ST's Spring refresh: Kyro II
With the exception of NVIDIA's GeForce3 GPU launch earlier this year, 2001 has been a relatively stale year for the 3D graphics market. The six-month product cycles of two years ago appear to be a thing of the past. Sure, NVIDIA has released its new line of GeForce2 MX boards, but the MX400 offers little (if any in some situations) gains over the original GeForce2 MX, and the GeForce2 MX 200 is aimed squarely at the sub-$100 market, where its Vanta line of TNT2 products has reigned. Clearly little has changed in the graphics market over the past six months.
While that statement would have been considered accurate a few months ago, STMicroelectronics (ST) has shaken things up a bit with its Kyro II graphics accelerator. If you recall, ST has partnered with Imagination Technologies to bring its PowerVR Series3 line of chips to the PC. In particular, Imagination Technologies' PowerVR division provides the technology, while ST is responsible for manufacturing and marketing the product. (ST had a somewhat similar arrangement with NVIDIA several years ago, so they're by no means new to the graphics market.) The first product to hit the retail market as a result of this deal was the Kyro graphics accelerator launched last summer.
Aimed squarely at the value segment, Kyro was a good start for ST but plagued early on by its drivers and lack of support from video card manufacturers. While several Taiwanese firms jumped onboard, none had a significant presence in the US retail market, relegating Kyro to niche status. With Kyro II however, ST has scored a significant design win with Hercules, arguably the most recognizable brand in 3D board manufacturers now that Creative has taken a step back from graphics to focus on its audio business.
Tile-based rendering: The technology behind Kyro/Kyro II
Unlike traditional architectures that draw the entire contents of a scene regardless of whether or not they're actually visible on the screen, the tile-based approach determines which areas are visible and which ones aren't. We'll take Sarju's smiley face example from our original Kyro article:
To generate the previous image, a traditional immediate mode renderer would draw both smiley faces and then cut out the right portion of the blue smiley face (behind the yellow smiley face). This cut out portion is known as overdraw, as it isn't used in the final scene. On the other hand, the Kyro II never draws this unused portion; instead it only draws the visible portions of a scene:
By drawing only the visible portions of a scene, the Kyro II doesn't perform the redundant work of conventional graphics accelerators, and most importantly doesn't perform additional texture fetches from memory, an important aspect that allows the Kyro II to make more efficient use of its available memory bandwidth. Our smiley face example is an extremely simplified example of the challenges facing today's graphics accelerators. With scene complexity in games rising each year, overdraw continues to suck in more bandwidth. Graphics manufacturers have dealt with this problem by moving to faster DDR memory, but this can be an expensive solution. Both ATI and NVIDIA have implemented their own solutions for dealing with overdraw in their RADEON and GeForce3 products respectively, but neither implementation is as aggressive as tile-based rendering. Is tile-based rendering the way of the future? If you ask Imagination Technologies, the answer is a resounding yes. ATI and NVIDIA think differently however.
This is a brief summary of how the tile-based rendering scheme implemented in the Kyro family of graphics cards works. For a more in-depth explanation, please refer back to our original Kyro article, or PowerVR's white papers, as this is a well-documented topic. Right now we want to get into what's new with the Kyro II.