| From: | Jean-Pierre |
| Date: | 5 Aug 2000 at 22:44:48 |
| Subject: | AMIOPEN: Arstechnica mentions Developerworks Interview |
http://www.arstechnica.com/
Interview on the new Amiga
Posted 08/04/2000 - 12:27am EDT
IBM developerWorks has a feature article that's an interview with Bill
McEwen, President of Amiga, Inc., and Fleecy Moss, Amiga's VP of
Development. The interview discusses some of the differences between the old
Amiga and the new Amiga, as well as giving a good overview of some of the
technology that's going into the new AmigaOS. Here's one part that I found
particularly interesting:
dW: How can you possibly get "very fast" performance from
platform-independent binaries?
Moss: This is one thing people don't understand, and they keep on saying,
"The new Amiga provides portable binaries. Portable binaries are slow." Now,
in the past they've been slow because portable binaries have always been
interpreted. But what happens with Tao's Intent is that they are dynamically
translated into actual native machine code. It only has to be translated
from VP code to native code once.
This perfectly exemplifies trends that I pointed out in my article, The
Future of x86. In fact, after that article I got a lot of mail from folks
who suggested I check out AmigaOS for just this reason.
Amiga is doing one interesting variant on the translation and caching scheme
that I discussed in my article: they're allowing AmigaOS tools to be
developed natively for specific hardware platforms, tools which then replace
Amiga's virtual-processor-based equivalents.
And one of the other gorgeous things about it is first you write the VP code
version [of the driver/program]. If you then decide that you want to add
some optimizations, say for the 3DNow! or Altivec instruction set, Intent
can't abstract that because it's chip-specific. But what you can do is have
new tools written in native code. So you can say, "here's all my VP code,
which does the job, but if you have a PowerPC running underneath, then here
are a few extra tools that replace these VP code ones and use the special
registers and instructions."
So they're not requiring that all AmigaOS software and tools be written to
the virtual processor, which is interesting because this at first seemed to
me to be a sort of fence sitting approach that would reduce the advantages
of translation -- why not just go ahead and use an OS that's completely
portable at the source code level, like Linux? When I thought about, though,
I realized that what Amiga wants is to release an OS in binary form that
runs on a variety of platforms from the start, and have people start moving
pieces of the OS into native binary form for their specific platform as they
see fit. This would be the opposite of a Linux-style approach, where you
initially release an OS in source code form that runs on one platform, and
then let people port the entire thing to their individual hardware.
If Amiga can really deliver on all they're promising, then I think that this
piecemeal porting model might represent a superior way to do things. Think
about it: instead of having to port the OS, libraries, etc. to a new
platform to get full functionality and application support, you just port
the translator and a few other critical pieces and you're up and going
full-bore from the start; you can then port the OS in piecemeal fashion at
your leisure. Props to TB for the interview link. -Hannibal
Subscribe/Unsubscribe: open-request@amiga.com
Amiga FAQ: http://www.amiga.com/faq.html