| Author |
Message |
Casey Hawthorne
Guest
|
Posted:
Sun Sep 25, 2005 8:15 am Post subject:
Is backward compatibility with x86 architecture holding back |
|
|
Is backward compatibility with x86 architecture holding back CPU
functionality?
Back in the 486 days, I wondered why newer computers couldn't have two
CPU's; a 486 for x86 compatibility along with a newer Pentium design.
Are the design issues/trade-offs to complicated to have a P4 and a
newer chip (X5?) on the same motherboard?
Hopefully, the OS's would be rewritten faster than the applications to
take advantage of the newer chip while older software would run off
the older x86 chip, until some rewrites!
--
Regards,
Casey |
|
| Back to top |
|
 |
Zak
Guest
|
Posted:
Sun Sep 25, 2005 8:15 am Post subject:
Re: Is backward compatibility with x86 architecture holding |
|
|
Casey Hawthorne wrote:
| Quote: | Back in the 486 days, I wondered why newer computers couldn't have two
CPU's; a 486 for x86 compatibility along with a newer Pentium design.
|
You could get the 486 and the i860 on a board. From
http://lair.xent.com/pipermail/fork/Week-of-Mon-20050207/033587.html :
"Hauppauge made a workstation (the 4860) with both an i860 and a 486.
The 486 was to take the interrupt load away from the i860."
Hauppauge does mostly TV cards now, I think.
Thomas |
|
| Back to top |
|
 |
Guest
|
Posted:
Mon Sep 26, 2005 4:15 pm Post subject:
Re: Is backward compatibility with x86 architecture holding |
|
|
Casey Hawthorne wrote:
| Quote: | Is backward compatibility with x86 architecture holding back CPU
functionality?
|
No, not really; since the days of the 486; x86 designers have added
about 300 instructions; gone from 32-bits to 64-bits, added 64-bit MMX
and 128-bit SSE instructions, and a host of other non instruction
related functionality.
| Quote: | Hopefully, the OS's would be rewritten faster than the applications
|
The is no evidence that software can be moved forward as fast as
hardware, and considerble evidence that it cannot. |
|
| Back to top |
|
 |
|
|
|
|