Feeds

ATI: all our GPUs will be 80nm by July 2007

And then it'll release 65nm parts...

Choosing a cloud hosting partner with confidence

ATI may have only just shipped its first 80nm graphics chip - the Radeon X1950 Pro, reviewed here - but it's already talking about the 45nm node. It's out to adopt such a process by 2008, a senior staffer has revealed.

Speaking at the Taiwanese launch of the X1950 Pro earlier this week, Edward Chou, marketing chief at ATI's Asia Pacific operation, said the company began sampling early 65nm GPUs during the middle of the year, according to a DigiTimes report.

He also noted 80nm samples started coming off TSMC's production line late 2005, so at that rate, ATI's first 65nm parts will appear next summer or thereabouts. By then, of course, the company will be owned by AMD and probably called AMD too.

During H1 2007, all of AMD/ATI's GPUs will be fabbed at 80nm, said Chou, adding that the process punches out chips that are 10-15 per cent cheaper than their 90nm equivalents. In other words, the process is effectively yielding 10-15 per cent more working GPUs per wafer, a figure that can only increase as the yield rises as the process matures. ®

Related review

PowerColor Radeon X1950 Pro

Security for virtualized datacentres

Whitepapers

A strategic approach to identity relationship management
ForgeRock commissioned Forrester to evaluate companies’ IAM practices and requirements when it comes to customer-facing scenarios versus employee-facing ones.
Storage capacity and performance optimization at Mizuno USA
Mizuno USA turn to Tegile storage technology to solve both their SAN and backup issues.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?
Beginner's guide to SSL certificates
De-mystify the technology involved and give you the information you need to make the best decision when considering your online security options.
Security for virtualized datacentres
Legacy security solutions are inefficient due to the architectural differences between physical and virtual environments.