this case scores match up with perception.
Click to expand...
Click to collapse
asenduk said:
Just curious, but how does your Nexus S score on linpack?
Click to expand...
Click to collapse
Like crap lol, ~20mflops
wow, really?
wonder why..
jlevy73 said:
Like crap lol, ~20mflops
Click to expand...
Click to collapse
asenduk said:
wow, really?
wonder why..
Click to expand...
Click to collapse
Not sure, could be that Linpack is optimized for the snapdragon because on every other cpu bench like smartbench or cpu bench, etc, the Nexus S scores significantly higher.
128 bit FPU vs. a 64 bit one in non snap CPUs. Not sure about Tegra 2.
adrynalyne said:
128 bit FPU vs. a 64 bit one in non snap CPUs. Not sure about Tegra 2.
Click to expand...
Click to collapse
With Tegra 2, not sure about the specs, but it lags more than one would expect (especially on the 2X). Homescreen requires multiple swipes at times to register. Where it really excels is of course gaming. There's obviously more to a device than gaming hence why the bolt is a beast. It's like walking around with a mini computer
jlevy73 said:
With Tegra 2, not sure about the specs, but it lags more than one would expect (especially on the 2X). Homescreen requires multiple swipes at times to register. Where it really excels is of course gaming. There's obviously more to a device than gaming hence why the bolt is a beast. It's like walking around with a mini computer
Click to expand...
Click to collapse
Really not fair to judge it when its running Moto's Failed Blur software.
AFAIK, LG's offerings are a ton faster and smoother than the Atrix.
asenduk said:
wow, really?
wonder why..
Click to expand...
Click to collapse
because the snap dragon chipset kicks the hummingbird's butt at certain types of computational tasks. example the old nexus one gets a 38mflops. but overall the hummingbird beats snapdragon with its better graphics etc.
RogerPodacter said:
because the snap dragon chipset kicks the hummingbird's butt at certain types of computational tasks. example the old nexus one gets a 38mflops. but overall the hummingbird beats snapdragon with its better graphics etc.
Click to expand...
Click to collapse
If by tasks, you mean crunching pi... I agree.
Sent from my ADR6400L using XDA App
I get 2050~2100.
1.2Ghz OCed on ProTekk kernel, stock rom.
eccenpix said:
I get 2050~2100.
1.2Ghz OCed on ProTekk kernel, stock rom.
Click to expand...
Click to collapse
1 ghz. My kernel. BAMF 1.4.
{
"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"
}
is that a flat black notification bar on Bamf I see??
which theme is that?
adrynalyne said:
1 ghz. My kernel. BAMF 1.4.
Click to expand...
Click to collapse
asenduk said:
is that a flat black notification bar on Bamf I see??
which theme is that?
Click to expand...
Click to collapse
Lol, I noticed the same thing

It's safe to say by now Tegra2 CPU owns Snapdragon 1.2

I have owned the G2X & Sensation also Atrix My sensation is by far the best looking device ever now before the Sensation got S-off I would find myself saying boy you just wait until we get roms this thing will fly like my g2x its funny boy was I wrong. I'm on the Senseless v1.2 Custom kernel and the Sensation is very fast now but still not on my g2x speed so I think its safe to say the Tegra2 is pretty much a better faster CPU period now I'm hoping and praying the upcoming roms and kernels prove me and US all wrong but as of now custom roms and s-off had done nothing to bring out this beast my Ayrix is now hitting 3700 quadrants. Time well let us know
Sent from my HTC Sensation 4G using XDA App
If you're using quadrant as evidence to prove that Tegra 2 is actually faster, you're standing on thin ice. Quadrant can easily be skewed with a few modifications.
I'm using the stick ROM with ADW EX and my Sensations UI is as fast as my G2X. I can also say that my Sensation has always been faster at opening and interacting with apps such as Google Maps which loads a lot faster on my Sensation. When it cones to the web browser there both eveb in terms of loading pages but my Sensation scrolls smoother. I actually like the looks of the G2X better but I mainly use the Sensation due to it loading and interacting with apps faster.
Just for information, my G2X is bug free.
Another piece of information. I have seen a difference with speed in the 4 Sensations I've used. I really don't know what it is but the second one I had was slow as hell. You must have a bad one because the Sensation should load stuff faster than the G2X.
Sent from my HTC Sensation 4G using XDA Premium App
Why are there dozens of posts like this everyday?
Does it matter? Does seeing another phone with more powerful SoC immediately kill the battery of yours? Does having a faster SoC get you laid? Why does Quadrant matter? Does your phone require you to finish a Quadrant run before you can answer your phone everytime it rings? Because Quadrant is hardly a benchmark. Does your phone run slower than the G1? Does the Atrix run so much faster than the Sensation that makes it feels like a Razor? If yes, why bother, just dump the Sensation and use the Atrix. If no, why bother, since both phones are fast enough for everything a normal human would do on a phone for now.
Back to your question. No. First of all, Tegra 2 is not a CPU. It's a system on a chip (SoC). The processor cores in Tegra 2 are Cortex A9, and yes, they are more advance than the Scorpion cores in the Snapdragon, which falls roughly between Cortex A8 and A9. So CPU raw power alone, the Tegra 2 should be faster except in a few situations, because Scorpion's strength is floating point calculation and the Cortex A9 in Tegra 2 (but not necessarily other Cortex A9 SoC) lacks the NEON instruction sets.
I'd also like to add that you are comparing our performance to devices that have had more time for development and improvements. Give it time. The sensation devs will do their thing and although they might not be able to close the gap completely, I have faith that they will get us close.
Sent from my HTC Sensation Z710e using XDA Premium App
well the g2x is stockish android gingerbread which is fast.
The sensation has a ton of bloat with sense 3.0
When CM7 is optimized, the sensation will edge out the g2x.
Compare apples to apples br0.
Wait for cm7, sense is still on all roms.
Sent from my HTC Sensation 4G using XDA App
chisleu said:
well the g2x is stockish android gingerbread which is fast.
The sensation has a ton of bloat with sense 3.0
When CM7 is optimized, the sensation will edge out the g2x.
Compare apples to apples br0.
Click to expand...
Click to collapse
No it won't. The a9 is just a better architecture. More cache, shorter pipeline, more efficient memory usage etc. If you overclock the scorpion 25% or so vs a9 it might be able to catch up. Amd and Intel had done research and they found that it doesn't make sense to use asynchronous cores as in most real world cases it didn't yield better battery life. They found it was better to use power gating to save power. Unfortunately qualcomm is company that makes majority of its net income from 20 year old patents so it is no surprise they made questionable decisions. They sunk all this money into the scorpion architecture and they have to use it as long as they did due to sunk costs. However Nvidia screwed up to not having 1080hp decoder or neon. At least I thought the adreno 220 would be a beast, but It doesn't seem to be much better than the 2 year old overclocked sgx540 on omap4430.
However I didn't buy the sensation just because of its SOC, but it is just a solid phone in terms on construction and you know xda will support it, and it isn't slow either. I wish HTC would switch to a better soc, but they are tied with qualcomm at the hip.
First off I own both phones... My wife's G2x is faster, but it's running stock android. Every phone I've ever owned running Sense is slower than stock Android...
It's not really comparable until we have a stock build of android running on the Sensation. Then compare.
magic_man said:
However I didn't buy the sensation just because of its SOC, but it is just a solid phone in terms on construction and you know xda will support it, and it isn't slow either. I wish HTC would switch to a better soc, but they are tied with qualcomm at the hip.
Click to expand...
Click to collapse
The next generation of Qualcomm CPUs are coming soon and those will move from Scorpion to Krait. The Krait is based off of the A15 architecture from what I've read.
Heh, my Samsung Galaxy S II is better than your Sensation, G2x, or Atrix. Got Quadrant scores of 4k. Run Quadrant 3 times to see accurate score. Oh yeah I underclocked it too to 1GHZ. You won't find another fine device like this phone. Running on Revolution rom with bloatware removed.
