Re: Work on improving/expanding 'HLA Adventure'

From: Me (debs-agent_at_dwiles.demon.co.uk)
Date: 12/10/04 

Date: Fri, 10 Dec 2004 00:14:33 +0000

On Thu, 09 Dec 2004 17:21:05 +0100, "The \\\\o//annabee"
<Faq(Sqeeeek!)@.@szmyggenpv.com> wrote:

>> I'm sure it'll look better than it is. I'm both a mathematician and a
>> hacker.
>
>Ah....a Mahatmahacker ? :)

Something like that :)
>
>> To me that means that I think logically and like my work to be
>> neat, and I tend to write the code before I design the app. I know
>> what I want it to do and I set about writing the code, designing each
>> individual element while I am writing it.
>
>Same as I do, mostly. However if its a challenge...pen and paper. If its a
>real challange, lie on bed, thinking first, then pen and paper, or talk a
>walk.

Hmm, now there's an idea. I'm in the process of decorating anyway, and
as I can't work too fast that should give me a year or two to write on
walls :)
>
>:-) Maybe something will turn up. Are you conserned about not having
>commersials at you site ?

That's not a major problem, but I will be cinsidering paying for some
ad free space on an annual basis until I can get a connection that
makes running a web server from home more realistic.
>
>> I doubt realistically that I'll have a distributed OS before I get a
>> single PC system up and running. I have to learn about networks from
>> the ground up before I can make use of them in such a manner. I'd also
>> have to plan it more than I do for other code.
>
>I guess so. Must be a real hell to to get anywhere and a real orgasm to
>have something finally working.

That makes it sound like a MUCH better idea for my OS :)
>
>> Boot loader
>> 2nd stage loader (set up pmode, initialise processor, etc)
>> Drivers for various standard hardware
>> Selection of graphics cards plus VBE support to start
>> Drives
>> Sound cards
>> File system support (this has to be done early on)
>> A few basic apps
>> Assembler
>> C Compiler
>> Text editor (to write my code)
>
>Oh, thats some order you've got there. Will be very interessting to watch.

Not sure it'll all happen in that order, but definitely something like
that. Basic apps may be written to run under DOS first and then used
to help develop the kernel (having an editor ready before I get the
kernel to boot to a command line would be nice).
>
>> After that I'll have to look into network support and some basic
>> internet stuff. As I'm planning to make it all open source I have a
>> lot of sample code to help in the areas which I find harder, which
>> will probably mainly be the network stuff. I may even compile nasm to
>> run on my OS initially...
>
>Cool. Its means I can learn a lot from it. Maybe even help. Who knows. But
>we need something a lot diffrent from both Windows and Linux. Something
>with a vision of its own, and targeted to endusers.

Definitely. I'm thinking of code that I can use in practical terms,
not just something aesthetically pleasing. When I write my system
calls I may even try working it in such a way that DOS and Linux
system calls will be translated by the OS into native calls (ie having
calls to Int 21h and 80h both directed to the relevant routines in my
code, which wilkl help while developiong my own system calling
methods).
>
>>> And hopefully ready interfaces to most of the
>>> hardware.
>
>> Naturally. No point writing a boot loader then finding you can't do
>> anything practical with the hardware :)
>
>;-) I meant, callable APIs for an application programmer, to ease the
>initial work.

Ah, yes. That's where the interrupt routines come in. Some will
initially use BIOS routines, but a good chunk should be creatable
without major problems. And using pmode does mean they'll all have to
be written eventually.
>
>> I'm not the smartest person on the planet, but I know I have the brain
>> power to do the work.
>
>Will and ability. Its the most wonderful thing.

I hope that will power can overcome all other obstacles...
>
>> The main stumbling block will be the fatigue I
>> suffer due to health problems (I won't go into that here, please don't
>> ask).
>
>What are theese Health-problems ? (Please do not answer :-))))

LOL

I'll probably post something about it when I'm having a real bad day,
but until then I'll spare everyone :)
>
>> That has always slowed me down, but at the same time it has
>> enabled me to do my work at least as well as those around me as I
>> don't rush it (my FDC programming pages are an example of what I am
>> capable of, although I never did write the driver, I gave away my
>> computer to go trasvelling before I got that far).
>
>Where are theese FDC programming pages ? I do not even know what FDC is .

FDC = Floppy Drive Controller.

Current link to the pages is:

http://debs.future.easyspace.com/Programming/Hardware/FDC/floppy.html

I think it's the most complete resource of it's kind online. I note
that it's been the first hit on all google searches for "Floppy Drive
Programming" for at least a couple of years, and I get emails from all
over the world about it!
>
>> Thanks. I'll appreciate testers as much as anyone.
>
>Thats sounds like a programmer to me. Good luck.

Thanks :)

Debs

A friend in need is a strain on the purse

To reply via email, remove "-agent" from return address

Relevant Pages

  • Re: newbie: capabilities of c++ in numerical computation
    ... >>1) Does object oriented programming introduce significant overhead as ... I personnaly think that arrays of doubles are better at low ... fortran or c library routines from your code. ... >>calling fftwlibraries? ...
    (comp.lang.cpp)
  • Re: C128 technical
    ... Is it common to use the ROM routines on the Spectrum? ... > with the single byte BRK instruction to generate a software interrupt. ... Rodnay Zaks also did 'programming the 6502'. ...
    (comp.sys.cbm)
  • Re: Spirit rover OS problems ( a reliable language )
    ... But when I coded up my work using the guide as an abosolute, ... More probably that came from mixing up structured programming and ... > keep the routines short. ... But the complexity measurement says "the more cases in a switch, ...
    (comp.arch.embedded)
  • Cockadoodledoo! (Was: Re: User-written system service, how to protect data?)
    ... As far as the original poster goes: Your main problem is you are programming ... in C where memory allocation appears to be a complete mystery to most people ... >:the private data area out of my routines, I find I can still modify the ...
    (comp.os.vms)