I'm trying to understand how a G2 that uses the same Snapdragon 800 and Adreno 330 could be faster than Nexus5.
Is seems that G2 is faster than N5.
Ridiculusly there are many test where Galaxy S4 beats N5.
https://www.youtube.com/watch?v=S3uZdVsND1E&feature=youtube_gdata_player
GS4 runs on a Snapdragon 600, how can be faster than N5?
Not sure why you are comparing benchmarks on phones http://www.anandtech.com/show/7384/
The only thing I care is real world performance and thermal throttling.
How is it running Real Racing 3 and other super-demanding games?
& this is why benchmarks are pure BS.
Here is a more realistic comparison in speed between the Galaxy S4 & the N5.
https://www.youtube.com/watch?v=p-F6bJ218Bc&feature=youtube_gdata_player
Sent from my Nexus 4 using XDA-Premium 4
ste1164 said:
Not sure why you are comparing benchmarks on phones http://www.anandtech.com/show/7384/
Click to expand...
Click to collapse
very funny, I lost that article.
Android manufacturers are known to optimize for specific benchmarks, Anandtech did an article on this. I will only trust real world performance and analysis by reputable tech sites like AT.
sblantipodi said:
I'm trying to understand how a G2 that uses the same Snapdragon 800 and Adreno 330 could be faster than Nexus5.
Is seems that G2 is faster than N5.
Ridiculusly there are many test where Galaxy S4 beats N5.
https://www.youtube.com/watch?v=S3uZdVsND1E&feature=youtube_gdata_player
GS4 runs on a Snapdragon 600, how can be faster than N5?
Click to expand...
Click to collapse
Google has always throttled their devices. So when you push it to the extremes in benchmarks for 5 min straight it throttles back the CPU therefore giving a lower score. Like said real world performance is what matters.
Sent from my Nexus 4 using XDA Premium 4 mobile app
Even a 'cheating' S600 phones shouldnt be near an S800 phone. The N5 does seem to have very aggressive throttling indeed, indeed my HTC One beats my N5 in just about all benchmarks. Something im sure the devs will fix soon enough.
ChrisM75 said:
Even a 'cheating' S600 phones shouldnt be near an S800 phone. The N5 does seem to have very aggressive throttling indeed, indeed my HTC One beats my N5 in just about all benchmarks. Something im sure the devs will fix soon enough.
Click to expand...
Click to collapse
It doesn't have any throttling at all.
They just don't play the benchmark game like OEMs do.
The N5 is noticeably faster than an S4 in all tasks yet the S4 scores higher on Antutu for example.
benchmark apps are pure BS.
Which is better - having the fastest smoothest phone available or being slower at everything yet scoring higher on a benchmark app?
People need to get their priorities right.
Sent from my Nexus 4 using XDA-Premium 4
chrisjcks said:
It doesn't have any throttling at all.
They just don't play the benchmark game like OEMs do.
The N5 is noticeably faster than an S4 in all tasks yet the S4 scores higher on Antutu for example.
benchmark apps are pure BS.
Which is better - having the fastest smoothest phone available or being slower at everything yet scoring higher on a benchmark app?
People need to get their priorities right.
Sent from my Nexus 4 using XDA-Premium 4
Click to expand...
Click to collapse
ALL phones have thermal throttling, thats not up for debate, what is up for debate is how aggressively they set the limits.
What many here fail to realise is that the 'cheating' that goes on is just thermal management tricks, nothing more than that. Samsung and the others have programs that detect benchmarks launching, and then set the thermal management to very light limits. In the case of the S4 they clock the GPU to the maximum rated limit and dont throttle it down (533MHz), whereas its normally limited to 480 for thermal management reasons. 533 is not an overclock, 480 is an underclock.
An N5 should be faster than an S4 even if the S4 is at 533, so either Google is heavily throttling the N5, or its got some serious optimisation work to do.
Benchmarks mean fcuk all and too be honest if that's all the op cares about them the nexus ain't for him.
Sent from my Nexus 5 using Tapatalk
ChrisM75 said:
ALL phones have thermal throttling, thats not up for debate, what is up for debate is how aggressively they set the limits.
What many here fail to realise is that the 'cheating' that goes on is just thermal management tricks, nothing more than that. Samsung and the others have programs that detect benchmarks launching, and then set the thermal management to very light limits. In the case of the S4 they clock the GPU to the maximum rated limit and dont throttle it down (533MHz), whereas its normally limited to 480 for thermal management reasons. 533 is not an overclock, 480 is an underclock.
An N5 should be faster than an S4 even if the S4 is at 533, so either Google is heavily throttling the N5, or its got some serious optimisation work to do.
Click to expand...
Click to collapse
Again - look at the video link I posted above against the S4.
Which is better - faster phone throughout or a nice pretty score in a free benchmark app?
When will people learn - these apps are absolute junk and in No Way do they reflect the speed of the device or the power of the internals inside.
It seriously sounds like you'd accept a slower less powerful phone as long as it scored higher on the pretty charts in these apps.
