Re: A89: Re: Re: Re: Re: CORRECTION: Shift+ON


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

Re: A89: Re: Re: Re: Re: CORRECTION: Shift+ON




> > Do you think that TIGCC will be actual at all after releasing
> > the SDK?

That's questionable - I haven't compared TIGCC and the SDK's compiler in
terms of program size and efficiency yet, although I personally think that
the SDK compiler may stick a little too close to following the flow the
original code rather than make any huge optimizations.  The one time I DID
compare output - on the interrupt installer - the SDK killed TIGCC, but hand
written code was about twice as good as either of them (but it was a strange
function to test with).  There's also the question of how well the inline
ASM format works out (can't elaborate) and how much the GCC extensions
contribute to TIGCCLIB's workings and whether kernel progs can be made
without too much trouble.  That, and there are apparently a few bugs =)

> It definitely will, since the free SDK has an 8 KB app size limit, if I
> remember that right.  It might have been changed 24 KB since AMS 2.04 also
> has the ASM program limit changed to that value, but there would still be
a
> limit inside the SDK.
> I think that you can only get rid of that limit legally if you buy the SDK
> for a lot of money, and I don't promote illegal things.  My point:  The
SDK
> cannot replace TI-GCC.

Geez, what's up with the rash of ill-informed posts on this list?  The size
limit is in SOFTWARE ONLY and only applies to ASM PROGRAMS and not
applications.  ASM programs, whether built by TIGCC, TI Flash Studio, or
a68k, have an 8k/24k limit (depending on AMS version) in software which has
been circumvented by certain programs.

The catch is that you CAN'T program apps without paying to have the signed,
even free ones.  The SDK can make ASM programs just as well as anything
else, but to use all the benefits of apps (including "legally" surpassing
the size limit) you'll have to pay for signings.

    -Scorr




Follow-Ups: