scs@itivax.iti.org (Steve C. Simmons) (07/22/89)
Is this newsgroup exclusively for MIPS systems, or for any systems which use their chip sets? We got a DEC 5810 coming, and I'd be interested. First posting? First at this site, anyway. -- Steve Simmons scs@vax3.iti.org Industrial Technology Institute Ann Arbor, MI. "Velveeta -- the Spam of Cheeses!" -- Uncle Bonsai
cliff@cpsc.ucalgary.ca (Cliff Marcellus) (07/24/89)
In article <2293@itivax.iti.org>, scs@itivax.iti.org (Steve C. Simmons) writes: > > Is this newsgroup exclusively for MIPS systems, or for > any systems which use their chip sets? We got a DEC 5810 > coming, and I'd be interested. > I personally would like to see this group be used for ALL systems derived from the MIPS chip sets. I recently benchmarked both the DECstation 3100 and the MIPS RS2030 systems (looking for more workstations, again...). THe two systems compared very closedly in performance, except in floating point operations where the RS2030 was roughly 30% faster. (curious, me thinks). I suspect (probably!) different compiler technology... nonetheless. the information exchange would be most useful for FUTURE purchase considerations and such.... (BTW. I bought two RS2030's... I'd also be very interested in hearing more about the DEC 5810 -- from an end-user site) Cliff Marcellus UUCP : {any backbone}!calgary!ssg-vax-a!cliff Institute for Space Research DOMAIN : cliff@ssg-vax-a.phys.UCalgary.CA Dept of Physics and Astronomy SPAN : CANCAL::CLIFF The University of Calgary "If it's not fun, don't do it!"
mash@mips.COM (John Mashey) (07/24/89)
In article <2293@itivax.iti.org> scs@itivax.iti.org (Steve C. Simmons) writes: > >Is this newsgroup exclusively for MIPS systems, or for >any systems which use their chip sets? We got a DEC 5810 >coming, and I'd be interested. I think this was originally aimed at MIPS systems, and I do think it would make sense to post questions about a DEC 5810 to comp.sys.dec, and about SGI boxes to comp.sys.sgi. For instance, it certainly makes no sense to post questions like "Does Ultrix release xxx on the 5810 do such and such?" here. On the other hand, it's probably not unreasonable to post questions here that are generic to the use of MIPS systems, software, or chips, and maybe sometimes crosspost them to the comp.sys.* group of the system you're using, if you are. Note: this is my OPINION; I didn't start this newsgroup.... -- -john mashey DISCLAIMER: <generic disclaimer, I speak for me only, etc> UUCP: {ames,decwrl,prls,pyramid}!mips!mash OR mash@mips.com DDD: 408-991-0253 or 408-720-1700, x253 USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086
leoh@hardy.harris.com (Leo Hinds) (07/24/89)
As far as I have been told ... MIPS is ONLY a compiler Co (these days) and lets others manufacture it chips ... based on that I would say this group should be open to all machines based on the MIPS chip. Leo Hinds
davidh@dent.Berkeley.EDU (David S. Harrison) (07/25/89)
> ... We got a DEC 5810 coming ...
Just to clarify things for me, I assume you mean a DECstation 3100.
This is the only announced product I know of from DEC that uses
the MIPS architecture.
David Harrison
UC Berkeley Electronics Research Lab
(davidh@ic.Berkeley.EDU, ...!ucbvax!ucbcad!davidh)trj@lll-lcc.UUCP (Terry Jones) (07/25/89)
In article <2293@itivax.iti.org> scs@itivax.iti.org (Steve C. Simmons) writes: > >Is this newsgroup exclusively for MIPS systems, or for >any systems which use their chip sets? We got a DEC 5810 >coming, and I'd be interested. > >First posting? First at this site, anyway. >-- >Steve Simmons scs@vax3.iti.org >Industrial Technology Institute Ann Arbor, MI. >"Velveeta -- the Spam of Cheeses!" -- Uncle Bonsai As the person that held the vote to create comp.sys.mips, I'll give you my opinion. The charter for the group was as follows: Provide a place for discussion of issues related to the products (including systems, chips, compilers, etc.) from MIPS Computer Inc. In addition, *relevant* topics about systems based on MIPS technology from other vendors are discussed. The intent of putting the word *relevant* in the charter was to keep the discussion relevant to products from MIPSco. Therefore, IMHO, questions regarding the instruction set, processor, and the like would be appropriate for comp.sys.mips; questions regarding the XMI bus and Ultrix would be more suited for comp.sys.dec. Is that clear as mud? terry ----------------------------------------------------------------------------- terry jones | lawrence livermore lab | POB 808 MS: L-67 | How much net work could a network work, Livermore, CA 94550 | if a network could net work? trj@lll-lcc.llnl.gov | trj@lll-crg.llnl.gov | - Jeffery DeMello (415) 423-9834 | -----------------------------------------------------------------------------
scs@itivax.iti.org (Steve C. Simmons) (07/25/89)
davidh@dent.Berkeley.EDU (David S. Harrison) writes: >> ... We got a DEC 5810 coming ... >Just to clarify things for me, I assume you mean a DECstation 3100. >This is the only announced product I know of from DEC that uses Uh...you're not reading the newspapers. It's about 18MIP, R-3000 based. So new we don't even have a delivery date yet. -- Steve Simmons scs@vax3.iti.org Industrial Technology Institute Ann Arbor, MI. "Velveeta -- the Spam of Cheeses!" -- Uncle Bonsai
mash@mips.COM (John Mashey) (07/25/89)
In article <15770@pasteur.Berkeley.EDU> davidh@ic writes: >> ... We got a DEC 5810 coming ... > >Just to clarify things for me, I assume you mean a DECstation 3100. >This is the only announced product I know of from DEC that uses >the MIPS architecture. They just announced 3 or 4 more products [2100, 5400, 5810/5820] July 11. I suspect comp.sys.dec has the info. -- -john mashey DISCLAIMER: <generic disclaimer, I speak for me only, etc> UUCP: {ames,decwrl,prls,pyramid}!mips!mash OR mash@mips.com DDD: 408-991-0253 or 408-720-1700, x253 USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086
mash@mips.COM (John Mashey) (07/25/89)
In article <1636@cs-spool.calgary.UUCP> cliff@cpsc.ucalgary.ca (Cliff Marcellus) writes: >I recently benchmarked both the DECstation 3100 and the MIPS RS2030 systems >(looking for more workstations, again...). THe two systems compared very >closedly in performance, except in floating point operations where the RS2030 >was roughly 30% faster. (curious, me thinks). I suspect (probably!) different >compiler technology... nonetheless. the information exchange would be most >useful for FUTURE purchase considerations and such.... Although the DS3100 and RS2030 both run at 16.7MHz, the 2030 has smaller caches, but (I think) a slightly faster write to memory, so they trade places on individual benchmarks. The 2030 does have newer compilers (2.0 vs 1.31). -- -john mashey DISCLAIMER: <generic disclaimer, I speak for me only, etc> UUCP: {ames,decwrl,prls,pyramid}!mips!mash OR mash@mips.com DDD: 408-991-0253 or 408-720-1700, x253 USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086
fritz@unocss.UUCP (Tim Russell) (07/26/89)
davidh@dent.Berkeley.EDU (David S. Harrison) writes: >> ... We got a DEC 5810 coming ... >Just to clarify things for me, I assume you mean a DECstation 3100. >This is the only announced product I know of from DEC that uses >the MIPS architecture. It seems the trade rags haven't made much of it, but.. We also have a 5810 coming. It's an R3000 based system in a Vax 6x00 cabinet with space for 2 RA90 drives in the top (1.2 gig each) and up to 256 meg of memory. 1 or two processors can be used, placed along with the memory and up to 5 VAXBI bus interfaces. All in all, it sounds like a mid-range screamer. I'll let you know how it performs when we get it. I can't wait! (My apologies if this is supposed to be a MIPS-only group.) -- ---------------------------------+-------------------------------------------- Tim Russell, Computer Operator | Internet: russell@zeus.unl.edu Campus Computing | Bitnet: russell@unoma1 University of Nebraska at Omaha | UUCP: uunet!zeus.unl.edu!russell
bin@primate.wisc.edu (Brain in Neutral) (07/26/89)
From article <1124@hcx1.UUCP>, by leoh@hardy.harris.com (Leo Hinds): > As far as I have been told ... MIPS is ONLY a compiler Co (these days) and lets > others manufacture it chips ... based on that I would say this group should be > open to all machines based on the MIPS chip. The machine on which this reply is being composed is a MIPS M/120 running RISC/os, the MIPS Inc., version of UNIX. I don't know whether MIPS actually made the chip or not, but the company certainly markets complete systems, not just chip technology. Paul DuBois dubois@primate.wisc.edu
frank@croton.dec.com (Frank Wortner) (07/26/89)
In article <2554@lll-lcc.UUCP>, trj@lll-lcc.UUCP (Terry Jones) writes: > > IMHO, questions regarding the instruction set, processor, and the > like would be appropriate for comp.sys.mips; questions regarding > the XMI bus and Ultrix would be more suited for comp.sys.dec. > Fair enough, but IMHO, questions regarding Ultrix would be more suited to comp.unix.ultrix.
