Oct 12 AT 11:01 PM Taylor Wimberly 45 Comments

3DMarkMobile GPU showdown: Adreno 205 vs PowerVR SGX540

Last month we got a look at some leaked benchmark scores and determined the HTC G2 was going to offer about the same performance as the Samsung Galaxy S. Now we finally got a G2 in-house so I thought it was time to put both devices head to head and see which is faster.


The G2 features the new Adreno 205 GPU from Qualcomm. This GPU is found in the second-gen Snapdragon chips including the MSM7x30, MSM8x55, and QSD8x50A. HTC is the main handset maker using these new Snapdragons and you will find them in their upcoming phones including the Desire Z, Desire HD, G2, and Merge.

The Samsung Galaxy S features the PowerVR SGX540 GPU from Imagination Technologies. You will find PowerVR GPUs in both Samsung and Texas Instruments processors. They are also found in the iPhone and all of Intel's processors (CE4100) for Google TV devices.

For this first round of benchmarks I will be using 3DMarkMobile produced by Rightware. If the name sounds familiar that is because 3DMarkMobile was formerly owned by Futuremark who makes the popular PC benchmarks.

The Test Devices

I have a lot of phones that I could have benchmarked, but I kept it to three to keep it simple. I chose the G2 since it is the only device shipping in the U.S. with the Adreno 205 GPU, the Nexus One since it is a 1st-gen Snapdragon, and the Galaxy S because it has the fastest GPU we have benchmarked.

  • HTC Nexus One: 1 GHz Qualcomm Snapdragon QSD8250, Adreno 200 GPU, 512 MB RAM, Android 2.2
  • Samsung Galaxy S (Epic 4G): 1 GHz Samsung Hummingbird S5PC110, PowerVR SGX540 GPU, 512 MB RAM, Android 2.1
  • HTC G2: 800 MHz Qualcomm Snapdragon MSM7230, Adreno 205 GPU, 512 MB RAM, Android 2.2

Both the Nexus One and G2 are running stock Android 2.2, while the Galaxy S (Epic 4G) is running Android 2.1 with Samsung's TouchWiz UI. All devices run at the same 800x480 resolution. Each phone was rebooted before testing and I left the radios on to simulate normal every day usage.

The Video

3DMarkMobile ES 2.0 V1 - Hoverjet and Taiji

3DMarkMobile ES 2.0 V1 results.


Taiji: The Taiji Girl test represents a game with human-like characters. The girl in the scene is skinned using 29 bones. The total amount of polygons in the scene is 70,000 of which approximately 50,000 are visible per frame on average. This test is designed to provide a heavy fragment load. The scene is set in a serene courtyard garden enclosed on all sides by stone walls. A larger context is implied to the viewer using a skydome texture and trees implemented as alpha-blended quads visible beyond the courtyard wall. All of the fragment operation intensive elements are located inside the courtyard in a small radius around the main character. We believe closed environments, such as the courtyard in this test, are a viable solution for real-world game design. Limited environments will enable the developer to concentrate on the important elements of the set (those that are visible) and avoid the need to generate large amounts of unseen graphical elements which do not add value to the story nor to the game play. As shadow mapping is used in this test, the depth texture extension is required in the hardware in order to get approved results.

Hover: The second graphic’s test represents a racing game including five racers and a racing track in a desert environment. The graphics load presented is balanced more towards polygons than pixel processing. The total amount of polygons in the scene is 300,000 of which on average 100,000 are visible per frame.

The difference between the two tests  is that the Taiji test has about 70k polygons and has a heavy pixel shader load while the hover test has 300k vertices (of which around 100k is visible per frame) and light pixel shader load. Thus, the two tests stress different sections of the graphics processing pipeline. The key point about benchmarking is that a score in a single device without comparison scores from other similar devices has very little meaning. You can use our benchmarks to compare similar devices to each other, but on a single device the relative scores between tests are not very informative. So, if a single device has better scores at Taiji than at Hoverjet it will give you very little valuable information. However, if you compare scores from at least two devices the differences in scores between individual tests are immediately quite telling. So, if device A beats device B in all tests then clearly A is the performance winner. On the other hand, if device A  beats B in taiji but loses in hover it tells you that A has a better performance at evaluating pixel shaders but does not handle high geometry loads as well as B. In that case, you need to analyze the application portfolio you wish to run on the device to decide whether you value a higher geometry load than pixel load or vice versa.

Taylor's comments: To my surprise the G2 came out on top in the pixel shader and geometry tests. I ran them several times to make sure everything was correct and the G2 always scored better. This could be caused by some optimizations in Android 2.2 (like the JIT compiler) so we will have to run this benchmark again when Sammy releases Android 2.2 for the Galaxy S.

