21264 information

[Bohn, Christopher A.]

		The article is
<http://slashdot.org/articles/980722/1459215.shtml>
http://slashdot.org/articles/980722/1459215.shtml - unfortunately, slashdot
appears to be down right now, so I cannot give you the URLs for the articles
that it has links to.  Here's the nuts'n'bolts:

-	64 bit processor
-	Can dispatch up to 4 integer & 2 floating point instructions
-	Out of order execution
-	4 different branch prediction algorithms; best result selected
on-the-fly
-	I've been told (but haven't found a hard reference) that when fully
"trained" the branch prediction scheme can get up to 98% correct predictions
-	Twice the performance of the 21164 at the same clock rate
-	64KB on-chip dcache; 64BK on-chip ichache
-	"up to 5GB bandwidth for L2 cache"
-	DEC claims it can provide broadcast quality & real-time
interactivity
-	Suuport for WinNT, UNIX, Linux, VMS
-	Expect systems to be introduced Q3/98
-	"Industry open standard socket - AMD K7 chipset support and Slot A
interchangeability on the same mohterboard"
-	"Scalable architecture based on cross-bar switch chipset
implementations"
-	Chipsets available from "more than four" leading-edge vendors
-	Uniprocessor, dual, quad, 24+
-	Pricing & Performance (quantities of 1000)
-	KP21264-600CN		$5,500		Est 40.0i SPECint95, 60.0
SPECfp95
-	KP21264-575CN		$3,850		Est 38.0i SPECint95, 57.5
SPECfp95
-	KP21264-500CN		$2,750		Est 33.3i SPECint95, 50.0
SPECfp95
-	KP21264-450CN		$2,200		Est 30.0i SPECint95, 45.0
SPECfp95

		*-*-*-*-*-*-*-*
		Capt Christopher A. Bohn
		Graduate Student, Electrical (digital) Engineering
		Air Force Institute of Technology     Phone (937)255-3636
(DSN 785)
		AFIT/EN638                              Lab x4606  
Voicemail x6638
		2950 P St, Box 4638                         email
cbohn@afit.af.mil
		Wright-Patterson AFB OH 45433-7765                
EngrBohn@aol.com
		               http://members.aol.com/EngrBohn/
		*-*-*-*-*-*-*-*



				-----Original Message-----
				From:	Capt Bohn, Christopher A.
[SMTP:cbohn@afit.af.mil]
				Sent:	Thursday, September 24, 1998 6:32 PM
				To:	Beowulf
				Cc:	Bascule Pendragon Overdrive
				Subject:	RE: 21264 information

			> Does anyone here possess some sort of objective
statistics regarding the
			> 21264?

			Hmm.  There was an article I had some weeks back.
I'll see if I can find it
			in the morning.

			> does anybody know the actual speed at which the
processor runs?  I
			> have been hearing everything from 500MHz to
1500MHz (which is a bit
			> outlandish)

			I seem to recall the low-end was 400 or 500MHz, and
the high-end is still
			shy of 1000MHz (800?)

			> Currently I am only able to
			> find the filth put out by marketing hype artists
(e.g. "Enormous cache,
			> unbelieveably fast bus speeds", "Superprocessor")

			As best as I can remember, the cache is rated at
impressive, but will be
			upgraded to unbelievable in about a year.  =)

			As I said, I'll look for that article in the
morning.
			Take care,
			cb
			*-*-*-*-*-*-*-*
			Capt Christopher A. Bohn
			Graduate Student, Electrical (digital) Engineering
			Air Force Institute of Technology     Phone
(937)255-3636 (DSN 785)
			AFIT/EN638                              Lab x4606  
Voicemail x6638
			2950 P St, Box 4638                         email
cbohn@afit.af.mil
			Wright-Patterson AFB OH 45433-7765                
EngrBohn@aol.com
			               http://members.aol.com/EngrBohn/
			*-*-*-*-*-*-*-*