Compared to other Snapdragon S4 Pro phones like the Nexus 4, Sony Xperia Z, and others, why is gaming performance and graphics benchmarking performance extremely slow compared to the scores on apps and games like Asphalt 8, 3DMark, and the newly released Anomaly 2 Benchmark? On Asphalt 8, sometimes my fps can dip all the way to 15 or so, on 3DMark I ran the offscreen "Unlimited" test and got over 6000 while the Nexus 4 got near 11k. I ran the Anomaly 2 Benchmark last night and got a bronze score on high of like 120k, and I was reading how everyone's HTC One and Nexus 4 were getting close to or over 200k. I understand the HTC One has a Snapdragon 600, and that the nexus 4 has a 720p screen in comparison to the DNA's 1080p, but from the looks of it, Asphalt 8 renders in 720p, and the 3DMark Unlimited test is off screen, so it directly tests the chipset. Does anyone have any explanation as to why the DNA performs the way it does? According to the 3DMark leaderboards, it has the second worst average Adreno 320 GPU score. Not to mention, the Droid Ultra has a WORSE chipset than the DNA's yet it gets a significantly higher 3DMark score?..
Also, if it helps, I'm running pio_masaki's CARBONrom build 9.23 Android 4.3 and using crpalmer's 4.3.3 kernel.
Thanks to any help and explanation!
rejectedjs said:
Compared to other Snapdragon S4 Pro phones like the Nexus 4, Sony Xperia Z, and others, why is gaming performance and graphics benchmarking performance extremely slow compared to the scores on apps and games like Asphalt 8, 3DMark, and the newly released Anomaly 2 Benchmark? On Asphalt 8, sometimes my fps can dip all the way to 15 or so, on 3DMark I ran the offscreen "Unlimited" test and got over 6000 while the Nexus 4 got near 11k. I ran the Anomaly 2 Benchmark last night and got a bronze score on high of like 120k, and I was reading how everyone's HTC One and Nexus 4 were getting close to or over 200k. I understand the HTC One has a Snapdragon 600, and that the nexus 4 has a 720p screen in comparison to the DNA's 1080p, but from the looks of it, Asphalt 8 renders in 720p, and the 3DMark Unlimited test is off screen, so it directly tests the chipset. Does anyone have any explanation as to why the DNA performs the way it does? According to the 3DMark leaderboards, it has the second worst average Adreno 320 GPU score.
Also, if it helps, I'm running pio_masaki's CARBONrom build 9.23 Android 4.3 and using crpalmer's 4.3.3 kernel.
Thanks to any help and explanation!
Click to expand...
Click to collapse
What are the scores ( that mean nothing ) on other roms and also when the phone is running the stock VZW rom?
RLGL said:
What are the scores ( that mean nothing ) on other roms and also when the phone is running the stock VZW rom?
Click to expand...
Click to collapse
Ive only ran 3DMark when I was on stock ROM, considering A8 and the A2 Benchmark weren't out then, and on the Extreme test on 3DMark I used to always hit 5.9k or below, now I get about 61xx. On other ROMs, especially the NOS M7 port, or any M7 port for that matter, I always got way worse performance through both gaming and benchmarks.
I thought the 2013 Droids and the Moto X had a modified S4 Pro chip? Not necessarily worse, but tweaked? That could explain the performance boost, as I believe it outperforms the S4 and HTC One's Snapdragon 600.
I would suspect actual gaming performance would take a hit because of the screen, but I cannot give a reason behind the offscreen benchmarks, as I do not know how they test the chip.
You brought up the Nexus 4 a lot (as I assume you have owned it), but what are numbers like on the Z or Optimus G?
raichur0xx0rz said:
I thought the 2013 Droids and the Moto X had a modified S4 Pro chip? Not necessarily worse, but tweaked? That could explain the performance boost, as I believe it outperforms the S4 and HTC One's Snapdragon 600.
I would suspect actual gaming performance would take a hit because of the screen, but I cannot give a reason behind the offscreen benchmarks, as I do not know how they test the chip.
You brought up the Nexus 4 a lot (as I assume you have owned it), but what are numbers like on the Z or Optimus G?
Click to expand...
Click to collapse
I've never owned a nexus 4, but I have seen many videos and results of it outperforming the DNA when it comes to gaming. The LG Optimus G performs lower in every benchmark in 3DMark compared to the DNA, while the Xperia Z performs faster in comparison to the DNA.
rejectedjs said:
I've never owned a nexus 4, but I have seen many videos and results of it outperforming the DNA when it comes to gaming. The LG Optimus G performs lower in every benchmark in 3DMark compared to the DNA, while the Xperia Z performs faster in comparison to the DNA.
Click to expand...
Click to collapse
1080p vs 720p displays.
mwl1119 said:
1080p vs 720p displays.
Click to expand...
Click to collapse
I'm sure there is a more in depth reason than that considering the Optimus G has a quad core processor w/ an Adreno 320, while the Droid Ultra has the same setup except with a dual core, and it performances nearly twice as fast than the OG in 3DMark. Both have 720p displays.
Slow?
I've owned the DNA since launch (a launch model and a replacement w/ the Verizon update pre-installed). My original was rooted with S-OFF. My current is stock. I have never heard the word slow and the Droid DNA in the same sentence. I have a 2013 Nexus 7, which I believe is also a S4 Pro, and when comparing the two, my Nexus has has more non responsive moments than my phone. My DNA has been a champ for close to a year now. The only drawback is battery life, but that's not the topic of this conversation.
You mention benchmarks are low on the DNA but does the phone feel slow? I'm also on Carbon with crpalmers kernel and although the benchmarks are lower than stock it feels so snappy and responsive, that's why I don't put much weight on benchmark scores..
You also mention slow gaming performance but didn't explain if this is based on benchmarks or actual experience.
He did mention lower framerates on Asphalt 8.
I think he's mainly concerned about gaming performance, as we now have some fairly demanding games on the market. I don't play games on my phone, so I can't really give any examples of my experience...
Otherwise, regardless of benchmark numbers, I think 2012+ phones have such advanced internals that day-to-day basic usage of the device will be fine.
orangechoochoo said:
You mention benchmarks are low on the DNA but does the phone feel slow? I'm also on Carbon with crpalmers kernel and although the benchmarks are lower than stock it feels so snappy and responsive, that's why I don't put much weight on benchmark scores..
You also mention slow gaming performance but didn't explain if this is based on benchmarks or actual experience.
