Re: Mike and Ward, what next?

Erik <groups@[EMAIL PROTECTED]
> wrote:

> Just to add to what I wrote above, I found this interesting quote from
> the comp.sys.mac.system newsgroup.  It regards Rosetta, the PowerPC
> emulator for the new Mac/Intel CPUS:
> 
> -------------------------------
> 
> Rosetta does not run the following:
> 
> - Applications built for Mac OS 8 or 9
> - Code written specifically for AltiVec
> - Code that inserts preferences in the System Preferences pane
> - Applications that require a G4 or G5 processor
> - Applications that depend on one or more kernel extensions
> - Kernel extensions
> - Bundled Java applications or Java applications with JNI libraries
> that can't be translated
> 
> -------------------------------
> If that's true then Mops has to change once the Intel CPUs are in full
> force.  Anything other than Mach-O format won't be sup****ted.

Jobs makes it sound like he has a smooth transition all mapped out, but
this is a much bigger change than 68k->PPC.  

Obvious deductions are that CFM Carbon will be gone with the release of
"leopard" as will Classic and CFM sup****t.

Jobs generously offers to sell mid and upper tier developers a 4.6 GHz
P4 Mac for $995 (plus $500 for Select member****p).  So we need yet
another machine on hand for developing and testing.


GForth is currently Mach-based.  But as Erik pointed out, the completely
different register model and optimizing methods likely will make
PPC-based Forths dog-slow when run under Rosetta (see page 67 of the
Universal Binary do***ent).  And Rosetta is not like the old PPC-based
68k emulator - not all PPC code will run under Rosetta.

Since Jobs really wants developers to convert to Universal Binary
format, other issues come up. To build a Universal Binary Forth, it
sounds like XCode will automatically make 2 code images, one for PPC and
one for x86.  This will make it really tough to do an interactive
compiler that can make a snapshot/turnkey that can run on the "other"
platform as well (assuming Apple even do***ents the format of a
Universal Binary).  So for those of us who are doing our own
compiling/optimizing, it means two versions as far as I can tell. 



.... maybe we should go back to a tokenized model with the token
displatching engine written as a Universal Binary and just abandon
including an assembler or attempting to do non-trivial optimizing...

My comment for the day is "ugh"

 -- w