Sent from my Nexus 4 using Tapatalk
Anandtech pointed out that OEMs like Samsung boost CPU & GPU clocks during benchmarks, that's why you get higher numbers.
For Gods sake, it's the same SoC.
sblantipodi said:
I'm trying to understand how a G2 that uses the same Snapdragon 800 and Adreno 330 could be faster than Nexus5.
Is seems that G2 is faster than N5.
Ridiculusly there are many test where Galaxy S4 beats N5.
https://www.youtube.com/watch?v=S3uZdVsND1E&feature=youtube_gdata_player
GS4 runs on a Snapdragon 600, how can be faster than N5?
Click to expand...
Click to collapse
Nexus 5 is absolutely faster than Galaxy S4 in real life usage, no doubt about it. When it comes to G2 its more even between the two.
chrisjcks said:
Again - look at the video link I posted above against the S4.
Which is better - faster phone throughout or a nice pretty score in a free benchmark app?
When will people learn - these apps are absolute junk and in No Way do they reflect the speed of the device or the power of the internals inside.
It seriously sounds like you'd accept a slower less powerful phone as long as it scored higher on the pretty charts in these apps.
Sent from my Nexus 4 using Tapatalk
Click to expand...
Click to collapse
Thats not what I said. Read again..
SOME benchmarks are purely about number crunching and the fact is the S800 should wipe the floor with the S600, if its not, something is going on.
While it doesnt matter if the device is smooth in real world usage it still points to the fact that the software needs a lot of optimisation to be done yet.
ChrisM75 said:
Thats not what I said. Read again..
SOME benchmarks are purely about number crunching and the fact is the S800 should wipe the floor with the S600, if its not, something is going on.
While it doesnt matter if the device is smooth in real world usage it still points to the fact that the software needs a lot of optimisation to be done yet.
Click to expand...
Click to collapse
Its not just about smoothness.
Go look at some comparisons of the N5 vs Galaxy S4.
N5 is faster at booting, browsing speed, smoothness, speed of loading of apps, gaming frame rates & loading speeds - basically EVERYTHING!
so you either believe the benchmark app or the actual speeds of the devices.
Simply put - you'd prefer a slower phone so long as it scores higher in these apps.
If you want one of these s600 phones like the S4 & ONE - go & get one! - but don't expect anything to be faster than the N5 just because these free benchmarking apps tell you so.
Sent from my Nexus 4 using XDA-Premium 4
You should change the title. " Galaxy S4 and G2 are faster than G2 "
Lol.
This is the Nexus 4 performance discussion all over again.
Other Manufacturers use specific Dalvic patches that grearly improve performance in benchmarks.
If you really want to compare performance of the SoC use something like Geekbench that runs native
code and not ontop of the Dalvic Virtual Maschine.
---------- Post added at 06:48 PM ---------- Previous post was at 06:46 PM ----------
aletto said:
You should change the title. " Galaxy S4 and G2 are faster than G2 "
Click to expand...
Click to collapse
It should be changed to "Galaxy S4 and G2 perform better then the N5 in useless Benchmarks that don't reflect real world performance"
The only thing obvious from the video comparing the N5 to the G2 is the on screen black levels.
Blacks seem blacker on the G2 in menus and in game, the N5's blacks are grayer.
Google doesn't seem to be interested in calibrating their Nexus line screens
Why does my golden retriever outperform my cat in the fishing dead ducks out of the pond test?
Sent from my Nexus 7 using Tapatalk 4
Related
hello guys..i heard that galaxy note and other samsung device are using an outdated GPU (Mali-400MP GPU)...so is it a little "fail" for our note to have an outdated GPU?plss give ur opinion.. thanks guys
..u can read the review about the GPU--> Here
It's so much faster than the sgx540 in the nexus it's ridiculous and since my choice was between those two I'm very happy with it.
Sent from my superior GT-N7000 using Tapatalk
Check out the real world performances. Mali 400 outclasses Adreno 220 easily.
The weakpoint of Mali is geometry performance, but it does not matter much with mobiles until now as mobile games are not geometry heavy.
On the other hand, the OpenGL ES 2.x performance and real world performance of Mali is excellent.
With the clock speed of exynos in Note which actually gives much better real world performance with Mali 400 than even SGS2, it runs circles around Adreno 220 powered devices like sensation and even SGX540 powered devices.
http://www.anandtech.com/show/4686/samsung-galaxy-s-2-international-review-the-best-redefined/17
The above review is of SGS2. And mind you the performance of note is much better than SGS2. It is one of the most balanced GPUs on market with great gaming as well as multimedia performance (which actually matters more to someone like me.)
Funkym0nkey said:
Check out the real world performances. Mali 400 outclasses Adreno 220 easily.
The weakpoint of Mali is geometry performance, but it does not matter much with mobiles until now as mobile games are not geometry heavy.
On the other hand, the OpenGL ES 2.x performance and real world performance of Mali is excellent.