Click to expand...
Click to collapse
Games like Asphalt 8, I can understand having a drop to a lower framerate, however, even when playing Dead Trigger on High on my current kernel and ROM setup, when turning the camera, I experience immense framerate drops, to even like, 11 or so. I honestly don't understand why the DNA is so under performing. I do a lot of gaming on my DNA, and early benchmarks when the phone was first released revealed better scores than the Nexus 4, but now it seems like the exact opposite. I can't even keep a decent frame rate on Real Racing 3.
rejectedjs said:
I'm sure there is a more in depth reason than that considering the Optimus G has a quad core processor w/ an Adreno 320, while the Droid Ultra has the same setup except with a dual core, and it performances nearly twice as fast than the OG in 3DMark. Both have 720p displays.
Click to expand...
Click to collapse
The amount of cores really does not matter unless you're using a bunch of intensive programs at once. Most of the time you're only using 1 or 2 cores.
mwl1119 said:
The amount of cores really does not matter unless you're using a bunch of intensive programs at once. Most of the time you're only using 1 or 2 cores.
Click to expand...
Click to collapse
You're failing to understand my point here. My main question in this thread is, if all these devices use the exact same GPU, and most of them, the exact same chipset, why do the Adreno 320s in phones like the Nexus 4, the Xperia Z, Droid Ultra, how do those outperform the GPU in the DNA by so much? Even in real world performance. In real racing 3, a nexus 4 can get like a good average 30 FPS, on my DNA, I'm lucky to see FPS Meter hit 30 once.
I don't know why you're mentioning the nexus 4 as having better benchmarks, the n4 tends to get pretty low scores with everything due to thermal throttling. Also, look at the benchmarks for the 2013 n7, they're also on the low side for an s4 pro.
Edit: never mind, I see you already addressed that.
As far as real world performance with gaming and such, I think it's a combination of the 1080p screen and the fact that sense is resource intensive. Also I think most s4 pros are clocked at 1.7 instead of 1.5 like us, don't quote me on that though.
On aosp Roms the problem is that we don't get to use the proprietary drivers that HTC does. Kind of like using Linux's open source graphics drivers vs proprietary and or nvidia ones. I haven't had very good gaming performance on any aosp ROM, especially the 4.3 ones.
Sent from my Droid DNA using Tapatalk 4
jamiethemorris said:
I don't know why you're mentioning the nexus 4 as having better benchmarks, the n4 tends to get pretty low scores with everything due to thermal throttling. Also, look at the benchmarks for the 2013 n7, they're also on the low side for an s4 pro.
Edit: never mind, I see you already addressed that.
As far as real world performance with gaming and such, I think it's a combination of the 1080p screen and the fact that sense is resource intensive. Also I think most s4 pros are clocked at 1.7 instead of 1.5 like us, don't quote me on that though.
On aosp Roms the problem is that we don't get to use the proprietary drivers that HTC does. Kind of like using Linux's open source graphics drivers vs proprietary and or nvidia ones. I haven't had very good gaming performance on any aosp ROM, especially the 4.3 ones.
Sent from my Droid DNA using Tapatalk 4
Click to expand...
Click to collapse
This is the type of information I'm talking about. I don't know more about Android other than flashing zips, but I never knew that we're basically running off of a generic driver. It's literally about the same as me installing a new GPU in my desktop and running off of Microsoft's OEM display driver? If so, how much better would gaming be if the upcoming HTC One Dev Edition 4.3 update was ported to the DNA as a flashable rom? I did notice that the NOS HTC One GE rom on 4.2.2 had a lot better performance than all other 4.2.2 roms. If the HTC One Google Edition's 4.3 was ported to the DNA, it'd be running off of Qualcomm proprietary drivers? If so, I absolutely cannot wait.
Yeah, it's kind of similar to that. Obviously the devs tailor to the specific device as best they can but the fact is that with the majority of phones AOSP roms will have lower performance in a lot of situations and there will be some features that will never work, such as HDMI on the GS3 for the reason I just stated. So it's kind of just a trade-off of whether you want buttery smooth performance or infinite customizability. I've been running NOS's GPE rom with Beastmode the past few days and the gaming performance is awesome, and yes it is running off HTC's proprietary drivers as will 4.3. I'll end up switching back to a CM-based rom though sooner or later though, nothing beats the customization... I really wish we could dual boot these things.
I hope I'm not spreading false information here, I'm not a developer, this is just my understanding of it through various things I've read on the subject.
jamiethemorris said:
Yeah, it's kind of similar to that. Obviously the devs tailor to the specific device as best they can but the fact is that with the majority of phones AOSP roms will have lower performance in a lot of situations and there will be some features that will never work, such as HDMI on the GS3 for the reason I just stated. So it's kind of just a trade-off of whether you want buttery smooth performance or infinite customizability. I've been running NOS's GPE rom with Beastmode the past few days and the gaming performance is awesome, and yes it is running off HTC's proprietary drivers as will 4.3. I'll end up switching back to a CM-based rom though sooner or later though, nothing beats the customization... I really wish we could dual boot these things.
I hope I'm not spreading false information here, I'm not a developer, this is just my understanding of it through various things I've read on the subject.
Click to expand...
Click to collapse
Huh..NOS GPE w/ Beastmode? I tried that kernel, only on pure M7 ports though, and gaming performance was absolutely awful. I'll try your setup, I'll report back to see how it is.
rejectedjs said:
Huh..NOS GPE w/ Beastmode? I tried that kernel, only on pure M7 ports though, and gaming performance was absolutely awful. I'll try your setup, I'll report back to see how it is.
Click to expand...
Click to collapse
Well... I might not be as much of a phone gamer as you so YMMV. I mostly use my PC and my nexus 7 for that.
Sent from my Droid DNA using Tapatalk 4
jamiethemorris said:
Well... I might not be as much of a phone gamer as you so YMMV. I mostly use my PC and my nexus 7 for that.
Sent from my Droid DNA using Tapatalk 4
Click to expand...
Click to collapse
On 3DMark, my scores are higher than the average for any device with an Adreno 320 GPU, on all three tests Thanks dude, I absolutely cannot wait for a 4.3 update!
Performance: Maxed out
Extreme: 7369
Unlimited: 10881
Answer
resolution has alot to do with graphic performance. Using a high resolution can have a big toll on a graphic chip and thats true with any kind of computer, tablet or smartphone. the reason the Droid ultra and Nexus 4 run smoother in games because the resolution is lower. Simple as that. Dont know about the Xperia Z. 720 vs 1080p. Thats a 25% increase in resolution. It wont be appearant on most games but on high gpu powered games you will start to notice fps drops.
