Post your New Ivy Bridge Processor Thermals, Type of Cooling, and First Impressions:

RamonGTP said:
In this case:

IB runs hotter (+1 for SB)
IB is consuming less power (+1 for IB)
IB is producing less overall heat (+1 for IB)

Thats 2/3 for IB, time to think outside of your little 216 mm² box.
Click to expand...

Also add:

Higher IPC (+1 for IB)
PCIe 3.0 (+1 for IB)
HD4000 IGP (+1 for IB)

Even if you doubt the benefit of the IGP for graphics, it is much faster for Quicksync, which enthusiasts use.
 
Still waiting on a few odds and ends to arrive next week and then I'll get everything installed.

jpbihx.jpg
 
Well, i know your're not really supposed to be calling people out specifically but I wish we could actually have a decent thread on IB without some idiot like shuttleluv feeling the need to just troll . . . troll . . . troll . . .

So just in case you missed it, the title of this thread is "Post your New Ivy Bridge Processor Thermals, Type of Cooling, and First Impressions:"

If you want to slam IB or debate it or whatever else, go somewhere else. If you decide you want to "Post your New Ivy Bridge Processor Thermals, Type of Cooling, and First Impressions:" then by all means stay.
 
Hm, running into a strange issue with my 3770K. Right now stable at 1.35v for 4.85 GHz but if I try and bump it up to 4.9 GHz or more, Win 7 will boot up but fails to display the login screen no matter how much more voltage I give the CPU. Temps are pretty decent at these settings, low 60's in Prime95. Seems like the Windows 7 login screen is creating a wall for my CPU at the moment for some reason. System in sig, any thoughts?
 
ShuttleLuv said:
That list is lame. Fire up Realtemp and it still shows 80c+ on water cooling half the time, which is pretty pathetic imo.
Click to expand...

Since you're repeating the same argument I responded to, would you like me to do a recount? I gaurantee you the pros and cons aren't going to swing the other way, no matter how many times you want to bring up the same argument.
 
Vega said:
Hm, running into a strange issue with my 3770K. Right now stable at 1.35v for 4.85 GHz but if I try and bump it up to 4.9 GHz or more, Win 7 will boot up but fails to display the login screen no matter how much more voltage I give the CPU. Temps are pretty decent at these settings, low 60's in Prime95. Seems like the Windows 7 login screen is creating a wall for my CPU at the moment for some reason. System in sig, any thoughts?
Click to expand...

Not really strange to hit an OC limit, I doubt winodws 7 is creating the wall. you can try bumping your PLL voltage.
 
RamonGTP said:
Not really strange to hit an OC limit, I doubt winodws 7 is creating the wall. you can try bumping your PLL voltage.
Click to expand...

Could be a Windows Power Control feature that is not working out and causing BSOD.

Not sure.
 
What termal compound are you guys using?

My idle temps are sitting around mid 30s while load temps(running prime) are high50s to low60s. All stock .
 
Last edited:
swatX said:
What termal compound are you guys using?

My idle temps are sitting around mid 30s while load temps(running prime) are high50s to low60s. All stock .
Click to expand...

What cooler and how hot is your room?

Here's my idle temps:
lammotz6oem.png

Ambient temperature was pretty low at floor level (core 1 and 2 even dropped down to 17c) usually those are ~20c. When doing encoding with latest x264 encoder, load temps are around 45c. I have Thermalright True Spirit cooler and use AC MX-3 (blob method) as TIM.
 
Tup3x said:
What cooler and how hot is your room?

Here's my idle temps:
snip
Ambient temperature was pretty low at floor level (core 1 and 2 even dropped down to 17c) usually those are ~20c. When doing encoding with latest x264 encoder, load temps are around 45c. I have Thermalright True Spirit cooler and use AC MX-3 (blob method) as TIM.
Click to expand...

I have the same cooler haha. TS140. Except i used the paste that came with it. I will try to find MX3 and see if that makes any diffrence. My computer is on my desk. Can't give you a exact # for my room since i dont know but it's chilly.

what case do you have?
 
swatX said:
I have the same cooler haha. TS140. Except i used the paste that came with it. I will try to find MX3 and see if that makes any diffrence. My computer is on my desk. Can't give you a exact # for my room since i dont know but it's chilly.

what case do you have?
Click to expand...

The one in my sig, Antec 300. I don't use the stock fan, though. I have 12cm fan version of True Spirit.
 
I made quick test, pumped turbo to 43 voltages at auto just to see how much voltage it gives under 100% load (again, x264 encoding):
//edit: forget this picture
As far as I know, those are pretty excessive volts so I probably should use negative offset.

EDIT:
Just tested something more reasonable voltages. At the moment, 1.152v under full load and temperatures between cores are around 55c when doing encoding (I encode 1080p FRAPS recoding, takes about 10 minutes). Has been stable with these voltages so far - doing another run before moving to 3DMark11. This is definitely cooler chip than my old 2600K, maybe it was crap.
 
