Yes. Non-extreme SB-E chips only OC to about 4GHz with turbo + Bclk adjustments.

 

So, the prior statement that SB-E is a good over clocker comes from reports from people who are using extreme processors? I would have imagined being a good over clocker = at least 4.6 or higher.

There are those BCLK straps that make non-extreme SB-E chips just as good OCer's as the 3960X

Even the I7 3820 with a "locked" multiplier (max 43) can (potentially) be overclocked passed 5GHz using a 1.25 gear ratio:

 

37 x 125 = 4.625GHz

There are those BCLK straps that make non-extreme SB-E chips just as good OCer's as the 3960X

Even the I7 3820 with a "locked" multiplier (max 43) can (potentially) be overclocked passed 5GHz using a 1.25 gear ratio:

 

37 x 125 = 4.625GHz

 

Has anyone done it reliably though?

Has anyone done it reliably though?

 

Yeah, there's a lot of them in every OC'ers forums.

Here's just one example

Yeah, there's a lot of them in every OC'ers forums.

Here's just one example

 

Fair enough. I know Bclk adjustments on every other platform are all but worthless, good to see x79 has real Bclk overclocking.

Fair enough. I know Bclk adjustments on every other platform are all but worthless, good to see x79 has real Bclk overclocking.

 

I believe it's not a real BCLK overclock, just some sort of combined CPU + RAM multi, but the BCLK stays at default 100MHz. Not 100% positive though, but in bechmarks it's showing no benefit over regular multi-only OC

I believe it's not a real BCLK overclock, just some sort of combined CPU + RAM multi, but the BCLK stays at default 100MHz. Not 100% positive though, but in bechmarks it's showing no benefit over regular multi-only OC

 

Ah, that makes sense. I saw the user in the post you linked describe a "Bclk multi".

I was having heat problems with my 3570K running an H70 for cooling. 4.4 GHz @ 1.25V was starting to climb into the low 80's, and showing no sign that was where it would stop. I went out and picked up an H100 and some 4x Scythe 1900 RPM fans (pushing 110 CFM). It's all installed, and after 25 minutes running Prime95, my temps are staying in the 50's. I think with propper cooling this chip could hit some ridiculous overclocks.

Interesting... I may just have to do that myself, presuming my wife and I can put up with the noise. How loud are those 1900RPM fans anyway?

 

Edit: I don't suppose you have results with just the stock fans for the H100 by any chance? Also, what version of Prime 95 are you running because if it doesn't feature AVX support you may as well not even be running it and reporting temps.

It really adds to the realism, it's like i have a jet here in the room with me! I hooked them up to a controller, and I turn them down when I'm not simming, and they get nice and quiet. I generally wear a headset when I'm flying, and don't notice the noise at all.

I should've made a new post to ask my question, so here it is.

 

Do you have temp measurements for the H70 and did you install the H100 with stock fans first to test or go straight to the 4x1900RPM Scythe fans? I'd like to see how your temps compare to mine with 2 stock H100 fans. Also, what version of Prime 95 are you running? If it's not a newer version with AVX support it's useless for temperature reporting, especially on Ivy Bridge. Also, how are you measuring temps (which program?)

 

Thanks

I don't have any logs or anything, but with the H70, @4.4 GHz w/1.25 Vcore, I was in the low 80's. I went directly to the Scythe fans with the H100, and that same OC was in the high 50's, and a moment or two with one core reaching 60-61. I actually gave away the Corsair fans that came with the H100 at Microcenter to another guy who was getting one, so that he could have a push-pull configuration (I have no need for extra fans).

 

I'm measuring my temps with coretemp, and running Prime95 v27.7, the latest version, which I downloaded yesterday. I've run both a custom blend utilizing 90% available memory to test my RAM stability, and the In Place Large FFT's, to generate maximum CPU heat.

 

You might try plugging the H100's stock fan's directly into molex, rather than the H100's controller. All the reviews I've seen say that the H100 throttles the fans pretty significantly. The specs on them say they will push 90 CFM at full power, though I've heard they're also loud.

I should've made a new post to ask my question, so here it is.

 

Do you have temp measurements for the H70 and did you install the H100 with stock fans first to test or go straight to the 4x1900RPM Scythe fans? I'd like to see how your temps compare to mine with 2 stock H100 fans. Also, what version of Prime 95 are you running? If it's not a newer version with AVX support it's useless for temperature reporting, especially on Ivy Bridge. Also, how are you measuring temps (which program?)

 

Thanks

 

Here are the non-AVX linpack libs for Linx Tech, in case you want to test without the excessive load AVX adds. You just need to download Linx and copy them to the same directory where the Linx.exe sits

I don't have any logs or anything, but with the H70, @4.4 GHz w/1.25 Vcore, I was in the low 80's. I went directly to the Scythe fans with the H100, and that same OC was in the high 50's, and a moment or two with one core reaching 60-61. I actually gave away the Corsair fans that came with the H100 at Microcenter to another guy who was getting one, so that he could have a push-pull configuration (I have no need for extra fans).

 

I'm measuring my temps with coretemp, and running Prime95 v27.7, the latest version, which I downloaded yesterday. I've run both a custom blend utilizing 90% available memory to test my RAM stability, and the In Place Large FFT's, to generate maximum CPU heat.

 

You might try plugging the H100's stock fan's directly into molex, rather than the H100's controller. All the reviews I've seen say that the H100 throttles the fans pretty significantly. The specs on them say they will push 90 CFM at full power, though I've heard they're also loud.

 

Thanks for the info and the tip! I have noticed the fans being fairly silent under normal conditions and getting quite loud under heavy load, maybe I'll hook them up to fan controller to dial them in at a reasonable medium.

I was reading on a few hardware sites including hardocp that noticed coretemp and real temp are in need of an update for IB since they are posting erroneous temps.

 

Here is the main page post from HardOCP -

 

 

"UPDATE: We have a bit of an update to this post. While breaking in a new i7-3770k system, we noticed that both Real Temp and Core Temp were giving us temperature values on our test system that were just flat out wrong. So, we reached out to ASUS and this is what the company had to say:

 

CoreTemp is in serious need of updating at this point for full compatibility with Ivy Bridge as is RealTemp (both latest releases do not include E1 stepping compatibility). Our TurboV is very good at monitoring the correct temps and voltages, but if you need third party verification we highly suggest the latest version of Aida64 Extreme (1938 beta or later) as they fully support Ivy Bridge and all current Z77 boards (not just from us). Besides working with the manufacturers, they also work directly with Intel to ensure proper readings and the stability test is actually quite good also."

 

http://www.hardocp.com/news/2012/05/18/pop_top_on_your_i73770k_for_better_temps63/

 

Just an FYI