Sent from my GT-I9100 using XDA App
kotaro_14 said:
Heh, my Samsung Galaxy S II is better than your Sensation, G2x, or Atrix. Got Quadrant scores of 4k. Run Quadrant 3 times to see accurate score. Oh yeah I underclocked it too to 1GHZ. You won't find another fine device like this phone. Running on Revolution rom with bloatware removed.
Sent from my GT-I9100 using XDA App
Click to expand...
Click to collapse
Epic troll?
Firstly, OP's English is terrible. No offence, but I had to put into notepad and punctuate it, just to understand what you were saying.
Secondly: the Sensation is on Sense, which definitely slows this phone done. I'd go as far as saying CM7 will be anything up to 2x as fast as Sense is.
Vandam500 said:
Epic troll?
Click to expand...
Click to collapse
More likely looking for an ego boost because his phone gets better results on a benchmarking tool that's severely outdated.
Dude, here is a sensible test.
1:- install a vanilla ROM on all phones
2:- get all devices to use the same resolution
3:- get a sensible graphics test app like nenamark 2
4:- test the devices and check the score.
This is the only way u will have to test.which phone is better or which processor is better
Cheers
VIPER
I played my friend's Atrix 4G,that screen,that ui,haha,i can't help to laugh
Half the people on these forums must be those who buy guys cars based solely on HP figures /rolleyes
twomix9900 said:
Half the people on these forums must be those who buy guys cars based solely on HP figures /rolleyes
Click to expand...
Click to collapse
LOL I know exactly what you're talking about. I'm on a VW forum (hence the avatar), and we had this SRT-4 owner try to troll us. We were all like "you can keep your 400 whp champ, we like our interiors lol". As to this thread, I don't think it's fair to say that either "owns" the other. They are pretty darn even.
aNYthing24 said:
More likely looking for an ego boost because his phone gets better results on a benchmarking tool that's severely outdated.
Click to expand...
Click to collapse
Meh... Troll... Lol...
Sent from my GT-I9100 using XDA App
{
"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"
}

[Q] Why isn't Optimus 3D as supported as Samsung Galaxy S2?

So, LG offers us a 2 yrs warranty, even if we root our amazing smartphone.
Samsung does not.
Everyone knows Optimus 3d has got better processor, but it still lags with every rom you actually put on it.
Also SGS2 has got very good hardware, but it's supported moooooore than our LG smartphone.
Why this?
What's wrong on our O3D??
It's just been released in USA as the Thrill.
Sent from my LG-P920 using Tapatalk
What are you talking about? LG O3D has been relased in the same SGS2 countries
the gs2 had from the start on a very huge dev community because of the sg1 and the other galaxy smartphones...
i think it only takes some time to get great devs for this phone
i currently have a galaxy 3 but i ordered the o3d and i can't wait to hold it in my hands!!
Because sgs2 is more popular than o3d
Sent from my LG-P920 using Tapatalk
@deox91: I had a sgs2 before having the o3d... I'd come back without any doubt
@dexxxterek: sure, it's popular, but how's it possible there isn't any unofficial (downloadable) gingerbread-based cooked rom (like happened with optimus 2x)? there isn't a decent kernel which fixes lags, there are no decent roms... nothing of nothing!
Gingerbread o3d update should be released in october, yes, but Ice Cream Sandwich is coming soon....
Delux9364 said:
Everyone knows Optimus 3d has got better processor
Click to expand...
Click to collapse
No, it doesn't. Exynos > OMAP4.
Killer Bee said:
No, it doesn't. Exynos > OMAP4.
Click to expand...
Click to collapse
Actually, it does.
In terms of performance, OMAP4 is better.
On the S2, Exynos was just overclocked, but when both are at 1ghz, OMAP4 is better.
I've underclocked my SGS2 to 1 Ghz to test benchmarks and it still does better than OMAP4. And for real world performance, I underclock my SGS2 to 800mhz (to save battery) and notice no difference between it and 1.2Ghz.
