Develop Bugs

[Q] Thoughts on 'weak' GPU?

Hi
Noob here. Was just wondering what the general consensus was on the 'old' GPU that the Galaxy Nexus will be supplied with. It isn't as powerful as say, the Galaxy S2 or the iPhone4s, but will this have an overall effect on how the phone performs in day to day use? Will it only effect the high end games that are currently available? I'm seriously tempted by this phone - mainly due to the lovely looking ICS but I'm concerned I may regret purchasing if there are serious issues with the GPU.
Cheers.

I am presuming the GPU is clocked all the way up to 384Mhz like the chip's specification says, if not then curses to Google.
TBH I believe it'll be fine, although it is an old GPU it is still quite a powerful one and can handle almost every game fine. Tegra 2 is generally a weaker GPU than SGX540 @ 200 and can still manage games just fine at 1280 x 800, I don't see why the SGX540 @ 384 can't do that. Although we'll never know for sure until we get the phone.

I'll quote myself from the other thread here:
Here's a lovely anecdote: I use an Eyefinity (three-monitor) setup on my gaming rig. It's a general rule of thumb that (compared to a single 1080p monitor) adding an additional 1080p monitor will reduce your performance by about 30%. A third 1080p monitor will reduce your performance to about 50% that of a single-screen setup.
Now consider, the Nexus Prime has about 2.4 times the number of pixels as the Nexus S. If the same formula as a desktop GPU holds true for mobiles, we could expect about a 40% loss in 3D performance. Now the GPU has been clocked up about 92%. It's throughput is now approaching double that of the Galaxy S, when it needed only make up a 40% defecit. Of course if you consider diminishing returns from clockspeed scaling, the [email protected] should perform at 720p about as well as it did at 200MHz and 480p. /shrug
The usual disclaimer: this was all conjecture on my part.
Click to expand...
Click to collapse
Like I said, that's just my theory, and it's got no real grounding (since I haven't used the new Nexus yet.)

Hope you guys are right, of course I'm not going to hold you to it, I just would like to have seen fresh architecture.
If we get a kernel, or I should say when we get a kernel that allows overclocking, does that only OC the CPU cores or will it OC the GPU even more?
Sent from my Droid using Tapatalk

And while you guys are skeptical of a gpu more powerful than the Geforce on the Tegra 2, which has its own games zone dedicated to it's well-known-to-be-awesome-or-atleast-marketed-well performance, I'm rocking an Adreno 200 powering a thoroughly shattered-yet-still-working-perfectly 4.3" WVGA standard LCD display. That powervr is probably more powerful than my Geforce3 ti 200 on my desktop.
I need a refresh.

Andreno200 < Adreno205 < [email protected] < [email protected] < or = Adreno 220
The Andreno 205 is 2X the 200, but the SGX is around 1.5X Adreno205, 220 is 2X Andreno205...So [email protected] is similar to Andreno 220 at same res, but slower at 720P?

I think it's stupid that people think it's weak because:
* It isn't brand new
* They've never seen it clocked like it is and/or matched with the OMAP processor it's matched with.
* Have never played a game optimized for it
* Can't name a game/movie/program that will run on something else but not the combination mentioned above
* Assume that superficial benchmark results mean much in real world applications
The entire conversation is like talking about a way to make your race car's top speed go from 210mph to 230mph on a track that is designed to make it impossible to go faster than 175mph.
For the last time, this is NOT the same GPU that is in the SGS.

Dragooon123 said:
I am presuming the GPU is clocked all the way up to 384Mhz like the chip's specification says, if not then curses to Google.
TBH I believe it'll be fine, although it is an old GPU it is still quite a powerful one and can handle almost every game fine. Tegra 2 is generally a weaker GPU than SGX540 @ 200 and can still manage games just fine at 1280 x 800, I don't see why the SGX540 @ 384 can't do that. Although we'll never know for sure until we get the phone.
Click to expand...
Click to collapse
You got it wrong there dude, SGX 540 @304 Mhz is equal or +1 to 2 % faster then the Geforce ULP GPU @800x480( Note that this can be because of the Dual channel memory the 4430 soc uses (optimus 3d)). Also the ULP Geforce does not work the same way as the SGX. Geforce ULP has the tendency to not get major performance hits when resolution gets bigger hence why all tablets use Tegra 2(Got a source for this however cant find it right now), it was Nvidias plan all along to grab the Tablet market.
I hope the extra Mhz helps the sgx 540 to perform well on the galaxy nexus when it comes to Games and so on. If it doesn't there are tricks to bypass things and get good performance in gaming however it up to google/samsung to implement them

I'm looking forward to try the phone myself when it hits the stores, and hope it'll be ok...

taxas said:
You got it wrong there dude, SGX 540 @304 Mhz is equal or +1 to 2 % faster then the Geforce ULP GPU @800x480( Note that this can be because of the Dual channel memory the 4430 soc uses (optimus 3d)). Also the ULP Geforce does not work the same way as the SGX. Geforce ULP has the tendency to not get major performance hits when resolution gets bigger hence why all tablets use Tegra 2(Got a source for this however cant find it right now), it was Nvidias plan all along to grab the Tablet market.
I hope the extra Mhz helps the sgx 540 to perform well on the galaxy nexus when it comes to Games and so on. If it doesn't there are tricks to bypass things and get good performance in gaming however it up to google/samsung to implement them
Click to expand...
Click to collapse
That might be the case but I saw SGX540 outperforming tegra at 720p, so even then at a tablet resolution the SGX540 doesn't fail to perform. Regardless, the gpu in galaxy nexus is nothing short of high end and should perform fine.
Sent from my GT-I9000 using Tapatalk

...and Tegra isn't that great either!

