Anandtech Review of Ivy Bridge is Up now! Other sites added

[Lonbjerg]

I'm still waiting for aan upgrade to my i7 920 @ 3.5 Ghz that makes sense.

I think I won't be happy to upgrade untill "Haswell".

 

[WhoBeDaPlaya]

And a useless "metric" to evalute a CPU on...when som many ovther factors are more important.

Admittedly, platform features (native USB 3.0, etc.) are also important (which is why I love my X58), but for the crowd you are addressing, overclockability is definitely a prime concern.

I'm still waiting for aan upgrade to my i7 920 @ 3.5 Ghz that makes sense. I think I won't be happy to upgrade untill "Haswell".

Well, 3.5GHz is a little more marginal (vs. my 4.1GHz) given that a 4.5GHz SB = ~5GHz Nehalem, but I see your point

 

[Lepton87]

Admittedly, platform features (native USB 3.0, etc.) are also important (which is why I love my X58), but for the crowd you are addressing, overclockability is definitely a prime concern.

Well, 3.5GHz is a little more marginal (vs. my 4.1GHz) given that a 4.5GHz SB = ~5GHz Nehalem, but I see your point

more like 5.3GHz Nehalem

 

[WhoBeDaPlaya]

more like 5.3GHz Nehalem

Rub it in, why don't you

 

[Lonbjerg]

Admittedly, platform features (native USB 3.0, etc.) are also important (which is why I love my X58), but for the crowd you are addressing, overclockability is definitely a prime concern.

Well, 3.5GHz is a little more marginal (vs. my 4.1GHz) given that a 4.5GHz SB = ~5GHz Nehalem, but I see your point

The only game right now that I play, that could use more CPU is ARMAII.
But only in singleplayer missions, the A.I. really hogs the CPU.

But most games run fine with my current rig...and I love my X58 too, gave me loads of fun with SAS drives in RAID before I went SSD.

 

[bryanW1995]

I'm still waiting for aan upgrade to my i7 920 @ 3.5 Ghz that makes sense.

I think I won't be happy to upgrade untill "Haswell".

2500k @ 4.5 will net you around 48% (1.15 * 4.5/3.5) improvement, plus a quieter and less power-hungry rig. I'm extremely happy with my switch, and my i7 920 was @ 3.95 instead of 3.5.

 

[pelov]

http://techreport.com/discussions.x/22859

Curious, we asked Intel about the interface between the Ivy Bridge die and the heat spreader. Intel has confirmed to TR that Ivy uses a "different package thermal technology" than Sandy Bridge. The firm stopped short of answering our questions about why the change was made and how the thermal transfer properties of the two materials compare. However, Intel claims the combination of the new interface material and Ivy's higher thermal density is responsible for the higher temperatures users are observing with overclocked CPUs.

Intel's way of saying

 

[Lonbjerg]

2500k @ 4.5 will net you around 48% (1.15 * 4.5/3.5) improvement, plus a quieter and less power-hungry rig. I'm extremely happy with my switch, and my i7 920 was @ 3.95 instead of 3.5.

It's not enough for me to build a new system.

 

[dma0991]

http://techreport.com/discussions.x/22859
Intel's way of saying

Intel sold me a gimped CPU!

 

[Lonbjerg]

Intel sold me a gimped CPU!

It dosn't live up to specs?

If it adhears to specs...cry me a river.

 

[dma0991]

It dosn't live up to specs?

If it adhears to specs...cry me a river.

It doesn't matter much to me since it has been my original plan to go with conservative clocks (4.5GHz) and I'm on water. If Intel had not gimped IB, I could probably do with 10C less heat into my room.

It's going to be a big problem for most air coolers whose ambient temps are in the range of 30C or more. Not quite sure of the motive behind it since it would potentially kill some sales among the small community of enthusiast.

 

[Lonbjerg]

It doesn't matter much to me since it has been my original plan to go with conservative clocks (4.5GHz) and I'm on water. If Intel had not gimped IB, I could probably do with 10C less heat into my room.

It's going to be a big problem for most air coolers whose ambient temps are in the range of 30C or more. Not quite sure of the motive behind it since it would potentially kill some sales among the small community of enthusiast.

Gimped?

lol

People whining over O.C. amuses me...they take out of specs operation..and make it sound like something they have a right too...instead of treating it like it really is:
Icing in the cake...spoiled posters indeed. :thumbsdown:

 

[Absolute0]

If Intel had not gimped IB, I could probably do with 10C less heat into my room.

you do realize that the core temp and the thermal energy dumped into your room are very different things?