Maybe you don't know, but Cyanogen is working for Samsung, and he's help is invaluabe ;]
Killer Bee said:
I've underclocked my SGS2 to 1 Ghz to test benchmarks and it still does better than OMAP4. And for real world performance, I underclock my SGS2 to 800mhz (to save battery) and notice no difference between it and 1.2Ghz.
Click to expand...
Click to collapse
I believe reason for that is the 2.2 vs 2.3 argument.
One has better dual core support and one... =/ well everyone knows the answer to that one. So just trying to tell you that underclocking one to 1ghz while its on 2.3 and the other is running 2.2 still doesn't say which is the better processor.
Also, benchmark results are very subjective to many things. I highly doubt you would have the exact same things running on both devices when you ran the tests. Even if the apps are the same, the stock apps that comes with the devices and other factors also affect the results. For example, I'm on stock rom and a very heavy user. Quadrant gives me a score of 2138. Yet, real performance is still much better than you would expect.
DevJackey said:
I believe reason for that is the 2.2 vs 2.3 argument.
One has better dual core support and one... =/ well everyone knows the answer to that one. So just trying to tell you that underclocking one to 1ghz while its on 2.3 and the other is running 2.2 still doesn't say which is the better processor.
Click to expand...
Click to collapse
I really doubt that GB will bring significantly better performance. Its all about optimizing the drivers to which Samsung did VERY well (compare Debut GS2 to a Released GS2). OMAP4 has a lot of potential imo, but I'll guess we'll see in the future with updates or the Nexus Prime.
Also, benchmark results are very subjective to many things. I highly doubt you would have the exact same things running on both devices when you ran the tests. Even if the apps are the same, the stock apps that comes with the devices and other factors also affect the results. For example, I'm on stock rom and a very heavy user. Quadrant gives me a score of 2138. Yet, real performance is still much better than you would expect.
Click to expand...
Click to collapse
We were talking about which processor was better, right? I don't know how else you can determine that besides benchmarks.
Killer Bee said:
I really doubt that GB will bring significantly better performance. Its all about optimizing the drivers to which Samsung did VERY well (compare Debut GS2 to a Released GS2). OMAP4 has a lot of potential imo, but I'll guess we'll see in the future with updates or the Nexus Prime.
We were talking about which processor was better, right? I don't know how else you can determine that besides benchmarks.
Click to expand...
Click to collapse
If you don't believe it, I can't change your mind.
We will see.
What performance boost does GB have? I remember Froyo had JIT but I don't recall anything like that being in GB.
Yes, but my aim wasn't to start a discussion about which processor is the best...
My aim was to let people able to do this, think about developing more on our amazing device.. :\
Gingerbread is faster... I have an SE X10-
From using highly customized froyo Roms to the bloated, stock gingerbread was a massive speed shift, and then once the custom Roms came out it was like having a whole new phone.
I'm here because my contract is up in a few months and I'm considering getting this phone, actually it's a toss-up between this and the SGS2.
sent from my Wolfbreak GingerBread x10 using XDA App
curious_yellow_eyes said:
Gingerbread is faster... I have an SE X10-
From using highly customized froyo Roms to the bloated, stock gingerbread was a massive speed shift, and then once the custom Roms came out it was like having a whole new phone.
I'm here because my contract is up in a few months and I'm considering getting this phone, actually it's a toss-up between this and the SGS2.
sent from my Wolfbreak GingerBread x10 using XDA App
Click to expand...
Click to collapse
Not true at all.
I have an SE X10 as well for a year and half now. The stock 2.3 ROM is A LOT slower than any custom 2.2 ROMs.
Only massively tweaked 2.3 ROMs like Wolfbreak or Trip's MIUI ROMs are fast. The stock 2.3 ROM is so slow and laggy.
I found it faster than scramble...
sent from my Wolfbreak GingerBread using XDA App
Killer Bee said:
I've underclocked my SGS2 to 1 Ghz to test benchmarks and it still does better than OMAP4.
Click to expand...
