Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!uwm.edu!rpi!crdgw1!crdos1!davidsen
From: davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr)
Newsgroups: comp.sys.ibm.pc
Subject: Re: 640K limit
Message-ID: <2025@crdos1.crd.ge.COM>
Date: 17 Jan 90 18:37:58 GMT
References: <4668.25aed7f2@uwovax.uwo.ca> <1468@blackbird.afit.af.mil> <28808@amdcad.AMD.COM> <729@jethro.Corp.Sun.COM>
Reply-To: davidsen@crdos1.crd.ge.com (bill davidsen)
Organization: GE Corp R&D Center, Schenectady NY
Lines: 25

In article <729@jethro.Corp.Sun.COM> acm@sun.UUCP (Andrew MacRae) writes:
| Simple.  IBM/MicroSoft *should* have used soft pointers to the I/O memory
| areas.  They made the same mistake that was made with CP/M, hardcoding
| areas of memory for specific uses.  For those who don't remember, CP/M
| was hardcoded so that the kernal *had* to reside in the top 4kb of a 64kb
| address space, forever limiting its applications to 60kb in size.

  1) your first statement is not true.
  2) your conclusion is meaningless

  CP/M required that the o/s sit at the top of available memory, not the
top of 64k. Early systems usually didn't have 64k, 16k of memory on 4k
static memory boards costing $200 each was about standard at the
beginning. It was also possible to put the o/s and most of the BIOS in
an alternate bank, allowing 63.5k of user memory (the highest and lowest
"page" of 256 bytes were dedicated).

  Since the o/s ran on a CPU with only 64k addressing, and the o/s was
at least 4k unless you played the games described above, applications
were limited in size no matter where the o/s was located. There never
was an equivalent of an extended memory standard to allow bank switching
(which is *very* similar to EMS).
-- 
bill davidsen	(davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen)
            "Stupidity, like virtue, is its own reward" -me