zalman@mips.com (Zalman Stern) (05/11/91)
In article <576@appserv.Eng.Sun.COM> lm@slovax.Eng.Sun.COM (Larry McVoy) writes: >umh@vax5.cit.cornell.edu writes: >[Marketing rumours about future ACE based hardware.] > [...] > >Let's suppose that I could get one [R4000] in my grubby little paws today. >I still have to build a motherboard, a case, all that glob. Call it >9 months of a very agressive schedule. >I still have to have an OS that runs on the thing. Call it another 9 >months on an extremely aggresive schedule. An appropriate word for companies which don't aggressively pipeline this process in the 90's is "roadkill." You can simulate all this stuff without working chips. Here's a quote from HP's "CMOS PA-RISC Processor for a New Family of Workstations", Mark Forsyth et al. (this years ISSCC proceedings): Extensive pre-release simulation using a highly evolved set of system level, chip level, and transistor level verification methodologies was employed to ensure fully functional first silicon. The HP-UX operating system was booted and running within four hours of recieving packaged first silicon CPU chips. If hardware is waiting for software, then something is very wrong. According to some hardware folks I talk to occasionally, doing the board and such is relatively easy. I don't know about Sun, but MIPS puts a lot of work into simulation. On the other hand, all this technology only helps after you figure out what to build in the first place... -- Zalman Stern, MIPS Computer Systems, 928 E. Arques 1-03, Sunnyvale, CA 94088 zalman@mips.com OR {ames,decwrl,prls,pyramid}!mips!zalman (408) 524 8395 "Never rub another man's rhubarb" -- the Joker via Pop Will Eat Itself
dboles@ccwf.cc.utexas.edu (David Boles) (05/12/91)
In article <3411@spim.mips.COM> zalman@mips.com (Zalman Stern) writes: >>Let's suppose that I could get one [R4000] in my grubby little paws today. >>I still have to build a motherboard, a case, all that glob. Call it >>9 months of a very agressive schedule. >>I still have to have an OS that runs on the thing. Call it another 9 >>months on an extremely aggresive schedule. > >An appropriate word for companies which don't aggressively pipeline this >process in the 90's is "roadkill." You can simulate all this stuff without >working chips. Here's a quote from HP's "CMOS PA-RISC Processor for a New >Family of Workstations", Mark Forsyth et al. (this years ISSCC >proceedings): I think you are both on the wrong track. The reason that the ACE time- table is unrealistic is software. Microsoft started started on OS/2 5 years ago and the system has only been really ready for a year. They announced DOS5.0 (right after Digital Research did their DRDOS5) about a year ago and we haven't seen it yet (and this is DOS !!). SysVr4 has been out for a while yet but almost nobody has it working on machines yet, and these machines are known quantities. I think that it will be late 1993 before you see OS/2 NT running on r4000's and late 1992 for ACE to have any machines on the proverbial shelves. Can anyone even name an OS that was completed on time? -- ------------------------------------------------------------------------------- David Boles Applied Research Laboratories dboles@ccwf.cc.utexas.edu DOS is severely brain-damaged, apas611@chpc.utexas.edu so just pull the plug and let it DIE. -------------------------------------------------------------------------------
lindsay@gandalf.cs.cmu.edu (Donald Lindsay) (05/12/91)
In article <48857@ut-emx.uucp> dboles@ccwf.cc.utexas.edu (David Boles) writes: >Can anyone even name an OS that was completed on time? For an OS, "completed" == "dead". If you mean, a version being shipped on schedule, then, yes. In particular, porting something from the R3000 to the R4000 (32-bit mode) shouldn't be the stuff of legends. If there's an OS schedule slip, it will more likely be for other reasons. -- Don D.C.Lindsay Carnegie Mellon Robotics Institute