Regardless of whether the phone is fast or not, there is the overwhelming feeling that it could have been better. I think most people wanted a 543MP2 or if it were possible, the 543MP4+ (it isn't) on th Vita.
Sent from my SPH-D700 using xda premium

There is no soc out yet apart from the A5 with the 543mp2, the lead time on a soc is huge, i mean they were designing the OMAP 4460 back in 2009 or earlier (first mentions in white papers of the 4460 where in Feb 2009) but i am sure they where working on it before then.
Sent from my GT-I9100 using xda premium

veyka said:
There is no soc out yet apart from the A5 with the 543mp2, the lead time on a soc is huge, i mean they were designing the OMAP 4460 back in 2009 or earlier (first mentions in white papers of the 4460 where in Feb 2009) but i am sure they where working on it before then.
Sent from my GT-I9100 using xda premium
Click to expand...
Click to collapse
True. We know that the new A15s have been in development since at least 2009.
The Omap 5430 has a 544MPx; we don't know how many cores.
I suppose there was no alternative except the Exynos?
Sent from my SPH-D700 using xda premium

sauron0101 said:
True. We know that the new A15s have been in development since at least 2009.
The Omap 5430 has a 544MPx; we don't know how many cores.
I suppose there as no alternative except the Exynos?
Sent from my SPH-D700 using xda premium
Click to expand...
Click to collapse
Well there is Exynos, OMAP or snapdragon for current generation soc's.
OMAP and exynos are S9 cores. Snapdragon is kinda A8 with extra SIMD performance.
That's generally why snapdragon gets out performed clock for clock by A9+neon designs (that's why a 1.5ghz snapdragon eg sensation xl gets or tmob USA sgs2 is out performed by a 1.2ghz exynos.
I am more happy with OMAP than snapdragon that's for sure.
Sent from my GT-I9100 using xda premium

A lot of people seem to bemoaning the fact that this phone doesn't have a 1.5Ghz Exynos 4212 or even the 4210. The big worry is that the chip may not run well at 1280x720, hence the "lag" we saw in the leak videos.
There is disagreement on if the Mali 400 or the SGX 540 is better (at this clock anyways), but there seems to be a consensus that the Exynos is a faster CPU than the OMAP 4. I suppose that a few were hoping for a ARM Cortex A15 with a 2-core SGX 554. No such a SOC currently exists sadly.
I am also hopeful that there have been some software optimizations in Ice Cream that could improve performance.
Part of me wonders if Google should do what Apple did - get its own semiconductor design department and outsource the actual fab. It seems to be offering Apple a competitive advantage of sorts.

my thoughts are that i don't care.
eric b

veyka said:
Well there is Exynos, OMAP or snapdragon for current generation soc's.
OMAP and exynos are S9 cores. Snapdragon is kinda A8 with extra SIMD performance.
That's generally why snapdragon gets out performed clock for clock by A9+neon designs (that's why a 1.5ghz snapdragon eg sensation xl gets or tmob USA sgs2 is out performed by a 1.2ghz exynos.
I am more happy with OMAP than snapdragon that's for sure.
Sent from my GT-I9100 using xda premium
Click to expand...
Click to collapse
Agreed. Better Omap 4 than Scorpion.
Apparently there are also a few people who were hoping for a Tegra 3. It might have been doable (and I stress the might), as the new Asus Transformer Prime is rumoured to carry Kal El.
Sent from my SPH-D700 using xda premium

TBH the GPU and CPU are more than capable off handling the gui, its not like they are pulling out a fully 3D gui, even if the resolution is bumped the hardware should still be able handle it without breaking a sweat. It's only the games where the doubt arises.

sauron0101 said:
Agreed. Better Omap 4 than Scorpion.
Apparently there are also a few people who were hoping for a Tegra 3. It might have been doable (and I stress the might), as the new Asus Transformer Prime is rumoured to carry Kal El.
Sent from my SPH-D700 using xda premium
Click to expand...
Click to collapse
I am not sure if kal el is ready yet, i dont think the transformer prime is due till q1 2012, and I'm sure if the smartphone Tegra 3 is ready as well.
And Tegra 2 doesn't even have neon!
Sent from my GT-I9100 using xda premium

Galaxy Nexus' Memory Bandwidth and efficiency puts it ahead of the competition

There have been a lot of people who have been doing comparisons of various phones (particularly the Galaxy S2) to the Galaxy Nexus. I recalled during the Samsung/Google event them saying they chose to use an industry leading hardware inside the phone, so I decided to look into this a little further. As I'm sure you're well aware, the first thing people tend to point at are benchmarks. The gpu benchmarks are particularly what have come under fire when people make their comparisons. Though the mali 400 does bench out higher than the SGX540 the higher performance on the mali isn't a tangible benefit as "end device applications have not yet caught up with the highest graphics performance delivered by these" (http://armdevices.net/2011/10/26/interview-with-the-texas-instruments-omap4-team/). In other words that's like having a road with a 300mph speed limit but the cars are only able to achieve 120mph. Driving on the road with the 300mph speed limit won't get you there any faster than driving on a road with a 200mph speed limit if the speed of the car is the same.
As for processors, the processors as they are now are roughly on par with each other with them both being 45nm A9's clocked at 1.2. The difference between them is that the exynos 4210 is clocked at it's true clock speed at 1.2, whereas the omap 4460 is actually underclocked to 1.2 and has a true clock speed of 1.5. Thus meaning the processor has more speed potential than that of the exynos.
One thing that does stand out as an advantage of the omap 4460 over some of the competition is it's memory bus bandwidth. For those that don't know, in simple terms, it's how fast information can be read from and stored to memory by the processor. In other words, you can have the fastest processor in the world but if you don't have enough memory bandwidth to accommodate the amount of information that needs to be transferred then that speed won't matter because it will be bottlenecked. For example, let's say you have a car that can reach 200mph and you want to drive that car at full speed. However the street you're driving on can only handle 20 cars at a time, and you're the 21st car, well in this case you're going to be stuck in traffic. Sure you have the raw potential of doing 200mph, but you won't ever get close to that because of traffic congestion. The same concept applies when we're talking about memory bandwidth. That being said, the memory bandwidth on the omap 4460 is 6.4GB/s, the exynos 4210 is 6.4GB/s, the iphone 4s is 6.4GB/s, and the Tegra 2 is 2.5 GB/s. Add all of this with the fact that the TI processor is underclocked to 1.2ghz (for power savings) as opposed to running at full strength and full power (ie. Galaxy S2), you have what is in my opinion the superior processor. Personally, I'd rather have an underclocked processor that delivers the same or better performance and saves me power, than to have a gpu that has excessive power that I can't even make use of. It's kind of like saying you have a rocket launcher for self defense, sure the rocket launcher is powerful but it isn't something you can really make use of. In closing I think AnandTech said it best when they said "Until Tegra 3 and Krait show up, the CPU side of the 4460 is as good as it gets."
Sources:
http://www.samsung.com/global/business/semiconductor/productInfo.do?fmly_id=844&partnum=Exynos 4210
http://www.anandtech.com/show/2911
https://docs.google.com/viewer?a=v&...mA2j6n&sig=AHIEtbTE8LvpHXPUcE4w_wGU5apdbGD0Eg
http://armdevices.net/2011/10/26/interview-with-the-texas-instruments-omap4-team/
https://plus.google.com/105051985738280261832/posts/2FXDCz8x93s
http://www.phonearena.com/news/Why-...Galaxy-Nexus-Android-ICS-poster-child_id23089
http://www.anandtech.com/show/5133/galaxy-nexus-ice-cream-sandwich-initial-performance

