Just had a look Noel, I couldn't find out much about them. Couldn't find any comparisons.

 

It may be that the units aren't as efficient as we might presume.

 

Must be cooler than air though I would have thought, or there's not much point in them.

 

It's an external water cooling solution. The rad, reservoir and LCD display mount on top of the case.

I'm playing devil's advocate to a degree hear. But perhaps, 4.7 would have been achievable for Ivy E with old fashioned solder or even Aa thin layer of liquid pro TIM

 

You do it so well ;o)

 

I think what they reported in the linked article about IB-E is pretty much what one might expect.  For starters, and I don't think you mentioned this, from the article, 'All tests were done w/ HT on...' or something very close to that comment.  Moreover, the temp reported was, '55C to 65C' in each or all cores, not '65C'.   With HT on w/ that kind of voltage my guess is that whatever material they are using is working, for all practical purposes as far as it relates to my needs, good 'nuf.  For starters, FSX in non-HT mode never generates that kind of temps in my SB-E.  It's right at 58C when the room is warm as it is during the summer, around 78C.  I don't know how many chips they tested, but they did say 4.5Ghz was easy to do.  Since I see no practical difference between 4.4 and 4.5, or ever 4.3 and 4.5, I think this chip is clearly an ok replacement candidate for my SB-E if and when it dies.   There is ample horsepower in both of these workhorses--now all we need is a better simulator which can better utilize all the power these things offer.  Put another way:  to risk prying the IHS off just to get another what, .2 or .3 Ghz is not something I would ever bother with.  If you really get off on high clockspeeds get a blithering liquid nitrogen solution or something else and get on with it!  This 'much ado about nothing' on a few tenths of a gigahertz is just that IMNSHO ;o)

Noel, fear not.

 

I'm working on a quantum chip in my garage. Alien tech.

 

When I've back engineered it and made it compatible with puny human technology I'll post to off to you.

 

It will even handle crummy archaic FSX.

You might find this interesting...

 

The sample I have here was able to reach 4444MHz, but to do so required much more voltage than I would like at 1.42v set in the BIOS while using High LLC. This required a switch from the Corsair H100 to a full on water cooling system to manage the thermal load put out by six cores under load. On Intel's 22nm architecture that's going to be a bit high for daily use. However it is what is needed to keep the processor running at that speed. Memory overclocking seems to be on par with the 3770K running my test system memory at over 2400MHz with some tweaking to the timings and voltages needed to run the number.

 

 

Overclocking on the Core i7 4960X was bit of a disappointment after pushing to over 4.7GHz on the Core i7 3960X Sandy Bridge Extreme with a core clock of only 4.44GHz possible on this chip fully Prime 95 stable. At 4.44GHz I had to push 1.42v through it to gain stability. A push to 4500MHz needed another boost in vcore that really was too high for daily use on Intel's 22nm architecture long term. It could be just my sample and is hopefully not the norm or average for the 4960X.

They had HT on of course.

 

http://www.overclockersclub.com/reviews/intel_core_i7_4960x/3.htm

 

 

At 4.44GHz I had to push 1.42v through it to gain stability. A push to 4500MHz needed another boost in vcore that really was too high for daily use on Intel's 22nm architecture long term. It could be just my sample and is hopefully not the norm or average for the 4960X.

 

Does Intel guarantee their chips under their insurance program for using that kind of voltage?  1.42v sounds like an early demise to me.  It does tempt me to pick up another SB-E while they last--this guy works good enough for FSX and doesn't need this kind of voltage.

They don't really say. Nothing in my 3770K insurance anyway. Nothing about voltage.

 

I guess most people would fib if Intel asked them what voltage they had used.

 

To be honest, I've not heard about anyone claiming, don't know how easy it is, whether Intel get awkward about it.