Replies to post #15189 on Advanced Micro Devices Inc (AMD)

[wbmw]

Blauboad, Re: Somebody here mentioned that they thought Alchemy might be eliminated.

People at Microprocessor Forum were wondering the same thing. So AMD already has Alchemy, right? But now they go and buy Geode - a lesser product with x86 compatibility - and now Webber is preaching that the world run x86. I think it speaks of an uncertain corporate strategy, if you ask me. AMD thinks they are the first one to advocate a single instruction set, and it was more than insulting to ARM and MIPS that AMD should proclaim AMD64 as the superior one. Both ARM and MIPS cracked a joke a Webber's expense when they took the podium for the high performance embedded portion of the conference. I didn't stay for the Q&A session, but I have to wonder if Webber was grilled as he should have been.

Meanwhile, Sun's keynote had a similar theme, but their solution was to solve the problem using software (i.e. Java). That seems to me a better solution than using x86 (or in AMD's case, their self-serving AMD64 advocacy), when it clearly doesn't come close to meeting the low power and high embedded performance of ARM/MIPS.

[CombJelly]

"I am skeptical about moving down."

So far, x86 hasn't made much of an impact in the low-powered embedded end of things because the dies are relatively larger and their power consumption is higher. In addition, for many types of embedded you have to do some, if not all, of the coding in assembly. And doing a significant amount of coding in x86 assembler is like eating ground glass. But I digress...

I am not sure there is all that much current x86 software that can be used for many embedded applications. And those that can be, like Linux, already have ports to other architectures. I suppose MS-DOS stuff can be used, I've worked on projects that used it, but the "why?" question kept coming up...

Which is not to say it can't be done, if but a fraction of the effort that went into, say, Willamette, were applied to embedded x86, something could be done. And if enough could be produced on a small enough process, they could be cheap. There is still the issue of programming in assembly, but if enough managers are sold on the idea of "all those tools, applications and programmers familiar with the architecture that are out there", it could even be successful.

I am guessing AMD is getting ready to release something new in the Geode line. Probably something that was developed by National before it punted the line. What this could mean for Alchemy, I dunno.

[yourbankruptcy]

I can be counted as amdroid, but even me, I doubt there is any big value in having x86 on handhelds computers.

Thouh I see good market for ultimate-portable notebooks, with a 7" screen size and 5x3.5x1" dimensions. The requirement is that it must have SVGA plug and several USB ports for keyboard and CD-rom.