XCore Exchange

Great article on the Financial Times this morning about XMOS with a super opening paragraph:

"Under the solemn gaze of a life-size toy moose and rhinoceros, David May and Terry Leeder are explaining their plans for turning the small business started by Prof May into a big part of the global semiconductor industry."

If you have ever visited the XMOS office that will probably bring back a memory or two.

Go check it out here: http://bit.ly/gZgE4u (the FT requires you to register for free to read it, but it is a good read none the less).

It was a good article, apart from a bit of hyperbole regarding Altera and Xylinx (sic). XMOS only competes with them on low-end applications, and it isn't its main market.

Can we have a photo of the moose and rhino?

The XMOS marketing I've seen pitches the product up against FPGAs and as Altera and Xilinx are the top dogs there, it makes sense to me. What do you consider to be the primary market then?

The main market is high-performance embedded systems.

..and high-performance embedded systems don't use Xilinx and/or Altera then, in your view?

They mostly use MCUs, like ARM chips. I suppose they could use an MCU like the Xilinx MicroBlaze in an FPGA, but that's expensive unless the FPGA is also needed for something else.

Well I'd guess most people using ARM these days would be going for SoCs, where the peripherals are already provided. I think the latest incarnation of mid-range ARM (Cortex M4) might be a competitor for XMOS. Certainly I'm torn between the two of these for my project - the biggest draw is the standard instruction set which makes the code eminently portable across different vendor's platforms. To me that's one of the biggest (maybe the only major) weaknesses of Xcores.