GPU Performance

GFXBench Offscreen – T-Rex

The T-Rex test is based on OpenGL ES 2.0 and includes textures, material, geometry and particle effective that were highly detailed at the time of release. The graphics rendering engine features planar reflection, specular highlights and soft shadows, providing a good workout even for flagship smartphones and tablets. Offscreen tests are run at 1080p, regardless of the device’s native resolution, and are best used to compare the performance between competing silicon, not competing devices.

The PowerVR G6430 inside the Atom Z3580 starts out with a mediocre result, a score of 1728 that is just barely faster than that of the Apple A7 GPU. Both of the Tegra K1 platforms and the Apple A8X are clearly faster on the GPU front.

GFXBench Offscreen – Manhattan

Manhattan was the first benchmark to utilize OpenGL ES 3.0 features and uses a nighttime setting with a lot of external illumination to stress the GPU. It uses a deferred rending engine with multiple render targets for the geometry pass, includes both diffuse and specular lighting, uses depth shadow maps, bloom, depth of field and quite a bit more. 

Things don’t really look any better after moving to the OpenGL ES 3.0 test here – the Atom Z3680 is slower than all four of the other tablets tested here today.

GFXBench Offscreen – ALU Performance

This test measures the pure shader compute performance using a fragment shader and rending a single full-screen quad. 

Raw ALU results keep the PowerVR graphics architecture where you’d expect it, slower than the Kepler-based graphics from NVIDIA and behind Apple’s latest architecture in the A8X.

GFXBench Offscreen – Alpha Blending Performance

Rendering semi-transparent quads with uncompressed textures allows this test to measure the alpha performance of the GPU directly. 

Alpha blending performance is similar between the A7, the Atom Z3580 and the Tegra K1 Denver part with only the A8X running away with a synthetic performance victory.

GFXBench Offscreen – Driver Overhead

This test renders a large number of very simple objects one-by-one, changing state with each item in a pattern consistent with real-world applications. This allows the benchmark to measure the CPU overhead of the OpenGL driver. 

Not a good result in the driver overhead result either where the benchmark attempts to measure how much compute overhead there is to manage the GPU pipeline.

GFXBench Offscreen – Fill Rate Performance

The portion of the test measures texture fill rate performance by rendering four layers of compressed textures, a very common scenario in gaming. 

Finally we get to the fill rate that finds the Dell Venue 8 7000 comfortable towards the bottom of our test results yet again. Clearly the PowerVR graphics at work on the Atom lineup isn’t capable of raw performance that impresses.

GFXBench – Native Resolution

All of the above tests were run at 1080p to compare the performance of the GPU systems on equal ground. But tablets are not equal ground – the displays that are part of the platform different from unit to unit and that resolution can and does directly impact perceived GPU performance. Take the 2560×1600 resolution screen of the Dell Venue 8 7000 and compare it to the 1920×1200 screen on the NVIDIA SHIELD Tablet. We have already determined that the SHIELD has a better GPU overall, so putting a higher relative load on the PowerVR graphics in the Atom Z3580 (via additional pixels to render) should result in an even bigger swing in these benchmarks.

That proves to be the case in the Manhattan test where the Dell Venue 8 7000 falls from a score of 743 frames to 436 frames, a drop of 70%. The SHIELD Tablet however drops from 1946 frames to 1868 frames, only a 4% drop. Both the iPad Air 2 and the Nexus 9 see larger drops as well with higher than 1080p native screen resolutions (33% and 44% respectively).

3DMark

Use 3DMark Ice Storm Unlimited for chip-to-chip comparisons of the hardware inside your device without vertical sync, display resolution scaling and other operating system factors affecting the result. In Unlimited mode the rendering engine uses a fixed time step between frames and renders exactly the same frames in every run on every device. The frames are rendered in 720p resolution "offscreen" while the display is updated with small frame thumbnails every 100 frames to show progress.

Ice Storm Graphics test 1 stresses the hardware’s ability to process lots of vertices while keeping the pixel load relatively light. Hardware on this level may have dedicated capacity for separate vertex and pixel processing. Stressing both capacities individually reveals the hardware’s limitations in both aspects. Pixel load is kept low by excluding expensive post processing steps, and by not rendering particle effects.

Graphics test 2 stresses the hardware’s ability to process lots of pixels. It tests the ability to read textures, do per pixel computations and write to render targets. The additional pixel processing compared to Graphics test 1 comes from including particles and post processing effects such as bloom, streaks and motion blur. The numbers of vertices and triangles are considerably lower than in Graphics test 1 because shadows are not drawn and the processed geometry has a lower number of polygons.

The purpose of the Physics test is to benchmark the hardware’s ability to do gameplay physics simulations on CPU. The GPU load is kept as low as possible to ensure that only the CPU’s capabilities are stressed. The test has four simulated worlds. Each world has two soft bodies and two rigid bodies colliding with each other. One thread per available CPU core is used to run simulations. All physics are computed on the CPU with soft body vertex data updated to the GPU each frame. The background is drawn as a static image for the least possible GPU load. The Ice Storm Physics test uses the Bullet Open Source Physics Library.

Here’s an interesting result – in 3DMark the Dell Venue 8 7000 is nearly on par with the iPad Air 2, but falls behind the Tegra K1 devices with the obviously superior Kepler graphics core.

Looking at the raw GPU scores though, the Atom Z3580 still has a fairly low result, barely edging out the Apple A7 in the iPad Mini Retina and scoring 87% lower than the Denver-based Tegra K1 Nexus 9.

But here is where the Atom processor and Silvermont architecture see some benefit – the CPU based physics test of 3DMark Ice Storm mobile suite puts the quad-core Intel x86 processor at the top of the stack, above even dual-core Denver CPU from NVIDIA.

Basemark X

Basemark X is the world’s most popular benchmarking tool for evaluation and cross-platform comparison of gaming and graphics performance between Android, iOS and Windows Phone 8 smartphone and tablets.

Basemark X is the only vendor-independent benchmark that utilizes the real-world game engine Unity which is very popular among game developers. This means that it scores correlate exceptionally well with real-life gaming performance.

Basemark X includes two game-like graphics tests: Dunes and Hangar. Both tests contain heavy graphics content rendered with detail and complexity, thus pushing the measured device to the limit. The polygon counts in test sequences are up to 911,000.

The Basemark X results put the Dell Venue 8 7000 at the bottom of the pack, even below the iPad Mini Retina.

Basemark OS II – Graphics

Basemark OS II is a system-level All-In-One benchmarking tool designed for measuring overall performance of smartphones and tablets from all platforms, including Android, iOS and Windows phone 8.

The benchmark features a comprehensive suite of tests including system, internal and external memory, graphics, web browsing, camera, battery and CPU consumption.

Finally, in the Basemark OS II Graphics sub-test we find the PowerVR graphics matching the performance of the A7 but falling woefully behind all of the other competitors.

« PreviousNext »