With the clock speed of exynos in Note which actually gives much better real world performance with Mali 400 than even SGS2, it runs circles around Adreno 220 powered devices like sensation and even SGX540 powered devices.
http://www.anandtech.com/show/4686/samsung-galaxy-s-2-international-review-the-best-redefined/17
The above review is of SGS2. And mind you the performance of note is much better than SGS2. It is one of the most balanced GPUs on market with great gaming as well as multimedia performance (which actually matters more to someone like me.)
Click to expand...
Click to collapse
thanks for this info sir
although mali has been here for a very long time, it was well ahead of its time. and it still is i guess
anjath said:
although mali has been here for a very long time, it was well ahead of its time. and it still is i guess
Click to expand...
Click to collapse
Yeah well Scott Adams is wayyyyyyy past his heyday (heck, even being relevant).... haven't read him since 2007 or so, when he started dabbling in intelligent design woo and sexist claptrap...
for being a heavy mobile gamer
i can assure you that the mali 400 on the note does very well with the latest games (asphalt7, dead trigger to name a few) despite having to compute for a much higher resolution display than other phones...
and with a little overclocking (tegrak app or gl notecore kernel) gpu performance can get sky high.
best phone i ever got :victory:
GAME ON said:
hello guys..i heard that galaxy note and other samsung device are using an outdated GPU (Mali-400MP GPU)...so is it a little "fail" for our note to have an outdated GPU?plss give ur opinion.. thanks guys
..u can read the review about the GPU--> Here
Click to expand...
Click to collapse
The note was released ten months ago but still its gpu is better than all the others except sgs3 and and maybe one x..
Btw do you even own a note?? Did you every notice any lag in any game??
Whiskeyjack4855 said:
The note was released ten months ago but still its gpu is better than all the others except sgs3 and and maybe one x..
Btw do you even own a note?? Did you every notice any lag in any game??
Click to expand...
Click to collapse
The NOTE's and SGS3's GPU are the same.
However, the S3 is built on a smaller 32nm die-size, so it means it uses less space and less power for same performance. Samsung uses this advantage to clock the frequency much higher than the NOTE (which is built on a 45nm die).
Also, the S3 implements a new, updated driver for the gpu and squeezes more performance out. This was a same move Samsung made with the SGX540, which is also a very fast gpu. The original SGS was clocked real-low and had outdated drivers... after stealing the driver sources from the LG with OMAP 4440 SoC, the SGS (with 4.0.3) was performing in the same league as the 2011/2012 devices.
Kangal said:
The NOTE's and SGS3's GPU are the same.
However, the S3 is built on a smaller 32nm die-size, so it means it uses less space and less power for same performance. Samsung uses this advantage to clock the frequency much higher than the NOTE (which is built on a 45nm die).
Also, the S3 implements a new, updated driver for the gpu and squeezes more performance out. This was a same move Samsung made with the SGX540, which is also a very fast gpu. The original SGS was clocked real-low and had outdated drivers... after stealing the driver sources from the LG with OMAP 4440 SoC, the SGS (with 4.0.3) was performing in the same league as the 2011/2012 devices.
Click to expand...
Click to collapse
I know that that both the note and sgs3 has same gpu.. But the one in sgs3 its more powerful cause you said it's overclocked and has better drivers..
Btw do you know why the mali in sgs3 gets so high benchmark scores even wih the 720p screen? I mean is it all due to oc and better drivers?
Whiskeyjack4855 said:
I know that that both the note and sgs3 has same gpu.. But the one in sgs3 its more powerful cause you said it's overclocked and has better drivers..
Btw do you know why the mali in sgs3 gets so high benchmark scores even wih the 720p screen? I mean is it all due to oc and better drivers?
Click to expand...
Click to collapse
Better drivers + a little O'C makes the overwhelming difference.
I mean the NOTE comes with *only* 2 cores and *slow* gpu... after I customized it, its running toe-to-toe with the HTC One X (Tegra3).
An easier way to understand is to look at the new RIM PlayBook.
It's got the same processor as the Gnex (Galaxy Nexus) however its much much faster, especially in browsing. It decimates it. It even decimates the ASUS Transformer Prime Infinity (O'C Tegra3 + ICS).... or the Nexus7 (U'C Tegra3 + JBean).
You are only as fast as your slowest component. In the case of Android, its the high-level (slow) implemented software.
= Getting a faster soc with more cores and more ram doesn't really increase performance that much.
Some serious thread necromancy going on here!
Regards,
Dave
Sent from my GT-N7000 using Tapatalk 2
Kangal said:
Better drivers + a little O'C makes the overwhelming difference.
I mean the NOTE comes with *only* 2 cores and *slow* gpu... after I customized it, its running toe-to-toe with the HTC One X (Tegra3).
Click to expand...
Click to collapse
By toe to toe with the one x do you mean benchmarks or real life perfomance..
Hey one thing more..aren't you a engadget reader?
Whiskeyjack4855 said:
By toe to toe with the one x do you mean benchmarks or real life perfomance..
Hey one thing more..aren't you a engadget reader?
Click to expand...
Click to collapse
Both.
But I don't live by the benchmarks. I mean have you tried some of the HD Apps from TegraZone. On stock TouchWizz, the NOTE really struggles. With a custom setup, I don't get much/any problems.