Nice post!
Gave me some more insight in the decision for the Omap processor and makes me feel i made a real good decision ordering the GN.
Maybe some paragraphs in your text would be nice, to make it easier to read.
Kind regards.

Thanks, now I feel even better for chosing my Nexus....

wow that was very informative. all that confusion the past couple of months about why the g-nex would be using the TI OMAP 4460 instead of, what was believed to be more powerful processors like the Exynos. but this post really cleared that up for me.
thanks!

Thanks !! Nice and simple.

Woah, huge walloftext.jpg
Anyway, hate to break it to you, but the processors in Galaxy Nexus are 4460, but they're probably binned 1.5ghz processors i.e. processors that couldn't run at the full 1.5ghz
https://twitter.com/#!/coolbho3k/status/140218721774997504
https://twitter.com/#!/coolbho3k/status/140214089183010819
Oh, and this: https://plus.google.com/105051985738280261832/posts
Galaxy Nexus runs at significantly higher res than Galaxy S II, and has to push many many more pixels...

Rawat said:
Woah, huge walloftext.jpg
Anyway, hate to break it to you, but the processors in Galaxy Nexus are 4460, but they're probably binned 1.5ghz processors i.e. processors that couldn't run at the full 1.5ghz
https://twitter.com/#!/coolbho3k/status/140218721774997504
https://twitter.com/#!/coolbho3k/status/140214089183010819
Oh, and this: https://plus.google.com/105051985738280261832/posts
Galaxy Nexus runs at significantly higher res than Galaxy S II, and has to push many many more pixels...
Click to expand...
Click to collapse
I think you're missing my point. Yes the full speed potential of the 4460 chip is 1.5ghz, but it is clocked at 1.2ghz. Whether or not you will be able to clock the chip back up to 1.5 isn't what I'm getting at. The point is it has comparable processing power at a lower clock speed that it is spec'ed at. Meaning it can give comparable processing power as it's competitors but use less power, which obviously is a good thing. As for your last statement, you do realize that person you linked to is one of the sources I cited . The fact that so many more pixels have to be pushed is exactly why having more memory bandwidth is so important. Had they gone with an exynos processor clocked at 1.2 instead, there's a chance the user experience may have suffered. Think about this, the galaxy not has the exynos, but they had to clock it to 1.5 (instead of the 1.2) AND give it a 2500mah battery so that it can get decent battery life. Obviously they wouldn't be able to fit a battery of that capacity (or even close to it) in the galaxy nexus, so if they put a exynos in the nexus clocked that high, there would be a serious battery problem...and if they put a 1.2 in there with the lower memory bandwidth, there could also be a potential for user experience issues. The overall point that I'm making is that the 4460 is actually a very good chip due to the high memory bandwidth and the fact that it's more power efficient.

"More power efficient"
More power efficient than what? Exynos is pretty power efficient itself, and Note doesn't have such a large battery to counter the clock speed of Exynos, it's because it's a huge fricking phone, and they can fit it inside.
OMAP4460 is rather decidedly the 3rd best smartphone SoC in the market. A5 and Exynos are ahead (well, not CPU on a5, but GPU more than makes up for it) but it's better than Tegra2, and snapdragon. At least that's something, eh?

Awsome! Well explained, thank you, make sticky please!
Sent from my X10i using xda premium

Rawat said:
"More power efficient"
More power efficient than what? Exynos is pretty power efficient itself, and Note doesn't have such a large battery to counter the clock speed of Exynos, it's because it's a huge fricking phone, and they can fit it inside.
OMAP4460 is rather decidedly the 3rd best smartphone SoC in the market. A5 and Exynos are ahead (well, not CPU on a5, but GPU more than makes up for it) but it's better than Tegra2, and snapdragon. At least that's something, eh?
Click to expand...
Click to collapse
I will keep my response to you brief, since it's obvious you didn't read what I posted (judging from the fact that you posted link to the same person that I already had a link to in my sources). That being said, explain why the Exynos and the A5 are ahead. Instead of making a generalized please use some facts to support what you state. If that is what you think, I'd love to read why.

