alexc@psu-cs.UUCP (Alex M. CHAN.) (01/18/88)
Hi,
I am currently looking for a used 8087-2 Co-processor
for my PC. I have check on some mail-order companies,
they are asking for about $ 160.00 for it. It seems to be
a bit too expensive for me, I would settle for a used one
and pay a bit less for the chip. If someone out there who have
such a chip ( 8087-2 8Mhz ) and willing to sell it for less,
please let me know. Thank you in advance.
Sincerely,
Alex.
-------------------------------------------------------------------------------
PATH=alexc@psu-cs
Wit, n.:
The salt with which the American Humorist spoils his cookery
... by leaving it out.
-- Ambrose Bierce, "The Devil's Dictionary"
-------------------------------------------------------------------------------dean@violet.berkeley.edu (Dean Pentcheff;4059 LSB;;39048;KG97) (01/19/88)
In article <479@psu-cs.UUCP> alexc@psu-cs.UUCP (Alex M. CHAN.) writes: > I am currently looking for a used 8087-2 Co-processor Intel used to have a program where they gave away cosmetic reject chips to academics. We've recently written (paper mail) to them about this program to see if they still do it. It's worth a try, anyway... Anyone know anything more up-to-date? -Dean Dean Pentcheff (dean@violet.berkeley.edu) "A university is a place where people pay high prices for goods which they then proceed to leave on the counter when they go out of the store." Loren Eiseley
ugfailau@sunybcs.uucp (Fai Lau) (01/20/88)
In article <6598@agate.BERKELEY.EDU> dean@violet.berkeley.edu.UUCP (Dean Pentcheff) writes: >Intel used to have a program where they gave away cosmetic reject chips >to academics. We've recently written (paper mail) to them about this >program to see if they still do it. It's worth a try, anyway... > Pardon me but isn't it true that once you've plugged the chip into your mother board and screw the case back on you'll never have to look at the chip again? Fai Lau SUNY at Buffalo (The Arctic Wonderland) UU: ..{rutgers,ames}!sunybcs!ugfailau BI: ugfailau@sunybcs INT: ugfailau@joey.cs.buffalo.EDU
berger@clio.las.uiuc.edu (01/21/88)
Cosmetic rejects aren't just ugly chips. They're dies that are
chipped, don't meet all QC tests, etc. Sometimes they work, but
only within a narrow voltage or temperature range. I think Intel
discontinued that program years ago.
Mike Berger
Center for Advanced Study
University of Illinois
berger@clio.las.uiuc.edu
{ihnp4 | convex | pur-ee}!uiucuxc!clio!bergerthollowe@oucsace.cs.OHIOU.EDU (Tom Hollowell) (01/13/89)
I know this is going to sound like a pretty lame question, but I was wondering if anybody out there could tell me what the 8087-2 does, and whether it is worth getting. I do know it is a math co-processor for floating point arithmetic, but does it affect speed? Thanks in advnace.
suitti@haddock.ima.isc.com (Stephen Uitti) (01/17/89)
In article <408@oucsace.cs.OHIOU.EDU> thollowe@oucsace.cs.OHIOU.EDU (Tom Hollowell) writes: >I was wondering if anybody out there could tell me what the 8087-2 >does, and whether it is worth getting. I do know it is a math >co-processor for floating point arithmetic, but does it affect >speed? Thanks in advnace. I recall the 8087-2 as the 8 MHz floating point chip for use with the 8088 processor. My XT clone (Leading Edge Model D) runs at 4.77 or 7.15 MHz, and I upgraded it with an 8087-2. There are several models of 8087, which run at different speeds. Don't assume that the -* number makes any sense. The slower chip (8087-3?) runs at 5Mhz, and for me didn't run at 7.15 (the system crashed on first access). If your application doesn't use any floating point, or doesn't look to see if the chip is out there & uses it (Turbo C applications can do this), you won't get any improvement. For example, the public version of EMPIRE (Wargame of the Century) was written in FORTran and compiled for a non-8087 environment. It uses sine & cosine functions with wild abandon to generate maps. It can take a half hour to generate a map. If it is recompiled for an 8087 environment, the time for map generation is about 3 minutes on the same machine (no I haven't done this, and I don't have the source at the moment, nor do I even have a FORTran compile for MSDOS). This is about a x10 improvement. I have generally seen from x8 to x45 improvements in speed for applications that do heavy floating point, and that are or can be compiled for using the chip. Similar results are obtained on Macs with 68881s, for example in a Mac II. For me, the only things that really move faster are my benchmarks. I don't do any spreadsheet work, but rumor has it that Lotus 123 goes much faster with one. Stephen.
jc58+@andrew.cmu.edu (Johnny J. Chin) (01/22/89)
The math co-processor is only used the help SPEED up calculations on programs
which will use the chip.
Use the following as a guide for which chip to use.
CPU speed Co-Processor
4.77-8MHz (8088,8088-2) 8087, 8087-3 (5MHz)
8MHz-10MHz (8088-2, 8088-1) 8087-2 (8MHz)
10MHz and up (8088-1) 8087-1 (10MHz)
If you have a V20, follow the specs for 8088-2.
Of course, the price will increase with the speed of the chip.
Generally, 5MHz 8087 costs about $90-$110, 8MHZ costs $110-$180, and
10MHz costs $175-$240.
I hope this info helps.
-- J. Chin (a.k.a. Computer Dr.)
xxxxxxxxxx
xxx xxx xx ------------------ Carnegie Mellon University ------------------
xxx xxx xx 4730 Centre Ave. #412 ARPAnet: Johnny.J.Chin@andrew.cmu.edu
xxxxxxxxxxxx Pittsburgh, PA 15213 BITnet: jc58@andrew.BITNET
x xxxxxxxx x (412) 268-8936 UUCP: ...!harvard!andrew.cmu.edu!jc58
xx xx ----------------------------------------------------------------
xxxxxxxxxx Smile! -- Mr. HappyWOWface -- (got this from the network)
Disclaimer: The views expressed herein are STRICTLY my own, and not CMU's.