gemini@homxb.UUCP (Rick Richardson) (12/01/85)
A month or two ago, Intel created a stir on this newsgroup by announcing
the Intel 386. They claimed that the message was sent from a live 386
running UNIX. In response, I requested that the Dhrystone benchmark
be run on the machine, and the results sent to me for inclusion in the
next posting. Well, it is 1.5 weeks to the next posting, and I haven't
seen anything from Intel, even though I've sent the source and request
twice.
I challenge you folks at Intel to back up your announcement with
a dhrystone run. Do it soon, and send the results directly to me.
I'll protect anonymity if so desired.
Rick Richardson, PC Research, Inc. (201) 922-1134
..!ihnp4!houxm!castor!{rer,pcrat!rer} <--Replies to here, not to homxb!!!mdm@ecn-pc.UUCP (Mike D McEvoy) (12/05/85)
In article <965@homxb.UUCP> gemini@homxb.UUCP (Rick Richardson) writes: >I challenge you folks at Intel to back up your announcement with >a dhrystone run. Do it soon, and send the results directly to me. >I'll protect anonymity if so desired. >Rick Richardson, PC Research, Inc. (201) 922-1134 I've sent requests to several Intel folks to do the same thing... No response. I know that new product introduction causes a loss of sleep, but alot of potential customers (read that real money) are interested in the Dhrystones, as well as floating point performance with the new 386/Weitek 1163/1164/1165 co-processor vs the 68020/68881. How about some results????? Mike McEvoy 317-497-0509
phil@amdcad.UUCP (Phil Ngai) (12/06/85)
The 386 machine which runs Xenix is based on a 286 system and as such,
has only a 16 bit data path. Kind of like plugging a 68020 into a MAC
and making it use its 16 bit bus mode. Give them time to build a 32
bit system. Two or three weeks is not an adequate amount of time.
--
Even lefties have rights!
Phil Ngai +1 408 749-5720
UUCP: {ucbvax,decwrl,ihnp4,allegra}!amdcad!phil
ARPA: amdcad!phil@decwrl.dec.commdm@ecn-pc.UUCP (Mike D McEvoy) (12/07/85)
In article <7205@amdcad.UUCP> phil@amdcad.UUCP (Phil Ngai) writes: >The 386 machine which runs Xenix is based on a 286 system and as such, >has only a 16 bit data path. Kind of like plugging a 68020 into a MAC >and making it use its 16 bit bus mode. Give them time to build a 32 >bit system. Two or three weeks is not an adequate amount of time. Last I heard (from an Intel Employee at WESCON) was that XENIX (and soon to be system 5) is running on Multibus II (32 bit bus!!!). I know Intel has a 386 card running on the bus (or so they say), I've got a picture of the thing in my hand. Based on comments I've recieved, the benchmarks have been run, and they are soon to be released. I guess the real question is "When will Intel Marketing say OK?" Soon, I hope. A few potential customers are waiting with baited breath.