Rawat said:
Anyway, hate to break it to you, but the processors in Galaxy Nexus are 4460, but they're probably binned 1.5ghz processors i.e. processors that couldn't run at the full 1.5ghz
Click to expand...
Click to collapse
Hate to tell you this, but if it is indeed a 4460, these are 1.5 GHz parts. Plain and simple, if they weren't they would have a wildly different part number (think of Intel CPU's...the new I7 39xx series are binned Xeon parts...) than the one shown in the pictures or on the IC's themselves. Why? To put it bluntly, false advertisement. Every single thing online states 1.5 GHz for the part. No literature (that I can find) says anything less than that. And yes, I know a few things have the wording 'up to', but that doesn't change the fact that it's still a 1.5 GHz part. It just isn't rated for higher than 1.5 GHz. It's similar to how Apple clocks the A5 down to 800 MHz for the Iphone 4S. Get some power savings at the price of a small bit of performance. Does this mean that the A5's in the Iphone 4s can't do 1 GHz? Probably not.
Got proof of the accused binning? Then maybe I'll start considering that belief. But until I see 'real' proof, I highly doubt that TI is selling binned parts that can't make 1.5 GHz. That would kind of be pointless to say the 4460 is a 1.5 GHz part, but sell it with a max of 1.2 GHz without atleast changing the part number in some way (ie 4450 for instance).

mysterioustko said:
I think you're missing my point. Yes the full speed potential of the 4460 chip is 1.5ghz, but it is clocked at 1.2ghz. Whether or not you will be able to clock the chip back up to 1.5 isn't what I'm getting at. The point is it has comparable processing power at a lower clock speed that it is spec'ed at. Meaning it can give comparable processing power as it's competitors but use less power, which obviously is a good thing.
Click to expand...
Click to collapse
This is faulty reasoning. You are claiming that because the OMAP4460 in the GN is underclocked from 1.5GHz to 1.2GHz, it must consume less power than a Exynos 4210 clocked at 1.2GHz. This is only true if the OMAP4460 at 1.5GHz consumes the same amount of power as the Exynos 4210 at 1.2GHz. But we have no evidence that this is the case. The OMAP4460 at 1.5GHz might simply have a higher thermal envelope than the Exynos 4210 at 1.2GHz and is able to draw more power. Thus the OMAP4460 at 1.2GHz might consume power comparable to the Exynos 4210.
darkhawkff said:
Hate to tell you this, but if it is indeed a 4460, these are 1.5 GHz parts. Plain and simple, if they weren't they would have a wildly different part number (think of Intel CPU's...the new I7 39xx series are binned Xeon parts...) than the one shown in the pictures or on the IC's themselves. Why? To put it bluntly, false advertisement. Every single thing online states 1.5 GHz for the part. No literature (that I can find) says anything less than that. And yes, I know a few things have the wording 'up to', but that doesn't change the fact that it's still a 1.5 GHz part. It just isn't rated for higher than 1.5 GHz. It's similar to how Apple clocks the A5 down to 800 MHz for the Iphone 4S. Get some power savings at the price of a small bit of performance. Does this mean that the A5's in the Iphone 4s can't do 1 GHz? Probably not.
Got proof of the accused binning? Then maybe I'll start considering that belief. But until I see 'real' proof, I highly doubt that TI is selling binned parts that can't make 1.5 GHz. That would kind of be pointless to say the 4460 is a 1.5 GHz part, but sell it with a max of 1.2 GHz without atleast changing the part number in some way (ie 4450 for instance).
Click to expand...
Click to collapse
From the kernel code of the Galaxy Nexus arch/arm/mach-omap2/id.c:
Code:
if (cpu_is_omap446x()) {
si_type =
read_tap_reg(OMAP4_CTRL_MODULE_CORE_STD_FUSE_PROD_ID_1);
switch ((si_type & (3 << 16)) >> 16) {
case 2:
/* High performance device */
omap4_features |= OMAP4_HAS_MPU_1_5GHZ;
omap4_features |= OMAP4_HAS_MPU_1_2GHZ;
break;
case 1:
default:
/* Standard device */
omap4_features |= OMAP4_HAS_MPU_1_2GHZ;
break;
}
}
There appears to be something in the OMAP hardware that designates whether it is a "high performance device" or a "standard device". A standard device can only operate at 1.2GHz, not 1.5GHz. It is unclear if "device" here refers to the SoC or the phone. If it refers to the SoC, then it would suggest that the SoCs are binned into high and low performance categories, with the low performance devices incapable of performing at 1.5GHz.
But here's some preliminary evidence that the SoC itself may be missing something that's required for 1.5GHz to work: http://forum.xda-developers.com/showpost.php?p=19931580&postcount=97
mysterioustko said:
That being said, the memory bandwidth on the omap 4460 is 7.5GB/s the exynos 4210 is 6.4GB/s, and the Tegra 2 is a mere 2.5 GB/s.
Click to expand...
Click to collapse
Where are you getting that the memory bandwidth of the OMAP 4460 is 7.5GB/s? The Galaxy Nexus uses the Samsung K3PE7E700M-XGC1 1GB memory package, which is a 400MHz, LPDDR2, 32-bit dual-channel memory package. This means it has a memory bandwidth of 400 * 2 (for DDR) * 32 * 2 (for dual-channel) = 51200Mb/s = 6.4GB/s, same as the Exynos 4210.
See: http://www.samsung.com/us/business/oem-solutions/pdfs/PSG2011_web.pdf for details on the memory package used in the GN.

mysterioustko said:
I will keep my response to you brief, since it's obvious you didn't read what I posted (judging from the fact that you posted link to the same person that I already had a link to in my sources). That being said, explain why the Exynos and the A5 are ahead. Instead of making a generalized please use some facts to support what you state. If that is what you think, I'd love to read why.
Click to expand...
Click to collapse
I actually read your whole post, but skipped over reading the sources you had linked to.
A5 and Exynos are widely regarded as the 2 best SoC on the market (not including Tegra3, which just launched). Don't take my word for it, see anandtech here and here. But of course, it's hard to directly compare A5 to another (non-apple) SoC, because they run on different OSes

Rawat said:
I actually read your whole post, but skipped over reading the sources you had linked to.
A5 and Exynos are widely regarded as the 2 best SoC on the market (not including Tegra3, which just launched). Don't take my word for it, see anandtech here and here. But of course, it's hard to directly compare A5 to another (non-apple) SoC, because they run on different OSes
Click to expand...
Click to collapse
I don't care what it's "widely regarded" as. I asked for concrete information that supports what you state. You support your argument by stating that people's opinion of it is that it's best....that's not exactly a compelling argument.

