The genre of mobile GPU rendering principles--IMR, TBR, and TBDR

The genre of mobile GPU rendering principles--IMR, TBR, and Tbdr

The mobile GPU is only a small child when compared to the desktop-level GPU, although children can have an advantage over adults on certain occasions (for example, acrobatics, contortion, etc.). But there are innate differences in power, and the main tables are now theoretical in performance and bandwidth.

Compared to the desktop GPU 256bit or even 384bit of bit width, 1.2-1.5ghz high-frequency memory. The mobile GPU not only shares memory bandwidth with the CPU, but also generally uses a dual 32bit bit-wide, lpddr2-800, or about 1066 memory system. The total bandwidth is generally within 10gb/s. The sad Tegra 3 uses a single-channel memory mode, with ddr3l bandwidth just 6.4gb/s.

The GPU's most powerful ipad 4 bandwidth is now just 17gb/s (image from AnandTech)

The highest memory bandwidth in the mobile processor is the ipad 3/4. Because they use the retina screen. 2048x1536 's high-resolution GPU bandwidth requirements are higher, but even if the two products, 17gb/s bandwidth and PC graphics on the 200gb/s more than the bandwidth is still pediatrics.

Without high bandwidth there is no high-volume texture data, and there is no quality. Although bandwidth is not the only factor restricting the development of mobile GPUs. But under the constraints of the moment. The first thing that a mobile GPU vendor cares about is how to improve GPU performance and image quality with the smallest possible bandwidth requirements, the texture compression described earlier is a method, and the other is to use different rendering methods. Mainly have IMR, TBR and TBDR and so on.

"Immediate rendering Mode"--IMR

IMR (Immediate Mode Rendering) is just as literal-the submission of every rendering requirement starts immediately, which is a simple and brutal idea. Strengths and weaknesses are very obvious. Assuming you don't have to worry about performance, this is a very convenient way to do it. However, the rendering of IMR is done without discrimination, and those parts of the masking process will still be rendered to the processor. This also leads to meaningless read and write operations in many other. Wasted a lot of performance and bandwidth.

In short, imr such a rendering method on the mobile GPU evaluation can only be "negative points, roll-coarse!"

”。

The clever "map rendering"--TBR

IMR silly approach is not advisable, then a clever way to--TBR (tile Based Rendering, map rendering), it will need to render the screen into a block (tile), each block of coordinates by the intermediate buffer in the form of a list in the system memory.

The advantage of this rendering method is that the relative IMR reduces unnecessary rendering tasks. The disadvantage is that the masking debris will still exist in small amounts and require an intermediate buffer.

TBR rendering divides the game screen into different chunks

Re-evolved rendering style--TBDR

TBR is much smarter than IMR, though. There are still a number of flaws, TBDR (Tile Based Deferred Rendering, map delay rendering), it is similar to the TBR principle, but using a deferred rendering (Deferred Rendering), combining the perfect pixel. The HSR (Hidden surface removal, hidden surface elimination) further reduces the need for rendering and reduces bandwidth requirements. In fact, these changes are somewhat similar to the rendering on a PC.

One key to tbdr rendering is deferred rendering

Several other manufacturers are using TBR technology, TBDR is mainly imagination in use, which is one of their biggest chips.

Support for TBDR is also mentioned in Microsoft's DX11.1 upgrade, as the Windows 8 system is also optimized for tablet and touch optimization, and it is also necessary to optimize the technology that is often used by mobile platforms such as TBDR.

The genre of mobile GPU rendering principles--IMR, TBR, and TBDR