indra@brahms.amd.com (Indra Singhal) (02/19/91)
Do 386 machines have a socket that allow using a 287?
For a 33 Mhz 386 machine, what speed 287 should be used?
Thanks!
--
iNDRA | indra@amd.com or {ames apple uunet}!amd!indra
| (Indra Singhal) (408) 749-5445; Advanced Micro Devices
| MS 167; Box 3453; 901, Thompson Pl., Sunnyvale, CA 94088mwizard@eecs.cs.pdx.edu (Craig Nelson) (02/19/91)
indra@brahms.amd.com (Indra Singhal) writes: >Do 386 machines have a socket that allow using a 287? >For a 33 Mhz 386 machine, what speed 287 should be used? >Thanks! >-- >iNDRA | indra@amd.com or {ames apple uunet}!amd!indra Hate to say it, but you get the prize of forking over some rude cash if you want a 387 for that machine. 80387DX Coprocessors run bucks. And No, to my knowledge jamming a 287 in just won't work. No Slot, wrong speed, get the picture ? Craigf (mwizard@eecs.ee.pdx.psu)
jporter@twaddle.dell.com (Jeff Porter) (02/19/91)
In article <1991Feb19.080902.21887@amd.com>, indra@brahms.amd.com (Indra Singhal) writes: |> Do 386 machines have a socket that allow using a 287? |> |> For a 33 Mhz 386 machine, what speed 287 should be used? |> |> Thanks! |> |> -- |> iNDRA | indra@amd.com or {ames apple uunet}!amd!indra |> | (Indra Singhal) (408) 749-5445; Advanced Micro Devices |> | MS 167; Box 3453; 901, Thompson Pl., Sunnyvale, CA 94088 I won't say that it's not possible, but I will say that it is very difficult to put a 80287 in a 80386-33 design. The 287 was used in some of the early 386-based machines before the 387 became available, but I seriously doubt that there are any 33 MHz 386s with a socket for a 287. Most 386-33s will have a socket for a 387 and/or Weitek coprocessor. (The only way to get a 287 to work in a 33 Mhz system would be to drastically slow down the clock during coprocessor cycles, a tricky engineering task.) Jeff Porter jporter@twaddle.dell.com
james@bigtex.cactus.org (James Van Artsdalen) (02/20/91)
In <1991Feb19.080902.21887@amd.com>, indra@brahms.amd.com (Indra Singhal) wrote: > Do 386 machines have a socket that allow using a 287? Not any more. > For a 33 Mhz 386 machine, what speed 287 should be used? Won't work. Intel dropped the 287 support from the 386 back with the D step I believe. All 33MHz parts are D0 steppings or later, so there's no way to get a 287 to work. Besides, a 287 never really worked right with a 386 anyway. -- James R. Van Artsdalen james@bigtex.cactus.org "Live Free or Die" Dell Computer Co 9505 Arboretum Blvd Austin TX 78759 512-338-8789
josephc@nntp-server.caltech.edu (Simplelogic) (02/21/91)
mwizard@eecs.cs.pdx.edu (Craig Nelson) writes: >indra@brahms.amd.com (Indra Singhal) writes: >>Do 386 machines have a socket that allow using a 287? >>For a 33 Mhz 386 machine, what speed 287 should be used? >>Thanks! > Hate to say it, but you get the prize of forking over some rude cash >if you want a 387 for that machine. 80387DX Coprocessors run bucks. And No, >to my knowledge jamming a 287 in just won't work. No Slot, wrong speed, get >the picture ? Not true. I have a 386-25 system that runs an 8MHz 287 as a "cheap way out" for those who don't want to spend the money for a more expensive 387, running at full system speed. (Cyrix 287 compatible are $99 each! Yeah!) BTW, more andmore motherboards I'm seeing have dual co-processor sockets (one for 287, one for 387; or 387/Weitek combo, though that's more exotic...) -- Joseph I. Chiu, Department of Computer Science, Calif. Inst. of Technology 1-57 Fleming House, Caltech, Pasadena 91126. (818) 585-0393 josephc@coil.caltech.edu ...Just another lost soul in the universe
plim@hpsgwp.sgp.hp.com (Peter Lim) (02/21/91)
/ james@bigtex.cactus.org (James Van Artsdalen) / 2:46 am Feb 20, 1991 / writes:
$ Won't work. Intel dropped the 287 support from the 386 back with the
$ D step I believe. All 33MHz parts are D0 steppings or later, so
$ there's no way to get a 287 to work. Besides, a 287 never really
$ worked right with a 386 anyway.