Related
What do you think about this review? GSII won in all of the tests
http://pocketnow.com/android/htc-evo-3d-vs-samsung-galaxy-s-2-video
Personal preference is what it comes down to for every phone and person in the world. There is no such thing as the best phone, just the best phone for each person.
That said, who cares about this review? I like my EVO3D, period. Might another phone be slightly better in some areas, sure. Does it matter, no.
Sent from my PC36100 using Tapatalk
It won by like one milla second in real world you will not know the difference
Sent from my PG86100 using XDA App
I'll admit that the processor in the sgs2 is better in benchmarks but man he sounded really bias.
What I love about all these Galaxy S II comparisons is the fact that the phone IS NOT YET AVAILABLE IN THE US!!! Carrier variants are known to have differences (HW & SW wise, especially in the US) so the small edge the test unit may have on Evo 3D can easily disappear once the REAL choice is available... just my $0.02
l33tlinuxh4x0r said:
I'll admit that the processor in the sgs2 is better in benchmarks but man he sounded really bias.
Click to expand...
Click to collapse
Is this due to synchronous vs asynchronous dual cores?
Does the "comparison" mention that the Samsungs build quality is the lowest of it's kind?
That the screens have these "weak spots" where it you press it too much they'll go black/dimmer?
They the charger port goes bad VERY quick and need to be replaced probably within the first 6 months.
That the speaker phone ALWAYS sounds like there's plastic film stuck in it?
I've had many Samsung phones, the only one I liked was the VERY first one I ever had back in 2002.
I'm no HTC nuthugger(because fk HTC for not releasing the unlocked bootloader) but HTC builds better phones. Simple.
Will have to revisit the test once the phone is actually available on the Sprint network, with the Sprint bloatware loaded on the Sammy. The GS2 is obviously a good phone, but this test isn't apples to apples.
Sammy sure knows how to make a gorgeous screen though
that superamoled + blows the superlcd on the evo3d away
the person who mentioned carriers will nerf **** too is correct
T-Mobile's Sensation has 7 something mb of ram, the evo's same hardware/phone almost but has 1 gb
Carriers lower the amount of ram to save money all the time, among other things, could result in a worse phone
I don't understand why Samsung insists on not making a good qHD screen tho with it....
Yes, the difference in benchmark scores is due to the differences between how the processors run. (synchronous vs. asynchronous)
on every day use, you cant see a difference between a really good single core vs dual core. Yes when you bench it, you see a difference. I mean we use our phones to use them everyday and bench them every day.
I really doubt you would see the difference between sgs2 and evo3d processor on everyday use.
This is like having a dual chip graphics card and running it on a 17inch monitor and not even use all the power. That is what android is right now, it is just not that demanding. I still highly doubt on every day use you will be able to see a difference between snapdragon dual core or samsung's new processor even with ICS.
I always though the benchmarks where due to the way the cores are set up/different. But I clocked my evo 3d to 1.8Ghz and ran quadrant on the performance governor. The highest I got was 2985. Now if I am correct, both cores should be maxed out due to the governor set by setcpu. If this is the case, the gs2 still beats the evo3d easily, even with the evo3d clocked to 1.8ghz.
I don't feel like this can be correct, but maybe it is? Maybe the exynos are that much better?
Or could it really be that sense 3.0 is a hog? I mean, with the hero, once it is on asop, you see a near 100% increase in scores.
mlin said:
Is this due to synchronous vs asynchronous dual cores?
Click to expand...
Click to collapse
Doubtful since benchmarks that utilize one or both cores still favor the sgs2
hydrogenman said:
Sammy sure knows how to make a gorgeous screen though
that superamoled + blows the superlcd on the evo3d away
the person who mentioned carriers will nerf **** too is correct
T-Mobile's Sensation has 7 something mb of ram, the evo's same hardware/phone almost but has 1 gb
Carriers lower the amount of ram to save money all the time, among other things, could result in a worse phone
I don't understand why Samsung insists on not making a good qHD screen tho with it....
Click to expand...
Click to collapse
depends on the person. I personally think that super amoled screens are over saturated, I looked at the droid charge at verizon store and I really didn't like how the colors were on the icons. Yes its a beautiful screen but I myself prefer slcd better, just because it seems like the colors are more naturals on it.
BlueGoldAce said:
I always though the benchmarks where due to the way the cores are set up/different. But I clocked my evo 3d to 1.8Ghz and ran quadrant on the performance governor. The highest I got was 2985. Now if I am correct, both cores should be maxed out due to the governor set by setcpu. If this is the case, the gs2 still beats the evo3d easily, even with the evo3d clocked to 1.8ghz.
I don't feel like this can be correct, but maybe it is? Maybe the exynos are that much better?
Or could it really be that sense 3.0 is a hog? I mean, with the hero, once it is on asop, you see a near 100% increase in scores.
Click to expand...
Click to collapse
sense probably has something to do with it. It is a resource hog, although it runs smooth, it does eat up a lot of resources.
BlueGoldAce said:
I always though the benchmarks where due to the way the cores are set up/different. But I clocked my evo 3d to 1.8Ghz and ran quadrant on the performance governor. The highest I got was 2985. Now if I am correct, both cores should be maxed out due to the governor set by setcpu. If this is the case, the gs2 still beats the evo3d easily, even with the evo3d clocked to 1.8ghz.
I don't feel like this can be correct, but maybe it is? Maybe the exynos are that much better?
Or could it really be that sense 3.0 is a hog? I mean, with the hero, once it is on asop, you see a near 100% increase in scores.
Click to expand...
Click to collapse
I had the SGS2 for 3 weeks and the benchmarks do NOT reflect real world performance. The SGS2 did download web pages faster...by like maybe 3-4 seconds.
The build btw, feels way better than the current ones out...stronger plastic, matted back.
Even the gorilla glass website says our device does not have Gorilla Glass on it.
Love the review.
nkd said:
on every day use, you cant see a difference between a really good single core vs dual core.
Click to expand...
Click to collapse
let me stop you there, and just say you're full of it
This phone has a dual core 1.2 ghz, just came from the Infuse on AT&T with a 1.2 ghz single core. it's fine most of the time, but has annoying lags when it's doing things (installing apps comes to mind, among some others)
The evo's really yet to ever stutter or lag like the Infuse did.
rockky said:
I had the SGS2 for 3 weeks and the benchmarks do NOT reflect real world performance. The SGS2 did download web pages faster...by like maybe 3-4 seconds.
