A82: Re: Crashes!!!!!
[Prev][Next][Index][Thread]
A82: Re: Crashes!!!!!
>I, being one of the few people with a link at my school decided to release
>ASH onto the general public today. Needless to say by the end of the first
>period, 3 people had crashed their calcs and by the end of the second,
>everybody who I gave a copy had crashed their calcs at least once. We need
a
>shell, like zshell, that is virtually uncrashable. Some of these people
The shells are just as uncrashable as ZShell (as far as i know), the problem
is that the program contains errors. No errors which causes crashes has been
found in any of the shell as far as i know.
>crashed their calcs just running the programs without running ash, others
did
>it while doing their physics or math analysis. One friend divided 8/7 and
her
>calc turned off. That is kind of weird. Anyway, we need a shell that is
user
The problem with some program is that they mess up the last entry buffer,
and therefore the calc crashes when you a accessing it (like when you do
8/7).
>friendly, I mean a shell that is idiot proof. Either the programs should
run
If you have any ideas on how to improve the user interface i would love to
hear about it,
>without using the shell, or that it should be impossible to access them
>outside the shell. When the shell is exited, all the buffers should be
Why should it not be possible to start the programs from the shell and
without the shell ?
>cleared and it should be hard as hell to fuck the calc up with a faulty
>program. I did release some questionable games, like space invaders (I am
not
The problem with this is that the function of most of the system part of the
ram is now know yet, and the shells can therefore not restore it with out
saving a copy of it. Saving a copy of the buffers in ram will make it
nessasary to have alot of free ram when you start an asm programs, and it
will make it impossible to change variables or pass any kind of information
from the asm programs to the TI-OS. I think this is would be great to have
in shell, and even if the buffers where saved it would not make the shells
crash proff.
>sure if it is the right version), tetris (which hasn't been thorougly beta
>tested), Same Game (which still in its delevomental stages, although is
>probably very safe), and the pong game I wrote (which is reasonably simple,
>yet there is countless oppurtunity to crash your calc in asm). The breakout
>type game that I released also hasn't been thoroughly tested with the 82.
>Games that I call safe are like tunnels, which has been thoroughly tested
by
>the user community and is relatively simple. Monday I will release only the
>safest games that I know of and we will see what happens.
The wrong version of Spaze Invaders causes exactly the problems you are
ralking about.
Dines
Follow-Ups: