Linked by Thom Holwerda on Wed 31st Jan 2007 22:06 UTC, submitted by novaburst
Amiga & AROS The new AROS Developer Robert Norris has been interviewed by the AROS Show. "One of the original goals for AROS was 100% compatibility with AmigaOS 3.1. This is a noble goal, but it's not particularly forward-looking. AmigaOS has moved on since then, and there will be no new m68k hardware, yet many AROS developers are intent on making sure everything they do can be made to work on the older systems. That's their perogative of course, but my concern is that by constantly looking backwards we're missing the opportunities in the future."
Thread beginning with comment 208060
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE: What's the Point?
by viton on Thu 1st Feb 2007 17:02 UTC in reply to "What's the Point?"
viton
Member since:
2005-08-09

but I don't see the point in trying to bring an old broken design to new hardware
So, what is broken actually?

The only bad thing i see is what OS internals are exposed to applications.

SAS memory protection could be added on 64 bit processors even without the software modifications. Same for resource tracking.
Multiuser support may require api modifications of course.

Reply Parent Bookmark Score: 1

RE[2]: What's the Point?
by Mike Pavone on Thu 1st Feb 2007 18:04 in reply to "RE: What's the Point?"
Mike Pavone Member since:
2006-06-26

SAS memory protection could be added on 64 bit processors even without the software modifications

You cannot have proper memory protection, even in a single address space, in Amiga OS without modifying the API. Now from what I recall, the people (person?) involved in the x86-64 port was interested in making those changes if he/they haven't already done so.
You can't move to multiple address space (which makes things like virtual memory and memory mapped files much more practical) without completely reworking the message passing system in an incompatible way.

If you're going to break API compatability why bother trying to be compatible at all? What does Amiga OS do better than the alternative I mentioned that makes it worth dealing with the legacy issues it presents?

Reply Parent Bookmark Score: 1

RE[3]: What's the Point?
by viton on Thu 1st Feb 2007 19:15 in reply to "RE[2]: What's the Point?"
viton Member since:
2005-08-09

Full protection may not be possible without finegrained memory access privilegies.
But most recent programs could be modified to do it.
Non-modified programs can be partially protected in this system to prevent code corruption at least.
The lost memory and global memory fragmentation issues also can be easily avoided.

Multiple address space is old-fashioned technology for memory constrained machines ;) Today, machines has a plenty of virtual memory to map everything into 64bit space.
In current desktop unux/windows environments this address space is just wasted.

Itanium have some hardware support for SASos, but too sad, software industry is too inert to use it.

---

Why to bother with it at all?
I remember using the AMIGA was fun =)
Maybe not so fun anymore, but all OSes i use today are slow, bloated and hard to configure/fix.
But too sad for me, they have the software i need.
I didn't tried syllable or Haiku, although i used Beos in the past and it was good.

BTW, i don't see why anyone need to build legacy-compatible API as OS foundation if one can run old software in a sandbox and share resoures between the environments, like the Mac.

Reply Parent Bookmark Score: 1