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.
Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?


To: Paul Engel who wrote (47849)1/29/1999 5:18:00 PM
From: RDM  Read Replies (1) | Respond to of 1571808
 
They may not put their pants on two legs at a time, but at least those Intels are not total fools. 18-24 months is a little far away, but may get their in time for their 800 Mhz and above CPUs.

I have seen references that stated the 200 MHz of of EV6/K7 is just the first generation and that the "design" indicates a 400 MHz possibility.

We all have our dreams I guess. For those not used to Geek Speak this 400 MHz at 64 bits wide is 2.56 Gbytes/sec. Also possible is 64 bits at 32 bits and 128 bits (1.28 Gbytes/sec and 5.12 Gbytes/sec). There is a considerable possibility that we will not be seeing these bandwidths in a mass-produced product for some time (< 3years).




To: Paul Engel who wrote (47849)1/29/1999 5:31:00 PM
From: Tenchusatsu  Read Replies (1) | Respond to of 1571808
 
Paul, NGIO is very, very different from any sort of bus-topology idea. NGIO is more of a switched fabric allowing devices to be connected outside the server chassis, much like Fibre Channel or a network. I don't know much about NGIO yet, but I imagine that the reason it's point-to-point is precisely because of this switched-fabric paradigm.

Plus, I imagine that NGIO uses very few pins or wires, but transmits data across the small number of wires at extremely high speeds. That's another benefit from the P2P: it allows fewer pins and wires.

Tenchusatsu