To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.off-topic.geekOpen lugnet.off-topic.geek in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Off-Topic / Geek / 289
288  |  290
Subject: 
Re: Perl rules!
Newsgroups: 
lugnet.off-topic.geek
Date: 
Tue, 20 Jul 1999 20:33:17 GMT
Viewed: 
1248 times
  
On Tue, 20 Jul 1999 18:27:57 GMT, "Todd Lehman" <lehman@javanet.com> wrote:

I'd say, back in 1983, the lack of virtual memory and the 640KB limit was
no big deal (in the PC industry).  By 1989, it was becoming unfortunate.
By 1991, it was getting really bad.  And by 1994, it was pure evil, IMHO.
Now, it's just laughable.

But not so much the lack of virtual memory as the lack of addressability
beyond 1 MB (or 640 KB, since devices started at A000:0000 in the PC.)
Sure, there were DOS extenders and protected-mode drivers and libraries, but
those didn't become ubiquitous until years after they were sorely needed.

And why did IBM stick all the ROM and system stuff in high memory (>640K),
rather than low memory?  If they had done that, it would have been (more)
possible to extend the address space without totally losing backwards
compatibility.

Or if Intel had made a processor which would do "old mode" as a secondary
option, rather than the bootup default.  Then old programs could be run by
dropping into stupid mode, rather than the system having to drag itself up
out of stupid mode every time it wanted to do 32-bit.

It just sucks that IBM chose the Intel 80x86 archtecture over, say, the
Motorola 68xxx architecture.  The 68xxx architecture was always 32-bit and
would have saved the PC industry years of segmentation grief.  And although
the 32-bit 80386 was released in 1985, it took nearly 10 years to stomp out
all of the old 8086 legacy garbage -- and it's not even totally gone today.

Yeah, whoever decided to do the segmented memory thing should have been
shot.  What a horrible idea.  At least, it was enough to scare me off of
learning 80x86 assembler.

Was the 8.3 file format originated with MS or DOS?  I thought it was
around before them.

I think MS-DOS got that from CP/M, but I'm not sure.  I do remember for sure
that some of my old CP/M programs in the late 70's all ended with .BAS and
had short names, but I'm not sure if CP/M originated 8.3 or whether it
actually was 8.3 and not something else close to that.

Hmm.  Commodore machines (PET, C64, VIC) used 8.3, but they got it from MS
(because MS wrote the in-ROM BASIC with file IO kernal).

But anyway, had Linux been used instead of DOS (and this is not to suggest
that James made a mistake), there certainly wouldn't be the 8.3 limit.

True.

IIRC, the old SVGA architecture used 16KB buffer windows below the 1MB
boundary for 16-bit apps to write to them, or you could make simple INT 10
BIOS calls (ayyyahhh!)...  Not much need for device-abstraction libraries
back then since the video hardware was still rather homogeneous, but the
big limitation was not being able to access 32-bit addresses directly
without doing really gross things like flipping in and out of 32-bit
protected mode.  :-(

Yeah, but what does that have to do with LDraw not supporting modes with
more than 4-bit color?

In that regard, thank god for MS-Windows, since it helped do away with most
of the mess created by itself and Intel earlier.

Well, that's one good thing.

Are they all done rediscovering the big-iron concepts that they didn't have
to worry about in the early 80's?  Or is there more to come?

I'm still waiting for the *real* OS to come along.  *nix will never appeal
to the masses--the knowledge requirement is too high.  WebTV (and similar
things) won't cut it as a real computer.  Windows is too bloated and
legacy.

So what will it be?

Steve



Message has 2 Replies:
  Re: Perl rules!
 
(...) Wait a minute. Feel free to correct me if I'm wrong, but the segmented memory architecture allowed you to create absolutely *tiny* programs that did wonderful things. And you could fit tons of these programs on an affordable *floppy* disk. I (...) (25 years ago, 20-Jul-99, to lugnet.off-topic.geek)
  Re: Perl rules!
 
(...) Actually, it would have been a hell of a lot easier, given that you at elast have homogeneous address space. But you do need to rewrite the kernel to be able to access it at all, and we all know how fast MS is at that sorta thing.. (...) And (...) (25 years ago, 22-Jul-99, to lugnet.off-topic.geek)

Message is in Reply To:
  Re: Perl rules!
 
[removed lugnet.off-topic.debate from crosspost list] (...) I'd say, back in 1983, the lack of virtual memory and the 640KB limit was no big deal (in the PC industry). By 1989, it was becoming unfortunate. By 1991, it was getting really bad. And by (...) (25 years ago, 20-Jul-99, to lugnet.off-topic.geek)

433 Messages in This Thread:
(Inline display suppressed due to large size. Click Dots below to view.)
Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact

This Message and its Replies on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR