Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: greenspirit who wrote (131261)	3/30/2001 12:31:44 AM
From: Paul Engel	Read Replies (2) \| Respond to of 186894

Michael - Re: "could you comment on this FPGA technology? Specifically, do you think it's viable, and who leads this effort?" Altera and Xilinx started making reprogrammable logic boards ("computers") about 8 or 9 years ago. They had some university help and thought that this would be a great way to utilize gobs of programmable logic. After a few years of poor acceptance, the idea was put on the back burner - but every couple of years it seems to resurface - as it has now. In principle, given "enough" logic, FPGA's/EPLDs can be programmed to implement sophisticated processors and DSPS, etc. With on chip memory in the FPGAs, even "cache" can now be emulated. In fact, certain DSP functions could be performed MUCH FASTER using this technique than conventional "hard wired" DSPs. However, as general purpose processors, these aren't as fast as dedicated chips from Intel or AMD or Alpha - they are generally built on a foundry wafer process - and that process is more optimized for density, cost and yield and performance is a little bit slower than dedicated processors. There was a company called QuickTurn systems that used to sell these hardware simulators - just stuffed with Xilinx FPGAs and controlled by Sun workstations - and Intel was an early customer of theirs - back in the early 1990s. Intel effectively built a Pentium simulator using racks of these FPGAs from Quickturm to prove out the logic for the original Pentium before the chip layout was even completed. So - this technique has merit - but it will be more in the domain of specialty research where new architectures can be simulated by reprogramming the FPGA chips, new instruction sets can be tried, etc. Paul