A report over The Inquirer web-site suggests that graphics cards makers are not satisfied with the strategy of NVIDIA Corporation for PCI Express x16 graphics cards. Aiming to improve gross-margins this year, hardware makers are complaining about additional costs NVIDIA’s special bridge chip brings.
According to an article, NVIDIA’s chip that allows AGP graphics processors to work on PCI Express x16 bus and otherwise costs $5, a tangible sum for makers of add-in graphics cards. The bridge is needed to make graphics processors without PCI Express support to work with the bus before chips with native PCI Express x16 controller emerge.
Please click to enlarge
Currently NVIDIA’s graphics chips work using AGP 8x bus, but once they are equipped with the company’s AGP-to-PEG bridge, they will bolster the transfer rate to the so-called AGP 16x speed between the graphics processor and the bridge. The latter will work with PCI Express for Graphics lane at some 4GB/s speed. Therefore, provided that the bridge itself does not cause tangible latencies, NVIDIA’s GeForce FX as well as next-generation graphics chips will see a speed boost from PCI Express.
Even though native PEG x16 (PCI Express for Graphics x16) support ATI promises to bring may be more advanced from technology point of view and provide higher performance and bandwidth, NVIDIA’s approach may probably help the company to better control its inventory, as GPU lineup will be unified for AGP and PEG platforms. In contrast, ATI will have to stock two separate lines of its VPUs. Such approach means ATI has to design, tape-out and manufacture two families of processors, while NVIDIA needs to make one lineup of its chips as well as manufacture bridges. However, the strategy that is cost-effective for NVIDIA appears to be pretty expensive for graphics cards makers, if the report is correct.
Actions of graphics cards manufacturers we may expect from them are not clear. Even though $5 is significant sum for entry-level graphics cards, it is not a lot for performance-mainstream or high-end solutions initial customers for PCI Express platform may install.
PCI Express platforms are likely to emerge in the second half of the year.
Representatives for NVIDIA declined to comment.
Comments currently:
2
Discussion started: 03/29/04 09:02:12 AM
Latest comment: 06/12/04 08:23:53 AM
[1-2]
1.
The problem is not only the cost of the chip (5$) but the design of the cards will also be more expensive (more pbc, interconnections, etc...)
" In contrast, ATI will have to stock two separate lines of its VPUs. Such approach means ATI has to design, tape-out and manufacture two families of processors, while NVIDIA needs to make one lineup of its chips as well as manufacture bridges"
When saying that it seems that Nvidia aproach doesnt have separe lines. Haven't they also the AGP and PEG version of the card?
Why will ATI make two families of the processor? isnt RV3x0 and R3x0 already working in AGP 8X, so why bother?
[
Posted by: I
|
Date: 03/29/04 09:02:12 AM]
2.
This is really starting to look more and more like the life of AGP is limited once PCI-E gets really mainstream, say 2005. It could cost so much to keep an agp variant around (ie on the shelves and in production) that few vendors will do it, and the corresponding cost will be noticeably higher to the end user. The extended effect of this is that gradually end users stop buying, cost goes even higher and fewer choices available and pretty soon agp is gone altogether.
[
Posted by: Anemone
|
Date: 06/12/04 08:23:53 AM]
[1-2]
Make Home Page
Add to Favorites
XbitLabs is using Twitter
Windows Live Alerts