Yeah, I do frequent engadget... also on heaps of other sites.
Kangal said:
Both.
But I don't live by the benchmarks. I mean have you tried some of the HD Apps from TegraZone. On stock TouchWizz, the NOTE really struggles. With a custom setup, I don't get much/any problems.
Yeah, I do frequent engadget... also on heaps of other sites.
Click to expand...
Click to collapse
Would you be kind enough to educate me about your setup?
Kangal said:
Better drivers + a little O'C makes the overwhelming difference.
I mean the NOTE comes with *only* 2 cores and *slow* gpu... after I customized it, its running toe-to-toe with the HTC One X (Tegra3).
An easier way to understand is to look at the new RIM PlayBook.
It's got the same processor as the Gnex (Galaxy Nexus) however its much much faster, especially in browsing. It decimates it. It even decimates the ASUS Transformer Prime Infinity (O'C Tegra3 + ICS).... or the Nexus7 (U'C Tegra3 + JBean).
You are only as fast as your slowest component. In the case of Android, its the high-level (slow) implemented software.
= Getting a faster soc with more cores and more ram doesn't really increase performance that much.
Click to expand...
Click to collapse
Exactly. Even though the Playbook has its many flaws (owned two both with screen/USB issues) it was a powerhouse. Multimedia was outstanding and web surfing was by far the fastest.
But the OS, QNX, is to thank for that. If the Playbook was running android it would be nothing out of the ordinary. As much as I love Android it really is not as efficiant as QNX
Sent from my GT-N7000 using xda app-developers app
anything on market today is outdated tommorow
Sent from my GT-N7000 using Tapatalk 2
Recently I had a doubt about if it's possible to unlock the 2 locked remaining cores in the Galaxy Note N7000? Because I realize that the Note only uses 2 of the 4 GPU cores... It's there a possibility to do this? How?
i think mali 400 is a good GPU because Note1 and Note2 using it. and really nice GPU for gaming
My buddy got a Droid DNA the other day. I immediately snatched it up to check it out. What a screen!
Anyway...ran a couple benchmarks and was AMAZED!
Quadrant = 8085
BenchmarkPi = ~280
0.0
Our phone's guts are the same! My Nexus 4 is even overclocked and doesn't come CLOSE...
Just out of curiosity...how is it that my rooted, ROM'd, overclocked Nexus gets its teeth kicked in by a stock Droid DNA?
rmp5s said:
My buddy got a Droid DNA the other day. I immediately snatched it up to check it out. What a screen!
Anyway...ran a couple benchmarks and was AMAZED!
Quadrant = 8085
BenchmarkPi = ~280
0.0
Our phone's guts are the same! My Nexus 4 is even overclocked and doesn't come CLOSE...
Just out of curiosity...how is it that my rooted, ROM'd, overclocked Nexus gets its teeth kicked in by a stock Droid DNA?
Click to expand...
Click to collapse
Does it really matter? I don't know about your n4 but mine has zero lag and handles every app and game I have thrown at it with ease. Benchmark scores mean nothing compared to actual use and performance. Dont worry about scores and enjoy your 4.2.2 phone.
Sent from my Nexus 4 using xda premium
That was quick. Yes, I love my phone, yes it runs extremely well, yes benchmarks aren't perfect.
Now that we got that out of the way...
Wrong section, also the Droid DNA screen is incredibly saturated screen..way more then the N4 thats for sure.
rmp5s said:
That was quick. Yes, I love my phone, yes it runs extremely well, yes benchmarks aren't perfect.
Now that we got that out of the way...
Click to expand...
Click to collapse
I mean no disrespect, I just don't understand all the infatuation with benchmark scores. If a phone has zero lag and handles everything one could ask beautifully then why does it matter if it scores low?
Sent from my Nexus 4 using xda premium
For god's sake, please stop using stupid Quadrant, makes no sense at all.
rmp5s said:
My buddy got a Droid DNA the other day. I immediately snatched it up to check it out. What a screen!
Anyway...ran a couple benchmarks and was AMAZED!
Quadrant = 8085
BenchmarkPi = ~280
0.0
Our phone's guts are the same! My Nexus 4 is even overclocked and doesn't come CLOSE...
Just out of curiosity...how is it that my rooted, ROM'd, overclocked Nexus gets its teeth kicked in by a stock Droid DNA?
Click to expand...
Click to collapse
quadrant runs bad on n4, no one seems to know why (prob 4.2.2 reason, 4.1.x seems ok). dunno about pi test.
try antutu, it's recently updated and not built for over year old chipsets. you will see your n4 runs fine
rmp5s said:
My buddy got a Droid DNA the other day. I immediately snatched it up to check it out. What a screen!
Anyway...ran a couple benchmarks and was AMAZED!
Quadrant = 8085
BenchmarkPi = ~280
0.0
Our phone's guts are the same! My Nexus 4 is even overclocked and doesn't come CLOSE...
Just out of curiosity...how is it that my rooted, ROM'd, overclocked Nexus gets its teeth kicked in by a stock Droid DNA?
Click to expand...
Click to collapse
quadrant hasnt been updated for years. using quadrant is like using a car from 60's and complaining the heater doesn't heat fast. use antutu and see what you get on both phones.
Exodian said:
I mean no disrespect, I just don't understand all the infatuation with benchmark scores.
Click to expand...
Click to collapse
You don't need to understand; OP wasn't directing his post at anyone in particular. Just like MPG, 0-30/60/1/4 mile times, HP, and torque ratings in car reviews help people compare different car models, benchmarks in mobile devices help people make similar comparisons. That's why noted review sites like GSMArena and Anandtech devote 1/3 of their 10+ page reviews to performance benchmarking. I totally agree that if a device's performance meets someone's needs that’s more important than benchmarks. But I also feel performance benchmarks, just like audio, display brightness/contrast, and camera tests serve a legitimate purpose. I've found people that don't like benchmarks are usually device owners on the wrong side of them. Clearly HTC did a better job of optimizing their s/w for S4 Pro than either Google or LG did which, to OP's point, is why the DNA benchmarks so much better. In another thread someone said individual user s/w influences benchmarks. Most of them are low-level tests or use emulators (Sunspider) so what's running (or not) on the device won't influence the result. It is what it is.
Here are some comments from Anandtech talking about the Optimus G's less than steller benchmark performance. It's not due to S4 Pro but it doesn't really matter because the net result is what people experience using the device. Great performance which benchmarks measure takes a combination of s/w and h/w as shown by the DNA comparison OP's made.
Sunspider is lightly threaded and thus doesn't see huge scaling going to four cores. In fact, in this case we're not seeing any real improvement over the dual-core Krait based devices from HTC. It's unclear how much of the Optimus G's performance is due to LG's browser/software stack vs. the underlying hardware.
BrowserMark doesn't look great and the Optimus G's performance is almost certainly due to LG's own browser code. Qualcomm's reference software stack can provide great performance, but it's up to the individual OEM to take advantage of it.
ooooh it has big numbers ...
HTC will always be laggy running sense!!!
cryshop said:
For god's sake, please stop using stupid Quadrant, makes no sense at all.
Click to expand...
Click to collapse
+1
mentioning quadrant should be an automatic beating with the ban-stick
Please use the Nexus 4 vs. Any other phone thread that is a sticky
Closed
Ok bad news, I use system tuner + stability test apps and the cpu slow down to 1119 Mhz. This explains low scores in some benchmarks.
jlmcr87 said:
Ok bad news, I use system tuner + stability test apps and the cpu slow down to 1119 Mhz. This explains low scores in some benchmarks.
Click to expand...
Click to collapse
Why is that bad news? I've set my cpu speed to 1200mhz and have been running chrome without hiccups. I don't think it actually affects performance that much.
kaywalker23 said:
Why is that bad news? I've set my cpu speed to 1200mhz and have been running chrome without hiccups. I don't think it actually affects performance that much.
Click to expand...
Click to collapse
But benchmarks man....Seriously though the phone is screaming fast so I say let it save some battery by throttling down.
Every phone, including the Iphone throttles. Unless there is evidence that N5 throttles a lot more than other S800 phone in non-benchmark tasks this is nothing to worry about imo.
Established behavior. Other phones detect benchmarking software and go full throttle to look better on paper than they are in use.
Sent from my Nexus 4
Ajfink said:
Established behavior. Other phones detect benchmarking software and go full throttle to look better on paper than they are in use.
Sent from my Nexus 4
Click to expand...
Click to collapse
Samsung ones especially....
Sent from my Nexus 5 using Tapatalk
I don't care what the numbers say... my phone flies through everything in real world use. This, is good news.
---------- Post added at 10:31 PM ---------- Previous post was at 10:30 PM ----------
I just had a little go on the wife's galaxy nexus... I thought something was wrong with it with how slow it felt
I have now played many games on my Nexus 5 and I haven't noticed any impact on the performance when playing for long periods of time. Whereas Nexus 4 would slow down dramatically once the throttling comes.
dannstarr said:
I don't care what the numbers say... my phone flies through everything in real world use. This, is good news.
---------- Post added at 10:31 PM ---------- Previous post was at 10:30 PM ----------
I just had a little go on the wife's galaxy nexus... I thought something was wrong with it with how slow it felt
Click to expand...
Click to collapse
Funny, my gnex is just as smooth as the new n5. Gnex is running 4.3.1, but I also loaded a couple kk roms and also just as smooth as the n5. Doesn't take much to run an android os, but other things go a helluva lot faster on the n5 with the 4 cores and faster clock speeds.
kaywalker23 said:
Why is that bad news? I've set my cpu speed to 1200mhz and have been running chrome without hiccups. I don't think it actually affects performance that much.
Click to expand...
Click to collapse
what app are you using?
incredicontrol, kernel tuner and nofrills arent working for me, you can set the max freq but the cpu ignores the limit
Yakandu said:
what app are you using?
incredicontrol, kernel tuner and nofrills arent working for me, you can set the max freq but the cpu ignores the limit
Click to expand...
