23-Apr-2024 18:43 GMT.		 UNDER CONSTRUCTION
Anonymous, there are 4 items in your selection
[News] Amiga's Strategy explainedANN.lu
Posted on 03-Apr-2001 10:10 GMT by Christian Kemp4 comments
View flat
View list
Gary Peake says that this posting [on Moo Bunny] does a really good job of summarising Amiga Inc.'s strategy.
Amiga's Strategy explained : Comment 1 of 4ANN.lu
Posted by Ian on 02-Apr-2001 22:00 GMT
"..not sure why they chose PPC not x86..."
Well, while the x86 is currently the popular one, some industry noises are being made about exactly how well the design philosophy fits into the future. Already the P4 has a big question mark over whether it is any better than a P3, or in fact slower. The PPC, especially G3 and up, are recent designs not harnessed to old ideas. Can't say for sure, but *maybe*, given enough market exposure, the PPC will prove to be the more powerful device.
Nice idea anyway, eh? :)
Amiga's Strategy explained : Comment 2 of 4ANN.lu
Posted by Federico Stein on 02-Apr-2001 22:00 GMT
>"..not sure why they chose PPC not x86..."
I think that a PPC architecture (or any architecture that doesn't carry any 'feature' that was designed in 1975 to last 2 years...) *IS* more powerful in computing power (and not just number crunching... WHY MY SYSTEM HALTS WHEN I USE THE F**KING FLOPPY DISK!).
But, and there is always a but... why, since I have a P3 with tons of memory+HDD space+peripherals+AmigaDE SDK should I have to:
a) Buy a new machine with similar computing power (AmigaONE).
b) Wait until a x86 version of the AmigaOS is released.
My question is: why they doesn't develop an AmigaOS for VP instead of PPC, and made only the critical parts (MP etc.) in native PPC *and* x86 code?
Amiga's Strategy explained : Comment 3 of 4ANN.lu
Posted by John Payne on 03-Apr-2001 22:00 GMT
> develop AmigaOS for VP code
Repeating caveat: Admittedly I was reading between the lines...
That's what I gather is in the works, probably incrementally,
with those services (cpu intensive) that benefit least from
being written directly to the hardware moving over to the VP first.
Maybe I'm taking Fleecy out of context, but one of his messages did
say that "hardware independence" would be introduced with AmigaOS 4.2
If the VP is as good as they say it is, and I have no reason to
believe otherwise, there's no hit at all from moving the system onto
it, so long as the code is all precompiled, which, of course, it would
be. The only time VP code runs slower than native code is the first
time; after that it's just as fast, and in some cases faster because
the VP can make optimizations for the specific hardware it's running
on that ordinary compilers can only make by producing very fat code,
with versions optimized for all contingencies.
Amiga's Strategy explained : Comment 4 of 4ANN.lu
Posted by John Payne on 03-Apr-2001 22:00 GMT
> with those services (cpu intensive) that benefit least from
> being written directly to the hardware moving over to the VP first
Anonymous, there are 4 items in your selection
Back to Top