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

234 Comments

View All Comments

  • doobydoo - Saturday, March 31, 2012 - link

    Lucian Armasu, you talk the most complete nonsense of anyone I've ever seen on here.

    The performance is not worse, by any stretch of the imagination, and lets remember that the iPad 2 runs rings around the Android competition graphically anyway. If you want to run the same game at the same resolution, which wont look worse, at all (it would look exactly the same) it will run at 2x the FPS or more (up-scaled). Alternatively, for games for which it is beneficial, you can quadruple the quality and still run the game at perfectly acceptable FPS, since the game will be specifically designed to run on that device. Attempting anything like that quality on any other tablet is not only impossible by virtue of their inferior screens, they don't have the necessary GPU either.

    In other words, you EITHER have a massive improvement in quality, or a massive improvement in performance, over a device (iPad 2) which was still the fastest performing GPU tablet even a year after it came out. The game developers get to make this decision - so they just got 2 great new options on a clearly much more powerful device. To describe that as not worth an upgrade is quite frankly ludicrous, you have zero credibility from here on in.
    Reply
  • thejoelhansen - Wednesday, March 28, 2012 - link

    Hey Anand,

    First of all - thank you so much for the quality reviews and benchmarks. You've helped me build a number of servers and gaming rigs. :)

    Secondly, I'm not sure I know what you mean when you state that "Prioritizing GPU performance over a CPU upgrade is nothing new for Apple..." (Page 11).

    The only time I can remember Apple doing so is when keeping the 13" Macbook/ MBPs on C2Ds w/ Nvidia until eventually relying on Intel's (still) anemic "HD" graphics... Is that what you're referring to?

    I seem to remember Apple constantly ignoring the GPU in favor of CPU upgrades, other than that one scenario... Could be mistaken. ;)

    And again - thanks for the great reviews! :)
    Reply
  • AnnonymousCoward - Wednesday, March 28, 2012 - link

    "Retina Display" is a stupid name. Retinas sense light, which the display doesn't do. Reply
  • xype - Thursday, March 29, 2012 - link

    GeForce is a stupid name, as the video cards don’t have anything to do with influencing the gravitational acceleration of an object or anything close to that.

    Retina Display sounds fancy and is lightyears ahead of "QXGA IPS TFT Panel" when talking about it. :P
    Reply
  • Sabresiberian - Thursday, March 29, 2012 - link

    While I agree that "Retina Display" is a cool enough sounding name, and that's pretty much all you need for a product, unless it's totally misleading, it's not an original use of the phrase. The term has been used in various science fiction stories and tends to mean a display device that projects an image directly onto the retina.

    I always thought of "GeForce" as being an artist's licensed reference to the cards being a Force in Graphics, so the name had a certain logic behind it.

    ;)
    Reply
  • seapeople - Tuesday, April 3, 2012 - link

    Wait, so "Retina Display" gets you in a tizzy but "GeForce" makes perfect sense to you? You must have interesting interactions in everyday life. Reply
  • ThreeDee912 - Thursday, March 29, 2012 - link

    It's basically the same concept with Gorilla Glass or UltraSharp displays. It obviously doesn't mean that Corning makes glass out of gorillas, or that Dell will cut your eyes out and put them on display. It's just a marketing name. Reply
  • SR81 - Saturday, March 31, 2012 - link

    Funny I always believed the "Ge" was short for geometry. Whatever the case you can blame the name on the fans who came up with it. Reply
  • tipoo - Thursday, March 29, 2012 - link

    iPad is a stupid name. Pads collect blood from...Well, never mind. But since when are names always literal? Reply
  • doobydoo - Saturday, March 31, 2012 - link

    What would you call a display which had been optimised for use by retinas?

    Retina display.

    They aren't saying the display IS a retina, they are saying it is designed with retinas in mind.

    The scientific point is very clear and as such I don't think the name is misleading at all. The point is the device has sufficient PPI at typical viewing distance that a person with typical eyesight wont be able to discern the individual pixels.

    As it happens, strictly speaking, the retina itself is capable of discerning more pixels at typical viewing distance than the PPI of the new iPad, but the other elements of the human eye introduce loss in the quality of the image which is then ultimately sent on to the brain. While scientifically this is a distinction, to end consumers it is a distinction without a difference, so the name makes sense in my opinion.
    Reply

Log in

Don't have an account? Sign up now