Jozef, "The scenario for success of hammer..."
"... depends on Microsoft"
I agree.
"Suppose the Win64 becomes available, and one of these niche applications, such as the Flask or Divx (or whatever) gets recompiled into a true x86-64 app, with use of all the extra registers, basically takes full advantage of the processor. Now someone posts the score, which is let's say 80% better than the next fastest processor.."
I am not familiar with algorithmic details of Flask or
Divx, but I am absolutely sure that expectations of
80% performance gain are out of any reality. Sorry.
"I think Hammer in full 64 mode will be on top of many of these benchmarks. ... And I didn't even mentioned the benefit of code in the Direct X and video drivers at some point being compiled to 64 bit mode."
I am not positive that there are many benefits from
64-bit mode for most of ordinary applications. Advantage
of having more register was already partially taken by
register renaming technique, so explicit coding for
extended registers will likely not lead to a breakthrough.
64-bit addressing will bloat the code and harm the decode
bandwidth.
Also, as you mention, all code has to be
recompliled, and probably will not run on Intel CPUs
(unless they copy AMD's x86-64 exactly). This will
create big problems for code maintenance for SW
people, with corresponding cost.
>"Hammer breaks 500 fps barrier" (in resolution 640x480x16bit),<
Looking at game benchmarking results, it is quite
obvious that CPUs are not a bottleneck there.
I agree that success of Hammer is in significant
performance lead. Somehow I think that it is not
going to happen. I think the best processors from
both rivals will continue to be in plus-minus 10%
range of performance. Given infrastructure deficiencies,
it will not be enough to beat Intel with one "left hand".
I'd be happy to be wrong on my predictions, but...
- Ali |