Last edited:
This is the least amount of voltage I can use to get 4.6ghz. Not to happy with the temps. Really didn't want to go over 75c on this thing. I'm running a H50 on it. Not the best setup right now the H50 is mounted inside of my rack monted case. But I'll be going to a custom water loop here soon. Might bump it down to 4.4 till then.


ivybridge2.jpg
 
Vega said:
Hm, running into a strange issue with my 3770K. Right now stable at 1.35v for 4.85 GHz but if I try and bump it up to 4.9 GHz or more, Win 7 will boot up but fails to display the login screen no matter how much more voltage I give the CPU. Temps are pretty decent at these settings, low 60's in Prime95. Seems like the Windows 7 login screen is creating a wall for my CPU at the moment for some reason. System in sig, any thoughts?
Click to expand...

Hi Vega. How did you get 4.85? Are you increasing BCLK also? I am in a similar situation where I have hit my "wall" with my Multi at 4.7GHz and 1.25vcore. 4.8GHz is not IBT stable with a vcore all the way up to 1.36. Would it be possible for me to increase the BCLK slightly to get a better overclock? My temps are very well in check for what I normally do (gaming) the cores never pass 55 degrees C. Burn test gets the hottest to 77. What would your recommendation be on increasing BCLK? Does this affect my GTX680? I am aware that BCLK affects the PCIE as well. Are there any voltages that are specifically tied to this setting that would need to be increased? Any help would be appreciated. I have an Asrock extreme4 and a 3570k.
 
chronicfx said:
Hi Vega. How did you get 4.85? Are you increasing BCLK also? I am in a similar situation where I have hit my "wall" with my Multi at 4.7GHz and 1.25vcore. 4.8GHz is not IBT stable with a vcore all the way up to 1.36. Would it be possible for me to increase the BCLK slightly to get a better overclock? My temps are very well in check for what I normally do (gaming) the cores never pass 55 degrees C. Burn test gets the hottest to 77. What would your recommendation be on increasing BCLK? Does this affect my GTX680? I am aware that BCLK affects the PCIE as well. Are there any voltages that are specifically tied to this setting that would need to be increased? Any help would be appreciated. I have an Asrock extreme4 and a 3570k.
Click to expand...


BCLK will give you that '4.85' for sure if you increase it a wee bit, but honestly, increasing BCLK even a little is more for a Dry Ice / LN2 run than it is for air 24/7. BCLK is harder on the memory controller so it will be a tad harder to stabilize than without it on air and frankly, Ivy Bridge has more than enough memory speeds to play with so thats maybe another reason not to touch BCLK on Ivy.
 
ShuttleLuv said:
BCLK will give you that '4.85' for sure if you increase it a wee bit, but honestly, increasing BCLK even a little is more for a Dry Ice / LN2 run than it is for air 24/7. BCLK is harder on the memory controller so it will be a tad harder to stabilize than without it on air and frankly, Ivy Bridge has more than enough memory speeds to play with so thats maybe another reason not to touch BCLK on Ivy.
Click to expand...

Thanks Shuttle. If you feel that it is hard on the motherboard than maybe it isn't such a great idea to acheive 100-200MHz like that with air cooling. I did try it once just for kicks yesterday at 105 and of course it didn't boot. But I just thought there was a voltage I was missing that needed to be increased somewhere. I am coming from a Q9550 which I had an original C1 stepping overclocked to 3.8 and I remember playing with a whole lot of diff voltages back then (2008 ish). This is totally different now.
 
chronicfx said:
Thanks Shuttle. If you feel that it is hard on the motherboard than maybe it isn't such a great idea to acheive 100-200MHz like that with air cooling. I did try it once just for kicks yesterday at 105 and of course it didn't boot. But I just thought there was a voltage I was missing that needed to be increased somewhere. I am coming from a Q9550 which I had an original C1 stepping overclocked to 3.8 and I remember playing with a whole lot of diff voltages back then (2008 ish). This is totally different now.
Click to expand...

Yeah you have to tweak the VCCSA / VCCIO among others when doing the BCLK route it can be a tricky thing. The Q9550 was a trickier thing to overclock with the GTL's and such so I'd compare those days more to Sandy Bridge E. Sandy Bridge Non E and Ivy Bridge are relatively more simple in overclocking terms with basic multiplier overclocking and maybe a few others unless you really push it.
 
I am finding that prime95 just isn't stressing these chips enough to give a good indication of stability.
And my previous Intel Burn Tests were without SP1, so no updated AVX instructions.

This is what I am seeing at 1.25v:

Prime95 max temps: ~70C
Intel Burn Test w/o avx: ~80C
Intel Burn Test w/ avx: ~90C
 
deltronik said:
I am finding that prime95 just isn't stressing these chips enough to give a good indication of stability.
And my previous Intel Burn Tests were without SP1, so no updated AVX instructions.

This is what I am seeing at 1.25v:

Prime95 max temps: ~70C
Intel Burn Test w/o avx: ~80C
Intel Burn Test w/ avx: ~90C
Click to expand...