The build btw, feels way better than the current ones out...stronger plastic, matted back.
Click to expand...
Click to collapse
Do you have the evo3d now? How would you compare/rate both phones?
BlueGoldAce said:
Do you have the evo3d now? How would you compare/rate both phones?
Click to expand...
Click to collapse
It really is preference. I am looking for camera quality and SGS2 is better there so I might be moving there. The speaker quality is really good on S2 also.
But some like the heavier/build quality of HTC better. Many don't like the super amoled Samsung screens. If you love Sense its Htc. (I use launcherPro). The dev community will be stronger undoubtedly for Evo...(although SG2 was rooted immediately.)
But...it is megalight, powerful and yep, sexy.
All that said...might not see it released for a while...some release issues were reported yesterday. Google around.
This might be a bit of good news. Not so sure how much the 720p LCD of the EVO 4G LTE will affect some of these test but I'm hoping not too far off from what the S gets. Sorry if this has been posted. Leaving work and didnt have much time to search.
http://www.engadget.com/2012/04/05/htc-one-x-vs-one-s/
Another good news. ATT version with S4 benchmarked! So higher screen res doesnt seem to hinder performance. Great news indeed.
http://phandroid.com/2012/04/10/att...k-gauntlet-dual-core-processor-holds-its-own/
Anyone see an article saying the Evo 4g LTE or whatever the evo one is named may actually be better than the one series phones?
Sent from my HTC EVO 3D X515m using xda premium
have u not seen the benchmarks of the s4 pwning tegra 3 single thread and multi thread but then again benches dont really matter..tegra is usually good for gaming
JoelZ9614 said:
have u not seen the benchmarks of the s4 pwning tegra 3 single thread and multi thread but then again benches dont really matter..tegra is usually good for gaming
Click to expand...
Click to collapse
So what about the heat? Which processor is the best? Overheat will effect gaming performance.
More comparison with video. I notice some very minor stuttering in the One S when he is swiping screens. Again I really hope this doesnt translate over to the Evo LTE. The T3 is really smooth on the other hand. I have a Transformer Prime and its very smooth as well.
http://www.intomobile.com/2012/04/05/htc-one-comparison/
Sent from my Transformer Prime TF201 using Tapatalk
More benchmark comparison
http://www.slashgear.com/htc-one-x-vs-htc-one-s-benchmarking-war-03221385/
Sent from my Transformer Prime TF201 using Tapatalk
aznmode said:
This might be a bit of good news. Not so sure how much the 720p LCD of the EVO 4G LTE will affect some of these test but I'm hoping not too far off from what the S gets. Sorry if this has been posted. Leaving work and didnt have much time to search.
http://www.engadget.com/2012/04/05/htc-one-x-vs-one-s/
Click to expand...
Click to collapse
The Evo 4g LTE/HTC ONE XL are using MSM8960 which i think is better than the MSM8260A used in the ONE S. So they will handle the 720p resloution just fine.
I might be wrong though.
JoelZ9614 said:
have u not seen the benchmarks of the s4 pwning tegra 3 single thread and multi thread but then again benches dont really matter..tegra is usually good for gaming
Click to expand...
Click to collapse
This man gets it!
GPU wise the Tegra3 will win the battle...most else is on the S4 side or at most a draw between the 2
---------- Post added at 09:09 AM ---------- Previous post was at 08:24 AM ----------
mnomaanw said:
The Evo 4g LTE/HTC ONE XL are using MSM8960 which i think is better than the MSM8260A used in the ONE S. So they will handle the 720p resloution just fine.
I might be wrong though.
Click to expand...
Click to collapse
Not exactly right here...
The MSM8260A and MSM8960 are the exact same top to bottom except in one area, and thats the Modem supported. The 8960 is more of a world more chip supporting LTE, GSM/EDGE/GPRS, EVDO TD-SCDMA, DC-HSPA+, among a few others...while the 8260A only supports UMTS(DC-HSPA+, TD-SCDMA)...
clearer break down of all their SoC's is here:
https://developer.qualcomm.com/sites/default/files/snapdragon-specs.pdf
Though comparing the performance of the OneS to the OneX is not really going to give us a solid idea of what to expect out of the EVO 4G LTE...Display differences b/w the two devices would effect performance some. The best idea will come from the OneXL.
wow...world band support...
too bad the EVO 4G LTE doesn't seem to use a sim card. otherwise i'd be ordering mine asaaaaaaap!
Sorry, but most benchmarks out there seem to be really weird.
Antutu reports what we would expect, namely almost twice as good results for the 4cores. That's integer and float. This result makes perfect sense.
Also, for tests that use only a single core, the two are head to head with an edge for the Qualcomm.
So how in the world should the Qualcomm keep up with the Tegra in other multicore benchmarks ?
Here is something to consider:
The benchmarks claiming the same multicore performance of the two only use two cores. Not all 4.
Why? Because at the time of programming them, 4 cores did not exist and the programs just did the easy thing. Making code multi-threaded is difficult by itself. But making code for an arbitrary number of cores is even more difficult.
Only antutu does it.
There is no other explanation.
Face it.
Sent from my Nexus S using XDA
hachiroku said:
wow...world band support...
too bad the EVO 4G LTE doesn't seem to use a sim card. otherwise i'd be ordering mine asaaaaaaap!
Click to expand...
Click to collapse
dont read into that much. its only the "supported" bands for the chip. still requires antennae's for each band and all that.
---------- Post added at 11:12 AM ---------- Previous post was at 11:09 AM ----------
fuzzifikation said:
Sorry, but most benchmarks out there seem to be really weird.
Antutu reports what we would expect, namely almost twice as good results for the 4cores. That's integer and float. This result makes perfect sense.
Also, for tests that use only a single core, the two are head to head with an edge for the Qualcomm.
So how in the world should the Qualcomm keep up with the Tegra in other multicore benchmarks ?
Here is something to consider:
The benchmarks claiming the same multicore performance of the two only use two cores. Not all 4.
Why? Because at the time of programming them, 4 cores did not exist and the programs just did the easy thing. Making code multi-threaded is difficult by itself. But making code for an arbitrary number of cores is even more difficult.
Only antutu does it.
There is no other explanation.
Face it.
Sent from my Nexus S using XDA
Click to expand...
