ATI Technologies, who is not expected to release its next-generation R520 visual processing unit shortly, is reportedly manufacturing its new graphics processors at Taiwan Semiconductor Manufacturing Company (TSMC) in quantities. NVIDIA Corp. also seems to make its future the chips commercially now. The volumes are not clear, but related packaging and testing orders to Advanced Semiconductor Engineering (ASE) from the two graphics companies are likely to impact ASE’s Q3 results, which may indicate that the chips would be made commercially.

ATI’s R520 graphics chips have been produced in small volumes at Taiwan Semiconductor Manufacturing Company (TSMC) using 90nm process technology and DigiTimes web-site indicated that ASE has received orders for related wafer sorting and flip-chip (FC) packaging. NVIDIA Corp. is also expected to place orders with ASE for FC packaging and testing for its forthcoming G70 chips.

The report means that both leading graphics companies are on-track to release their next-generation visual processing units commercially soon enough. Both NVIDIA and ATI are expected to showcase their next-generation GeForce and RADEON GPUs at Computex Taipei 2005 trade-show behind closed doors, but Santa Clara, California-based NVIDIA is likely to be ahead with actual release of its GeForce 7800-series product. Still, both R520 and G70 graphics cards may become available widely only in the second half of Q3, when chips are tested, packaged, installed onto graphics cards and sent to stores. Nevertheless, according to some reports, the first commercial NVIDIA GeForce 7800-series products would emerge in June.

NVIDIA GeForce 7800 graphics processor is expected to feature 24 pixel and 8 vertex pipelines and to operate at 430MHz, HKEPC web-site recently informed. It is unknown whether the G70-series of the graphics processing units will get any feature-set upgrade compared to the GeForce 6800-series that supports Shader Model 3.0. It remains to be seen whether the G70 is to be called NVIDIA GeForce 7 or receive a different brand-name.

Some sources suggest that NVIDIA’s code-named G70 chip would be made using 0.11 micron process technology at TSMC. By contrast, NVIDIA’s present top-of-the-range chip GeForce 6800 Ultra is manufactured at IBM’s facility in East Fishkill, NY, using 0.13 micron fabrication process.

Specifications of ATI’s code-named R520 VPU, which was originally scheduled for release in June,  are unclear at this time. Some sources suggest that the chip may have up to 32 pixel pipelines and up to 350 million transistors, which makes the processor extremely complex. However, given that a new fabrication process is to be used for the manufacturing of R520, it is unlikely that the visual processing unit will be tremendously large in terms of transistor count and complex in terms of the number of pipelines. Fabless semiconductor designers tend to balance complexity of their chips for new fabrication processes. For instance, since 2002, ATI has not launched manufacturing of high-end graphics chips using a new process technology unless the technology was tested on mainstream chips. Still, even on relatively new manufacturing processes, ATI has set pretty high clock-speeds for its VPUs.