Yes AVX will charbroil your cpu no doubt about it, but what software really uses AVX? Think about that for a minute. Even OCCT the stress program leaves it unchecked by default and says it's not as good otherwise.
 
Higher temps doesn't necessarily mean the program is better at testing stability.
 
RamonGTP said:
Higher temps doesn't necessarily mean the program is better at testing stability.
Click to expand...

Ok, let me rephrase that. My computer is verifiably less stable at the same core speed and voltage using IBT w/ avx.

The same seems to be true for Prime95 vs IBT. My computer is still burning away at p95 while I sit at work. It will be around 20 hours when I get home, we'll see, but I bet it's still running stable even though IBT crashes.
 
Makeroflostsouls said:
This is the least amount of voltage I can use to get 4.6ghz. Not to happy with the temps. Really didn't want to go over 75c on this thing. I'm running a H50 on it. Not the best setup right now the H50 is mounted inside of my rack monted case. But I'll be going to a custom water loop here soon. Might bump it down to 4.4 till then.
Click to expand...

I wouldn't be too worried about 77C - remember that Tjunction is a little higher for Ivy, so that's the equivalent to 70C with Sandy (not that they are directly comparable, but you get the idea). You won't see anything close to that temp in games.

chronicfx said:
Thanks Shuttle. If you feel that it is hard on the motherboard than maybe it isn't such a great idea to acheive 100-200MHz like that with air cooling. I did try it once just for kicks yesterday at 105 and of course it didn't boot. But I just thought there was a voltage I was missing that needed to be increased somewhere. I am coming from a Q9550 which I had an original C1 stepping overclocked to 3.8 and I remember playing with a whole lot of diff voltages back then (2008 ish). This is totally different now.
Click to expand...

I agree that BCLK adjustments just aren't worth it. That one or three percent isn't worth the increased stability issues (considering it impacts more components).
 
The only stability I really care is real life usage scenarios, like encoding video and gaming. If Prime fails in less than hour, then something is clearly wrong but that's about it in my opinion. After all, some times your system can run hour and hours stress test but then just fails in real life usage cases.
 
Tup3x said:
The only stability I really care is real life usage scenarios, like encoding video and gaming. If Prime fails in less than hour, then something is clearly wrong but that's about it in my opinion. After all, some times your system can run hour and hours stress test but then just fails in real life usage cases.
Click to expand...

In my experience, if that's the case, then there are other problems with the system causing the instabilities rather than the processor itself.
 
Tup3x said:
The only stability I really care is real life usage scenarios, like encoding video and gaming. If Prime fails in less than hour, then something is clearly wrong but that's about it in my opinion. After all, some times your system can run hour and hours stress test but then just fails in real life usage cases.
Click to expand...

In my experience, if the system isn't prime stable, you get other weird things going on that may not be "game-breaking" but annoying nonetheless. Things like the odd Internet explorer tab not responding and needing to be reloaded or windows hanging on the splash screen on the first boot after a cold boot. I'll gladly drop 100Mhz or raise the voltage if I can do so safely to avoid that.
 
For me if it's not 24 hours for Prime 95 stable it's not stable.. I want to be able to use it like the CPU came stock at whatever speed I have it at. Install windows OC, Fold , game and do whatever and not worrie about it crashing. I'm at 9+ hours now on my 4.4ghz OC and counting. Playing WOW, Surffing the net and other stuff. at the same time just to put some extra stress on it.
 
I'm currently running IB with a megahalems, and want to OC some. Would a DH14 be worth it? Assuming I can OC my CPU enough to get a reasonable amount of heat, that is :)
 
Makeroflostsouls said:
For me if it's not 24 hours for Prime 95 stable it's not stable.. I want to be able to use it like the CPU came stock at whatever speed I have it at. Install windows OC, Fold , game and do whatever and not worrie about it crashing. I'm at 9+ hours now on my 4.4ghz OC and counting. Playing WOW, Surffing the net and other stuff. at the same time just to put some extra stress on it.
Click to expand...
What I'm saying is that I do not trust one program only. I have seen that some are able to run Prime or something else hours and hours yet it will crash in certain game. It gives some data but ultimately real world usage in the best stress test in my opinion.
 
rampantandroid said:
I'm currently running IB with a megahalems, and want to OC some. Would a DH14 be worth it? Assuming I can OC my CPU enough to get a reasonable amount of heat, that is :)
Click to expand...

In my case I hit my wall at 4.7GHz way before temps were a problem. Max temp I get from regular usage is about 55 degrees. I have a D-14 but I probably could have gotten away with keeping the xiggy S1283 I had from my q9550 build and acheived the same. And I have a feeling too that the heat pipe direct touch coolers will do better with these chips because people are saying its the transfer of the heat and not the amount of it.
 
I'm going to Prime the system overnight. So far @1.184V, the temps are fairly decent during the Intel Burn Test. No issues so far. I have an H100 pulling cold air from the ouside (~68-70 degrees F) in. Fans are set to 2000 RPM's.

IntelBurnTest.jpg
 
You must log in or register to reply here.
Back
Top