IB core gets hotter but it generates less heat because it uses less watts.

 

[dma0991]

Gimped?

And I suppose Intel is in the right even though they made changes that would affect it in a bad way, keep quiet and pretend like nothing happened.

Why limit the temp issue for out of spec operations alone? Even within spec users would benefit from having lower temps.

you do realize that the core temp and the thermal energy dumped into your room are very different things?

IB core gets hotter but it generates less heat because it uses less watts.

Maybe I should have omitted the end. That sounds about right.

 

[Dadofamunky]

You know we are having good hardware times, when the main grief agianst a CPU boils down to overclocking...make me smile....luxury problems in the developed world indeed.

I know, quite the first-world problem, eh?

 

[BFG10K]

I saw a die shrink that was accompanied by a heat increase, I was reminded of Prescott. I was surprised that everyone acted as if this phenomenon hadn't occurred before.

What I originally said was:

This is the first time in recent memory that we’ve seen a die-shrink causing higher temperatures that couldn't be explained away with differences in coolers or from a higher TDP.

In the case of Prescott, higher temperatures were caused by the higher TDP. This is unlike IB which has a lower TDP but higher temperatures with the same cooler.

 

[ShintaiDK]

Prescott was essentially a new uarch. The pipeline went from 20 stages in Northwood to 31 stages in Prescott.

So it was not a shrink of the current uarch. Nor was 90nm flawed as we saw with the very successful Pentium-M series.

 

[Lonbjerg]

Prescott was essentially a new uarch. The pipeline went from 20 stages in Northwood to 31 stages in Prescott.

So it was not a shrink of the current uarch. Nor was 90nm flawed as we saw with the very successful Pentium-M series.

Hey, don't come here and confuse people with facts...

On another note I also find it funny that people whine over OC's...on a CPU with a retarded IGP.

Pull out that IGP, most people that OC...also have a GPU...rendering much of the CPU diespace...useless.

 

[Termie]

TechReport's investigation of the TIM issue: http://techreport.com/discussions.x/22859

 

[Concillian]

If Intel had not gimped IB, I could probably do with 10C less heat into my room.

^ does not understand the difference between heat and temperature.

Heat is energy and what gets transferred to your room. Temperature is a result from heat and the materials that create the thermal gradient. 10C temperature difference on the CPU at the same speed and voltage has no real effect on how much heat is transferred from the CPU to your room.

And I suppose Intel is in the right even though they made changes that would affect it in a bad way, keep quiet and pretend like nothing happened.

When has Intel ever said anything directly to the incredibly small portion of users who are the enthusiast / overclocking community? They are more communicative with our group than I remember them ever being. I don't think of it so much as pretending it never happened as just not mentioning it because they never mention anything to us anyway.

However, I do understand your point. I think the fact that the K series processors exist now and didn't before... and they are directly profiting by this ~5% tax for overclocking that they should have some accountability for changes that directly affect overclockability now more than at any time in the past.

From their perspective they were probably able to save some bucks by switching to TIM from indium bonding as a direct result of the processor designs that allowed them to lower the TDP to 77W. This doesn't affect any of the non-k series processors in the slightest, but does significantly affect overclockability. The presence and extra cost of the k series gives more credence to arguments against cutting overclockability in this way.

 

[Concillian]

TechReport's investigation of the TIM issue: http://techreport.com/discussions.x/22859

It appears that, all other things being equal, Ivy Bridge runs hotter. That almost certainly means the cooling is less efficient.

*The thermal gradient or thermal density (or combination thereof) is what must be worse, cooling efficiency of the system is worse, but since they're using the same cooler, they can further narrow down the problem.

What I want to see is for someone to get a HSF or waterblock indium bonded directly to the CPU. At work, we occasionally need things indium bonding and there are any number of facilities in the bay area setup for indium bonding. Bonding a waterblock to a CPU should be trivial for someone who has the knowledge and tools for bonding... however, usually being business to business, the costs would likely be pretty high for a hobbyist. Still it'd be nice to get some data for direct indium bonding of the CPU to whatever cooler just to see if this really makes the difference between IVB and SB go away.

Hmm... looks like you can get single quantity 1" squares of indium foil for under $100. Melting temp is <350F, I bet it could be bonded in a conventional oven.