Intel "Haswell" Quad-Core CPU Benchmarked, Compared Clock-for-Clock with "Ivy Bridge"

[Ikaruga]

Anton Shilov and his findings on the Interwebs..... I better wait for something more creditable, even if he is right this time somehow.

 

[qubit]

Maybe I am wrong, but probably this next gen will be some minor tweaking over prev gen, an increase in transistor count and maybe higher frequencies. This is how Intel plans for idiots to change their mobos into new one. Dark deal made with the mobo manufacturers.

Very well said. At this rate, I'll be sticking to my trusty 2700K Sandy Bridge CPU.

 

[Ikaruga]

Very well said. At this rate, I'll be sticking to my trusty 2700K Sandy Bridge CPU.

If I have to guess, I think it will be about much lower power consumption instead, since that's the area Intel is focusing on the most lately.

 

[tacosRcool]

So power consumption is the only thing Haswell got vs Ivy Bridge since there is not a big performance difference

 

[Aquinus]

I'm feeling pretty good about investing in a skt2011 rig right now with IVB-E down the road.

 

[FreedomEclipse]

No need to upgrade from my 2500k i guess.... money well spent!

 

[iO]

This is either a fake, a very early ES or Intel goes the Microsoft route and says "Screw you desktop, them all want mobiles!"...

 

[Aquinus]

... or Intel goes the Microsoft route and says "Screw you desktop, them all want mobiles!"...

A number of confirmed changes to Haswell could support this. Intel definitely is playing the power consumption card and they're going to beet it to death. Intel's CPUs are plenty fast already. I think they're working on the easier things to improve at this point because you can only get clock speeds and your IPC so high before you run into the diminishing returns problem.

If Intel can get a CPU to consume less power but perform just as well, that's a win.

 

[Crap Daddy]

Can't say I'm too surprised. The desktop era is coming to a close and fast. Haswell has to offer competitive TDP and power consumption in the war x86 vs. ARM. It's the future man. Everybody has gone insane with the mobile stuff. Intel has to deliver very soon chips that will make the ultrabooks and surfaces or whatever smack the ipads and nexuses on the head from different points of view than sheer performance (which is unquestionable).

 

[Melvis]

So no point for anyone to upgrade to this CPU/Socket unless your running a Socket 775 or AM2 still?

Good chance for AMD to catch up then i guess if this is true?

 

[Frick]

So no point for anyone to upgrade to this CPU/Socket unless your running a Socket 775 or AM2 still?

Good chance for AMD to catch up then i guess if this is true?

If you're an avarage, "normal", user still no point.

 

[Melvis]

If you're an avarage, "normal", user still no point.

To true, im talking about high end junkys and gamers more so

 

[radrok]

At this point the only exciting release will be the new Ivy lineup for socket 2011.
I mean I'm all for refining and cutting power consumption but as an hardware addict that's just not enough, I want performance on top of it.

Let's just hope Intel goes wild on the core count on skt 2011.

 

[phanbuey]

just wait and see- i mean superpi and wprime are not exactly all-encompassing benchmarks. My sandy bridge laptop gets close to those numbers in superpi, but i guarantee you it would get stomped by a haswell or ivy quad in everything else.

 

[Easy Rhino]

am i missing something? take cpu A at 2.8 ghz. take cpu B which can do much faster than that and bring it down to the speed of cpu A. how is that a good comparison of the two cpus? After all you are spending your money on what the processor can do... It's not like i am going to buy cpu B and downclock it and then act disappointed at the results...

 

[qubit]

am i missing something? take cpu A at 2.8 ghz. take cpu B which can do much faster than that and bring it down to the speed of cpu A. how is that a good comparison of the two cpus? After all you are spending your money on what the processor can do... It's not like i am going to buy cpu B and downclock it and then act disappointed at the results...

It's a clock for clock comparison to show the architectural improvements, so it makes sense to do this. Only if the new architecture has something up its sleeve with higher clocks will it offer any advantage to performance enthusiasts (us lot).

According to Intel's slides a while back, Haswell has some wicked overclocking features, so that might be enough for us to upgrade our SB/IB to it if it clocks significantly higher. It'll be on the same 22nm process however, so I wouldn't be surprised if it doesn't.

 

[BrooksyX]

Its funny. I used to buy budget CPU's and would upgrade almost every year. But this time around I decided to go with the high end (2500k) and I really see no reason to upgrade my cpu at least for another 1.5~2 years.

 

[Easy Rhino]

It's a clock for clock comparison to show the architectural improvements, so it makes sense to do this. Only if the new architecture has something up its sleeve with higher clocks will it offer any advantage to performance enthusiasts (us lot).

According to Intel's slides a while back, Haswell has some wicked overclocking features, so that might be enough for us to upgrade our SB/IB to it if it clocks significantly higher. It'll be on the same 22nm process however, so I wouldn't be surprised if it doesn't.

i understand that. we want to see if the new line has architectural improvements. but all my wallet cares about is how much faster is it going to load programs and perform mathematical computations.

 

[qubit]

i understand that. we want to see if the new line has architectural improvements. but all my wallet cares about is how much faster is it going to load programs and perform mathematical computations.

Indeed, it's a bit like the old Athlon XP / Pentium 4 situation from a decade ago, isn't it? The Athlon was more IPC efficient, but the P4 clocked higher, so it won even though it was so inefficient.

The answer you're looking for (and so is everyone else, lol) will be answered when the official reviews come out. It's just that to me, I think the fact it's on the same 22nm process means it will perform similarly to IB.

 

[FreedomEclipse]

I think the fact it's on the same 22nm process means it will perform similarly to IB.

or a superclocked SB

 

[cadaveca]

That's a different point entirely. Obviously if they are the same speed as each other at the same clocks, then the one clocked higher is faster. However, this is when other elements of the benchmark come into effect, i.e. does it clock higher? Has there been any power revisions or die shrink?

In the case of various phenom II revisions they could all pretty much be clocked to the same area of 3.8/4Ghz but effectively(as they were the same architecture) gave the same performance at the same clocks.

C2C is best used when it's between two different architecures.

Not in this instance. For IVB, CPU cache speed is directly linked to core clock...they run the same speed.


SO by downclocking an IVB chip, you are not reporting actual performance. you are reporting a gimped performance, with L3 running at a lower speed than intended.




Haswell breaks this design, and has L3 clocked independently, so C2C compare at low clocks doesn't tell you anything, but what a broken IVB does vs a non-broken Haswell.



Which makes this compare stupid, and that's why it was allowed. It's not a "real" performance compare.

 

[qubit]

@Easy Rhino

Looks like Cadaveca's answered your question nicely - this performance test isn't valid.

 

[Jorge]

If those benches are accurate, Haswell is a bust just as IB was.

 

[cadaveca]

If those benches are accurate, Haswell is a bust just as IB was.

That's what I think, and have thought, for many many months.


This is not the first time Haswell has been shown running publically.

However, I need a board and to clock a chip myself before I am 1000% on that.

 

[TheHunter]

I dont think its a bust, I mean look at what Haswell brings to the table,
http://www.anandtech.com/show/6355/intels-haswell-architecture/6
if all these changes translate in a lousy 5-10% increase then they need to do some serious work.



But then again, like someone said why improve a dead x87 code anyway. Imo those leaked benches mean squat. I say bring on real applications and games that will love bigger registers, more execution branches, faster single threaded optimizations and what not.