3DMarkMobile ES 2.0 V2 - User Interface and Navigation

3DMarkMobile ES 2.0 V2 - User Interface test results

User Interface: The User Interface test represents a realistic user interface including eight main icons and their sub-icons in a futuristic environment. Main and sub icon shells use Cook Torrance shading with cube map reflections to achieve a realistic metal effect. The center of the icon features a texture distorted fragment shader. Environment uses per pixel Blinn-Phong shading with diffuse and specular maps. Fake ambient occlusion is achieved by baking it in the colors of the vertices.

Bloom and Depth-Of-Field post processing effects are used in the entire scene. The blurring of the scene is done only once, and it is used by both of the post processing effects; this conserves system’s computing resources.

Taylor's comments: Once again the G2 comes out on top. This could be caused by Android 2.2 like I mentioned above. Notice the large performance gap between the first-gen Snapdragon (Adreno 200) and the second-gen Snapdragon (205).

3DMarkMobile ES 2.0 V2 - Navigation test results

3DMarkMobile ES 2.0 V2 - Navigation test results


Navigation: The Navigation test represents a realistic workload of a Navigation application. Buildings and roads are modeled to look like generated map objects based on real data. Lighting is created by using one point and one directional light to achieve dynamic daytime lighting. The road shader is dynamic. It draws the path of route while the position indicator moves. The Indicator shader uses backlighting to highlight the indicator from the background. The Water shader uses normal and cube mapping to strengthen the effect of the water area. Basic blur post-processing effect is used to blur the route, which is seen behind the buildings, to avoid the path from disappearing out of view.

Taylor's comments: Finally we see the Galaxy S (Epic 4G) take the lead on something. I will be honest and say I'm not the smartest when it comes to all these shaders and blurring techniques so I can't explain why this benchmark is the opposite of the others. The Galaxy S has a faster clocked CPU than the G2 (1000 MHz vs 800 MHz) so maybe this test is CPU limited.


This debate is far from settled. We can clearly see Qualcomm has come a long way from their first Snapdragons and made great progress with the Adreno 205 GPU. I was starting to worry that Qualcomm was in trouble with their Adreno GPU family, but it holds its own against PowerVR and now I'm pretty excited about the Adreno 220 GPU coming in future dual-core Snapdragons.

Right now I'm not sure if the PowerVR SGX540 is still the fastest GPU available in an Android phone. This 3DMarkMobile benchmark is still new so I'm going to run the latest GLBenchmark 2.0 suite on it tomorrow and see how that turns out.

For all we know, the G2 could be scoring better since it has Android 2.2 and the Galaxy S is limited by Android 2.1. It really is unfair to compare different firmware versions, but that is all I have to work with. Hopefully Samsung rolls out Android 2.2 to the carriers this year and we can benchmark these phones on an even level.

If anything, I think this demonstrates how up-to-date firmware can sometimes be more beneficial than whatever the latest hardware is. (Yeah I'm looking at you Samsung. Why are you slacking with Android 2.2?)

I'm sure quite a few people that are smarter than me will read this post , so please post your educated feedback in the comments.

Source: Rightware

Taylor is the founder of Android and Me. He resides in Dallas and carries the Samsung Galaxy S 4 and HTC One as his daily devices. Ask him a question on Twitter or Google+ and he is likely to respond. | Ethics statement

    Most Tweeted This Week

  • http://Website Daniel

    I have to say, the difference in Android version alone could invalidate the hardware performance tests. It’d be a good idea to include Nexus numbers running 2.1, so that we can tell how much of a difference it might make.

    • KaiserJay

      Shouldn’t make much difference, as 2.2 improves the CPU bound performance and not the GPU performance.

      But I’m guessing the reason Taylor used a Nexus with 2.2 on this benchmark was to negate the usual gripes about it being “unfair because you’re comparing 2.1 v 2.2″.

      • http://Website watbetch

        The N1 did see a jump in benchmark graphics performance when updated to 2.2.

        I don’t understand the point here besides getting a bunch of G2 and Galaxy S fans riled up. The PowerVR SGX540 is better than the Adreno 205, yet the G2 has better benchmarks in this test by a frame or two (despite the huge scale, there is about 1-2 frames in difference between the two devices except for the Navigation test). Then the Epic isn’t on 2.2, this is only one benchmark, etc.

      • http://Website Daniel

        These tests are not entirely GPU-based, though. Even in the most GPU-intensive ones, there’s still a bit of CPU work going on, and the JIT could skew that.

