[comp.sys.ibm.pc] IBM PS/2 clock speed and others

cjnston@pnet02.CTS.COM (Chuck Johnston) (09/05/87)

marb@sphinx.uchicago.edu (Marty Billingsley) writes:
>
>
>I was recently asked this question and haven't been able to find a
>good answer:
>
>Is there any way to change the clock speed in an IBM PS/2 model 30?
>Either a hardware or software solution would be acceptable.
>The problem: the PS/2 runs at 8 MHz and so software (games, educational
>programs) that was designed to run at 4.77 MHz is too fast. (I guess the
>guy who asked me this question is a frustrated asteroids player)
>
>It has been suggested that he could program the 1C interupt, which is
>called every now and then by the OS (18 times/sec?), to delay for a set
>ammount of time.  Does anyone know how to do this???  
>
>Does anyone have any other suggestions???
>
>Please send email, and I'll post a summary.  Thanks!
>
>
>Marty Billingsley         (marb@sphinx.UUCP or marb@sphinx.UChicago.edu)
>
>----------------------------
>| SD (standard disclaimer) |
>----------------------------

        There are several public domain programs available to slow down turbo
XT's for games and such, they are usually very short and I would upload one if
someone could explain this ENCODE DECODE format to me. 

Now for my question, I recently installed an 8087 in my Epson Equity I (a
standard 4.77 Mhz clone) and thought that since I had to rip the would thing
apart to do so, I would give it a complete lobotomy and change the processor
to a NEC V20. Well, before the change it had an expected Norton SI of 1.0.
Afterwards it registered a 1.8!!  At first I thought this was due to the 8087,
but no, the real XT's at work still get a 1.0 with 8087's. Can anyone explain
this phenonema??  Should I pop a 10 mhz chip in and see what happens??

UUCP: {hplabs!hp-sdd!crash, seismo!scgvaxd!cadovax}!gryphon!pnet02!cjnston
INET: cjnston@pnet02.CTS.COM

dboyes@uoregon.UUCP (09/07/87)

The reason that you see a .8 or so increase in the Norton SI rating
after installing a V-20 is that although the NEC chip is doing exactly
the same things with the instructions in your program, the chip
architecture has been streamlined to a great degree, thus making the
processor more efficient. You're still running at only 4.77 Mhz, but the
V-20 is making more efficient use of the cycles.
 
 BTW, the 8087 will only affect programs that use it, i.e. heavy
 floating point & transcendentals. Try compiling anything that does
 graphics (use QuickBASIC on the programs that come on the DOS
 supplemental disk) with the 8087 support and compare -- it's much nicer
 on the 8087, especially when circles are involved.

-- 
David Boyes                   ARPA: 556%OREGON1.BITNET@WISCVM.WISC.EDU
Systems Division              BITNET: 556@OREGON1
University of Oregon Computing Center   UUCP: dboyes@uoregon.UUCP