SI
SI
discoversearch

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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?


To: DRBES who wrote (200707)6/6/2006 5:27:50 PM
From: PetzRead Replies (1) | Respond to of 275872
 
re: My guess is there are algorithms that x86 (probably most others too ... ) blows chunks for, that you can quickly fabricate a specialized chip for it in Verilog or VHDL, and speed things up 1000%

LOL, typical snot-nose teenager punk kid talk.

(but true)

Petz



To: DRBES who wrote (200707)6/6/2006 5:34:13 PM
From: MagratheaRead Replies (1) | Respond to of 275872
 
Don't forget "depending on how easy it is to PROGRAM this stuff." The tools will come.

I'm a believer. I'm buying it hook-line-and-sinker.

It is going to lower the barrier to entry into silicon processor design. It is going to allow people to build new design without worrying about backwards compatibility. You let the Opteron/A64 do the grunt work and admin and you let your little jewel of silicon perform magic on the 10% of the code that does 90% of the work.

-Magrathea



To: DRBES who wrote (200707)6/6/2006 5:56:27 PM
From: AK2004Respond to of 275872
 
re: very specialized matrix linear algebra chip

was not there very efficient and very cheap data compression coprocessor (for wide variety of military and civilian applications) developed few years back in Houston?