Breaking 640 w/o Extender 
Author Message
 Breaking 640 w/o Extender

Dose there exist a DOS C/C++ compiler that will handle memory transparently?  
In other words, it will transparently handle huge globaly and locally declared
arrays.  If there is no such product, then which one comes closer?  
Any comments on the product will be highly appreciated.

Borland C++ 4.0/ PowerPack?
Watcom?
Symantic?
NDP?
--

Salem



Sat, 22 Feb 1997 05:51:42 GMT  
 Breaking 640 w/o Extender

   Dose there exist a DOS C/C++ compiler that will handle memory transparently?  
   In other words, it will transparently handle huge globaly and locally declared
   arrays.  If there is no such product, then which one comes closer?  
   Any comments on the product will be highly appreciated.

Yes.  djgpp, an MS-DOS port of GCC, a C, C++, and Objective C compiler
that is portable to many other machines.  It requires and takes full
advantage of at least a 386 processor.  No need for 'far' and 'huge';
a pointer is a pointer.  It does require a dos extender, but what do
you know, it comes free with that too.

No, it's not a full IDE like BC, and it doesn't do MS Windows.
At least, I didn't see anything in the documentation about MS Windows.
If you want more of an IDE environment, pull down the 'boxer' editor, which
can be configured to somewhat look&feel like BC for DOS.

djgpp is available at oak.oakland.edu in pub/msdos
boxer is available at oak.oakland.edu in pub/msdos/editors

Ed Karrels



Mon, 24 Feb 1997 02:20:51 GMT  
 Breaking 640 w/o Extender

Quote:

>Dose there exist a DOS C/C++ compiler that will handle memory transparently?  
>In other words, it will transparently handle huge globaly and locally declared
>arrays.  If there is no such product, then which one comes closer?  
>Any comments on the product will be highly appreciated.

>Borland C++ 4.0/ PowerPack?
>Watcom?
>Symantic?
>NDP?
>--

>Salem

Try PHAR LAP's 286|DOS-Extender.
I use it and it works. The only thing I've noticed is that my applications run
slower with it.

==============================================================================
       _       _
      / |     / |                                  _/    _/ _/_/_/_/ _/_/_/_/
     //||    //||                                 _/    _/ _/    _/ _/
    // ||   // || Maxwell Laboratories           _/_/_/_/ _/    _/ _/_/_/
   //  ||  //  ||                               _/    _/ _/    _/ _/
  //   || //   ||_____________________________ _/    _/ _/_/_/_/ _/_/_/_/_____
_//    ||//
       |_/        S-CUBED Division
______________________________________________________________________________
                                      |


  S-CUBED Division                    |   S-CUBED  : (619) 637-7472
  8808 Balboa Avenue                  |   FAX      : (619) 637-7402,
  San Diego, CA 92123                 |              (619) 637-7404
______________________________________|_______________________________________



Tue, 25 Feb 1997 09:28:06 GMT  
 Breaking 640 w/o Extender
: Dose there exist a DOS C/C++ compiler that will handle memory transparently?  
: In other words, it will transparently handle huge globaly and locally declared
: arrays.  If there is no such product, then which one comes closer?  
: Any comments on the product will be highly appreciated.

: Borland C++ 4.0/ PowerPack?

Using '32 bit DPMI' Mode you can access 4GB linear.  There are no more far or
huge modifiers necessary.  

Note:  With the 16 bit DOS extender you still have the problem that the
DGROUP segment (data segment) cannot be larger than 64kb.  This bothers me
so I am using the 32 bit DOS extender.  Unfortunately the 32 bit compiler
is slower than the 16 bit thing.  The 32 bit De{*filter*} doesn't allow Hardware
breakpoints without Windows NT :-(   But TurboVision programs run as text
applications under Windows NT. (at least that's what Borland says)

: Watcom?
never use it.  Borland should be more comfortable.

: Symantic?
I run a few tests but not with the DOS extenders.  The IDE is nice but
Borland's IDE is better.
: NDP?
never heard anything about this product.

ciao, Markus
--
   _    _ _  _





Tue, 25 Feb 1997 16:49:47 GMT  
 Breaking 640 w/o Extender

Quote:

>>Dose there exist a DOS C/C++ compiler that will handle memory transparently?  
>>In other words, it will transparently handle huge globaly and locally declared
>>arrays.  If there is no such product, then which one comes closer?  

Chapter 18 of Al Williams book "DOS 6, A Developer's Guide" contains
code for accessing up to 4 Gbytes in real mode. It includes some ASM
stuff and C routines for interfacing to it. Note that you can only use
the 4 Gig memory for storing data.
Chris Chlap
University of Canberra, Australia.
Quote:
>>Any comments on the product will be highly appreciated.

