Int 13h buffer 64k boundaries (was: 8085 Dissasembly?

[Charles Anthony]

On Wed, Apr 18, 2018 at 9:20 PM, Fred Cisin via cctalk <
cctalk at classiccmp.org> wrote:

>
>>>
> You certainly did the right thing, narrowing it down to load address.  The
> final conclusion would have been to systematically try many/all load
> addresses, and see whether it was consistent for given ones, and what the
> failing ones had in common.
>
>
I was coming from a DEC environment (VMS, RT-11/ RSTS) --  produce a
repeatable demonstration of a error and send it off; a few weeks later a
fix appears. MS's failure to respond confirmed my suspicions; they were a
commodity software company.  (That is a perfectly valid business model, but
staking your own business success on them their behavior is a real risk.0



>
>
> I was not invited to be in the 3.11, nor WIN95, Betas.
> They wanted cheerleaders, not testers, anyway.
>

We received the SDK for the very first Windows. We looked at it and laughed
and laughed. You can still see the design flaw in today's Windows.


Much later, having fled the MS ecosystem, I was working on a SGI
application. Attended a MS seminar for the first NT release -- it featured
OpenGL support and they wanted all of the SGI developers to migrate. I
spend an entire day listening to them telling my how wonderful NT was. What
I learned could have been covered in three minutes:

   *  OpenGL support was a port of the reference driver, slow and buggy.
Hardware manufacturers would provide "real" OpenGL drivers. Of course, they
would focus on the gaming market, not data visualization, so the chances of
a usable implementation was low, plus the cost of losing customers who made
a poor hardware selection based on the manufacturer's claim that they are
OpenGL compliant. It is possible to write an OpenGL compliant driver for a
light bulb; it is no guarantee that you will be able to do data
visualization.

  * Despite the fact that (as I have heard quoted, but cannot give a
citation) 80% of an application is UI, there was no support for X11
semantics, so we would have to rewrite all of that code for the [patently
broken] Windows semantics.

We did not migrate, and I still waiting for MS to apologize for wasting a
day of my time with their delusions of grandeur.

-- <grumpy old> Charles