Click to collapse
SetCpu works like a charm.
I've only made 3 conditions for it.
when locked: 300/300
chrome: 300/1200
otherwise: 300/800
What's amazingly surprising is the phone is still very snappy with these frequencies. Sometimes I disable the profiles just to play around but when I'm looking for battery savings I enable them.
With stability test only 30 seconds for the CPU to drop to 1119 Mhz
jlmcr87 said:
Ok bad news, I use system tuner + stability test apps and the cpu slow down to 1119 Mhz. This explains low scores in some benchmarks.
Click to expand...
Click to collapse
I only care about throttling if it affects overall performance. So far users report smooth browsing and gaming in the most demanding situations, so it's cool. Nexus devices never fared well at benchmarks anyway.
Another dissapointing... I suppose this is what they were talking about when said "power saving".
PD: Cooltool + setcpu
http://www.youtube.com/watch?v=4GRcTxaWd24
check that if you want to see the effect of throttling in gaming
kaywalker23 said:
Why is that bad news? I've set my cpu speed to 1200mhz and have been running chrome without hiccups. I don't think it actually affects performance that much.
Click to expand...
Click to collapse
Do you pull half the fuel injector harnesses out in your car to?
The bigger factor is that all games look like they only use 2 cores even at lower speeds they would run much better if properly threaded
Sent from my Nexus 5 using XDA Premium 4 mobile app
noobdeagle said:
The bigger factor is that all games look like they only use 2 cores even at lower speeds they would run much better if properly threaded
Sent from my Nexus 5 using XDA Premium 4 mobile app
Click to expand...
Click to collapse
Yes that's true ,
Sent from my Nexus 5 using Tapatalk
That's the problem with having more cores, developers are still making things for older (and more prevalent) hardware. Same thing happened in the early days of multi-core CPUs in computers as well. To begin with you got far worse performance on a dual core 1.5Ghz machine than on a single core with 2Ghz because the OS and applications didn't use the second core. Fortunately phone evolution is way faster and I expect games and apps to take advantage of all the cores much sooner than it happened on PC.
As on the nexus 4, I have no doubt that we will be able to turn off thermal throttling with custom kernels
Sent from my Nexus 4 using Tapatalk
So I was kinda worried when reading benchmark reviews of the note 4. They would show that the iPhone 6's gpu was superior and to prove that they used GFX bench.
Sure after trying it myself I noticed something pretty obvious.
They are comparing the note 4's 2k res against the 720p res of the iPhone 6.
When you look at the note 4's offscreen performance it's better or equal.
I made a screen show to show you all.
Anyone else want to show their results?
Its not misLeading at all onscreen and offscreen tests 2 different things. And onscreen is more important cuz shows real world performance(if games support 2k ofc)
Biggest misleading information was when all reviews done benchmarks werent updated to support new iphone resolutions. All was running on 640p which was 5s res.
They published reviews with 18 fps ofscreen and 30 fps onscreen for 6 plus and didnt even think about what was wrong . Both score should be same cuz both tests done in 1080p for 6 plus. Even reviews said "apple did a great job with 6 plus. While having higher res screen it still beat iphone 6 in onscreen tests." It was totaly wrong and misleading stuff.
Now all benchmarks updated to support new screens and all gpu benchmark scores lowered iphone 6 has like 40% better performance in onscreen than 6 plus. But they just didnt update their reviews and still has wrong information for people. And still comparing new phones with wrong scores from iphones.
I don't know why people are so obsessed with benchmarks. I would never buy a phone because it is faster in some worthless synthetic benchmark
Skickat från min iPhone med Tapatalk
easycure1974 said:
I don't know why people are so obsessed with benchmarks. I would never buy a phone because it is faster in some worthless synthetic benchmark
Skickat från min iPhone med Tapatalk
Click to expand...
Click to collapse
Benchmarking gives a user an idea if their system is running normally. We do it with our pc's all the time. The note 4 has new hardware compared to the note 3. The new gpu adreno 420 has also a new architecture.
My point of the post isn't about the note performing low or anything but simply pointing out the lies about the iPhone being better when they obviously didn't point out that the note 4 is running 2k res vs the 1080p from the iPhone 6+. Even the s5 plus with the same hardware as the note 4 has higher scores because of its lower res.
Either way don't look at benchmarks to say your device is top dog but also don't Base your judgement on a device from inappropriate testing.
jetbruceli said:
Benchmarking gives a user an idea if their system is running normally. We do it with our pc's all the time. The note 4 has new hardware compared to the note 3. The new gpu adreno 420 has also a new architecture.
My point of the post isn't about the note performing low or anything but simply pointing out the lies about the iPhone being better when they obviously didn't point out that the note 4 is running 2k res vs the 1080p from the iPhone 6+. Even the s5 plus with the same hardware as the note 4 has higher scores because of its lower res.
Either way don't look at benchmarks to say your device is top dog but also don't Base your judgement on a device from inappropriate testing.
Click to expand...
Click to collapse
When comparing PCs, they're (almost) always using the same CPU architecture and operating system. It's kind of a difficult comparison on phones, since they're often nothing alike.
I also don't understand why it's so important it's better than the iPhone. It's like if it's worse than the iPhone, it's utter ****. Which is not the case.
3D Benchmarking on phones/tables is pointless when comparing to other devises 9 times out of 10 the screen res is not same and they never take this into factor.
Sent from my Nexus S using xda premium
---------- Post added at 11:38 AM ---------- Previous post was at 11:26 AM ----------
Mafle93 said:
When comparing PCs, they're (almost) always using the same CPU architecture and operating system. It's kind of a difficult comparison on phones, since they're often nothing alike. .
Click to expand...
Click to collapse
You clearly don't benchmark on PC
---------- Post added at 11:39 AM ---------- Previous post was at 11:38 AM ----------
jetbruceli said:
Benchmarking gives a user an idea if their system is running normally. We do it with our pc's all the time. .
Click to expand...
Click to collapse
Yea but on PC we have option to set a fixed res to compare hardware.
demo23019 said:
3D Benchmarking on phones/tables is pointless when comparing to other devises 9 times out of 10 the screen res is not same and they never take this into factor.
Sent from my Nexus S using xda premium
---------- Post added at 11:38 AM ---------- Previous post was at 11:26 AM ----------
You clearly don't benchmark on PC
---------- Post added at 11:39 AM ---------- Previous post was at 11:38 AM ----------
Yea but on PC we have option to set a fixed res to compare hardware.
Click to expand...
Click to collapse
Do you often see benchmarks with 2 completely different types of RAM, GPU, CPU, OS and monitor resolution when doing HW comparisons?
Does your Note 4 perform how YOU want it to perform? If so, then who cares about benchmarks? This benchmark obsession has gotten completely out of control. I never run them, I benchmark my phones myself and decide if it's running to MY liking.
WizeGuyDezignz said:
Does your Note 4 perform how YOU want it to perform? If so, then who cares about benchmarks? This benchmark obsession has gotten completely out of control. I never run them, I benchmark my phones myself and decide if it's running to MY liking.
Click to expand...
Click to collapse
I completely agree. Benchmarks are a good as an indication of your device's performance. Nothing more. The software "feeling" is just as important for the overall experience, and it can't be measured as easily.
Mafle93 said:
I completely agree. Benchmarks are a good as an indication of your device's performance. Nothing more. The software "feeling" is just as important for the overall experience, and it can't be measured as easily.
Click to expand...
Click to collapse
Yeah exactly. It's almost like letting an app decide if the food you just ate tasted good or not. Just judge it for yourself, screw all these synthetic benchmarks lol.
Benchmarks are only useful to find out performance differences, not which is the best or the better, like when you OC your phone.
From personal experience, benchmarks resulted as fake, because my own device was outperforming the one in their list or table not slightly, but 10 to 20%.
I am talking about Antutu benchmark.
I had the same issue wtih Quadrant.
I didnt bother anymore because I didnt want to stress test my phone any longer.
Mafle93 said:
Do you often see benchmarks with 2 completely different types of RAM, GPU, CPU, OS and monitor resolution when doing HW comparisons?
Click to expand...
Click to collapse
All the time people who generally benchmark build their own system .
You almost never see the same exact specs.
And when gpu benchmarking involved rules are set to run x resolution to get accurate comparison when comparing
Sent from my Nexus S using xda premium
Not all games run native 2k. Hi quality 3d games will most likely upscale a 1080p to 2k on your phone. Don't worry about it. This monster is fast even at 2k.
demo23019 said:
All the time people who generally benchmark build their own system .
You almost never see the same exact specs.
And when gpu benchmarking involved rules are set to run x resolution to get accurate comparison when comparing
Sent from my Nexus S using xda premium
Click to expand...
Click to collapse
Ok so here is the scenario for the pc, lets say you have the same mobo, gpu, cpu, ram and etc. You benchmark your completely new system and notice its not performing as well. Would using a benchmark then be helpful? YES
Second our devices share common specs with other devices. Galaxy s5 and Xperia z3, Note 4 and Nexus 6. These systems also produce different benchmarks. Of you own a Nexus 5 you will notice it doesnt play everything the same as the note 3. The note 3 was faster and smoother than the nexus 5. i would know I had both. I was so mad when I gave up my note 3. The nexus 5 ran poorly on benchmarks and would throttle instantly. I actually had to put it in the freezer for it come atleast close to the performance of the Note 3.
For me its really saying my note 4 is better than yours or your nexus 6 but it does suggest that I might get performance getting the note 4 over the G3
So after seeing all these benchmarks, how the heck is the iPhone 6 plus faster than the nexus6? We've got 4 2.7ghz processor even though it is 32-bit compared to a dual core 1.4ghz 64-bit for the iPhone 6 plus... One of the reviews with benchmark figures is here
http://www.phonearena.com/reviews/Google-Nexus-6-vs-Apple-iPhone-6-Plus_id3854/page/2
How the heck is this possible, in graphics tests the iPhone has like 50% better performance according to this as well as other sites I looked at.