        Perhaps more importantly, though, the JIT wasn’t the only change in Android 2.2. There were a number of minor changes in various other areas, and they can add up to very significant performance gains (or even loss, possibly!). Tweaks to memory management or the kernel’s scheduler, even marginally small configuration changes, can have very drastic results in how certain applications perform, and if I recall Android 2.2 does have a slightly more strict memory reclaiming policy.

    • Davids mom

      Just wanted to say:in CM6 i turned of JIT and scores HARDLY went down

  • http://Website James

    WOW!!! I NEver knew the new Adreno 205 would actually out do the hummingbird!

    • Nicko01

      Now not only are we comparing processing speeds of the CPU’s, but also the GPU’s! I love where this is going.

  • http://Website kidphat

    Any benchmarks for NVIDIA Tegra II?

  • http://Website Aaron

    Can’t we just flash a 2.2 Rom on the galaxy s?

    • http://Website Derek

      No. Samsung hasnt released a 2.2 ROM yet. Also, there is no 2.2 AOSP that will work on Samsung hardware. Samsung has not released all the drivers for their hardware.

      There’s some hacked up versions of AOSP 2.2 on XDA that can be flashed to the Galaxy S, but half the features (like data or voice) dont work. For all the bragging that android fanbois like to do, Android is NOT open. Just get a Galaxy S phone and try to flash different ROMs on it, you’ll quickly find out how “open” Android really is. Its crap. Trust me, I know, I’m stuck in a 2yr contract with a garbage AT&T Captivate.

      • phill

        um android is open its the device manufacturers that lock and fuck everything

  • http://Website jim2point0

    The adreno 205 chips do seem to be newer than the hummingbird. This sort of thing is expected. Technology movies at a pretty quick pace. By this time next year, the adreno 205 will be a dog.

  • http://Website Sbob

    @Aaron: They are a few firmwares with 2.2, just google them.

    And i dont think 2.2 for Galaxy will bring such big performance-boost until its isnt optimized for JIT like the snapdragon-phones.

    • Taknarosh

      I don’t think JIT compiler has anything to do with the CPU it’s running on. In fact, Qualcomm and Samsung use the same ARM architecture to fabricate their CPUs.

  • Taknarosh

    The real problem with the Galaxy S is the archaic FAT base I/O system on it. Fix that + 2.2 JIT and I think the results aren’t that close.

    • http://Website watbetch

      Samsung is really into their RFS. I don’t think they will switch it up any time soon. Besides after receiving JI6 on my Vibrant, I and others agree there is no need for any lag fix.

  • http://Website Sergio

    Taylor, if you benchmark using quadrant, check the 3d scene – you will notice that power vr is better than andreno there.

  • http://Website Bagera

    This might not be a fair comparison due to 2.1 vs 2.2 but that’s how you will experience it in real life.
    So ok, says nothing about potential gpu performance but that doesn’t matter to the user.

    • http://Website Daniel

      In this case, it’s a G2 vs Galaxy S October 2010 comparison, not a GPU comparison at all.

  • http://handheldusers.com Amir Rafieian

    Samsung enable the Vsync on Galaxy S , so the refresh rate is lock on 56fps, but the real power of PowerVR SGX540 is far better than this …

    Even on the PC , the Vsync is on because of lower power consumption ….

    • http://Website gee

      You might have had a point if any of the tests actually hit 56FPS, none even came close.

      • http://Website Antwan

        The purpose of Vsync is to sync the framebuffer to whatever the display hardware is being used. Any energy saved would simply just be a side effect.

  • http://Website freddie

    have the text with samsung galaxy tab.have android 2.2 froyo

  • http://Website Norm

    Where does one obtain the Android version of this benchmark?

  • http://Website orangearrows

    Samsung —- you listening to all this. Now do something about it.

  • http://Website Shanekwa Johnson

    Does anyone else get turned on by this fine piece of white chocolates voice. Damn baby, talk dirty to me, say Snapdragon one more time.

  • http://Website Mik

    Perhaps an updated test, now that sgs has 2.2¿

  • http://www.lmfao.com lmFao


    remember the old days with 3dmark.

    wow that just threw his credibility out the window.

    Futuremark still makes 3dmarks…..
    what a joke

  • http://Website Jorgen

    This is a laugh, I’m loving these biased comments. Especially watbetch: “SGX 540 is better, because it just is”.

    Why don’t you just go and get a HTC Desire with a 2.1 Android and run the test on it. Then upgrade to 2.2 and run the test again to see the difference. Stop this whining and put some facts to the table.

    • http://Website Fishy…

      I’m calling bs on this, because on another opengl es 2.0 benchmark called nenamark, any galaxy s model rapes the g2. Just look through the comments for the program called “nenamark” in the android market.

