RE: LZ: PSOII Libraries
[Prev][Next][Index][Thread]
RE: LZ: PSOII Libraries
>Micah/Richard Brodsky wrote:
>>
>> PSOII ... Passive Shell Overlay/Independant Implementation.
>
>IMHO, this is the best library-implementation strategy thought of so
>far. Why? Because it's *shell-independent* without taking up too much
>space.
Why, thankyou;) After I develop a few more experimental libraries, I will test
them with all available shells, excluding UShell (which it should work with,
but UShell crashes my calc). Then I will release my source code. Since this
specification is really simple, there's little to enforce. At the moment
>Space, of course, is a most challenging restriction to the TI-85
>programmer/user. I have been silent throughout the whole USGARD
>introduction, but I have been watching very carefully. Andreas, please
>don't take offense to the following--these are just my personal
>thoughts.
>
>It seems to me that USGARD is a *great* idea--if you had 64K or 128K of
>RAM. There is just not enough memory on the average non-ESF-equipped
>person's calc to put enough games AND the USGARD shell AND the required
>libraries. Remember: we have less than 32K at best in which to work.
>We only have room enough for something simple--and even good-ole CShell
>is too large to store more than a few games. It's just too cramped in
>there. Plus, USGARD's implementation strikes me as being a little
>complicated. Ever wonder why no one ever writes games anymore? They're
>too busy worrying about their shells or how to program for them.
>
>So, IMHO, USGARD is a great idea that will need some ver aggressive
>marketing (Andreas, you are doing a very good job) to make users and
>programmers see past the memory constraints and complexity. Ohhhh and
>as far as PSOII lacking interrupt mechanisms, well, you can write a
>PSOII library to do it, right?
Technically, a library for interrupts cannot be PSOII (unless it's willing to
sacrifice stability), because it must have a fixed address for the interrupt
table. So far, only shell based and shell extending (stable) interrupt
handlers have been successfully developed. I have been toying with the idea of
using my Self-Tweaking-Code theme of PSOII to make an interrupt handler that
determins how it could extend any shell by looking at its end address. So far,
no progress, but some ideas.
However, it is possible to design a shell extender to implement interrupts to
work with PSOII libraries, but this violates the basic concept: Shell
independance. One would have to be made for each shell. PSOII currently should
be no-hassle for transfering between calcs, even with different shells.
>As for me, I'll still use CShell. Go, Mike/Keith!
>
>Jeff
>--
>#include <std_disclaimer.h>
>--
>####################################################### J
># Jeffrey Scott Sharp // A.K.A. Kyrie Eleison # SAVESu2
># // mailto:kyrie@telepath.com # S
># // (F)lamerz will be irradiated # U
>####################################################### S
># Why is it the Pentium and not the 586? Intel used #
># the first prototype to add 100 to 486 and they got # JSS+CLB
># ZILCH! # TLA 4E
>#######################################################
>
NOTE: I will be away, starting on the 29th of June, for 3 weeks. During that
time, I will not be able to develop more for PSOII.
--MZB (micahbro@msn.com)