Can someone please explain this?
Sent from my Nexus 6 using XDA Free mobile app
nexus 6 vs iphone 6 vs note 4 speed test.. nexus 6 wins http://youtu.be/DV2X51kAVfc
Speed tests aside ... I had a 6 for 4 days to see if I would possibly like to switch. I didn't. I put it through it's paces and it is not even close to my N6. Now ... my previous N5 is slower but that's to be expected.
From that YouTube video it really seems like they were all the same, but it defiantly seems like the nexus 6 is slower at loading apps... Didn't really go over graphics performance which I thought was strange.... More detailed tests would be better...
I'll never buy an apple product in my life but I'm amazed at how that dual core processor can keep up with this quad core that is clocked twice as fast
Sent from my Nexus 6 using XDA Free mobile app
jwleonhart said:
From that YouTube video it really seems like they were all the same, but it defiantly seems like the nexus 6 is slower at loading apps... Didn't really go over graphics performance which I thought was strange.... More detailed tests would be better...
I'll never buy an apple product in my life but I'm amazed at how that dual core processor can keep up with this quad core that is clocked twice as fast
Sent from my Nexus 6 using XDA Free mobile app
Click to expand...
Click to collapse
In the end of that video the Nexus 6 catches way up because of it's RAM, the iPhone 6 has to reload the apps entirely because it dumps them due to low memory.
Anyway, Android software has to support a very large range of hardware which lowers efficiency, and just the way it handles certain things about the OS is less efficient in general. iOS is much more streamlined for specific hardware and specific APIs, so it can do things very well for the hardware it has. Plus the graphics part of the iPhone6 processor is pretty beastly, no denying that. It is easily faster in raw power than the stuff in Snapdragons.
jwleonhart said:
From that YouTube video it really seems like they were all the same, but it defiantly seems like the nexus 6 is slower at loading apps... Didn't really go over graphics performance which I thought was strange.... More detailed tests would be better...
I'll never buy an apple product in my life but I'm amazed at how that dual core processor can keep up with this quad core that is clocked twice as fast
Sent from my Nexus 6 using XDA Free mobile app
Click to expand...
Click to collapse
You're forgetting that its a 720p display vs a 1440p display...
iOS runs everything compiled. ART helps bridge that gap somewhat, but not completely. The apps are run compiled, since they are compiled when installed, but they are not fully optimized. A good iOS app is compiled and optimized for the very limited variations of hardware it will run on. There is an expectation in the iOS world of separate iPad and iPhone apps. On the Android side, the expectation is that every app will run on every Android device ever produced, lol. So we are counting on a generic compiling. There is simply no way it will be as effective. Its one reason Samsung TouchWiz camera apps are always so much better than the generic ones--they were built and optimized for a particular device.
Also, you are putting too much emphasis on the processor. Apple learned some hard lessons in the powerPC Mac days. The GPU often has more influence over the user experience than the main processor, and the GPUs that Apple uses are top shelf.
So where does this adreno GPU lay? Doesn't nvidia offer a tegra or something as a GPU? I'd assume that is probably killer.. This is a top shelf phone I'd just expect more... And the battery life is horrid I can't believe it. Gonna have to keep a charger with me at all times its that bad. This charge just last me about 6 hrs on medium usage
Sent from my Nexus 6 using XDA Free mobile app
and remember, the iPhone is a 1.5ghz dual core, the n6 2.65ghz quad core. that said, id stll want my device to be an android.
jwleonhart said:
So after seeing all these benchmarks, how the heck is the iPhone 6 plus faster than the nexus6? We've got 4 2.7ghz processor even though it is 32-bit compared to a dual core 1.4ghz 64-bit for the iPhone 6 plus... One of the reviews with benchmark figures is here
http://www.phonearena.com/reviews/Google-Nexus-6-vs-Apple-iPhone-6-Plus_id3854/page/2
How the heck is this possible, in graphics tests the iPhone has like 50% better performance according to this as well as other sites I looked at.
Can someone please explain this?
Sent from my Nexus 6 using XDA Free mobile app
Click to expand...
Click to collapse
Who cares
Sent from my Nexus 6 using XDA Free mobile app
jwleonhart said:
So after seeing all these benchmarks, how the heck is the iPhone 6 plus faster than the nexus6? We've got 4 2.7ghz processor even though it is 32-bit compared to a dual core 1.4ghz 64-bit for the iPhone 6 plus... One of the reviews with benchmark figures is here
http://www.phonearena.com/reviews/Google-Nexus-6-vs-Apple-iPhone-6-Plus_id3854/page/2
How the heck is this possible, in graphics tests the iPhone has like 50% better performance according to this as well as other sites I looked at.
Can someone please explain this?
Sent from my Nexus 6 using XDA Free mobile app
Click to expand...
Click to collapse
Notice the multi core performance. The nexus 6 easily beats it. And the Adreno 420 GPU blows everything out of the water. It has the same frame rates as the iPhone 6 even while its running in 1440p!
Sent from my Nexus 6 using XDA Free mobile app