As rival AMD was preparing to snap up graphics chipmaker ATI Technologies, Intel was considering topping AMD's offer or going after Nvidia, according to one of the company's top executives.
In an interview with The Inquirer, Pat Gelsinger, senior vice president and general manager of Intel's digital enterprise group, said Intel looked "pretty closely" at making a play for Nvidia or ATI, the two largest graphics chip companies in the world. Obviously, that never happened, as AMD closed its acquisition of ATI last year and Nvidia continues on as a standalone company.
Intel's Pat Gelsinger addresses attendees of the company's Fall 2007 Intel Developer Forum.
(Credit: Stephen Shankland/CNET News.com)Intel had some unique concerns that checked its ambitions, according to Gelsinger. "One issue was that we didn't know if we could because, if number one buys number two or three, what happens regulatory-wise?" Intel is the leading supplier of graphics technology for PCs because of its integrated graphics chipsets, and if it were to acquire a dominant share of the graphics market to augment its dominant share of the PC processor market, the U.S. government (well, perhaps the next administration) might have sat up and taken notice. And European regulators, currently hounding Intel on that continent, would almost assuredly have objected to the deal.
But graphics processors aren't just about rendering pretty pictures anymore. One of the real reasons graphics technology is attractive to both Intel and AMD is because graphics chips are very good at processing a stream of instructions at high speeds. That's why AMD bought ATI, and it has plans to integrate graphics chips directly onto a PC processor in 2009, a project known as Fusion.
Right now, those chips are designed to handle graphics data, but there's no reason why they couldn't be used for other applications that require high-performance computing, as long as the industry can figure out a way to program for those chips.
"The key transition (we're going through now) is in the graphics programming model," Gelsinger told The Inq. "The issue (GPU makers) have is making the pipelines more programmable, and we have the most programmable model on the planet--IA." IA (Intel Architecture) is Intel's term for the x86 instruction set; the company likes to remind everyone whenever possible that it came up with that idea.
Instead of teaching programmers how to exploit graphics chips, Intel's plan is to develop a project called "Larrabee" that will build a x86-compatible chip with the performance of a graphics chip. "Larrabee ends the debate on GPGPUs (general purpose graphics processing units)," Gelsinger said at the Beijing Intel Developer Forum in April. "This is what developers want." Both Fusion and Larrabee won't turn into products for a long time, so developers will have plenty of time to decide which model will prevail.
Check out the rest of The Inq's entertaining interview with Gelsinger, as well as the first two parts posted earlier in the week and the final part scheduled for tomorrow.
PALO ALTO, Calif.--AMD's pretty sure we all want better graphics on our PCs, and knowing us, they're pretty sure we don't want to cough up a lot of money to get it.
Phil Hester, AMD's chief technology officer, stopped by the Hot Chips conference here at Stanford University on Tuesday to talk a little more about Fusion, AMD's plan to integrate a graphics processor and PC processor onto the same chip. By the time the chip is ready around 2009, Hester thinks the growing explosion of video and 3D graphics on PCs these days will require an affordable chip that still delivers great graphics performance.
"It's not about the silicon, it's about the applications," Hester said. He stepped back in chip history to liken the Fusion project to Intel's decision to integrate a floating-point processor into the 486 chip. There's always a cost trade-off when you integrate something new into a processor that was once done separately. But when enough applications need the extra performance, it's easier to justify adding some cost to dramatically improve performance.
That's the plan for Fusion. It's not going to replace high-end discrete graphics chips coveted by gamers, and it's not going to deliver the ultimate in CPU performance, Hester said. But AMD thinks that integrating the GPU will be essential around the end of the decade because so many applications--games and videos, for starters--will want to latch onto the GPU architecture and because the relative performance of a GPU is way beyond the CPU right now, he said.
GPUs and CPUs have traditionally been designed with different priorities in mind. GPUs are designed to sling code in and out as quickly as possible and are good at working with parallelized code, the current issue for multicore processors. Traditional CPUs tend to focus more on code quality and solving problems in sequential order. There's a distinct advantage to having both types of processors on a single chip, so long as AMD can ensure that developers can write code for Fusion and the company clears the integration hurdles, Hester said.
Lots of decisions need to be made. Should PCs with chips like Fusion use DDR memory or graphics DDR memory? AMD's integrated memory controller architecture will require a decision one way or another. How will AMD open up the GPU so more applications can be written to take advantage of its parallel-friendly architecture? AMD hasn't worked out all the details just yet, but open-source elements could help bridge the gap between applications and the hardware.
It doesn't make sense to integrate everything onto a chip, but sometimes it's worth it to take the plunge, Hester said. AMD is betting that it can improve overall graphics performance and still give developers a way to take advantage of the GPU architecture for other tasks, while holding down the fort with traditional application performance. It has two years to work out the details.
SUNNYVALE, Calif.--AMD's Fusion chip will come in two varieties, one for PCs and servers and another for consumer electronics devices.
Bulldozer is the code name for the Fusion chip that will be designed for everything from servers to handhelds, said Phil Hester, AMD's chief technology officer. Bobcat is the name for a sub 10-watt x86 chip that AMD believes can power ultramobile PCs, cell phones and existing consumer electronics chips using the ARM or MIPS architectures.
You call in a bulldozer when you need a lot of earth moved in a short amount of time, Hester said. That's the idea for Bulldozer, in that it's the design that AMD wants to form the basis of its server and PC chips by the end of the decade. Bulldozer will be part of the "Falcon" PC platform that also includes an integrated memory controller, a graphics processor, cache memory and a PCI Express controller.
Bobcats, however, can be found in back yards and smaller spaces where you don't want to use a shovel, but you can't get away with a bulldozer. (I've always wanted to tool around in a Bobcat for a few hours.) These chips represent AMD's hope for getting x86 chips into handheld devices.
While x86 chips rule the PC market, it's really hard to find one in a smart phone. Both Intel and AMD are very interested in figuring out how to get their silicon inside this fast-growing part of the tech industry, and Hester thinks x86's time will come as software for handsets grows more powerful and the chips themselves become more power-efficient.
Hester also revealed a few more details about Sandtiger, the code name for AMD's 2009 server chip disclosed earlier in the morning. Sandtiger will use between 8 and 16 Bulldozer cores, but AMD might build smaller versions to take advantage of certain cases in which 4 cores make more sense, he said.
- prev
- 1
- next
