Ideas that didn't come off

Many many years ago, I invented a system to make my turbo pascal
programs run faster.

I had figured out how the overlay code worked, then did some stuff
with delayed disk reads, and how I could modify the call stack so that
overlayed programs used bank switching rather than disk reads to get the
required code into memory.

At the time, I thought I had a world beater: as long as there was
enough RAM in the target computer, the 64K barrier would be completely
broken for my TP programs.

I even considered submitting this technique to Borland as part of a
conference paper. Then the conference notes were published, and with
Anders talking about 32 bit flat mode, I knew that my idea was both way
behind the times and totally screwed.

I no longer even have any of the source for my idea. It was basically
a stack walker with an interrupt handler, and relied on being able to
modify memory in code segments so that the code could be changed based
on whatever was loaded into RAM at any particular time. I would replace
a far call with an interrupt instruction and use the bits left over to
give my system a chance of figuring out which code was required. While
this sounds like the overlay system already built into TP, my invention
allowed for stuff to be loaded from RAM instead of from disk.

Obviously, neither my idea nor my code ever saw the light of day.

I'm so hugely impressed at how the FastCode project people have been
able to create such a high performance memory manager for Delphi, that
thinking about my overlay manager from TP days makes me cringe at how
ineffective and ultimately futile my efforts would have been.

I've not been daunted, and I still manage to invent things that turn
out to be ridiculous, but sometimes I will create something that works.

I guess I'm asking to hear from others that have come up with bright
ideas, to admit their failure, and tell us what, why and how.

I do have a Delphi-based invention that I am not willing to share due
to commercial constraints, but I figure that there are lots of Delphi
inventions which sounded good on paper, but which (like mine) didn't
work in practice.

Will you share the gory details with us?

Scout
.

Relevant Pages

  • Re: Proper breakthroughs/inventions.
    ... >> oscillator with the bulb over the development of the transistor. ... >> on a higher scale the invention of virtual memory. ... An invention that would have no use whatsoever if PC ... The real world is always going to have more data than RAM available to ...
    (sci.electronics.design)
  • O. T Laptop RAM
    ... Toshiba's site to make sure which RAM it took they sell Kingston, ... Kingston seems to be a mid price module. ... the greatest invention in the history of mankind is beer. ... Oh, I grant you that the wheel was also a fine invention, but the wheel does ...
    (alt.marketing.online.ebay)
  • Re: Project Dionysus.
    ... > As could be predicted by anybody with an understanding of economics. ... It must have been the most hyped "invention" related to Delphi I had ...
    (borland.public.delphi.thirdpartytools.general)