      And people who are biased are stupid. cuz new technology always comes out and its usually better so eventually something will be better than the galaxy s.

      But as of now, and taking benchmark results from other benchmarks, it doesnt follow the other patterns, and i call bs, until another reliable tech site does the benchmark.

      • http://Website Tomas

        My result on galaxy S is 50,6… and G2?

  • http://Website Chris

    lol also he picked the epic lol which is probably the slowest of the four galaxy s phones from me using them the other three are much faster if he threw like the vibrant or fascinate i think results would be different because my fascinate scores higher

    • chris0101

      The Epic is actually considered the fastest of all of the 6 galaxy S phones around in terms of speed. The other 5 variants (international, AT&T, Verizon, T-Mobile, and US Cellular) are slower variants. One serious drawback though is that the Epic has less internal RAM, which may make it faster.

      • Fulaman

        They all have the same amount of RAM (512mb) with 128 mb going to the GPU. That said, The Epic 4G may be the fastest due to the least bloatware. They are all exactly the same when rooted and running MTD based roms.

  • http://Website Samurai

    ok, have you ever heard that the galaxy S have 90M/t speed (wll this MAY be true), but do you know also that galaxy S dont have the required RAM speed which is about 10MB/s. So the phone cant handle those 90M/t which Samsung announce very loudly .
    Adreno 205 have 22M/t but G2 takes them with RAM speed of 10MB/s (the same like in Galaxy S), so its not the GPU and CPU there is also and RAM.

    I will give an example how Galaxy s look in the PC world:
    You go to the shop and decide build home PC with separated parts. So you going and buying the best video card (some Nvida dual chanel) , then motherboard, then 8GB RAM , hard disk etc. etc. and at the end you buying an old Pentium 4 at 2GHz . Open youtube and try to play HD video, not 1080p, try 720p. You will watch something like 10-12 fps.

    • http://Website Nenamark

      Try Nenamark and Neocore on both. Galaxy S rapes G2 on both. Who knows what this benchmark is about? I think it has alot to do with optimizations for Snapdragon or Android 2.2 JIT.

  • http://Website Jimmy

    I hate all the fanboy stuff, I also dislike reviews, benchs, etc.. when not all the phones are tested under equal conditions, thats´s just ridiculus.

    Testing should be done on same os to start with, having people giving their own opinions on how much 2.1 vs 2.2 will affect and be noticable in the tests is of no intereset fpr anyone?? People do alwasy have opinions but seem to forget that real world often give a damn about those opinios, so test on perfect contitions and let the winner speak for it self. Doesn´t really matter anyway since they´r all enough for me, also towmorrow there´s something much faster on the store shelfs anyway.

  • http://Website D79_

    2.2 puts the benchmarks alot higher than 2.1 as i tested my sgs with 2.1 then with 2.2 and the difference between scores was totaly different. 2.2 is much faster within cpu gpu etc so there defo needs to be another test here to compare scores properly. Me personaly thinks the PowerVR will out do the adreno 205 but and this is a big but…. Wats the point in makin phones even faster than the sgs wen they dont even make games to really test the sgs, so how about get the software first then consider makin faster phones. The only thing i can see wat will help with faster phones is the camera at 1080p with not much compression etc…

  • http://Website Shocker

    PowerVR SGX540 is faster, not sure what the hell they are using to test these but there are quite a few good apps on the market that confirm this.

    Nenamark does a decent job, also try Dungeon Defenders, in their own recommendations Galaxy S is classed as High End and the Desire HD is mid range, I wonder why.

  • http://Website Eric

    guys, this article is outdated and extremely unreliable. If you want to see much better accurate results then check this link:

    and BTW, the epic is on 2.2 and is even compared with the nexus s(on 2.3)

  • http://Website shinji ikari

    to include a dedicated graphics processor if you have one of 1 gig is enough for the device you just have to integrate the language into the chip with that lower costs and save battery when you are using the games to run the game software that operating system will use a minimum of processor or memory load on the system files needed to leave the software to play the order of commands locate the files on the memory of that, you have more processing power for running and provide better quality of graphics

    • http://Website shinji ikari

      daniel blackman write that coment fron Panama


    And why? its because android doesnt support Gpu (hardware) acceleration and the results are cpu-native.

    Google android doesnt natively support hw acceleration, because some low end devices would caused problems with it and even made them slower. Btw the only reason why Google bought Motorla Mobile is implementing GPU acceleration for all oncoming Motorola phones.

  • Qre

    I don’t really know wich gpu is better. The only thing I see is that my Xperia Ray (with adreno 205) is much faster then my friend’s Galaxy S, both phones with gingerbread.