Click to collapse
Pics or it didn't happen!
Delux9364 said:
Yes, but my aim wasn't to start a discussion about which processor is the best...
My aim was to let people able to do this, think about developing more on our amazing device.. :\
Click to expand...
Click to collapse
I was correcting the statement you made in the OP with the intent to answer the question you asked in the Title.
I also want more devs to develop for the O3D and a LOT less devs to develop for the SGS2 (there are 2 freaking sections for development).
mangelok said:
Pics or it didn't happen!
Click to expand...
Click to collapse
1000GHz:
{
"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"
}
Antutu:
CF-Bench:
Linpack:
Nenamark 2:
Quadrant:
Vellamo:

Benchmark shock

Hi all
My wife (iPhone 4 user ) mailed me this link today and I must say I'm shocked ... it seems, in some respects, even the iPhone 3GS walks all over the Sensation ... how is this possible?
http://www.redmondpie.com/iphone-4s...-as-galaxy-s-ii-and-all-other-android-phones/
gpu in snapdragon soc's will never be as good as the SGX gpu in Apple phones.
even the Samsung Mali GPU is faster then Adreno 250.
Sadly HTC is Snapdragon only so unless Qualcomm improve it a lot HTC phones will always lag behind in benchmarks but benchmarks are not everything.
One more thing that should be taken in consideration is that Apple designs their hardware and software which means it is optimized for dual-cores.
When we get Ice Cream Sandwich is when we will get real dual-core support and be able to use the full power of our phones.
Sent from my HTC Sensation 4G using XDA App
Just ran these benchmarks myself and it looks a lot better than they make it seem, sure I'm an update up from their benchmark Sensation, but technically iOS 5 is newer than Android 2.3 anywayso it's negligible.I would like to see how Ice Cream Sandwich will compare to the 4GS (iOS 5).
Sunspider - 3532
Browsermark - 56065
In iPhone it is used in games only but generaly the IOS is very boaring, Android is much better.
the nexus one is faster than the sensation
Personally, I don't think its that the GPU is bad, it should be better than the mali400 by quite a bit in theory ( look at the Qualocomm reference board @1.5GHz, beats everything by a nice margin ) It's just that the kernel and driver software is ****.
IGstarr said:
the nexus one is faster than the sensation
Click to expand...
Click to collapse
Lol
Sent from my HTC Sensation 4G using XDA App
My scores are:
SunSpider - 2444 ms
Browsermark - 7900
Don't believe this stock using benchmark. I'm sure that SGS2 can do better as well.
Just ran the tests and got
Sunspider 1844 opera
Browsermark 43865 opera, and using the standered browser 71114
Not really a good test as its browser inflorescenced.
Sent from my HTC Sensation XE with Beats Audio using xda premium
Come on guys !!!!!!!!!!!!!!!!!!!!!!!!!!!!!
You know Apple is best at MARKETING!!!!!
1) APPLE IPHONE 4S PROPAGANDA, that can be seen on many website for a few days
2) THE REALITY, my HTC Sensation running ARHD 3.6.2 with bricked kernel , at 1.72 ghz (my max is 1.78)
with CPU1 forced online, I hit about 60000 at stock 1.2ghz CPU speed
Some will say i've always hated Apple's marketing, already fighting Apple fanboys in early 2009 with my Samsung Omnia i900 and FpseCE...
....well...that's true
EDIT: same config, I do 2664 at Sunspider, so just behind the iphone 4S, and GLbenchmark refuses to compute my score, saying my GL drivers are modified!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Preventing people who use custom ROMS, kernels or drivers is an easy way to win! (I mean, I was not cheating using Chainfire3D drivers to reduce textures or anything!)
Apple sucks!
thats the problem, all these benchmarks doesn't support what our devices (beast) has. so, it always end up like that. but, mind you. on sensation, it has the 2.3.3 version and with that and all the crap on stock, its really gonna be slow
Here what I got on my stock sensation these results are definately bias for the iPhone release =\ I mean I'm sure its a impressive piece of hardware but I doubt its that much better than other phones out on the market now
{
"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"
}
Sent from my HTC Sensation 4G using XDA App
Nekromantik said:
gpu in snapdragon soc's will never be as good as the SGX gpu in Apple phones.
even the Samsung Mali GPU is faster then Adreno 250.