>>Borland C++ 4.0/ PowerPack?
>>Watcom?
>>Symantic?
>>NDP?
>>--

>>Salem
>Try PHAR LAP's 286|DOS-Extender.
>I use it and it works. The only thing I've noticed is that my applications run
>slower with it.
>==============================================================================
>       _       _
>      / |     / |                                  _/    _/ _/_/_/_/ _/_/_/_/
>     //||    //||                                 _/    _/ _/    _/ _/
>    // ||   // || Maxwell Laboratories           _/_/_/_/ _/    _/ _/_/_/
>   //  ||  //  ||                               _/    _/ _/    _/ _/
>  //   || //   ||_____________________________ _/    _/ _/_/_/_/ _/_/_/_/_____
>_//    ||//
>       |_/        S-CUBED Division
>______________________________________________________________________________
>                                      |


>  S-CUBED Division                    |   S-CUBED  : (619) 637-7472
>  8808 Balboa Avenue                  |   FAX      : (619) 637-7402,
>  San Diego, CA 92123                 |              (619) 637-7404
>______________________________________|_______________________________________



Tue, 25 Feb 1997 22:24:08 GMT  
 Breaking 640 w/o Extender


Quote:

>Using '32 bit DPMI' Mode you can access 4GB linear.  There are no more far or
>huge modifiers necessary.  

Not for a couple of years anyway! :-)

--
-----------------------------------------


-----------------------------------------



Wed, 26 Feb 1997 18:51:16 GMT  
 Breaking 640 w/o Extender
: Borland C++ 4.0/ PowerPack?
: Watcom?
: Symantic?
: NDP?

I can comment onm Borland C++ / the PowerPack. It works damn good. I've
successfully compiled some really big stuff in both 16- and 32-bit extended
modes. It's transparent, and I really couldn't ask for it to work any other
way. When you order the PowerPack from Borland, they'll send you a free
CD-ROM update to 4.02. Note: You've gotta have 4.02 for the DOS PowerPack to
work; it won't work with 4.0.

'Sides, 4.02 offers a lot of nice updates including smaller IDE files (but,
of course, it'll still load 4.0, 3.1, 3.0, 2.0 ones) and it works with all
those 32-bit OS's you keep hearing about (Chicago betas, blah, blah...)




Wed, 26 Feb 1997 23:55:37 GMT  
 Breaking 640 w/o Extender

Quote:
>: Borland C++ 4.0/ PowerPack?
>: Watcom?
>: Symantic?
>: NDP?
>I can comment onm Borland C++ / the PowerPack. It works damn good. I've
>successfully compiled some really big stuff in both 16- and 32-bit extended
>modes. It's transparent, and I really couldn't ask for it to work any other
>way. When you order the PowerPack from Borland, they'll send you a free
>CD-ROM update to 4.02. Note: You've gotta have 4.02 for the DOS PowerPack to
>work; it won't work with 4.0.
>'Sides, 4.02 offers a lot of nice updates including smaller IDE files (but,
>of course, it'll still load 4.0, 3.1, 3.0, 2.0 ones) and it works with all
>those 32-bit OS's you keep hearing about (Chicago betas, blah, blah...)

OK they did a great job on the build side, but what about the support tools?
I cannot get the de{*filter*} to break - ctrl, break - ctrl, alt, sysreq etc...
dont stop the program so if you get into a loop, you must reboot the machine
(if you terminate the de{*filter*} in windows, it trashes the system).  This would
not be that bad if it weren't for the fact that they got rid of the profiler
as of 4.0.  Since 4.0 my profiling has been done by breaking the program and
checking the stack to find where most time was being spent.  Now this too does
not work .  I called their 900 number and I was told that nobody knew about
this (yeah riggght) and that they would get back to me with a workaround (it's
been three weeks and counting) and that I would still have to pay for the 10
minutes it took for the tech to build and run a test program. (Thanks guys)

It look like the war between microsoft and borland has wounded borlands
quality control.


Commercial Computer Systems Inc.               ag chemical management software
         home of the                                   1-800-340-7525
 Application Management System                           St Pete FL



Mon, 03 Mar 1997 22:06:53 GMT  
 
 [ 8 post ] 

 Relevant Pages 

1. Fast VESA (640*400*256) pupixel routine

2. Saving/Restoring screen in DOS VGA (640 x 480)

3. 640 K Barrier

4. lint on Altos 2000 is BROKE BROKE BROKE!

5. Best way to display at 640 by 480 in 24bit.

6. More than 640 kb memory access in Borland C++ 3.1 ?

7. Broken code or broken compiler?

8. break points - don't break??

9. Multiple extender providers?

10. Property options with Extender/Custom Designers

11. Extenders/Custom Designers (tooltip "control")

12. MSC 6.01a Pharlap 286 Extender and DS != SS

 

 
Powered by phpBB® Forum Software