• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Ivy Bridge E to come in Q3 2013

O

oneofusjustin

Member
"Sandy Bridge E 3970 to live another year


Sandy Bridge E was launched in Q4 2011 and for Intel, it looks like the enthusiast platform should last at least two years. This is definitely a slowdown from a yearly update cycle that was in effect until Ivy Bridge E, but Intel probably has a good reason for it.

Back in the fourth quarter of 2011 Intel launched the Core i7 3960X, an Extreme Edition part with six cores 3.3GHz clock speed and 3.9GHz Turbo, 15MB of cache and multithreading. The 3970 a faster version with 3.5GHz clock is schedule for launch in Q4 2012 and will extend the life of a platform for a good year.

Ivy Bridge E, a successor of this enthusiast platform is now scheduled for Q3 2013, roughly a year from today with a good chance of launching by September 2013, IDF event or around it. We also learned that Ivy Bridge E fits existing X79 motherboards and will continue to use LGA 2011 socket making it potentially an easy swap for enthusiasts and overclockers in 2013.

Many had hoped to get 22nm Ivy Bridge high performance processors in time for the 2012 holiday season, but the current plan is to offer a slight clock increase of existing Sandy Bridge E and call it Core i7 3970 and let the enthusiast wait until Q3 2013 for the new Ivy Bridge E flagship. " http://fudzilla.com/home/item/28192-ivy-bridge-e-to-come-in-q3-2013




can someone explain this to me like im a 5 year old? im confused what about haswell? does this mean we have to wait a year for an ivy revision the fix the heat issue?
 
What heat issue?

Ivy Bridge decreased die size by a significantly larger amount than it decreased power consumption. It produces less heat than Sandy Bridge, but that lower amount of heat is confined to a much smaller area. Ivy Bridge doesn't have a "heat issue."

Also, what about Haswell? It will be released on LGA 1150, just as it was always planned to.
 
clearly you haven't opened up an ivybridge and re applied paste like some of us have and seen huge gains in cooling.....

so intel is going to be releasing haswell q1-2 on a new socket then go back to an old socket in Q3 *scratches head*
 
There in no power consumption issue with IvyBridge, but there is a temperature issue, that is it operates at higher temperatures than SandyBidge.
 
oneofusjustin said:
clearly you haven't opened up an ivybridge and re applied paste like some of us have and seen huge gains in cooling.....

so intel is going to be releasing haswell q1-2 on a new socket then go back to an old socket in Q3 *scratches head*
Click to expand...

And IB runs fine. The amount of temperature reduction can be anything from 0 for a core to 10C. But it runs fine, at stock and at good OCs.

Haswell-DT gets released on LGA1150. And Haswell-E/EN/EP on their new socket. People buying LGA2011 and successor should know they buy a workstation/server platform.

Rvenger said:
So Haswell-E is out in 2 years now? wow
Click to expand...

Yep 2014. Thats also why it will use DDR4.
 
Last edited:
I am willing to bet that the heat issue is fixed with Ivy-E. Imagine a 6 core Ivy @ 5ghz? And Ivy has heat issues by design, not by the nature of the smaller cores being condensed. when has a die shrink ever produced heat issues until now? This is a garbage excuse and people believe it. Learn to embrace a little conspiracy people.
 
There are no heat issues. Ivy Bridge cores are warmer than Sandy Bridge: why does that matter? I've never heard of anyone's stock Ivy Bridge hitting 105 C and throttling. Even with a stock cooler, 4.4 GHz shouldn't be too difficult to achieve. You'd probably get to pushing 85-95 C, but that doesn't really matter.

Also, Prescott ran hotter than Northwood. It was more than a die shrink, but it was still the same architecture.
 
Last edited:
dangerman1337 said:
I wonder why Intel are going foward with Ivy-E instead of Haswell-E since it's a near two year wait.
Click to expand...

That would totally kill buyer confidence if they did that. Who wants to invest in a big, special enthusiast class setup only to have it totally thrashed by a new one a year or two later?
 
moonbogg said:
That would totally kill buyer confidence if they did that. Who wants to invest in a big, special enthusiast class setup only to have it totally thrashed by a new one a year or two later?
Click to expand...

If its anything like how the 980x stacked up vs a 4 core sandy it will probably get thrashed by a 4c haswell.
 
Maximilian said:
If its anything like how the 980x stacked up vs a 4 core sandy it will probably get thrashed by a 4c haswell.
Click to expand...

A Core i7-2600K has nothing on a Core i7-980X. The 2600K may perform better in games, but that's only because the 980X isn't properly utilized. As for overclocking potential, the 980X could actually hit the 4.2-4.5 GHz range relatively frequently. The end result? Each Westmere core may only be 87% as fast as an equally clocked Sandy Bridge core, but the Westmere CPU has 50% more cores.
 
pantsaregood said:
A Core i7-2600K has nothing on a Core i7-980X. The 2600K may perform better in games, but that's only because the 980X isn't properly utilized. As for overclocking potential, the 980X could actually hit the 4.2-4.5 GHz range relatively frequently. The end result? Each Westmere core may only be 87% as fast as an equally clocked Sandy Bridge core, but the Westmere CPU has 50% more cores.
Click to expand...

Agreed. Imagine an 8-core IB-E up against a 4C Haswell. The wild-card will be OC ability of each as well. Both will be 22nm, and it will be interesting to see what will clock higher, reliably.
 
Haswell will use the same process as Ivy Bridge, though it will be a bit more mature. If you look at the history of Penryn/Nehalem and Westmere/Sandy Bridge, then you can assume the tock will scale the same or slightly better than the previous tick. As it is now, Ivy Bridge can probably hit 5.0 GHz pretty consistently at "safe" voltages, it just happens to be thermally limited. The process will be more mature by the release of Haswell, so we may see units regularly hitting 5.0 GHz.

That is, of course, unless Haswell introduces some drastic change to the architecture that impacts its scalability. P6 -> Core -> Nehalem -> Sandy Bridge have scaled pretty predictably.
 
oneofusjustin said:
clearly you haven't opened up an ivybridge and re applied paste like some of us have and seen huge gains in cooling.....

so intel is going to be releasing haswell q1-2 on a new socket then go back to an old socket in Q3 *scratches head*
Click to expand...

If heat issues are due to that, they don't need a new processor to fix the issues, they just need to change the final package manufacturing stageso a new CPU release wouldn't make any difference since they could just end up applying paste as they are now.
 
Well, whether this is true or not, I do hope they go ahead and come out with it soon, or at least before not too long as to make it obsolete already.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top