I didn't read a word seeing as how it's all big one wall of text but I got the gist of it from the title. Yay my Nexus is even further ahead of the competition now than it was 20 seconds ago

mysterioustko said:
I don't care what it's "widely regarded" as. I asked for concrete information that supports what you state. You support your argument by stating that people's opinion of it is that it's best....that's not exactly a compelling argument.
Click to expand...
Click to collapse
Clock speeds aside, the A5 is clearly a better SoC. The CPUs in all three are pretty much the same, dual-core Cortex A9s on a 45nm process. It comes down to the GPU. The A5 has a better GPU simply because it's pretty much the multicore version of the GPU in the OMAP4460. Between the OMAP4460 and the Exynos 4210, it's more difficult to say. The PowerVR540 and Mali400 have different strengths and weaknesses, so I won't speculate here.
I would also suggest that you modify your original post. It contains quite a bit of misinformation and clearly many people have read it and taken it to heart. That's not a good thing, and I hope you will do the responsible thing and try to reverse the misinformation that you've spread.

Chirality said:
Clock speeds aside, the A5 is clearly a better SoC. The CPUs in all three are pretty much the same, dual-core Cortex A9s on a 45nm process. It comes down to the GPU. The A5 has a better GPU simply because it's pretty much the multicore version of the GPU in the OMAP4460. Between the OMAP4460 and the Exynos 4210, it's more difficult to say. The PowerVR540 and Mali400 have different strengths and weaknesses, so I won't speculate here.
I would also suggest that you modify your original post. It contains quite a bit of misinformation and clearly many people have read it and taken it to heart. That's not a good thing, and I hope you will do the responsible thing and try to reverse the misinformation that you've spread.
Click to expand...
Click to collapse
Omap vs Exynos ? The latter doesn't support HSPA+ or LTE.

Dmw017 said:
Omap vs Exynos ? The latter doesn't support HSPA+ or LTE.
Click to expand...
Click to collapse
This is hardly anything to do with "performance".....and not to mention that ICS is a TON more dependant on GPU renders....hardly a place it has any room to fall short.
Sent from my SAMSUNG-SGH-I777 using XDA App

Dmw017 said:
Omap vs Exynos ? The latter doesn't support HSPA+ or LTE.
Click to expand...
Click to collapse
Exynos will do 21Mbps, HSPA+
Sent from my GT-I9100 using Tapatalk

Chirality said:
Where are you getting that the memory bandwidth of the OMAP 4460 is 7.5GB/s? The Galaxy Nexus uses the Samsung K3PE7E700M-XGC1 1GB memory package, which is a 400MHz, LPDDR2, 32-bit dual-channel memory package. This means it has a memory bandwidth of 400 * 2 (for DDR) * 32 * 2 (for dual-channel) = 51200Mb/s = 6.4GB/s, same as the Exynos 4210.
See: http://www.samsung.com/us/business/oem-solutions/pdfs/PSG2011_web.pdf for details on the memory package used in the GN.
Click to expand...
Click to collapse
You are correct, I mistakenly read the the 4470's bandwidth when researching the 4460 (the 4470's bandwidth was referenced in the same TI interview where they discussed the 4460).

[Q] Is exynos worth buying?

The snapdragon version isn't available in my country, so I will have to buy the exynos (Pretty cheap right now $500 equivalent). The thing is reviews say the snapdragon doesn't lag a bit while exynos is made for a large device.
Is the performance really this bad? I'm not into eons right now by the way.

No its not worth buying the snapdragon version. My s4 is faster than my note...
Sent from my GT-I9505 using xda app-developers app

With HMP enabled there is no comparison between the two, exynos is up to 50% faster and potentially more efficient. With HMP disabled (as things currently are) then qualcomm is the slightly better chip, but I'm not convinced that the difference is enough to prefer one soc over the other...
In short Exynos 5420 is artificially neutered to seem worse than qualcomm, yet -even so- going either way won't make much of a difference...

Do you have any benchmarks to prove your claim of a speed bump of 50 %?
to OP
There are a lot of threads about Exynos vs snapdragon, long story short
Exynos , tad better cpu
Snapdragon tad better gpu
I've had both, ended with exynos , because I didn't need 4g, but needed 32 GB ( in scandinavia 4 G seems to be 16 gb only)
Lag was more or less the same
I felt the battery time on the exynos was a tad better
They felt equally as snappy when they needed to
BUT!!!
App support was a tad better on Snapdragon, ie more apps in the plastore worked with the snapdragon version, a few more games etc... no big deal for me, but still get me ticked of when I noticed a few apps I bought weren't compatible ( yet?!) with the new exynos chip ( but worked with my sammy S3 also exynos chip, older )

Exynos is fine. I've played with both and from a UI and app use perspective you can't tell the difference. Adreno's a bit faster than Mali but no so much as to drastically alter performance. Some games are better optimized for Adreno so depending on your choice of games it could make a difference. As for app compatibility it's more likely the 2,560x1,600 display that's causing the issue not the specific SoC. If there were huge differences between Exynos and S-800 or drastic app performance differences and app compatibility issues it would be all over the N3 forum and it's not.

