POWER7 Spec

By: Linus Torvalds (torvalds.delete@this.linux-foundation.org), February 10, 2010 9:17 am
Room: Moderated Discussions
someone (someone@somewhere.com) on 2/9/10 wrote:
>
>Nehalem beats Montvale by more than the frequency ratio?

Yup.

>You mean a processor that connects directly to 3 channels
>of DDR3 beating a processor that uses a far narrower pipe
>to memory through a front side bus, chipset (and in some
>cases DIMM expander bridge chips) on a memory intensive
>benchmark suite by more than the frequency ratio is a fault
>at the core level?

Yes. But your argument that this is somehow special is
total crap, because if you had actually used your
brain and been honest for one second, you would also
have realized that SpecInt doesn't have that excessive
a memory footprint, and the 9MB L3 (per core) of that
Itanium is certainly quite large for the load.

So.. A regular desktop Core 2 also beats Itanium senseless.
Including on a cycle-to-cycle basis.

That's no "3 channels of DDR3", that's the traditional
Intel FSB.

How hard is it for you to just admit that Itanium sucks
from a performance standpoint. The frequency has been
anemic, but despite that low frequency, the actual work
done per cycle has also been anemic.

>You have an amazing ability to disregard the obvious and
>claim ridiculous crap if the payoff is a gratuitous bash at IPF.

Well, at least I have my eyes open to reality, unlike you.

Yes, I'm biased against Itanium. I think the architecture
was created by retarded monkeys on crack. I also think that
the implementation of the architecture has also been pretty
sub-par.

But at least I can argue with real numbers, unlike you.
Your argument always seems to be ".. but wait until the
next one".

Go and look at the actual performance numbers. Here's a
relevant example:

- 403.gcc: Core 2 Quad Q9650 (3GHz): 23.9/20.5
- 403.gcc: best Itanium (1.66Ghz): 12.9/11.7

and notice how that POS Itanium number is actually lower
cycle-for-cycle than that Core 2 FSB number.

And btw, you don't need to pick some high-end Core 2 like
I did above. I picked that one because it has 6MB/two cores
of L2 cache, but a regular Core 2 Duo E6700 gets pretty
close to the same work-per-cycle as that high-end server
chip. And runs at a 60% higher frequency.

So what's your point again?

Admit it, you were wrong. Itanium sucks from a performance
perspective. The architecture is pure crap, the idea of
in-order wide execution as a "performance platform" is
total mindless drivel, and the execution on the micro-
architecture has always been pitiful.

How f*cking retarded do you have to be to still continue to
make excuses for it? When do you just admit that it didn't
work out? How many more failures do you need?

(And yes, it probably depends on the benchmark. I've used
403.gcc above, partly because I think it's a relevant one,
and partly so that you can't come up with your regular
"but but but libquantum" excuse. The point is, Itanium is
absolutely not all that great per-cycle either!)

Linus
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
POWER7 SpecParadox2010/02/08 11:05 AM
  POWER7 SpecThu Nguyen2010/02/08 12:58 PM
    POWER7 SpecIan Ameline2010/02/08 09:22 PM
      POWER7 SpecThu Nguyen2010/02/08 11:54 PM
        POWER7 SpecIan Ameline2010/02/09 06:46 AM
          POWER7 SpecLinus Torvalds2010/02/09 07:57 AM
            industry-standard single-threaded performance benchmarks absentMichael S2010/02/09 08:09 AM
              industry-standard single-threaded performance benchmarks absentParadox2010/02/09 08:33 AM
                industry-standard single-threaded performance benchmarks absentMichael S2010/02/09 09:30 AM
                  industry-standard single-threaded performance benchmarks absentParadox2010/02/09 10:52 AM
                    industry-standard single-threaded performance benchmarks absentJesper Frimann2010/02/09 11:33 AM
              industry-standard single-threaded performance benchmarks absentLinus Torvalds2010/02/09 08:48 AM
                industry-standard single-threaded performance benchmarks absentMichael S2010/02/09 09:26 AM
                  industry-standard single-threaded performance benchmarks absentLinus Torvalds2010/02/09 09:58 AM
                    industry-standard single-threaded performance benchmarks absentJesper Frimann2010/02/09 12:17 PM
                      industry-standard single-threaded performance benchmarks absentLinus Torvalds2010/02/09 12:54 PM
                        industry-standard single-threaded performance benchmarks absentJesper Frimann2010/02/09 02:10 PM
                    industry-standard single-threaded performance benchmarks absentParadox2010/02/09 01:22 PM
                industry-standard single-threaded performance benchmarks absentanon2010/02/09 10:21 AM
            POWER7 SpecJesper Frimann2010/02/09 12:30 PM
            POWER7 SpecDoug Siebert2010/02/09 05:38 PM
              POWER7 SpecLinus Torvalds2010/02/09 07:28 PM
                POWER7 SpecLinus Torvalds2010/02/09 07:28 PM
                  POWER7 SpecIan Ameline2010/02/09 08:02 PM
                    POWER7 SpecDoug Siebert2010/02/09 10:18 PM
                POWER7 Specsomeone2010/02/09 08:20 PM
                  POWER7 SpecLinus Torvalds2010/02/10 09:17 AM
                    POWER7 SpecLinus Torvalds2010/02/10 09:46 AM
                      POWER7 SpecEmil2010/02/10 11:06 AM
                    POWER7 SpecIan Ameline2010/02/10 10:13 AM
                    POWER7 Specsomeone2010/02/10 11:01 AM
                      POWER7 SpecLinus Torvalds2010/02/10 11:32 AM
                        POWER7 SpecMichael S2010/02/10 12:30 PM
                          POWER7 SpecLinus Torvalds2010/02/10 01:25 PM
                            POWER7 Specmpx2010/02/10 02:58 PM
                        POWER7 Specnemlis2010/02/11 12:24 AM
                          POWER7 Specnone2010/02/11 12:52 AM
                            POWER7 Specnemlis2010/02/11 01:52 PM
          POWER7 Specmpx2010/02/09 08:18 AM
          POWER7 SpecMichael S2010/02/09 09:08 AM
        POWER7 SpecJesper Frimann2010/02/16 02:29 AM
          POWER7 SpecJamie Lucier2010/02/16 06:35 AM
            POWER7 Specanon2010/02/16 07:16 AM
              POWER7 SpecJamie Lucier2010/02/16 10:46 AM
                POWER7 Specanon2010/02/16 03:42 PM
                  POWER7 SpecJamie Lucier2010/02/18 06:07 AM
              POWER7 SpecThu Nguyen2010/02/16 01:23 PM
            POWER7 SpecParadox2010/02/16 09:57 AM
              POWER7 SpecJamie Lucier2010/02/16 11:22 AM
                POWER7 SpecParadox2010/02/16 02:00 PM
            POWER7 SpecJesper Frimann2010/02/16 02:14 PM
              POWER7 SpecJamie Lucier2010/02/17 10:41 AM
  POWER7 SpecJesper Frimann2010/02/09 04:43 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell tangerine? 🍊