More Sandy Bridge performance numbers (+Rumoured Pricing)

[-Sweeper_]

Compared to Bulldozer there already is a nice collection of benchmark numbers for Sandy Bridge. For example those posted by Coolaler, a few BOINC benchmark results and a video with a mobile Sandy Bridge running Cinema 4D. The video analysis done in the Planet3DNow forums resulted in a deciphered score of 19641, confirmed by the measured run time (44 s). This means, the tested mobile Sandy Bridge processor was as fast as a Core i7-975 Extreme. Another comparison could be done by using a recently published Geekbench result of a 1.6 GHz Sandy Bridge CPU. So I compared it to a Core i7 also running at 1.6 GHz and made following table with overall results and a diagram showing the differences in detail.



So the average performance increase with those CPUs at the same base clock, but with different Turbo Boost implementations, is about 20%.

...see the rest in dresdenboy's blog: http://citavia.blog.de/2010/08/05/mo...mbers-9128712/

edit: rumoured pricing on some dual/quads

[informal]

"So the average performance increase with those CPUs at the same base clock, but with different Turbo Boost implementations, is about 20%"

What is different with SB Turbo mode?More aggressive?

[-Sweeper_]

What is different with SB Turbo mode?More aggressive?

turbo is already pretty agressive in mobile chips.. maybe one extra multiplier for SB...

anyway, it seems some people here were wrong with their previsions of nearly zero gains in IPC....

[Movieman]

What is different with SB Turbo mode?More aggressive?

When you engage Turbo mode on Sandy Bridge the first thing you notice is that the case magically doubles in size, heat starts to double every second,your kill-a-watt meter goes insane and melts, and then you see that 20% increase!

[informal]

turbo is already pretty agressive in mobile chips.. maybe one extra multiplier for SB...

anyway, it seems some people here were wrong with their previsions of nearly zero gains in IPC....

I actually said upto 10% in IPC.Looks like with the new Turbo,the difference is 20%,what a surprise! Not.

When you engage Turbo mode on Sandy Bridge the first thing you notice is that the case magically doubles in size, heat starts to double every second,your kill-a-watt meter goes insane and melts, and then you see that 20% increase!

Hahah you made me laugh pretty good

[-Sweeper_]

I actually said upto 10% in IPC.Looks like with the new Turbo,the difference is 20%,what a surprise! Not.

dont be shy, I'm sure it was a surprise for you...

there is a difference between 0-10% and a 15-20% jump..

[informal]

IPC or Turbo boost? There sure is 15-20% combined jump,there is no doubt about that.And that is what the buyer gets which is pretty good.

[-Sweeper_]

IPC or Turbo boost? There sure is 15-20% combined jump,there is no doubt about that.And that is what the buyer gets which is pretty good.

maybe a little bit less than ~20% IPC increase considering they could be running at different clock speeds (what... one extra multiplier for SB?), but we could see even prettier results until launch..

23% and 24% gains in int and FP are impressive..

and the new desktop quads shall work at higher clockspeeds than lynnfield...

[Movieman]

Some thoughts:
Unless Intel has some really bad arse bin's in mind for SB they may have issues with keeping up with the last generation at least at the top end.
I don't see anything so far that will match a dual X5670 or X5680 running at 4200 on air in a EVGA SR2 board for pure computational power with the exception of a quad socket Beckton X7560 32 core machine and thats over $20,000.00 so that takes it out of the picture.
Now I know thats probably less than 1/10 of 1% of potential sales but this XS and we look for the best available.

[-Sweeper_]

Some thoughts:
Unless Intel has some really bad arse bin's in mind for SB they may have issues with keeping up with the last generation at least at the top end.
I don't see anything so far that will match a dual X5670 or X5680 running at 4200 on air in a EVGA SR2 board for pure computational power with the exception of a quad socket Beckton X7560 32 core machine and thats over $20,000.00 so that takes it out of the picture.
Now I know thats probably less than 1/10 of 1% of potential sales but this XS and we look for the best available.

hexa-core and octo-core models (high-end client & server 1p/2p) are coming later, more like Q3/2011...

[Movieman]

hexa-core and octo-core models (high-end client & server 1p/2p) are coming later, more like Q3/2010...

Point, but at least as it applies to the dualies unless they have speed bins of close to 4GHz they will be sucking hind teat compared to the current dual westmere's..

[haylui]

the performance gain is scary "to AMD"

[informal]

SB are supposed to have 100Mhz base clock ,right?This gives intel a broad range of multies for very flexible up/down clocking.

[Mechromancer]

Extreme performance without extreme pricing. Sounds good to me.

[zalbard]

Very attractive pricing to say the least.
32nm i7 for under 300$? Yes, please!
The i5 offer is just great, seriously.

[mAJORD]

Unfortunatly there is no indication of actual clockspeeds (turbo boost) so the tests are a bit fruitless..
just 1 bin out at low clockspeeds could = 10% performance delta

In fact I'd say as fruitless as trying to extract BD performance from the "50% faster" blanket statement in another thread

[-Sweeper_]

Very attractive pricing to say the least.
32nm i7 for under 300$? Yes, please!
The i5 offer is just great, seriously.

that i5 2400 could be an awesome CPU for gamers, specially at the right price

[richierich]

Will Intel be lowering prices? The i5-2400 is 3.1GHz and will cost $180, while the current i5-760 is 2.8GHz and priced at $209.99.

[Bobsama]

I'm guessing that LGA1155 and LGA2011 are the planned sockets for these? And the former is dual-channel with the latter is quad-channel RAM, right?

[-Sweeper_]

I'm guessing that LGA1155 and LGA2011 are the planned sockets for these?

yes

[xVeinx]

The pricing looks to be pretty similar to what we have currently to be honest. I don't see anything new in that regard. What would be interesting to know is whether current linux kernels (ie, 2.6.32/35) will be sufficiently optimized to be able to fully exploit the newer turbo boost.

[Johnny87au]

I'm guessing that LGA1155 and LGA2011 are the planned sockets for these? And the former is dual-channel with the latter is quad-channel RAM, right?

Yes thats right, Cant wait to see the high end chips!

[god_43]

yeah i figured it would be only around a 20% increase over now. good for them, sb is more of a revision than a new uarch right?

[Sn0wm@n]

dont be shy, I'm sure it was a surprise for you...

there is a difference between 0-10% and a 15-20% jump..

20% and most of it is acredited to turbo mode ... so why is it suposed to be good ???

[-Sweeper_]

20% and most of it is acredited to turbo mode ... so why is it suposed to be good ???

we dont know anything about this new turbo mode, and I really doubt the frequency is much different from clarksfield w/ turbo when stressing all cores..

23/24% faster in int/FP at the same base clock is good either way..