Re: A82: Ash and Kurai's problems examined
[Prev][Next][Index][Thread]
Re: A82: Ash and Kurai's problems examined
At 07:28 PM 12/6/97 EST, you wrote:
>[snip very long post]
>
>I believe the problem lies in relocation. Think about it: how can you
>transfer more than half of the calc's mem? I am going to try and make a
>different relocator that does not mess up. Ash 4 should definately have
>relocation!!!
>
>~Adamman
Yep, according to what I've seen about these relocation problems, you are
just moving big chunks of memory around. How about using a relocator that
moves single bytes around from where the program began to the relocation
spot and then push everything down, one byte at a time until the program is
copied. Sure this would be a slow method, but it would relocate large
programs without any problems showing up. For smaller programs, you could
just use the faster method of relocation. Most people won't mind since the
initialization of the program isn't what matters...the actual gameplay
speed is. People don't want the shell to interact with the game (i.e.
PROGRAM_ADDR and slow calls) because that just slows down the game. So,
relocation of large and small programs at the fastest possible speed is
necessary.
NOTE: Moving one byte at a time isn't feasible. Move 50-75 instead making
it faster but keeping it small enough for the TEXT_MEM.
Thomas J. Hruska -- thruska@tir.com
Shining Light Productions -- "Meeting the needs of fellow programmers"
http://www.geocities.com/SiliconValley/Heights/8504
http://shinelight.home.ml.org
Spam-Fighter code is: 14806560 (New technique to fighting spam)
Enter "Spam-Fighter: XXXXXXXX" in the body of a personal message
to thruska@tir.com. Otherwise the message will be filtered as spam.
References: