How high will ivy bridge overclock

[SickBeast]

The preview did not discuss this. Have any of you seen any leaked info about this? Will 5ghz be easy for these chips?

 

[lau808]

i think the ocing is still under nda. im hoping for the golden chips to hit 5.5-5.6, and the avg chips to hit 5.0-5.2 but i have not seen anything yet.

 

[Puppies04]

From what I have seen of leaked ES results they require quite a bit less voltage to hit higher clocks but they also get hotter as the cores are more tightly packed together due to the smaller manufacturing process. Also nobody has data on how well the 3d transistors cope with higher than stock voltage and what the degradation is like.

TL : DR
All info availible atm is to be taken with a grain of salt.

 

[SickBeast]

When will the NDAs be lifted?

The higher temps don't make sense if there is considerably less voltage. I'm sure it's possible but it seems illogical to me.

 

[Puppies04]

Think about it, you have more transistors in a smaller space which means the cores will need to have more cooling.

 

[nyker96]

From what I have seen of leaked ES results they require quite a bit less voltage to hit higher clocks but they also get hotter as the cores are more tightly packed together due to the smaller manufacturing process. Also nobody has data on how well the 3d transistors cope with higher than stock voltage and what the degradation is like.

TL : DR
All info availible atm is to be taken with a grain of salt.

I also seen some initial reports of this chip ES version hitting close to 5 but gets very hot. But considering it's ES hard to say what the final product will be like, however, I have a feeling for people expecting 5+ overclocks, you need water.

 

[Puppies04]

I also seen some initial reports of this chip ES version hitting close to 5 but gets very hot. But considering it's ES hard to say what the final product will be like, however, I have a feeling for people expecting 5+ overclocks, you need water.

From what I have seen, keeping in mind there is nothing concrete out there yet I am apt to agree. Some people seem to think you are still going to be able to pump 1.5v+ through IB chips with no side effects. In reality the chip might only require 1.4v to hit around 5ghz but it is going to require some serious cooling not to hit its tjmax.

 

[SickBeast]

Think about it, you have more transistors in a smaller space which means the cores will need to have more cooling.

The heat will be more concentrated but with proper cooling it should actually make the chips run cooler overall. There will be less heat, it will just be focused on a smaller space. Perhaps Intel will engineer a special heat spreader.

 

[Puppies04]

The heat will be more concentrated but with proper cooling it should actually make the chips run cooler overall. There will be less heat, it will just be focused on a smaller space. Perhaps Intel will engineer a special heat spreader.

Why will there be less heat????

More transistors = more heat

Smaller process = more heat generated per mm2

Ok maybe intels 3d transistors will be lower leakage than their existing ones but that is in no way a given.

This is pretty much backed up from an ES test I read showing the chip stable at 4.8ghz @ 1.35v while hitting 90c+

 

[SickBeast]

Why will there be less heat????

More transistors = more heat

Smaller process = more heat generated per mm2

Ok maybe intels 3d transistors will be lower leakage than their existing ones but that is in no way a given.

This is pretty much backed up from an ES test I read showing the chip stable at 4.8ghz @ 1.35v while hitting 90c+

Anand said his sample ran cooler with lower power consumption.

When they use a new smaller process there tends to be less heat and less power. I don't think the leak you read is accurate. I don't see why overclocked it would all of a sudden run hotter than Sandy Bridge.

 

[Smartazz]

Anand said his sample ran cooler with lower power consumption.

When they use a new smaller process there tends to be less heat and less power. I don't think the leak you read is accurate. I don't see why overclocked it would all of a sudden run hotter than Sandy Bridge.

Whenever a smaller process is employed, you get less heat and power relative to the performance, not in absolute terms. Power consumption has been going up steadily with each process shrink.

 

[SickBeast]

Fair enough.

I still think that Ivy will run cooler than Sandy Bridge. It's not like they've added a lot of new functionality. It's pretty much a die shrink.

 

[rgallant]

"This is pretty much backed up from an ES test I read showing the chip stable at 4.8ghz @ 1.35v while hitting 90c+ "
- but heat sink was cool per the tester.
-on a z68 board haha

-have a look at the other one[same page] on a fan less cooler
http://www.xtremesystems.org/forums/showthread.php?279602-thirty-seven-seventy-......-K
-only hit 80c

-but even if true , I have big water
-so unless IVY-B is a dud , I'll be getting it.
-if it is a dud , I'll go the 3820\x79.
-z77 /ib offers real PCI-3.0 ,anything sb still is not rated by intel for 3.0 [yet?]
-want to see how the new PLX 3.0 chips work for sli.
I would not buy any sb board with nf200 junk on it or any sb\z68 .
-it's out dated regardless what people say[mostly sb owners]
-didn't need it last year and don't need it this year.
-time will tell.

 

[Jman13]

All you have to do is look at heatsinks nowadays to see that cooling needs continue to increase. When I built my Athlon XP system in 2001...I used one of the best coolers on the market, an all-copper block of a heatsink from Thermalright...it was massive, and it's about 1/2 the size of the CoolerMaster TX-3 I have on my current chip...and that's certainly not one of the bigger HSFs out there.

 

[SickBeast]

All you have to do is look at heatsinks nowadays to see that cooling needs continue to increase. When I built my Athlon XP system in 2001...I used one of the best coolers on the market, an all-copper block of a heatsink from Thermalright...it was massive, and it's about 1/2 the size of the CoolerMaster TX-3 I have on my current chip...and that's certainly not one of the bigger HSFs out there.

I just don't know what they're going to do soon. My heatsink extends right to the side of my case. All they could do is make it wider, but even then they're going to hit the back of the case and other components soon.

We might all be forced to use water cooling in the future. :thumbsdown::thumbsdown:

 

[jimpatrick]

why ask about overclocking capabilities then refused info given by people who posted some results?weird.

 

[SickBeast]

why ask about overclocking capabilities then refused info given by people who posted some results?weird.

I'm just saying that it doesn't make sense. It's not logical. The only concrete info we have is from Anand's preview.

How would you explain more heat when overclocked but less heat at stock speed?

I'm not going to throw logic out the window and blindly agree with speculation based on rumors.

 

[jimpatrick]

yeah thats why its either you wait for it to be released or be satisfied w/ "rumors".

 

[SickBeast]

ok.

 

[Tempered81]

April 19th for your info. I've seen 6.4 and 7.1 on LN2, 5.0 on about 4 different samples on air and water, but all talk about temps. IMO, 4.6-4.7 on air, higher with water, 5.0 for 2600/2700. High temps are because of a new type of transistor.

 

[rgallant]

just saying

-I have run my 920 mostly at 4.2 and never went past 1.325v
-sb =a lot of people run sb at 1.45+ 7\24 per their post so why didn't the [apox. 30-40-%temp spike ]per mm of die not come up then[45nm vs 32 nm]
-could it be sour grape's on something faster coming out:
-comments like
you don't need any thing faster than a 2500k [like I have]
ib is only =+ 10% IPC & +10% overclock's , who cares my 2600k is fast enough ,but they upgraded for [+20%]that when they did.

-wait till the real deal hits the benches
-checked a while ago but very few sb ES could hit any high clock's[sub 4.8 w\ high v] Google your self if you want.

 

[SickBeast]

April 19th?

And when does Haswell come out?

 

[BD231]

My bets are on Ivy brdige not being that much more overclockable than sandy due to thermal limitations. If any of the high temp rumors are true this could definitely be a let down for people seeking 5ghz.

Not sure if that will stop me from buying when it comes out though, it still has some awful tempting power figures for the performance you get. Adding performance to what I have now while cutting power consumption in half is definitely enticing.

 

[Absolute0]

Ya doesn't look like IB will clock much further than SB on air/water, prolly scales better with cold, will break all WR's and bench well but for 24/7 use doesn't look like it will get much past 5 Ghz.

So people hoping to run 5.3+ Ghz on air... not happening! The best SB will do 5 Ghz on air, I can see IB going about 100-150 Mhz further. I don't see it exceeding 5.2 Ghz on water so that's not much difference. Maybe the very best IB will prime at 5.3 on water but at the very least they will provide more consistent 5 Ghz chips with less volts. Best SB chips can break 5.1 Ghz and the very best can prime at 5.2. I can prime at 5.1 if I take the side of my case off to let my radiator breath a bit better lol.

 

[Hulk]

So if I'm reading this correctly, compared to Sandy Bridge, Ivy Bridge will have less heat rate (Watts) but more heat flux (W/m^2) right? This would seem to indicate that good contact (lapping) and high thermal conductivity (Cu cooler or at least the CPU contact area) for optimal cooling/overclocking. And/or water to increase heat transfer.