Sadly HTC is Snapdragon only so unless Qualcomm improve it a lot HTC phones will always lag behind in benchmarks but benchmarks are not everything.
Click to expand...
Click to collapse
Lol, there's no way the PowerVR SGX535 is faster than the Adreno 220. It's just iOS is way more optimized than Gingerbread + Sense 3.0 ...
The Snapdragon S4 SoCs should blow everything away when they come out, though.
Cm7 alpha8 tiamat OCd @ 1.83ghz
Sent from my Sensation using XDA App
You can't compare an overclocked phone running god knows which kernel (It's easy to cheat by optimizing the kernel for specific benchmarks) with a stock phone running at stock speeds. That doesn't make any sense.
It's like trying to prove that Toyota is faster than a Lexus by strapping a jet engine on to the Toyota: Unfair, irrelevant and proves nothing: Except that a Toyota with a jetengine strapped on to it iis faster than a Lexus...
So yeah, the iPhone 4S does beat whatever HTC has out by a significant margin even. I have an Evo3D and a SGSII and it is pretty annoying that the HTC is so poorly optimized. Time for HTC to lock some engineers in a basement, and not let them out until they modified Sense to be as fast as iOS...
vszulc said:
You can't compare an overclocked phone running god knows which kernel (It's easy to cheat by optimizing the kernel for specific benchmarks) with a stock phone running at stock speeds. That doesn't make any sense.
It's like trying to prove that Toyota is faster than a Lexus by strapping a jet engine on to the Toyota: Unfair, irrelevant and proves nothing: Except that a Toyota with a jetengine strapped on to it iis faster than a Lexus...
So yeah, the iPhone 4S does beat whatever HTC has out by a significant margin even. I have an Evo3D and a SGSII and it is pretty annoying that the HTC is so poorly optimized. Time for HTC to lock some engineers in a basement, and not let them out until they modified Sense to be as fast as iOS...
Click to expand...
Click to collapse
Of course we can! That's why we have an Android and not an iphone: we can customize and improve it as far as we can, that's the point here.
But even with a stock ROM, bloatwares removed, at 1.2ghz, we reach 60000 at browsermark, so Apple is clearly posting fake figures here.
Don't wait for HTC to solve your problems man, go for a custom ROM and a custom kernel, the Sensation has good hardware inside, you should try using it decently.
PS: my kernel is Bricked kernel from Show-p, you can't say it's "god knows which kernel"
Wouldnt screen size make a difference as well?
I've read an article about s2 and 4s benches. Basically, sw runs differently, Android does something vertically as well, when iphone doesn't, so you could double Android point to actually show the true value. I'll try to dig that article.
Sent from my HTC Sensation Z710e using XDA App
vszulc said:
You can't compare an overclocked phone running god knows which kernel (It's easy to cheat by optimizing the kernel for specific benchmarks) with a stock phone running at stock speeds. That doesn't make any sense.
It's like trying to prove that Toyota is faster than a Lexus by strapping a jet engine on to the Toyota: Unfair, irrelevant and proves nothing: Except that a Toyota with a jetengine strapped on to it iis faster than a Lexus...
So yeah, the iPhone 4S does beat whatever HTC has out by a significant margin even. I have an Evo3D and a SGSII and it is pretty annoying that the HTC is so poorly optimized. Time for HTC to lock some engineers in a basement, and not let them out until they modified Sense to be as fast as iOS...
Click to expand...
Click to collapse
I agree with this. That's why I did my test completely stock. And didn't remove bloatwear or anything and still got significantly better scores than what they posted.