Click to collapse
28nm Cortex-A15 vs 45nm Cortex-A9...
part of what you say is true but dont discount the S4 stomping the Tegra3 in some areas. b/c its not about the # of cores, theres a bunch more to it than that...
I'm more worried about battery life and how it handles being in standby
FreEvo 2.3d
Sony Brava Engine d-_-b
dannyboyex said:
I'm more worried about battery life and how it handles being in standby
FreEvo 2.3d
Sony Brava Engine d-_-b
Click to expand...
Click to collapse
Battery life is better than evo 3d IMO, both tegra and s4 have better battery life than s2 and s3...................................................
People... the S4 is fast; deal with it.
I really have no doubt that either phone will be fast. Definitely will be faster than the Evo 3d. In my case I'm more concern of how smooth its going to perform when it comes to scrolling. I think thats one area where android has not excelled yet and the tegra 3 IMO is the closest I've seen it, based on experience with my prime. Lots of talk went into how fast the Evo 3d processor was too before it came out and it was the reason I bought it and never bothered waiting for other devices. But the Evo 3d stock wasn't what I expected. A few months later the Samsung Epic touch came out and it was fast and smooth and handled multiple flash video on 1 page at a time without any problem. The other concern I have with Evo LTE is I will be most likely be using the device over hdmi. Not sure how well the S4 will perform. The demo Qualcomm had at CES was laggy (went to CES). My primes tegra 3 have proven to me that it can handle that very very well. It's super smooth and fast and has no degradation in performance and smoothness whatsoever over HDMI.
But I posted the OP to show that S4 is fast in hopes that the 720p screen will not hinder performance or smoothness as much. I'm really hoping so.
Ok now I'm convinced S4 will be on par with T3. See link below. The only thing we wont see are Tegra 3 optimized games.
http://phandroid.com/2012/04/10/att...k-gauntlet-dual-core-processor-holds-its-own/
From my understanding, the Tegra 3 is not LTE compatible. That seems to be the driving factor.
http://www.theverge.com/2012/2/29/2832546/nvidia-tegra-3-lte-problem-mwc-2012-report
coal686 said:
From my understanding, the Tegra 3 is not LTE compatible. That seems to be the driving factor.
http://www.theverge.com/2012/2/29/2832546/nvidia-tegra-3-lte-problem-mwc-2012-report
Click to expand...
Click to collapse
Correct that's why ATT and sprint will get s4. I really wanted a t3 because my asus prime has it and its the fastest and smoothest tablet ive seen. But after seeing more videos of the s4 in action i think itll be just as great.
Sent from my Transformer Prime TF201 using Tapatalk 2
Nice battery comparison. But I wonder how much the s4 screen came in to play. Roughly 1 hour and 22 minutes of test left the S with 84% battery and the One X with 69%.
url]http://www.youtube.com/watch?v=VPP-CSkMrf8[/url]
Nice battery comparison. But I wonder how much the s4 screen came in to play. Roughly 1 hour and 22 minutes of test left the S with 84% battery and the One X with 69%.
http://www.laptopmag.com/review/tablets/google-nexus-10.aspx
Some of my concerns seem to be true, the enormous resolution zaps a lot of the Mali 604s GPU power in games and graphics benchmarks making it at or below 3d performance of the older tf700. I just ask myself is such a high resolution so necessary? 1080p could've done it IMO, and it would have increased performance a lot in games. What's the point of a crystal clear resolution if the games are choppy?
The Exynos CPU score is great though and it'll be a good tablet for people wanting just a nice internet experience and viewing movies I'm sure will be great. Just disappointed about the 3D performance.
Sent from my Nexus 7 using Tapatalk 2
Toyeboy said:
http://www.laptopmag.com/review/tablets/google-nexus-10.aspx
Some of my concerns seem to be true, the enormous resolution zaps a lot of the Mali 604s GPU power in games and graphics benchmarks making it at or below 3d performance of the older tf700. I just ask myself is such a high resolution so necessary? 1080p could've done it IMO, and it would have increased performance a lot in games. What's the point of a crystal clear resolution if the games are choppy?
The Exynos CPU score is great though and it'll be a good tablet for people wanting just a nice internet experience and viewing movies I'm sure will be great. Just disappointed about the 3D performance.
Sent from my Nexus 7 using Tapatalk 2
Click to expand...
Click to collapse
We have a review thread already. All reviews are done on pre-release SW therefore not giving final scores.
BoneXDA said:
We have a review thread already. All reviews are done on pre-release SW therefore not giving final scores.
Click to expand...
Click to collapse
I don't buy that. Why would pre-release software be slower or return poorer benchmark results when there are only 10 days to go before launch?
Kookas said:
I don't buy that. Why would pre-release software be slower or return poorer benchmark results when there are only 10 days to go before launch?
Click to expand...
Click to collapse
Firstly, I do computer science at uni and depending what software design methodology Google uses, software optimization/debugging is generally the last step, getting everything working is a lot more important. Also, for a company like google, 10 days can make a huge difference, but we don't know how old that prerelease software and hardware is. For example andy rubin was using a prerelease nexus 4 which had lock screen widgets and he stressed that it wasn't the final product. Yet these reviews devices don't have lock screen widgets among other features which says these review units are older than the one andy rubin was using last week. Therefore using deduction, we could guess that these devices that the reviews are using could be running software from a week or two ago at a min, and add 10 days from now till the 13th. Lets just say 20 days is a hell of a long time in software development terms, especially for a company with the resources of google. Hope that helps
Yeah I think the reviews are going to be very close to the release versions which are most likely already packaged and ready for retail considering they're releasing November 13th. I hope gpu performance improves honestly or maybe if developers can run games at a lower resolution and upscale or something.
Still performance will be smooth and crystal clear outside of demanding games.
Sent from my Triumph using Tapatalk 2
Toyeboy said:
Yeah I think the reviews are going to be very close to the release versions which are most likely already packaged and ready for retail considering they're releasing November 13th. I hope gpu performance improves honestly or maybe if developers can run games at a lower resolution and upscale or something.
Still performance will be smooth and crystal clear outside of demanding games.
Sent from my Triumph using Tapatalk 2
Click to expand...
Click to collapse
Yeah, that's true. You could use 4 pixels for every 'real' pixel and still have a sharper image than any other tab out there.