rich@sendai.sendai.ann-arbor.mi.us (K. Richard Magill) (07/26/89)
In article <23924@winchester.mips.COM> mash@mips.COM (John Mashey) writes:
I think this was originally aimed at MIPS systems, and I do think
it would make sense to post questions about a DEC 5810 to
comp.sys.dec, and about SGI boxes to comp.sys.sgi. For instance,
it certainly makes no sense to post questions like "Does Ultrix
release xxx on the 5810 do such and such?" here.
Then I'd like to suggest that:
a) the name of this group be changed to comp.sys.mipsco. We use
mipsco around here to distinguish the company which is Mips.Com from
the too many other overloaded definitions for the word mips. (the
magazine, the chip architecture, the company, the units (there are
several of them) of cpu horsepower, etc.)
b) someone yell at DEC and tell them that their machine isn't a
"mips". Specifically, the DEC distributed C compiler defines "mips",
"MIPSEL", and "ultrix", which is your best clue to the fact that this
is a decstation, but machine(1) reports that the decstation is a
"mips".
--
rich.mash@mips.COM (John Mashey) (07/26/89)
In article <1124@hcx1.UUCP> leoh@hardy.hdw.harris.com (Leo Hinds) writes: >As far as I have been told ... MIPS is ONLY a compiler Co (these days) and lets >others manufacture it chips ... based on that I would say this group should be >open to all machines based on the MIPS chip. MIPS is not a chip company: we do VLSI architecture and implementation, but we have semiconductor partner arrangements with 5 companies that have the right to produce the chips we design and variants thereof. We expect to do VLSI design for a long time, but I doubt we'll ever have a fab. We certainly do compilers and sell them, and we'll do that for a long time, but: We "also" sell systems; also = 70% of our revenue, last year, this year, and what much of our sales force actually does.... -- -john mashey DISCLAIMER: <generic disclaimer, I speak for me only, etc> UUCP: {ames,decwrl,prls,pyramid}!mips!mash OR mash@mips.com DDD: 408-991-0253 or 408-720-1700, x253 USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086
aps@granite.dec.com (Armando P. Stettner) (07/30/89)
In article <RICH.89Jul25160644@sendai.sendai.ann-arbor.mi.us> rich@sendai.ann-arbor.mi.us writes: ..... >Then I'd like to suggest that: > .... > >b) someone yell at DEC and tell them that their machine isn't a >"mips". Specifically, the DEC distributed C compiler defines "mips", >"MIPSEL", and "ultrix", which is your best clue to the fact that this >is a decstation, but machine(1) reports that the decstation is a >"mips". >-- >rich. Our view is that machine(1) and the "mips" defination in the C compiler are indications of the architecture; not the specific model in a family. For example, you don't get an indication that the machine is a 780 or 750 or 6200; just that it is a VAX. If you have to know which machine it is, you are probably writing kernel code and there are other methods used there to differentiate models and characteristics. As for the "MIPSEL" and "ultrix", the former is an indication from the compiler as to which endianism the target machine is and the later indicates the target operating system. Both of these are defaults in the compiler and MIPSEL is also used by other (all?) MIPS C (other?) compilers. This complaint/comment should have gone to net.whatever for Digital systems and/or Ultrix. Sorry to take up the news message, John. armando. decwrl!aps aps@decwrl.dec.com