***Official Intel Haswell Thread***

[smilertoo]

I'm pretty sure all the cpus with more than 4 core actually run slower than 4 core equivalent unless the extra cores are used, then you see a massive jump.

 

[Kaapstad]

I'm pretty sure all the cpus with more than 4 core actually run slower than 4 core equivalent unless the extra cores are used, then you see a massive jump.

There is not much difference in speed (maybe 100mhz) between what you can get from a SB 2600/2700k and a SB-E hex core. Also when you take into account memory speeds obtainable the SB-E is quite a bit ahead of its mainstream counter part.

 

[Mr. May]

Will there ever be any extreme eight core versions ? of IB-E as that would destroy haswell.

 

[Martini1991]

Will there ever be any extreme eight core versions ? of IB-E as that would destroy haswell.

Doesn't seem it, IB-E seems to be 6.

 

[smelly_legend]

I'm thinking this too. Saving up

Yeah. I'm also interested in low TDP chips for the purpose of low heat, low powered, passively cooled m-ATX for the studio recordings, too.

0dB computer - yes please.

 

[Yaayuh!]

Will there ever be any extreme eight core versions ? of IB-E as that would destroy haswell.

I was hoping 8 but seems to be 6 core 12 threads again to keep the tdp down.

 

[Kaapstad]

Will there ever be any extreme eight core versions ? of IB-E as that would destroy haswell.

6 cores of SB-E or IB-E is more than enough to wollop Haswell.

 

[Yaayuh!]

6 cores of SB-E or IB-E is more than enough to wollop Haswell.

(providing multicore coding of applications is done nicely)

 

[Yaayuh!]

High-end i7 mobile chips in April : http://www.xbitlabs.com/news/mobile...l_Processors_for_Notebooks_in_April_2013.html

Considering extreme performance provided by Intel Core i7-4930MX, 4900MQ and 4800MQ central processing units and their 47W – 57W thermal design power, it is likely that they will primarily find them homes insider desktop replacement (DTR) laptops as well as all-in-one (AIO) desktops.

The first Intel Core i-series “Haswell” chips will be released in April, 2013; mainstream solutions powered by Haswell micro-architecture will be launched sometimes in Q3 2013, whereas ultra low-power Haswell products are projected to reach the market in Q4 2013.

 

[~Mike~]

What's the latest news on Haswell?

 

[topbanana]

What's the latest news on Haswell?

It's all been called off

 

[Nelly]

What's the latest news on Haswell?

June 2013.

 

[WingZero30]

Haswell vs Ivybridge benchmark results. Not sure how true it is though

http://www.xbitlabs.com/news/cpu/di...tel_Core_i_Haswell_Processor_Hit_the_Web.html

 

[~Mike~]

It's all been called off

Thanks for your helpful post.

Haswell vs Ivybridge benchmark results. Not sure how true it is though

http://www.xbitlabs.com/news/cpu/di...tel_Core_i_Haswell_Processor_Hit_the_Web.html

Not sure what to make of that.

 

[CAT-THE-FIFTH]

Haswell and Broadwell are all about saving more power for laptops,and improving the IGP even further. By reducing the TDP and power consumption of the CPU even further,more of this is made available for the IGP.

 

[HighlandeR]

High-end i7 mobile chips in April : http://www.xbitlabs.com/news/mobile...l_Processors_for_Notebooks_in_April_2013.html

thanks interesting details but conflicting, that Xbitlabs info from december says
"the first chips that Intel Corp. will release on its base will be high-end offerings for desktops and laptops in April"

But going by the general info it seems june is the new delayed release date ?

 

[Akagi]

Looks like my SB system will soldier on through this generation of chips too

 

[wazza300]

cant really tell from engineering samples,they could add more cache or higher stock frequencies ect yet

one thing would be nice to see is the temps

 

[Akagi]

8 Pack just wants to know how many gigajiggles he can get out of them

 

[Martini1991]

As it appears based on the findings of the enthusiasts, Haswell is slightly slower than Ivy Bridge on the same clock-speed in single-threaded SuperPi 1M, but is noticeably faster (11 minutes 27.505 seconds) compared to its predecessor (11 minutes 49.094 seconds) in long SuperPi 32M.

Unless I'm blind, I can't see what it's referring to.

Nor do I think Intel are going to push out an inconsistent CPU, but if they do, Haswell bar the IGP will receive a rather luke warm reception.

I'd be surprised if it wasn't a ~10% consistent improvement over Ivy in CPU performance, but who knows.