Sent from my HTC Sensation 4G using XDA App

G2x is better than GS2

I was over on the GS2's forum and saw that oc to 1.78ghz they are only getting in the high 6ks to 7500s on antutu benchmark where the G2x is getting between 7500s and 7900s. I am really glad that I didn't make the switch to the GS2. Just thought I would put that out there if anyone was curious and thought about switching.
Sent from my LG-P999 using xda premium
tidewaterns said:
I was over on the GS2's forum and saw that oc to 1.78ghz they are only getting in the high 6ks to 7500s on antutu benchmark where the G2x is getting between 7500s and 7900s. I am really glad that I didn't make the switch to the GS2. Just thought I would put that out there if anyone was curious and thought about switching.
Sent from my LG-P999 using xda premium
Click to expand...
Click to collapse
The SG2 has a better GPU, and better I/O. O:
Look at mine
This is a custom kernel made by Morfic (1.6GHZ)
{
"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"
}
nitrogen618 said:
The SG2 has a better GPU, and better I/O. O:
Look at mine
This is a custom kernel made by Morfic (1.6GHZ)
Click to expand...
Click to collapse
Nice. I was looking at some scores over on the SGS2 forums and they were showing a lot lower scores. Good to see that the phone is a little better.
Sent from my LG-P999 using xda premium
I have both of these. SGS2 is definitely more responsive but G2X with Tegra 2 is better at games.
..so because of some scores you saw on a benchmark you think that makes a phone better? In short, NO it does not..not even close. I'll give you a very simple test that will show you plenty, go to http://www.hdtrailerz.com/movie-trailers/men-in-black-3 and play the video on the GS2 and watch how it plays nicely, then play it on the G2X and see the difference.... that's something your benchmarks don't show..real world performance, and THAT'S what really matters, not some meaningless score.
Well said mystrioustko. Well said.
Teapots are the work of the devil. Long live taco sauce.
mysterioustko said:
..so because of some scores you saw on a benchmark you think that makes a phone better? In short, NO it does not..not even close. I'll give you a very simple test that will show you plenty, go to http://www.hdtrailerz.com/movie-trailers/men-in-black-3 and play the video on the GS2 and watch how it plays nicely, then play it on the G2X and see the difference.... that's something your benchmarks don't show..real world performance, and THAT'S what really matters, not some meaningless score.
Click to expand...
Click to collapse
Aahhh, but Nvidia has CUDA, and once developers start utilizing it, they will tap into a huge potential of parallel processing that will DESTROY other CPU/GPUs.
jacob808 said:
Aahhh, but Nvidia has CUDA, and once developers start utilizing it, they will tap into a huge potential of parallel processing that will DESTROY other CPU/GPUs.
Click to expand...
Click to collapse
Yes... But only GPU's it'll destroy are probably Qualcomm's S3 processors. It's not even close to being as good as the SGS2's Mali-400 GPU.
But to the OP, benchmarks mean nothing. By no means am I trashing the G2x, I actually quite love the phone. I have CM7 on it with Faux's bangin' kernel, but even with that set up, my gf's stock AT&T SGSII ****s on my phone. Everything about it is faster and better. No lag, handles 1080p videos much better. Browsing is butter smooth and handles flash content 100x better than my G2x. And it takes way better pictures, camera on that thing is a beauty.
mackster248 said:
Yes... But only GPU's it'll destroy are probably Qualcomm's S3 processors. It's not even close to being as good as the SGS2's Mali-400 GPU.
But to the OP, benchmarks mean nothing. By no means am I trashing the G2x, I actually quite love the phone. I have CM7 on it with Faux's bangin' kernel, but even with that set up, my gf's stock AT&T SGSII ****s on my phone. Everything about it is faster and better. No lag, handles 1080p videos much better. Browsing is butter smooth and handles flash content 100x better than my G2x. And it takes way better pictures, camera on that thing is a beauty.
Click to expand...
Click to collapse
Yeah, my wife has the T-Mo version in white w/the Qualcomm proc and it is nowhere close as far as everyday use. Can't even imagine one with the Exynos like the AT&T version.
jacob808 said:
Aahhh, but Nvidia has CUDA, and once developers start utilizing it, they will tap into a huge potential of parallel processing that will DESTROY other CPU/GPUs.
