SPECint_rate2006 (outdated?) scores for Barcelona

Quote:

---

Originally Posted by alayashu

no one in the industry who buys these CPUs gives a damn about how well
a single core performs. System performance is what matters period.

---

Except the same cores will be used for the desktop market, and where the non-rate SPEC tests are more relevant than the rate tests.

Quote:

---

your bashing of the _rate benchmarks and aproval of the nonrates is really
hard to understand and makes no sense.

---

I don't think it's a bashing of the rate benchmarks specifically, merely a desire for more benchmarks that paint a more complete picture of the performance of Barcelona, vs AMD's continual usage of only SPECfp_rate for months to hype the performance of Barcelona while being silent on the many other available server benchmarks.

Quote:

---

For your information the _rates
run the same nonrate benchmarks but on all the cores of the system, thus
giving us a good clue of how the whole system works on certain type of
applications. hard to pick your logic... seriously...

---

Running multiple copies of single-threaded applications is not really how most server applications work.

This whole debate reminds of when the P4 first made its appearance, at that stage memory bandwidth was the buz word and how important it was for Quake 3. Strange how times have changed.

Quote:

---

Originally Posted by Periander6

Those Specint2006 and Specfp2006 scores are very very weak, well behind what Clovertown offers on a clock for clock basis. This bodes quite badly for what K10 will offer in terms of desktop performance.

---

And you know what are the scores of Clowerton @1.9 GHz? Please share them with us!

Quote:

---

Originally Posted by Nedjo

And you know what are the scores of Clowerton @1.9 GHz? Please share them with us!

---

The best published scores for a 1.86GHz/1066MHz FSB E5320 is
SPECint2006 13.6
SPECfp2006 11.1

http://www.spec.org/cpu2006/results/...806-01697.html
http://www.spec.org/cpu2006/results/...514-01054.html

Intel Xeon processor 3040, 1.86GHz:

SPECfp®2006 = 12.5

SPECfp_base2006 = 12.3

http://www.spec.org/cpu2006/results/...625-01332.html

Dual core, though.

Quote:

---

Originally Posted by Shintai

So when we stop using bandwidth/rates benches. K10 suddenly drops behind...how funny...and how predictable.

---

And when we stop using benches that fit in the cache C2D suddenly drops behind... it's also funny...and predictable:
http://blogs.zdnet.com/Ou/images/4p-...amd-server.png

Look how L7345 beats the crap out of the E7320 despite being 200mhz slower... if that isn't funnier then any result K10 can produce i don't know what is.

But i guess it's ok to use cache, but completely wrong to use bandwidth... go figure.

Quote:

---

Originally Posted by DoubleZero

And when we stop using benches that fit in the cache C2D suddenly drops behind... it's also funny...and predictable:
http://blogs.zdnet.com/Ou/images/4p-...amd-server.png

Look how L7345 beats the crap out of the E7320 despite being 200mhz slower... if that isn't funnier then any result K10 can produce i don't know what is.

But i guess it's ok to use cache, but completely wrong to use bandwidth... go figure.

---

Since there is only 4MB DC Xeons and 8MB Quad Xeons..go figure it out. There is no 4MB quad Xeon. Its a wrong posting.

http://processorfinder.intel.com/det...px?sSpec=SLA69

But I would guess they use a single Quad there. And rest is dual.

again, (as i've said ad naseum), unless folks are using the production stepping vs. the B1 stepping tomorrow, you're going to see some pretty interesting results.

ask a lot of questions and try to understand what they're testing on. I've got no doubts that AMD wants the reviews to be positive, but given the huge populace of idiots in the reviewing world, someone's going to fubar it.

dave

Quote:

---

Originally Posted by Shintai

Since there is only 4MB DC Xeons and 8MB Quad Xeons..go figure it out. There is no 4MB quad Xeon. Its a wrong posting.

http://processorfinder.intel.com/det...px?sSpec=SLA69

But I would guess they use a single Quad there. And rest is dual.

---

That's a typo, E7320 is only 4mb.
http://www.hardwareschotte.de/hardwa...0+BX80565E7320

Don't you think the score is a bit high for a single Quadcore? :)

Quote:

---

Originally Posted by dave_graham

again, (as i've said ad naseum), unless folks are using the production stepping vs. the B1 stepping tomorrow, you're going to see some pretty interesting results.

ask a lot of questions and try to understand what they're testing on. I've got no doubts that AMD wants the reviews to be positive, but given the huge populace of idiots in the reviewing world, someone's going to fubar it.

dave

---

You mean run something other than SpecFP2006? Sorry... couldn't help it... this is going to be the bench people focus on.

Quote:

---

Originally Posted by Nedjo

And you know what are the scores of Clowerton @1.9 GHz? Please share them with us!

---

Clovertown 1.86 GHz best scores
http://www.spec.org/cpu2006/results/...0611-01262.pdf
http://www.spec.org/cpu2006/results/...0305-00613.pdf


SpecINT2006 peak 14.2, base 12.9
SpecFP2006 Peak 10.9, base 10.8

If you like, take the 2.0 clovertown results to
http://www.spec.org/cpu2006/results/...0806-01673.pdf
http://www.spec.org/cpu2006/results/...0301-00593.pdf

SpecINT2006 peak 15.5, base 14.0
SpecFP2006 Peak 14.0, base 13.6

Quote:

---

Originally Posted by DoubleZero

That's a typo, E7320 is only 4mb.
http://www.hardwareschotte.de/hardwa...0+BX80565E7320

Don't you think the score is a bit high for a single Quadcore? :)

---

Arh you are right. Ye could be so. Cache would have a large impact on the rates. Specially in 4 sockets.

Quote:

---

Originally Posted by JumpingJack

Clovertown 1.86 GHz best scores
http://www.spec.org/cpu2006/results/...0611-01262.pdf
http://www.spec.org/cpu2006/results/...0305-00613.pdf


SpecINT2006 peak 14.2, base 12.9
SpecFP2006 Peak 10.9, base 10.8

If you like, take the 2.0 clovertown results to
http://www.spec.org/cpu2006/results/...0806-01673.pdf
http://www.spec.org/cpu2006/results/...0301-00593.pdf

SpecINT2006 peak 15.5, base 14.0
SpecFP2006 Peak 14.0, base 13.6

---

Don't want to knock the results but why is the libquantum results so high? I didn't realise there was that kind of architecture improvement in clovertown?