But danielsf has a point, and knows a lot more about it than I do. 20 days is a lot for a team of 20-odd people working a 9-5er for 16 of those days.
via Tapatalk
Who says they're even working 9-5? Many software companies set a deadline for their developers and the developers work however they want so long as it is done. My friend works writing tax software and he will work 30 hour weeks for most of the year but towards October the usual programmer is working 50+ hour weeks so that they'll hit their deadline in mid December.
crash822 said:
Who says they're even working 9-5? Many software companies set a deadline for their developers and the developers work however they want so long as it is done. My friend works writing tax software and he will work 30 hour weeks for most of the year but towards October the usual programmer is working 50+ hour weeks so that they'll hit their deadline in mid December.
Click to expand...
Click to collapse
True, that does sound like a 'big tech firm' thing to do, especially one with a personal project funding thing.
via Tapatalk
Toyeboy said:
What's the point of a crystal clear resolution if the games are choppy?
Click to expand...
Click to collapse
So the adults can have a high end device and use it for its intended purpose.
People should stop feeding into benchmarks so much. I bet it'll still play 3d games just fine. If my lower clocked tegra3 nexus7 can play 3d games fine(no chop or slowdown) then nexus 10 will push those games with ease(even with higher resolution).
I never base my purchases off of benchmarks. I take in all the reviews then try for myself. The best indicator of performance is you using the device in your usage scenarios.
Toyeboy said:
http://www.laptopmag.com/review/tablets/google-nexus-10.aspx
Some of my concerns seem to be true, the enormous resolution zaps a lot of the Mali 604s GPU power in games and graphics benchmarks making it at or below 3d performance of the older tf700. I just ask myself is such a high resolution so necessary? 1080p could've done it IMO, and it would have increased performance a lot in games. What's the point of a crystal clear resolution if the games are choppy?
The Exynos CPU score is great though and it'll be a good tablet for people wanting just a nice internet experience and viewing movies I'm sure will be great. Just disappointed about the 3D performance.
Sent from my Nexus 7 using Tapatalk 2
Click to expand...
Click to collapse
I don't trust these new benchmarks.
GLBench is somewhat more accurate, in my opinion.
Look at those scores, they should be much more indicative.
Hemidroids said:
So the adults can have a high end device and use it for its intended purpose.
Click to expand...
Click to collapse
who's to say what it's intended purpose is? It's different for everyone and games happen to be very important for me. I look at every aspect of a device before I buy.
Just curious, but why does benchmarks show lower score for this SoC on tne N10 than eg. the Chromebook with the same SoC?
Correct me if I am wrong.
Sent from my GT-I9100 using xda app-developers app
borchgrevink said:
Just curious, but why does benchmarks show lower score for this SoC on tne N10 than eg. the Chromebook with the same SoC?
Correct me if I am wrong.
Sent from my GT-I9100 using xda app-developers app
Click to expand...
Click to collapse
Difference in chrome os and android os & power constraints on a tablet vs a notebook.
slide83 said:
Difference in chrome os and android os & power constraints on a tablet vs a notebook.
Click to expand...
Click to collapse
Could we hope for a more optimised Android code increasing performance?
borchgrevink said:
Could we hope for a more optimised Android code increasing performance?
Click to expand...
Click to collapse
Yes. Often this is the case, particularly with a device that is changing the paradigm as much as the Nexus 10 is.
My attitude toward these pre-release benchmarks is this. Take a casual look at them and note what they say, but don't read too much into them. Wait for the full release version and see how it actually performs. If there are a lot of people *****ing about something after it is released, assess how easy that would be to be fixed and the likelihood that it will be. If it is something that will likely be addressed in the near future, great. If not, then take a closer look before purchasing.
While I would love to have a Nexus 10 yesterday, I'm willing to wait and let the early adopters take a few arrows to see how this device actually performs before deciding if I will spend some of my hard earned money on it. :victory:
I think I'm just concerned about how it will affect the fluidity of the user interface. All the resolution in the world isn't worth it if you can't pinch and zoom or scroll a webpage/magazine app without choppiness.
They made a big deal about Project butter I hope this isn't a regression because the gpu can't keep up.
Sent from my SGH-T999 using Tapatalk 2
xManMythLegend said:
I think I'm just concerned about how it will affect the fluidity of the user interface. All the resolution in the world isn't worth it if you can't pinch and zoom or scroll a webpage/magazine app without choppiness.
They made a big deal about Project butter I hope this isn't a regression because the gpu can't keep up.
Sent from my SGH-T999 using Tapatalk 2
Click to expand...
Click to collapse
I agree, but I've heard its pretty smooth n the menus.
Sent from my Nexus 7 using Tapatalk 2
Maybe Google/Samsung intentionally set the expectation low and positively surprise us on release day like turning off debugging which tends to zap performance.
mi7chy said:
Maybe Google/Samsung intentionally set the expectation low and positively surprise us on release day like turning off debugging which tends to zap performance.
Click to expand...
Click to collapse
Don't think they'd intentionally do such a thing, but they did release these beta devices for testing knowing the final 4.2 SW is not ready yet. It's all a bit rushed so I don't expect significant improvement out of the box (but an overall smooth performance in both at browsing and gaming), and since Samsung is doubling GPU cores for the S4, I feel the Nexus 10's bottleneck will always be the quad-core Mali T604, despite performance optimizations surely to come. That 4M screen is a performance hungry beast.
Hi guys, just wondering how good it is in terms of games?
Just about to buy one and just wanted your opinion. Thanks for you time.
seventhworl said:
Hi guys, just wondering how good it is in terms of games?
Just about to buy one and just wanted your opinion. Thanks for you time.
Click to expand...
Click to collapse
Well, I'm a daily player with a number of games on my GPad, and for now it has been running everything flawlessly. I'm currently playing Asphalt 8 in high details, and Amazing Spiderman 2 : both are just perfect and beautiful. I expect it will be some time before you find something thaht lags on it.
Have you tried need for speed most wanted on it. Cause that game is very laggy on mine, Rest work fine.
Sent from my GT-I9505 using XDA Free mobile app
5warafat1 said:
Have you tried need for speed most wanted on it. Cause that game is very laggy on mine,
Click to expand...
Click to collapse
Weird, it's really, really buttersmooth on mine.
Modern Combat 4, Deus Ex 2, Frontline Commando 2, Real Racing 3, Call of Duty Strike Team - works great.
Mercvtio said:
Weird, it's really, really buttersmooth on mine.
Click to expand...
Click to collapse
That game was smooth on mine as well.
Though I only had the LG G Pad a few days before returning it (did not like the quality of sound output and the display was blinking - a fault?)