Click to expand...
Click to collapse
It also does not have neon AND it has less than half the memory bandwidth of other A9 processors. Like I said, a real world example of it's performance is to play that HD trailer I posted. Even on the TMo version of the GS2 (which doesn't even have the exynos) it plays smoothly, yet the same cannot be said for the G2X. The difference is quite ostensible. I'm sorry, but I don't think we'll be seeing nvidia "destroy" any other processor any time soon. Even their next gen processors, already appear to be following the same pattern of the tegra 2...that is, falling short when compared to the competition. I'm really just not all that impressed with the tegra 2.
The S2 have H/W Acc. built on to the browser.
nitrogen618 said:
The S2 have H/W Acc. built on to the browser.
Click to expand...
Click to collapse
Opera Mobile has hardware acceleration built in as well. Its pretty sweet. Mini doesn't but mobile is really nice
Sent from my LG-P999 using xda premium
Oh really? fndksjfhksjdhf
TIME TO DL
Edit:
It's a lot smoother than Dolphin Browser HD with a page full of images.
It's better because I have a G2x! F+*K ALL The wanna be's
This phone is awesome and does not need to be compared to another. It plays games sgs2 can't period. No benchmark or video can show u that only the apps them self. Have fun sgs2 but u will never be nvidia. This may be the first nvidia type but it evolve to something that we will never expect.
Nvidia always!
jetbruceli said:
It's better because I have a G2x! F+*K ALL The wanna be's
This phone is awesome and does not need to be compared to another. It plays games sgs2 can't period. No benchmark or video can show u that only the apps them self. Have fun sgs2 but u will never be nvidia. This may be the first nvidia type but it evolve to something that we will never expect.
Nvidia always!
Click to expand...
Click to collapse
They have already evolved into something...unfortunately what they evolved into is a mid tier processor. The games that you mentioned the gs2 not being able to play would be due to compatibility not due to it being incapable. The movie trailer link I posted, the G2X struggles with it due to incapability not incompatibility. That's a big difference. The tegra 2 is not the worse processor out there, but it definitely isn't one of the best.
Lets put it this way, the GS2 has more GPU power than an Xbox 360. I'm not trashing the G2X, I love my phone. I'm just giving people a scale of things.
Sent from my G2X
its pretty hard to compare anything and just decide which is better, it all depends on what you value more.
Sent from my LG-P999 using Tapatalk
mysterioustko said:
..so because of some scores you saw on a benchmark you think that makes a phone better? In short, NO it does not..not even close. I'll give you a very simple test that will show you plenty, go to http://www.hdtrailerz.com/movie-trailers/men-in-black-3 and play the video on the GS2 and watch how it plays nicely, then play it on the G2X and see the difference.... that's something your benchmarks don't show..real world performance, and THAT'S what really matters, not some meaningless score.
Click to expand...
Click to collapse
I know that scores don't matter. I know that att version is WAY better. I was only saying that the G2x is not as bad of a phone as some people make it out to be. Oh and your HD video you posted played very smoothly on my G2x. So if my OP got some people mad, then I am sorry because it was not meant for that.
Sent from my LG-P999 using xda premium
There's serious lag on my G2X. In Angry Birds, sometimes, when a bird has been "launched", it progresses in starts and pauses until it hits. Also, there's lag with button response, screen touch response, and browser use. Android's inherent weakness is screen drawing which doesn't get priority over everything else as it does in Apple's iOS. That's a known Android flaw which Android application developers criticize continuously.
Definitely comes down to a matter of personal preference. I LOVE gaming on my phone, and my g2x has PWNED every game I presented to it without any lags. Not to mention how well it handles when I feel like overclocking. I put it up to 1.4 today just for ****s and played Modern Combat 3 (which is only .99 cents for a limited time) and it killed! My only qualm is the battery life. 1500mAh just isn't enough for any dual core phone.
Sent from my LG-P999 using xda premium

Categories

Resources