Haswell 4c8t @2.6 Fritz Chess Benchmark <Update:2.8G Haswell>

[csbin]

http://bbs.pceva.com.cn/thread-80481-1-2.html

VS Piledriver

2.8G :twisted:

 

[ShintaiDK]

Thats around 10% faster than a 3770K@2.6Ghz.

 

[mikk]

Thats around 10% faster than a 3770K@2.6Ghz.

Source for 3770k@2.6?

 

[Gideon]

Source for 3770k@2.6?

Yeah, it seems oddly optimistic. Does anyone know, what version of the software were they using ? Regadless by all examples I could find, IF 3770K scales linearly it should be closer to 4-6%

Score @3.5 Ghz

14666 - Fritz 13 - http://www.techspot.com/review/523-ivy-bridge-intel-core-i7-3770k/page5.html
14017 - Fritz 12 - http://www.hardwarecanucks.com/foru...-intel-i7-3770k-ivy-bridge-cpu-review-11.html
13989 - Fritz Chess Benchmark 4.3 - http://www.behardware.com/articles/...-review-intel-core-i7-3770k-and-i5-3570k.html
13869 - (leaked hardware) http://news.softpedia.com/newsImage/Intel-s-Ivy-Bridge-Core-i7-3770K-CPU-Benchmarked-6.jpg/

 

[inf64]

Yeah, it seems oddly optimistic. Does anyone know, what version of the software were they using ? Regadless by all examples I could find, IF 3770K scales linearly it should be closer to 4-6%

Score @3.5 Ghz

14666 - Fritz 13 - http://www.techspot.com/review/523-ivy-bridge-intel-core-i7-3770k/page5.html
14017 - Fritz 12 - http://www.hardwarecanucks.com/foru...-intel-i7-3770k-ivy-bridge-cpu-review-11.html
13989 - Fritz Chess Benchmark 4.3 - http://www.behardware.com/articles/...-review-intel-core-i7-3770k-and-i5-3570k.html
13869 - (leaked hardware) http://news.softpedia.com/newsImage/Intel-s-Ivy-Bridge-Core-i7-3770K-CPU-Benchmarked-6.jpg/

Thanks for those links. We can take the average number of all of those since they are pretty close to each other even though versions differ a lot. Average is ~14135 or so.
Do we know if Haswell sample had Turbo enabled? And does 3770K run at 3.5Ghz fixed clock in Fritz? Maybe IDC can test his chip and monitor the clock ?
If it indeed runs at 3.5Ghz and if Haswell had no Turbo enabled than 2.6Ghz IB would score ~14135*2.6/3.5=10500pts. I think that 3770K does run this test @ >3.5Ghz so actual difference(advantage) can be ~5% in favor of Haswell at the same clock.

 

[Idontcare]

Thanks for those links. We can take the average number of all of those since they are pretty close to each other even though versions differ a lot. Average is ~14135 or so.
Do we know if Haswell sample had Turbo enabled? And does 3770K run at 3.5Ghz fixed clock in Fritz? Maybe IDC can test his chip and monitor the clock ?
If it indeed runs at 3.5Ghz and if Haswell had no Turbo enabled than 2.6Ghz IB would score ~14135*2.6/3.5=10500pts. I think that 3770K does run this test @ >3.5Ghz so actual difference(advantage) can be ~5% in favor of Haswell at the same clock.

I disable turbo for all my tests because it is not deterministic, it depends on one's ambient temps and cooling (because power draw is temperature dependent, so current limits and so forth are likewise dependent).

When I set the 3770k to 3.5GHz I get 13,270 with Fritz Chess Benchmark 4.3.

If the Haswell chip is running 2.6GHz then I doubt turbo is enabled.

 

[FlanK3r]

nice, finally some screen of performance

 

[Gideon]

I disable turbo for all my tests because it is not deterministic, it depends on one's ambient temps and cooling (because power draw is temperature dependent, so current limits and so forth are likewise dependent).

When I set the 3770k to 3.5GHz I get 13,270 with Fritz Chess Benchmark 4.3.

If the Haswell chip is running 2.6GHz then I doubt turbo is enabled.

Cool

Now would it be too much to ask to also run it @2.6 ?

 

[inf64]

Thanks IDC! So your 3770K would score ~9857pts or ~9% lower than the Haswell part(if its Turbo is deactivated that is).

 

[Idontcare]

Cool

Now would it be too much to ask to also run it @2.6 ?

Thanks IDC! So your 3770K would score ~9857pts or ~9% lower than the Haswell part(if its Turbo is deactivated that is).

Confirmed the 3770k scores 9,866 at 2.6GHz (no turbo).

 

[ShintaiDK]

What a surprise

Thats 10.3% faster to be precise.

 

[ChronoReverse]

Looking good but just one benchmark. My plans to upgrade my C2Q to Haswell looks like a winner.

 

[FlanK3r]

Have someone here Sandy Bridge i7? For comparison clock to clock?

 

[MisterMac]

Was hoping for more.

(Geuss i can look forward to 30% increase from piledriver on Steamroller - it'll own haswell then ^^ )

But - i geuss well see with newer workloads, if there's a significant difference.

I don't expect x87 codework to gain much from haswells incrementals.

 

[Arachnotronic]

Have someone here Sandy Bridge i7? For comparison clock to clock?

Got an i7 2600K. Will run tests when I get back home.

 

[Arachnotronic]

Was hoping for more.

(Geuss i can look forward to 30% increase from piledriver on Steamroller - it'll own haswell then ^^ )

But - i geuss well see with newer workloads, if there's a significant difference.

I don't expect x87 codework to gain much from haswells incrementals.

Given the transistor budget, power budget, and the relatively weaker design team(s) at AMD, I would SERIOUSLY doubt that a 28nm "Steamroller" would "own" Haswell.

Do AMD believers understand the basics of the semiconductor industry?

 

[AtenRa]

Do AMD believers understand the basics of the semiconductor industry?

Do you understand that SR design has started before the layoffs ??

 

[Idontcare]

Have someone here Sandy Bridge i7? For comparison clock to clock?

Confirmed the 3770k scores 9,866 at 2.6GHz (no turbo).

Popped out the 3770k, popped in the 2600k, same exact hardware config for everything all around, and the 2600k scores 9,823 at 2.6GHz (no turbo).

 

[piesquared]

Looking good but just one benchmark. My plans to upgrade my C2Q to Haswell looks like a winner.

There was also 2 other benchmarks previously, and combined with this look a little, well, underwhelming. I was expecting much more. As said, maybe other benchmarks will show better results.

 

[FlanK3r]

there is my i7-3770K at 2600 MHz

So, if is this:

Haswell 2600 MHz = 10 884p
Ivy Bridge 2600 MHz = 9 907p
Sandy Bridge 2600 MHz = 9 823p

Nice performance improve! around 990 points!

 

[Durvelle27]

 

[Arachnotronic]

Do you understand that SR design has started before the layoffs ??

Do you understand that Intel's sheer R&D power, coupled with a nearly 2-generation lead in process tech, is likely to ensure that AMD doesn't come out with a superior high performance, performance/watt focused architecture?

 

[ShintaiDK]

Do you understand that SR design has started before the layoffs ??

If thats so, then so was Phenom, Phenom II, Bulldozer and Pilediver. Look where that got them.

 

[Rvenger]

This again is not an AMD vs. Intel who is better thread. We are talking about Haswell performance, not AMD and Intel's R&D.


If you want to have a VS discussion go find a thread from 2006.

 

[pablo87]

Impressive results though that drip marketing drives me insane. at least its not netburst marketing.