DeBoX said:
Do you have any benchmarks to prove your claim of a speed bump of 50 %?
to
Click to expand...
Click to collapse
HMP for 8 cores have not yet released but look at Note 3 Neo, it uses 2 less large cores and it posts the same antutu score as our note, so by adding two more large cores you can expect the score to be about 50% more. As I said that is only true were all 8 cores would be used at the same time and they are not throttled (that is why I said "up to").

Stevethegreat said:
Look at Note 3 Neo, it uses 2 less large cores and it posts the same antutu score as our note
Click to expand...
Click to collapse
Not really. It also has a 267 PPI display which is benefitting its graphics scores in AnTuTu compared to the SGS4 at 441 PPI and N3 at 386 PPI.
http://www.nairaland.com/1597298/samsung-budget-galaxy-note-neo
S-800 vs. Exynos on the N3...

BarryH_GEG said:
Not really. It also has a 267 PPI display which is benefitting its graphics scores in AnTuTu compared to the SGS4
Click to expand...
Click to collapse
I was more properly referring to CPU scores which are the only ones benefitted from HMP.
I ran a quick AnTuTu (cpu) test to my Exynos 5420 equipped note and here are the results: http://i.imgur.com/zD32DZQ.png
Notice how remarkably similar they are to note neo's cpu score:
http://www.gsmarena.com/showpic.php3?sImg=newsimg/14/01/sgn3n-leak/gsmarena_006.jpg&idNews=7538
Note that note neo has only two large cores which are clocked lower by 10% compared to exynos 5420 and it still posts almost the same score merely by employing the help of the small cores. Now add two large cores more and you'd get 50% more performance, it's simple math really...
Now I'm not saying that it would be a performance that we would actually see in most occasions , it would either be throttled or -even- not supported by most apps but still it's potentially there (which was my point by saying "up to").
What will *definitely* be there if HMP is to be enabled is better battery -though- as it would make more efficient use of the small cores. Since exynos 5422 is also on 28nm yet has HMP enabled leads me to believe that we lack HMP for strategic reasons (so that samsung will sell more exynos 5422 / qualcomm equipped machines)

Stevethegreat said:
I was more properly referring to CPU scores which are the only ones benefitted from HMP.
I ran a quick AnTuTu (cpu) test to my Exynos 5420 equipped note and here are the results: http://i.imgur.com/zD32DZQ.png
Notice how remarkably similar they are to note neo's cpu score:
http://www.gsmarena.com/showpic.php3?sImg=newsimg/14/01/sgn3n-leak/gsmarena_006.jpg&idNews=7538
Note that note neo has only two large cores which are clocked lower by 10% compared to exynos 5420 and it still posts almost the same score merely by employing the help of the small cores. Now add two large cores more and you'd get 50% more performance, it's simple math really...
Now I'm not saying that it would be a performance that we would actually see in most occasions , it would either be throttled or -even- not supported by most apps but still it's potentially there (which was my point by saying "up to").
What will *definitely* be there if HMP is to be enabled is better battery -though- as it would make more efficient use of the small cores. Since exynos 5422 is also on 28nm yet has HMP enabled leads me to believe that we lack HMP for strategic reasons (so that samsung will sell more exynos 5422 / qualcomm equipped machines)
Click to expand...
Click to collapse
You can't divorce the impact of display area size and PPI from CPU performance. The GPU doesn't absolve the CPU's role in graphics output. An i3 PC with a killer graphics card will perform worse graphically than an i7 PC with a lesser card because most computational (not rendering, texture mapping, vectoring, and decoding) work is still done on the CPU. So I have no idea what AnTuTu's testing to come up with a CPU rating in isolation but if it's a real-time performance test the CPU's role in graphics output is impacting it. So comparing the Neo with a 5.5" display and 267 PPI against the N10.1-14 with a 10.1" display and 299 PPI isn't going to get you a relevant CPU comparison. That's why I used the N3 and SGS4 as comparisons because only the PPI is off. And the Neo would be well behind the SGS4 in the cumulative AnTuTu test if it had the same PPI because the lower workload of the lower PPI is artificially enhancing its score. At the end of the day an isolated CPU number is pretty meaningless. It's like bench horsepower in a car vs. horsepower to the wheels. A higher bench rating means nothing because none of us drive an engine, we drive a car. The total AnTuTu number (AKA: drive train loss) is more relevant even though it doesn't support the point you're trying to make about HMP.
http://en.wikipedia.org/wiki/Graphics_processing_unit#Computational_functions

BarryH_GEG said:
You can't divorce the impact of display area size and PPI from CPU performance. The GPU doesn't absolve the CPU's role in graphics output. An i3 PC with a killer graphics card will perform worse graphically than an i7 PC with a lesser card because most computational (not rendering, texture mapping, vectoring, and decoding) work is still done on the CPU. So I have no idea what AnTuTu's testing to come up with a CPU rating in isolation but if it's a real-time performance test the CPU's role in graphics output is impacting it. So comparing the Neo with a 5.5" display and 267 PPI against the N10.1-14 with a 10.1" display and 299 PPI isn't going to get you a relevant CPU comparison. That's why I used the N3 and SGS4 as comparisons because only the PPI is off. And the Neo would be well behind the SGS4 in the cumulative AnTuTu test if it had the same PPI because the lower workload of the lower PPI is artificially enhancing its score. At the end of the day an isolated CPU number is pretty meaningless. It's like bench horsepower in a car vs. horsepower to the wheels. A higher bench rating means nothing because none of us drive an engine, we drive a car. The total AnTuTu number (AKA: drive train loss) is more relevant even though it doesn't support the point you're trying to make about HMP.
http://en.wikipedia.org/wiki/Graphics_processing_unit#Computational_functions
Click to expand...
Click to collapse
Maybe so, but the benchmark in question runs off screen. So while in real life resolution matter in Antutu Cpu score, or super pi , or, or, it doesn't. HMP will make the Cpu 50% faster in multi threaded operations, I never claimed it makes the total machine faster by the same amount. For example an HMP equipped note 2014 will score around 40000 in Antutu , NOT 49500. I don't see where we disagree, I merely think you misunderstood my initial claim

If you live for real world use, the Exynos Note is a wonderful tablet. If you live in the world of needing the highest quadrant and antutu scores you should pass.
Sent via Tapatalk and my thumbs.

