GPU Performance

All of our discussions around the new iPad and its silicon thus far have been in the theoretical space. Unfortunately the state of Android/iOS benchmarking is abysmal at best today. Convincing game developers to include useful benchmarks and timedemo modes in their games is seemingly impossible without a suitably large check. I have no doubt this will happen eventually, but today we're left with some great games and no way to benchmark them.

Without suitable game benchmarks, we rely on GLBenchmark quite a bit to help us in evaluating mobile GPU performance. Although even the current most stressful GLBenchmark test (Egypt) is a far cry from what modern Android/iOS games look like, it's the best we've got today.

We'll start out with the synthetic tests, which should show us roughly a 2x increase in performance compared to the iPad 2. Remember the PowerVR SGX 543MP4 simply bundles four SGX 543 cores instead of two. Since we're still on a 45nm LP process, GPU clocks haven't increased so we're looking at a pure doubling of virtually all GPU resources.

GLBenchmark 2.1—Fill Test

GLBenchmark 2.1—Triangle Test (White)

GLBenchmark 2.1—Triangle Test (Textured, Fragment Lit)

Indeed we see a roughly 2x increase in triangle and fill rates. Below we have the output from GLBenchmark's low level tests. Pay particular attention to how, at 1024 x 768, performance doubles compared to the iPad 2 but at 2048 x 1536 performance can drop to well below what the iPad 2 was able to deliver at 10 x 7. It's because of this drop in performance at the iPad's native resolution that we won't see many (if any at all), visually taxing games run at anywhere near 2048 x 1536.

GLBenchmark 2.1.3 Low Level Comparison
  iPad 2 (10x7) iPad 3 (10x7) iPad 3 (20x15) ASUS TF Prime
Trigonometric test—vertex weighted
35 fps
60 fps
57 fps
47 fps
Trigonometric test—fragment weighted
7 fps
14 fps
4 fps
20 fps
Trigonometric test—balanced
5 fps
10 fps
2 fps
9 fps
Exponential test—vertex weighted
59 fps
60 fps
60 fps
41 fps
Exponential test—fragment weighted
25 fps
49 fps
13 fps
18 fps
Exponential test—balanced
19 fps
37 fps
8 fps
7 fps
Common test—vertex weighted
49 fps
60 fps
60 fps
35 fps
Common test—fragment weighted
8 fps
16 fps
4 fps
28 fps
Common test—balanced
6 fps
13 fps
2 fps
12 fps
Geometric test—vertex weighted
57 fps
60 fps
60 fps
27 fps
Geometric test—fragment weighted
12 fps
24 fps
6 fps
20 fps
Geometric test—balanced
9 fps
18 fps
4 fps
9 fps
For loop test—vertex weighted
59 fps
60 fps
60 fps
28 fps
For loop test—fragment weighted
30 fps
57 fps
16 fps
42 fps
For loop test—balanced
22 fps
43 fps
11 fps
15 fps
Branching test—vertex weighted
58 fps
60 fps
60 fps
45 fps
Branching test—fragment weighted
58 fps
60 fps
30 fps
46 fps
Branching test—balanced
22 fps
43 fps
16 fps
16 fps
Array test—uniform array access
59 fps
60 fps
60 fps
60 fps
Fill test—Texture Fetch
1001483136 texels/s
1977874688
texels/s
1904501632
texels/s
415164192
texels/s
Triangle test—white
65039568
triangles/s
133523176
triangles/s
85110008
triangles/s
55729532
triangles/s
Triangle test—textured
56129984
triangles/s
116735856
triangles/s
71362616
triangles/s
54023840
triangles/s
Triangle test—textured, vertex lit
45314484
triangles/s
93638456
triangles/s
46841924
triangles/s
28916834
triangles/s
Triangle test—textured, fragment lit
43527292
triangles/s
92831152
triangles/s
39277916
triangles/s
26935792
triangles/s

GLBenchmark also includes two tests designed to be representative of a workload you could see in an actual 3D game. The older Pro test uses OpenGL ES 1.0 while Egypt is an ES 2.0 test. These tests can either run at the device's native resolution with vsync enabled, or rendered offscreen at 1280 x 720 with vsync disabled. The latter offers us a way to compare GPUs without device screen resolution creating unfair advantages.

Unfortunately there was a bug in the iOS version of GLBenchmark 2.1.2 that resulted in all on-screen benchmarks running at 1024 x 768 rather than the new iPad's native 2048 x 1536 resolution. This is why all of the native GLBenchmark scores from the new iPad are capped at 60 fps. It's not because the new GPU is fast enough to render at speeds above 60 fps at 2048 x 1536, it's because the benchmark is actually showing performance at 1024 x 768. Luckily, GLBenchmark 2.1.3 fixes this problem and delivers results at the new iPad's native screen resolution:

GLBenchmark 2.1—Egypt (Standard)

GLBenchmark 2.1—Pro (Standard)

Surprisingly enough, the A5X is actually fast enough to complete these tests at over 50 fps. Perhaps this is more of an indication of how light the Egypt workload has become, as the current crop of Retina Display enhanced 3D titles for the iPad all render offscreen to a non-native resolution due to performance constraints. The bigger takeaway is that with the 543MP4 and a quad-channel LP-DDR2 interface, it is possible to run a 3D game at 2048 x 1536 and deliver playable frame rates. It won't be the prettiest game around, but it's definitely possible.

The offscreen results give us the competitive analysis that we've been looking for. With a ~2x die size advantage, the fact that we're seeing a 2-3x gap in performance here vs. NVIDIA's Tegra 3 isn't surprising:

GLBenchmark 2.1—Egypt—Offscreen 720p

GLBenchmark 2.1—Pro—Offscreen 720p

The bigger worry is what happens when the first 1920 x 1200 enabled Tegra 3 tablets start shipping. With (presumably) no additional GPU horsepower or memory bandwidth under the hood, we'll see this gap widen.

The Impact of Larger Memory A5X vs. Tegra 3 in the Real World
POST A COMMENT

232 Comments

View All Comments

  • mr_ripley - Friday, March 30, 2012 - link

    It's a shame some people argue that against the workers when over a hundred of them have committed suicide over the working conditions. How can you still say that they are being offers a better deal here??

    On the other hand, it is also unfair that Apple is being singled out here. The world of Chinese manufacturing is a dirty one and all major corporations have a part in it. I'd trust Apple over most other companies to make a difference in that regard, and I'm happy to see something is being done in that regard. Ever heard McDonalds CEO touring the slaughterhouse of the meat packing companies??
    Reply
  • name99 - Friday, March 30, 2012 - link

    Reporting suicides as a number not as a rate shows you to be either a fool or a deliberate liar. How many people, over how many years, comprise the pool from which this suicide number is drawn? Everything I have read says that the actual suicide rate is not only lower than the average rate for China, it is lower than the average rate for the US. Reply
  • mr_ripley - Friday, March 30, 2012 - link

    In 2010, 18 workers attempted sucide, 14 succeeded. To me even one in a whole year is not acceptable. If you think that is ok I hope that statistic turns out to be you!! Reply
  • name99 - Saturday, March 31, 2012 - link

    The argument was NOT that suicide is a tragedy, it was a claim that FoxConn employees specifically tied to Apple production have such lousy lives that they commit suicide in higher numbers that other people around the world.

    You have done NOTHING to prove this claim; all you have done is bring up a very different issue.
    Reply
  • mr_ripley - Saturday, March 31, 2012 - link

    There is no disputing the fact that these deths are related to working conditions. I'm pretty sure this has been well established and documented. However, I did say in my previous post that Apple is unfairly singled out. It could have been any other company.

    Comparison between suicide rates is irrelevant. Higher sucide rates elsewhere does not justify this problem. Again the fact remains that many people have died and it is directly related to the working conditions.

    Apple happens to be in a position to directly influence their lives and make it better, after all they profit in billions from the work these people do. Corporations typically place little value over human life and living conditions (IBM sold equipment to the Nazis to track the Jews in concentration camps). Somehow, I feel Apple is different.
    Reply
  • doobydoo - Sunday, April 01, 2012 - link

    Dude, sorry but you're talking no sense at all.

    First of all, pretty much any product you want to buy, electronics wise, uses parts from China where conditions are far worse on average, than Apples factories. So if you actually factored working conditions into the product review, it would look favourable for Apple.

    Secondly, your argument that comparison between suicide rates is irrelevant, is absurd. Higher suicide rates where legislation is such that no jobs suffer such terrible conditions that suicide is the only option, such as is the case here, prove that even if working conditions are refined, you still get some depressed people. Your argument, therefore, is with the people who committed suicide. You say it is 'directly related to the working conditions' but where have you evidenced this, at all? You simply haven't. The fact that the suicide rates at Apple factories are lower than some American ones further backs up my point on this.

    Every company is in a position to change lives and make them better. You too, are in a position to do this. But guess what. You, just like companies, can do WHATEVER YOU LIKE with your OWN MONEY and have NO OBLIGATION WHATSOEVER to solve the worlds problems. Apple already has amongst the best conditions of factories in China. The amount of profit they make is absolutely irrelevant, if you say Apple should be putting money into this then a lot more manufacturers should also put a lot more money into this. It's very easy to decide what other people 'should' do with their money now, isn't it?

    Corporations don't have to adhere to moral values - they are not people. They are there solely to make money. Nothing else. Don't confuse them with people. And I hope you donate every single spare penny to charity and spend every spare second of your time working to build homes in the 3rd world. Oh wait, you're on here crying that other people should do it instead.

    Get a hold of yourself you illogical fool.
    Reply
  • mr_ripley - Sunday, April 01, 2012 - link

    Like I have said before it is a shame some people argue with great zeal against others who are suffering and devalue human life. Fortunately, Tim Cook is not one of them.

    If scores of people killing themselves citing poor working conditions is not enough proof what is? If your claim that there are work environments in America that have higher suicide rates because of working conditions is true that needs to be investigated as well and rectified.

    You give charity to people who are in need and cannot earn for themselves. If you think giving someone fair amount of compensation for hard work is charity you are delusional.

    If working in those factories is such a pleasent experience I suggest you try it out for yourself. Maybe the experience might broaden your perspective.

    Although, I don't see the point I will attempt to educate you. Legally, a corporation is considered as a person, that is right just like a live human being. Regardless of that corporations are run by people and actions of a corporation reflect upon the morality of the people running them.

    I will stop here as there is no point in continuing but you can respond with more insults and accusations of what I do or have done which frankly is no concern of yours.
    Reply
  • PeteH - Monday, April 02, 2012 - link

    I've not seen a single report of people killing themselves and citing "poor working conditions" as the reason. Can you provide a link?

    There have been reports of people killed because of unsafe working conditions, but that's a different issue. Maybe you're confusing the two.
    Reply
  • mr_ripley - Tuesday, April 03, 2012 - link

    Here's a Wikipedia link: you can read some of the circumstances and judge for yourself.

    They may not have said it in so many words but it is clear they were unhappy with ther work environment.

    Imagine your boss coming and beating you up because you lost an iPhone prototype!!!
    Reply
  • mr_ripley - Tuesday, April 03, 2012 - link

    http://en.wikipedia.org/wiki/Foxconn_suicides Reply

Log in

Don't have an account? Sign up now