Badger, your joke was good, but I fail to recognize
who would be those two hikers.
Then, <How many times has AMD had to release a
fix or patch for a problem that occurs only when
you try to run an application on an AMD chip?>
And exactly how many times AMD has discovered a
hidden defect in your software?
You seem to be a programmer. Let me ask you, why would
programmers put several NOPs here and there, or use
JMP to next instruction, or use OUT anything to
a dummy I/O register? Because otherwise it
DOES NOT WORK ON PENTIUMS! This is a hack around
Intel's processor deficiencies, or a tribute for
"compatibility" with older, same Intel processors.
What makes you think that an alternative x86 execution
engine like AMD K6 is more flawfull than your
Pentium if it executes NOPs or LOOPs somewhat
faster? I think programmers must be thankful
to AMD for making their software products cleaner.
<It's a perception problem that AMD has to overcome,>
This is an education problem you
programmers have to overcome and to learn how
to write processor-independent software. The
Windows95 I/O initialization failure has also
been observed on higher speed P-II as well.
<and the only way to do that is to come out with
dependable, reliable, compatible silicon on a
timely basis.> You mean compatible with all
FOOF, FDIV, "Flag", and other 70+ "errata" on
Pentiums? |
