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

233 Comments

View All Comments

  • zanon - Wednesday, March 28, 2012 - link

    In the article:
    Alternatively, we're used to a higher resolution enabling us to see more on a screen at one time. In the case of the new iPad, the higher resolution just makes things look sharper.

    The higher resolution does make smaller fonts readable. For something like an SSH session, that really will mean significantly more stuff can be on a screen at once.
    Reply
  • MobiusStrip - Thursday, March 29, 2012 - link

    A more useful change would be abandoning the ridiculous glossy screens. It's sad that Apple takes its cues from the plastic schlock being peddled at Best Buy, and participates in this fraud of shoving glossy screens down customers' throats. Reply
  • repoman27 - Thursday, March 29, 2012 - link

    The plastic schlock at Best Buy has a glossy plastic film applied to a cheap TN panel. Apple puts a piece of glass in front of their much more expensive IPS panels to protect them. The only way to make that glass (or the glass of the LCD panel itself) matte would be to apply an antiglare plastic film coating to the glass. These films have drawbacks (they block and scatter light making small details and text blurry.) The drawbacks become more exaggerated the farther the front surface of the glass is from the plane of the actual LCD.

    But you're right, it's probably Apple copying the design language of sub $500 laptops in order to somehow defraud the general public and force their customers to buy the products they actually produce.

    And seeing as how this discussion is about the new iPad screen, I'd like to point out that you're complaining about the lack of an antiglare coating on a touchscreen device... Strong work.
    Reply
  • Sabresiberian - Thursday, March 29, 2012 - link

    How is it fraud? Apple isn't, like, saying their screens are anti-reflective and then giving you totally reflective glossy screens.

    Many people prefer a glossy screen and simply aren't bothered by background reflections.

    ;)
    Reply
  • Henk Poley - Monday, April 02, 2012 - link

    Yes, Apple really should use Schott Conturan/Amiran/Mirogard antireflective technology.

    btw, not-glossy does not mean matte. Air is not matte either. Glass can be see-through too ;)
    Reply
  • Watwatwat - Monday, April 02, 2012 - link

    Nope, steve gibson has tested even using screen protectors on the new ipad vs not, it seems to affect the resolution at that level, matte might not be a good idea at all for high density display. Reply
  • KoolAidMan1 - Thursday, March 29, 2012 - link

    I wasn't initially blown away, but then after a day of using it every other display seemed bad in comparison. It is one of those things you didn't realize was needed until using it, now I want very high DPI in all of my monitors. Reply
  • menting - Wednesday, March 28, 2012 - link

    Is it just me, or do Shadowgun and GTA screenshots look more detailed in Transformer Prime than in the iPad? Reply
  • menting - Wednesday, March 28, 2012 - link

    nm..i just noticed that it's scaled up in new ipad, so it's definitely not as sharp.
    However, how can fps be fairly compared in this case then?
    Reply
  • TheJian - Wednesday, March 28, 2012 - link

    Basically because of the way Nvidia and Apple approach games so far, you can expect games on Tegra3 to just look better as they seem to aim for more graphics and fewer games (they spend money on fewer projects that produce better results), as opposed to apple who spreads the wealth but just ends up with more cannon fodder if you ask me :) You should get more variety on Apple I'd guess, but a better experience with fewer choices on Tegra3/Android. I like QUALITY over QUANTITY personally and hope Apple leans the way of Nvidia in the future. I would rather have 10 games that I'd play for weeks or months (if I'm playing on my hdtv through one of these I want better water, buildings etc) rather than games I fire up for less than 20 minutes as their just another angry birds variant and arguably useless on your TV.

    I want these devices to KILL the consoles next year and make MS/Nintendo etc give it up in 2015 or whenever the next revs should come. I hope they just realize we won't buy them anymore. DirectX11 on my phone/tablet and probably standard 25x14 resolutions by then (won't all be retina by 2015?) make a console purchase STUPID. This could be the merging of console/pc we need since phones/tablets rev yearly like pc's instead of 10yr console's stuck in stone stagnating gaming. Your phone as a portable console with xbox/ps3/pc gamepad support would be excellent. Pump it out to a monitor and keyboard/mouse setup and you have a notebook replacement too...LOL Now if they'd just put in a few extra cores by then that will disable if on their own screen but turn on when on a larger display like TV/Monitor and we have exactly what we want in both cases :)

    Pipe dreams? Retina is here now, and gamepads sort of. Next stop cores that only turn on depending on display output :) Awesome battery on the road, and great power in the dock at home pushing your 27in monitor. :) The 28nm versions by xmas of everyone's chips should come close to console power or surpass them. Interesting times.
    Reply

Log in

Don't have an account? Sign up now