Stevethegreat said:
With HMP enabled there is no comparison between the two, exynos is up to 50% faster and potentially more efficient. With HMP disabled (as things currently are) then qualcomm is the slightly better chip, but I'm not convinced that the difference is enough to prefer one soc over the other...
In short Exynos 5420 is artificially neutered to seem worse than qualcomm, yet -even so- going either way won't make much of a difference...
Click to expand...
Click to collapse
How did you enable HMP? My note 3 snap dragon is so much faster than my note.
Sent from my SM-N900T using XDA Premium 4 mobile app

Stevethegreat said:
HMP for 8 cores have not yet released but look at Note 3 Neo, it uses 2 less large cores and it posts the same antutu score as our note, so by adding two more large cores you can expect the score to be about 50% more. As I said that is only true were all 8 cores would be used at the same time and they are not throttled (that is why I said "up to").
Click to expand...
Click to collapse
It will never be released for Exynos 5420 either, unless Samsung want alot of complains about fried Exynos 5420 chipsets. Also they already said it wont release HMP for Exynos 5420 cause of the heat.

dt33 said:
It will never be released for Exynos 5420 either, unless Samsung want alot of complains about fried Exynos 5420 chipsets. Also they already said it wont release HMP for Exynos 5420 cause of the heat.
Click to expand...
Click to collapse
Once again, that's not the reason that they won't release it, if anything the chip would be cooler because more use of A7 cores would be possible and if all 8 cores are needed Samsung could choose to throttle the thing. The reason that they don't release it is the Exynos 5422 which is the same chip but with all 8 cores enabled (also 28nm)...
So no fried socs, lesser profits more like

Tab pro or tab s

Hi. I have purchased 8.4" tab pro. Love it. It has good natural screen and is fast. But then I saw online the tab s. Is the screen worth the extra money and is the battery life much better? I am also concerned of the small bezels. Where should I hold it without touching the screen

In my opinion (which I'm quite fond of) the Tab S is worth the extra money.
If I already had a Tab Pro 8.4 that would be hard to say.
The bezels are slightly (mm's) slimmer on the Tab S from Tab Pro. If you were fine with the Tab Pro you shouldn't be a lot worse off with the Tab S.

I just returned my Tab Pro and picked up the Tab S. I recommend the 10.5 model just because you can read websites in portrait easier.
I was worried the amoled screen would be too much but it's great and I love the blacker text.

mitchellvii said:
I just returned my Tab Pro and picked up the Tab S. I recommend the 10.5 model just because you can read websites in portrait easier.
I was worried the amoled screen would be too much but it's great and I love the blacker text.
Click to expand...
Click to collapse
So did I. Went from 8.4 pro to the 10.5 S. So far I am very happy that I did.

The Tab Pro 8.4" is a quadcore, whereas the Tab S 8.4 is a octacore. So Tab S in this case would be far superior, plus the ultra power saving mode, finger print lock, and Super AMOLED screen is way better too on the "S" series for the 8.4" Tab comparisons.
Sent from my SM-P900 using xda premium

Debatable which processor is faster. Really depends on what you're doing with it. I had a chance to see them both side by side and the SAMOLED screen on the Tab S looked better to my eye, but again that's subjective. IMHO, it comes down to price. They were running the Tab Pro 8.4 for $249 last week. At that price, the Tab S just isn't worth it. That said, I walked out today with a Tab S that was a return from a customer who got the wrong color. Box was unopened, but the guy gave me a decent discount because there was a big sticker that had been ripped off the side of it. Box looked pretty rough, but the contents were untouched.
Sent from my SAMSUNG-SGH-I337 using XDA Free mobile app

snideguy19 said:
The Tab Pro 8.4" is a quadcore, whereas the Tab S 8.4 is a octacore. So Tab S in this case would be far superior, plus the ultra power saving mode, finger print lock, and Super AMOLED screen is way better too on the "S" series for the 8.4" Tab comparisons.
Sent from my SM-P900 using xda premium
Click to expand...
Click to collapse
It's not superior I wish they'd put the snap 801 instead of the exynos chip as only four cores can work at a time. Unless they can find a way to unlock. Plus it may take longer to root or get custom roms. I had the tab pro and a root was out in a couple of days.

My one knock on these super amoled screens
Reds tend to look a bit orangish.

will6316 said:
It's not superior I wish they'd put the snap 801 instead of the exynos chip as only four cores can work at a time. Unless they can find a way to unlock. Plus it may take longer to root or get custom roms. I had the tab pro and a root was out in a couple of days.
Click to expand...
Click to collapse
Wrong
All 8 cores can work simultaneously.

xRevilatioNx said:
Wrong
All 8 cores can work simultaneously.
Click to expand...
Click to collapse
I read the anandtech review it sounds there like it's only 4 at a time though the wording is confusing...
http://www.anandtech.com/show/8197/samsung-galaxy-tab-s-review-105-84inch/4
The WiFi only variants of the Galaxy Tab S all feature Samsung’s own Exynos 5 Octa (5420). Internally we’re dealing with four ARM Cortex A15s and four ARM Cortex A7s in a big.LITTLE configuration, with a maximum of four cores of the same type being active at any given moment. The Cortex A7 cluster can run at up to 1.3GHz while the Cortex A15 cluster maxes out at 1.9GHz.

cranch said:
I read the anandtech review it sounds there like it's only 4 at a time though the wording is confusing...
http://www.anandtech.com/show/8197/samsung-galaxy-tab-s-review-105-84inch/4
The WiFi only variants of the Galaxy Tab S all feature Samsung’s own Exynos 5 Octa (5420). Internally we’re dealing with four ARM Cortex A15s and four ARM Cortex A7s in a big.LITTLE configuration, with a maximum of four cores of the same type being active at any given moment. The Cortex A7 cluster can run at up to 1.3GHz while the Cortex A15 cluster maxes out at 1.9GHz.
Click to expand...
Click to collapse
it won't be the first time anandtech made a major error. I've caught a few gaffes over the years. Since then, I don't trust their reviews. I treat it like going to a doctor. You get a diagnosis but always go to several other physicians for a more informed opinion.
Here's some videos..
"Samsung Exynos 5 Octa (big.LITTLE technology
Heterogeneous Multi-Processing (HMP) Explained

xRevilatioNx said:
it won't be the first time anandtech made a major error. I've caught a few gaffes over the years. Since then, I don't trust their reviews. I treat it like going to a doctor. You get a diagnosis but always go to several other physicians for a more informed opinion.
Click to expand...
Click to collapse
I was under the assumption that the Exynos 5420 didn't have HMP enabled. Samsung released a vague statement in the past saying that they could update the kernel on devices running the 5420 so that HMP works, but I'm not sure if anything came from it. Of course, I could be totally wrong and Samsung got HMP working on these devices.
According to an Anandtech article (I know, sorry) on the newer 5422 processor back in February:
"The 5422 supports HMP (Heterogeneous Multi-Processing), and Samsung LSI tells us that unlike the 5420 we may actually see this one used with HMP enabled. HMP refers to the ability for the OS to use and schedule threads on all 8 cores at the same time, putting those threads with low performance requirements on the little cores and high performance threads on the big cores."

"Assume nothing and Check Everything "
It's enabled on their flagship device..

xRevilatioNx said:
"Assume nothing and Check Everything "
It's enabled on their flagship device..
Click to expand...
Click to collapse
Any links to back that up? Everything I read says no 5420 has HMP turned on and that the Tab S only uses one set or the other depending on performance needs, never both together.

xRevilatioNx said:
"Assume nothing and Check Everything "
It's enabled on their flagship device..
Click to expand...
Click to collapse
Hmm, you're right. I really should've checked more sources. One minute Google research doesn't get you really far. In that same article Anandtech's chart lists the max number of active cores for the 5420 as: 4 (?) .

Do what I did. Speak to technical support at Samsung.
I'll go one step further. What we do know is that the 5420 in the Tab S is scoring 35,000+ in Antutu...
It seems that our Note 2014 versions usually score about 32,000+, So that lead me to suspect that the newer Galaxy Tab S 5420 was HMP enabled due to the higher Antutu scores.*
And again, had confirmed when I spoke with Samsung directly.

xRevilatioNx said:
Do what I did. Speak to technical support at Samsung.
I'll go one step further. What we do know is that the 5420 in the Tab S is scoring 35,000+ in Antutu...
It seems that our Note 2014 versions usually score about 32,000+, So that lead me to suspect that the newer Galaxy Tab S 5420 was HMP enabled due to the higher Antutu scores.*
And again, had confirmed when I spoke with Samsung directly.
Click to expand...
Click to collapse
My note pro 8.4 with snapdragon 800 scores 35000 as well, I wouldn't call that proof on HMP. When you read the specs for the 5422 they state first chipset with HMP. Color me dubious that it's been turned on for the 5420. I hope I'm wrong

You're comparing apples to oranges. We have an AMOLED screen using the Exynos Octa 5. The Snapdragon on an inferiorly smaller screen (without AMOLED) will naturally get better CPU/GPU scores on Antutu

xRevilatioNx said:
You're comparing apples to oranges. We have an AMOLED screen using the Exynos Octa 5. The Snapdragon on an inferiorly smaller screen (without AMOLED) will naturally get better CPU/GPU scores on Antutu
Click to expand...
Click to collapse
The Pro 8.4 has the same resolution8.4 1600 x 2560 display) minus the amoled, and 1gb less RAM, same Android version.
I do agree any Antutu score is meaningless to prove HMP or not.

gottahavit said:
The Pro 8.4 has the same resolution8.4 1600 x 2560 display) minus the amoled, and 1gb less RAM, same Android version.
I do agree any Antutu score is meaningless to prove HMP or not.
Click to expand...
Click to collapse
You left out the smaller screen. Apples and Oranges.

[Q] Can Samsung Galaxy Tab S SM-T700 use all 8 cores at same time?

Can Exynos 5 OCTA 5420 use all 8 cores at same time?
How is Samsung Galaxy Tab S SM-T700 performance compared to Samsung Galaxy Tab S SM-T705( with Qualcomm Snapdragon 800)?

The Exynos cpu performs admirably against my htc one m8 snapdragon cpu and ranks highly with other top performing cpu's.
Even better with Skyhigh kernel.

Worth mentioning these Exynos units suck the battery dry much faster than the Qualcomm S800 without delivering more performance.

mrcet007 said:
Can Exynos 5 OCTA 5420 use all 8 cores at same time?
How is Samsung Galaxy Tab S SM-T700 performance compared to Samsung Galaxy Tab S SM-T705( with Qualcomm Snapdragon 800)?
Click to expand...
Click to collapse
According to the wikipedia article, the Linux kernel does have a scheduler mode that allows to use all cores or switch between them as needed. I don't know if the Samsung stock kernel or others actually use that mode.
One important issue is that you should not be carried away with the idea that having more cores is always better for performance. It takes a lot of work to write software that can actually load four or eight cores. Moreover, a lot of algorithms are still bottle-necked by one core and there is no way to change that. As a result, a typical PC with a quad core Intel i5 CPU is usually faster than a PC with a six or eight core AMD CPU, thanks to intel's much better individual core performance. This truth is even more relevant on tablets, which are effectively single-user machines, usually running only one big application at a time. I wouldn't lose a minute of my sleep over having only two working cores instead of four or eight.