$
I am not too sure about this. My 386 is a 25 MHz model (non-cache). It
has a 287 socket as well as 387 / Weitek socket. According to the manual,
the 287 should run at its own pace at 8 MHz whereas the 386 and 387 runs
at 25 MHz. I have never used the 287 socket, but a friend of mine, who
has an identical board uses a 287 and he says it works.
My board is more than one year old though. And since then, I have not
seen a 386 board which accept both 287 and 387 --- may be 387 has
become cheap enough for 287 to be ignored (no ??).
ps: If I remember correctly, my 386 chip is D0 stepping. But I have not
look at it for a long time, so I could be using an older chip.
Regards, . .. ... .- -> -->## Life is fast enough as it is ........
Peter Lim. ## .... DON'T PUSH IT !! >>>-------,
########################################### :
E-mail: plim@hpsgwg.HP.COM Snail-mail: Hewlett Packard Singapore, :
Tel: (065)-279-2289 (ICDS, ICS) |
Telnet: 520-2289 1150 Depot Road, __\@/__
Singapore 0410. SPLAT !
#include <standard_disclaimer.hpp>davidsen@sixhub.UUCP (Wm E. Davidsen Jr) (02/24/91)
In article <15285@uudell.dell.com> jporter@twaddle.dell.com (Jeff Porter) writes: | I won't say that it's not possible, but I will say that it is very difficult | to put a 80287 in a 80386-33 design. The 287 was used in some of the early | 386-based machines before the 387 became available, but I seriously doubt that | there are any 33 MHz 386s with a socket for a 287. Most 386-33s will have | a socket for a 387 and/or Weitek coprocessor. (The only way to get a 287 | to work in a 33 Mhz system would be to drastically slow down the clock during | coprocessor cycles, a tricky engineering task.) Whoops! I agree that no one is providing a 287 socket anymore, but not because it's hard to do. The 287 never ran at CPU speed, and always supported async operation. This system runs a 386 at 16MHz and the 287 at 10. It was designed before the specs were available for the 387, as were the early Compaq's. Once the 387 spec was availbale that became the chip of choice. The 287 is vastly slower than the 387, but was better than no FPU at all. The old 8087 ran with the 8086 with some strange timing constraints (the IBM PC ran the CPU at 6MHz, and the FPU at 4MHz, as I recall), but the 287 and later run more or less independently. -- bill davidsen - davidsen@sixhub.uucp (uunet!crdgw1!sixhub!davidsen) sysop *IX BBS and Public Access UNIX moderator of comp.binaries.ibm.pc and 80386 mailing list "Stupidity, like virtue, is its own reward" -me
campbell@dev8n.mdcbbs.com (Tim Campbell) (02/25/91)
In article <1991Feb19.080902.21887@amd.com>, indra@brahms.amd.com (Indra Singhal) writes: > Do 386 machines have a socket that allow using a 287? > > For a 33 Mhz 386 machine, what speed 287 should be used? > > Thanks! > -- Some 386 machines have this. I had some AST 386 machines that had them. These machines actually came with 2 sockets for math processors. One socket could take the 287 and the other could take the 387. (I think it also took a weitek (sp?). Math processors (ANY math processor) GREATLY increasing floating point speed (I easily observe a 10x speed increase). And 387-20's are expensive compared to 287-10's. (the machine I had was a 20Mhz 386, but it took a 287-10Mhz math chip - this still gave it a heck of a performance boost. Read the manual for the specific machine you wish to use to see if it has the capability to do this. It's not available on all 386 boxes. --------------------------------------------------------------------------- In real life: Tim Campbell - Electronic Data Systems Corp. Usenet: campbell@dev8.mdcbbs.com @ McDonnell Douglas M&E - Cypress, CA also: tcampbel@einstein.eds.com @ EDS - Troy, MI CompuServe: 71631,654 Prodigy: MPTX77A P.S. If anyone asks, just remember, you never saw any of this -- in fact, I wasn't even here.
