[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MiNT] strange memory violation

On Tue, 2005-11-08 at 19:08 +0100, Patrice Mandin wrote:

> The problem is that it is currently not that easy to make clean code.
> And I think the cause is that most people don't want to code multiple
> code paths for mint/magic/tos/whatever.


The biggest problem of Atari programming is that you have to support at
least three ways of doing everything, more if you choose to claim that
MiNT without Virtual Memory is a valid fourth alternative (I don't).
There isn't a single API to code to.  Then add NVDI vs fVDI vs GDOS vs
no-GDOS.  AES vs XaAES vs MyAES vs MagiC.  As the saying goes, "The best
thing about standards is there are so many to choose from!".

I've often said the best course for improvement is to agree on a single
API, but the only logical course for that would be make that API be
MiNT, and no one wants to leave Tos/MagiC users out in the cold.

I don't see an easy solution, but I definately think MiNT offers the
cleanest API when it comes to multitasking safety and abstraction of
hardware and it has the best chance of lasting.  Its also the only one
being updated.  There aren't enough programmers to support so many APIs,
and the resistance to change and the special needs of what is left of
the user-base only complicate matters.

-- Evan