Hope this topic gets more informative. Games are the only thing that is preventing me from buying this tablet. I haven't personally run any benchmark on the device but when I go to gfxbench.com and compare this tablet with other devices is seems like it will have a very short future on gaming. Because not only Galaxy S4 with the same chipset scores higher on the graphical benchmarks but even also Nexus 7 2013 is very much ahead of this tablet.
In the beginning I was going between two tablets; Tab Pro 8.4 and G Pad 8.3 because of Tab Pro's superior screen (everyone was saying G Pad was the dimmest screen they have ever seen). After seing that G Pad has actually better color accuracy and no downsides at all, I wanted to go for G Pad. However benchmarks are really scaring me on the GPU side (I don' care about CPU benchs because they are unreliable however GPU ones are very relistic).
Since I haven't even personally see a G Pad outside of a store I have no chance to run some benchmarks or measure any potential. So, I am trying to understand how even Nexus 7 can do better than G Pad. Could it be that the tests were made from the time of 4.4.2? Could any of you possibly run some tests on CM 11 and latest stock environments to share results? I know Adreno 330 has some technological advantages but it should not be that much better on the Tab Pro 8.4's resolution against G Pad..
Forget synthetic benchmarks, they are just for epeen bragging rights, and a lot of the time they aren't even accurate (see Samsung and HTC faking theirs) what games do you want tested?
I've never cared about benchmarks either and I must say I did not experience any problems with games I threw at LG G Pad.
With that said - and you guys are going to hate me for this - as far as tablet gaming is concerned, nothing beats the iPad. Whatever we may think about Apple's 'walled garden' ecosystem, the variety, the sheer amount of games available there is astounding. And don't get me started on optimization and regular updates to games....
Please understand that I'm not trolling, I've used many various Android tablets and loved them all. But for pure gaming purposes, I'd be hard-pressed to find a better tablet. Sorry.
Nothing to be sorry about, devs go where the money is, the Android ecosystem is different and follows the freemium approach more. You can make more money from less sales on the iStore with a $3 game than more android installs of a freemium game if barely anyone buys the extras.
I don't think that this tablet is a good performer in games.
My friend has a galaxy s3 and in almost every game his framerates and graphics are better than my g pad.
A test sample for me is carX drift (that i really love playing), he is able to run it without any framerate drops on high graphics
while my g pad is running it on low with some framerate drops (if i switch to medium or high graphics it is still playable,but the framerate is sub 25fps)
And on amazing spiderman 2 it is not smooth by any mean, it is just playable (20fps seems to my eyes)
alekos1992 said:
I don't think that this tablet is a good performer in games.
My friend has a galaxy s3 and in almost every game his framerates and graphics are better than my g pad.
A test sample for me is carX drift (that i really love playing), he is able to run it without any framerate drops on high graphics
while my g pad is running it on low with some framerate drops (if i switch to medium or high graphics it is still playable,but the framerate is sub 25fps)
And on amazing spiderman 2 it is not smooth by any mean, it is just playable (20fps seems to my eyes)
Click to expand...
Click to collapse
That seems like a problem. I know Google would never let its own device go under a mission impossible so I believe GP Edition of the tablet has either better optimised or newer drivers for GPU. I mean, look at Nexus 7 2013, even it has better performance when it comes to 3d gaming. This device is sold on Goole's vey own web page with Google's name on it. There is no way it should be lagging in games in the Google's own world without even turning 1 year old..
So it is either worse driver or underclocked GPU?
Has anyone tried OCing or using CM 11 since they "may" have better/newer drivers for GPU?
---------- Post added at 03:25 PM ---------- Previous post was at 03:20 PM ----------
BTW;
I persdonally don't think GPU benchmarks are inaccurate. We want to see how our devices would treat games and they give us a scientifically equal test environment.. If we were to compare two devices with the same game, two person -side by side- would have to play excatly the same way which is impossible. So consider the 3d benchmarks as playing and measuring a cinematic part of a game..
Of course this idea is valid until some *clever* company starts cheating..
Still, something is not with G Pad's GPU, it can't even beat Galaxy S4..
alekos1992 said:
I don't think that this tablet is a good performer in games.
My friend has a galaxy s3 and in almost every game his framerates and graphics are better than my g pad.
A test sample for me is carX drift (that i really love playing), he is able to run it without any framerate drops on high graphics
while my g pad is running it on low with some framerate drops (if i switch to medium or high graphics it is still playable,but the framerate is sub 25fps)
And on amazing spiderman 2 it is not smooth by any mean, it is just playable (20fps seems to my eyes)
Click to expand...
Click to collapse
You do realise the GPad has a much higher resolution than the S3?
I own both the GPad and an S3, I will try this out tonight and let you know.
---------- Post added at 01:37 PM ---------- Previous post was at 01:32 PM ----------
_delice_doluca_ said:
I persdonally don't think GPU benchmarks are inaccurate. We want to see how our devices would treat games and they give us a scientifically equal test environment.. If we were to compare two devices with the same game, two person -side by side- would have to play excatly the same way which is impossible. So consider the 3d benchmarks as playing and measuring a cinematic part of a game..
Of course this idea is valid until some *clever* company starts cheating..
Click to expand...
Click to collapse
It depends how you are testing, just running a timed demo on a commercial game would be more accurate than running 3DMark or something along those lines.
Companies have been caught out and basically letting the phone/tablet/whatever run flat out when it detects it's running a benchmark app, which they normally wouldn't do due to heat and battery problems.
Tachikoma_kun said:
You do realise the GPad has a much higher resolution than the S3?
I own both the GPad and an S3, I will try this out tonight and let you know.
---------- Post added at 01:37 PM ---------- Previous post was at 01:32 PM ----------
It depends how you are testing, just running a timed demo on a commercial game would be more accurate than running 3DMark or something along those lines.
Companies have been caught out and basically letting the phone/tablet/whatever run flat out when it detects it's running a benchmark app, which they normally wouldn't do due to heat and battery problems.
Click to expand...
Click to collapse
You are right. However, Google is not one of those companies. So this tablet -at least- should have gotten on par with the Nexus 7 2013 level (of course with the same software version). Like I said, I have never seen a Google Play Edition of G Pad with my own eyes and never seen a benchmark run on it either.. But I believe if run, the benchmarks should have been at least on par or higher than Nexus 7 2013, depending on S4 Pro and S600 differences. I know Nexus 7 has a special version of S4 Pro and it probably effects the Adreno 320's performance too but still this tablet has an S600 chipset.
If you check gfxbench.com, and compare Nexus 7 2013, Samsung Galaxy Tap Pro 8.4 and G Pad 8.3 to each other you definetely can say G Pad should not result like that (talking about "onscreen" values, the others are just tricky values).
If I may ask, would you mind running some tests on your tablet to see how it is on, onscreen tests on gfx bench (If you are on KitKat of course)?
And also, If someone has some time to run the same benchmark on the CM 11 would be very much welcome and I would be very much thankfull If the results are so different than stock LG software's then we will know that something is wrong on the driver side (outdated/misconfigured etc..)
Edit: bad English
_delice_doluca_ said:
If I may ask, would you mind running some tests on your tablet to see how it is on, onscreen tests on gfx bench (If you are on KitKat of course)?
Click to expand...
Click to collapse
As long as it doesn't cost me money, I am happy to run whatever you want. I'm using stock rooted KitKat, LG launcher, pretty much as standard as it can get.
Let me know what you want me to run and I will do it over the weekend (if I remember :laugh
Tachikoma_kun said:
As long as it doesn't cost me money, I am happy to run whatever you want. I'm using stock rooted KitKat, LG launcher, pretty much as standard as it can get.
Let me know what you want me to run and I will do it over the weekend (if I remember :laugh
Click to expand...
Click to collapse
Thank you
I guess 2 apps would be more than enough.
One is gfxbench 3.0, after you download (Wi-Fi recommended) and pass the welcome screen just click on "tests" and select everything except "off screen" ones then let it run. It will take like 5 mins. (You may want to try battery life test but it is not necessary)
You can go to gfxbench.com and chose your device to compare with your results. For example my Nexus 4's results are very much same with the web page's but my previous Note 2 was so different.
The other is 3D mark. A cross platform test so we can compare the tablet to any tablet or phone.. Simply run the tests and it will display results side by side.
I know I am asking for too much but please run all tests twice and run them as all other apps are closed. If previous results are significantly lower than seconds then it is a driver or stability issue, if the opposite there is thermal throttle, if they are close enough then those are the results we need to achieve..
Thank you so much for helping. That is too important to me.
Sent from my Nexus 4 using XDA Free mobile app
_delice_doluca_ said:
Thank you
I guess 2 apps would be more than enough.
One is gfxbench 3.0, after you download (Wi-Fi recommended) and pass the welcome screen just click on "tests" and select everything except "off screen" ones then let it run. It will take like 5 mins. (You may want to try battery life test but it is not necessary)
You can go to gfxbench.com and chose your device to compare with your results. For example my Nexus 4's results are very much same with the web page's but my previous Note 2 was so different.
The other is 3D mark. A cross platform test so we can compare the tablet to any tablet or phone.. Simply run the tests and it will display results side by side.
I know I am asking for too much but please run all tests twice and run them as all other apps are closed. If previous results are significantly lower than seconds then it is a driver or stability issue, if the opposite there is thermal throttle, if they are close enough then those are the results we need to achieve..
Thank you so much for helping. That is too important to me.
Sent from my Nexus 4 using XDA Free mobile app
Click to expand...
Click to collapse
Gfxbench done
3dmark done
There does seem to be some thermal throttling going on.
Tachikoma_kun said:
Gfxbench done
Click to expand...
Click to collapse
Thanks a lot.
Weird, there is noticible change between two tests. GPU benchmarks should not change that much unless there are thermal issues.
Another thing is the results are higher than gfxbench.com says, obviously 4.4.2 brings some boost over 4.2.2.. This way it is on par or close-enough with Nexus 7 2013.
Thank you again
Just updated with 3d mark, does look a lot like there's thermal throttling going on. The device does get warm under load but I out that down to having a metal back.
:>
Gangster vegas works fine :>
My question is why? 4 Gigs of RAM, 6 core 808 processor...this device should be smoking fast. Is it all because of the ultra-high PPI display?
It's because of stock, that's why.
My Nexus 9 kicks some benchmark behind, but its a laggy piece of junk, so yeah, who cares what benchmarks say. I'm more interested in real life performance and every video I've seen shows a very smooth phone with minimal lag.
chrisrotolo said:
It's because of stock, that's why.
Click to expand...
Click to collapse
My G3 had incredible benchmark scores when tested stock. "Because stock" doesn't make sense to me. (No offense.)
yourbrotherrex said:
My G3 had incredible benchmark scores when tested stock. "Because stock" doesn't make sense to me. (No offense.)
Click to expand...
Click to collapse
Yet the G3 is a very laggy device. I'm typing this using one. What's the fascination with a high benchmark when it doesn't translate into everyday use? Would you rather have a device that benchmarks insanely like a Nexus 9 and lags like crazy or a device that benchmarks poorly and doesn't lag at all like a nexus 5? I think the last phone I owned that I actually ran a benchmark on was an HTC Evo. The numbers have been shown to be useless in real world performance of a device.
Sent from my LG-D851 using Tapatalk
? My G3 doesn't lag whatsoever.
People still care about benchmark scores? Who's winning prizes from this? Its almost 2016, either your phone is running good or its not. Benchmarks have been doctored in the past anyway
yourbrotherrex said:
? My G3 doesn't lag whatsoever.
Click to expand...
Click to collapse
Then I'll just say we have a different definition of lag. Our dept issued the G3 as the workphone, and in setting up 48 of the devices, I can say without pause that every single one is a laggy phone. Grab a Nexus 5 and play with that, then go back to your G3 and say it has no lag whatsoever. Is it unusable? Not at all, but it certainly lags and to say no lag whatsoever, we clearly have different meanings of lag.
---------- Post added at 06:18 AM ---------- Previous post was at 06:17 AM ----------
bigstunta101 said:
People still care about benchmark scores? Who's winning prizes from this? Its almost 2016, either your phone is running good or its not. Benchmarks have been doctored in the past anyway
Click to expand...
Click to collapse
It's all about the e-peen man! My Nexus 9 blows the iPad away in benchmarks!!!!!11 Put the two side by side and see which one operates like a champ and which one operates like a chump. I'll give you a spoiler, the N9 is the chump.
I guess manufacturers do a huge effort to have best marks in the most common benchmark apps. It's like Volkswagen: Software optimized for test situations. I wouldn't care too much about it. After all I've seen, it's quite snappy.
Lol great comparison @DeGon
What scores have you been getting?