ted@helios.ucsc.edu (Ted Cantrall) (02/27/91)
In article <1670@pdxgate.UUCP> mwizard@eecs.cs.pdx.edu (Craig Nelson) writes: >indra@brahms.amd.com (Indra Singhal) writes: >>Do 386 machines have a socket that allow using a 287? >>For a 33 Mhz 386 machine, what speed 287 should be used? > >to my knowledge jamming a 287 in just won't work. No Slot, wrong speed, get --------------------------------- I suspect that this was a special design, but we have a 386 machine that has a 287 in it. This is an early 386, so there may not have been any 387's at that time. -ted- ------------------------------------------------------------------------------- ted@helios.ucsc.edu |"He has showed you, O man, what is good; and what does the W (408)459-2110 |Lord require of you but to do justice and to love kindness H (408)423-2444 |and to walk humbly with your God?" Micah 6:8 (RSV)
amichiel@rodan.acs.syr.edu (Allen J Michielsen) (02/27/91)
In article <12816@darkstar.ucsc.edu> ted@helios.ucsc.edu (Ted Cantrall) writes: >In article <1670@pdxgate.UUCP> mwizard@eecs.cs.pdx.edu (Craig Nelson) writes: >>indra@brahms.amd.com (Indra Singhal) writes: >>>Do 386 machines have a socket that allow using a 287? >>>For a 33 Mhz 386 machine, what speed 287 should be used? >>to my knowledge jamming a 287 in just won't work. No Slot, wrong speed, get >I suspect that this was a special design, but we have a 386 machine that >has a 287 in it. This is an early 386, so there may not have been any 387's If anyone still cares, MANY early 386's did come with either or sockets for 80387's or 80287's. Usually they also had a speed jumper for either or both sockets to select appropriate production (projected) chips, like a 16 Mhz 387 or 12 Mhz 387 or 8/10 Mhz 287 or 6 Mhz 287. ANY as a matter of fact, INTEL themselves MADE and SOLD a 80287 to 80387 conversion socket so you could plug in a 80287 into a 80387 socket, AND it had a speed jumper on it so you could use production rated speeds for the 287. 99.999% of all software won't ever know the difference between a 287 in a 387 socket & a 387. However, it is possible and especially now, with the prolif- eration of 386 systems, programmers MAY not be reading the fine print or remembering these little details anymore. If I had a choice, I wouldn't consider a 287 .... al -- Al. Michielsen, Mechanical & Aerospace Engineering, Syracuse University InterNet: amichiel@rodan.acs.syr.edu amichiel@sunrise.acs.syr.edu Bitnet: AMICHIEL@SUNRISE
davidsen@sixhub.UUCP (Wm E. Davidsen Jr) (03/01/91)
In article <3289@sixhub.UUCP> davidsen@sixhub.UUCP (bill davidsen) writes: | Whoops! I agree that no one is providing a 287 socket anymore, but not | because it's hard to do. Well, I was at most 2/3 right on that one, Intel makes more on the 387 than the 287, so (I'm told) recent mask versions of the 386 have been "enhanced" to require a real 387. Since I'm posting this from a machine with a 287, obviously I have an old chip. -- bill davidsen - davidsen@sixhub.uucp (uunet!crdgw1!sixhub!davidsen) sysop *IX BBS and Public Access UNIX moderator of comp.binaries.ibm.pc and 80386 mailing list "Stupidity, like virtue, is its own reward" -me