Interesting Micron slide at Comdex

So how long until Intel starts selling me a P4 with 512MB 256bit RAM onboard a card that plugs into a slot on the MB like video cards? :) Seriously, that's a huge performance gap, and I think that would be a very interesting step. The mos speed is required between the CPU and memory; everything else on the system could communicate just fine on the slower bus the CPU would be limited to if it had to be attached to a slot on the MB. Sure, for consumers the initial investment would be higher, as the RAM would be nonupgradable, but imagine the speed increase!

(Obviously there are a whole host of technical issues and systemwide considerations that I haven't discussed, and the cost of change would be not only high, but it would probably alienate RAM makers. But who cares? Imagine a CPU with 20GB/s bandwidth! :))
 
This makes it clear that now R420 or NV40 future version will ever get to use 512 Bit memory BUS .

We'll have to wait until 2006 for that .
 
Who needs 512b busses when you can have 800MHz GDDR3 on 256b jalopies? :)
 
I like the prospect of 100Gb/s+ in a videocard in the not extremely distant future... what was that often repeated bandwidth requirement for real time Toy Story now again? ;)
 
So how long until Intel starts selling me a P4 with 512MB 256bit RAM onboard a card that plugs into a slot on the MB like video cards?
You mean like Slot1/A?
But newer & with RAM rather than L2 cache.
512mb of L3 cache?
 
Ahhh....memories of Talisman...published in 1996 IIRC:

http://research.microsoft.com/MSRSIGGRAPH/96/Talisman/Html/fundforce.htm

Memory bandwidth is a key indicator of system cost. The left hand two columns indicate where current 3D accelerators for the PC are falling. A full up SGI RE2—a truly impressive machine—boasts a memory bandwidth of well over 10,000 MB per second. Its quite clear that SGI has nothing to fear from evolving PC 3D accelerators, which utilize traditional 3D pipelines, for some time to come.....These charts suggest that achieving high-quality imagery using the conventional graphics pipeline is an inherently expensive enterprise. Those who maintain that improvements in CPU and VLSI technology are sufficient to produce low-cost hardware or even software systems that we would consider high-performance today, have not carefully analyzed the nature of the fundamental forces at work.

:D
 
Back
Top