I5-750 Overclocking and Foxconn socket

[knuppel]

http://www.anandtech.com/mb/showdoc.aspx?i=3661

I was recently convinced to change my planned amd budget gaming rig to an i5-750 build. Part of this build being powerful enough was the ability to overclock. I was ready to pull the trigger on this build until I read this article.

If it seems like there is a decent chance it might give me trouble I will look more into an i7-920 build.

Can I get a headcount of people using an OC'd i5-750 at around 4ghz and has anyone had trouble as described in the article?

Can you also tell me which motherboard you are using?

 

[cory1234]

Using a gigabyte p55m-4. 3.8@1.26v. 100% stable, 59c @ 100% load.

Not using a high end heatsink tho: 212+ hyper.

I hit 4.2 ghz, but the temps were at 70c on load. This card is very overclock friendly. I hit 4.2 on my first try and never had the computer crash..just prime failed once or twice.

 

[knuppel]

Thanks for the info man. Anyone else want to help me collect info on this problem?

 

[cory1234]

Intel purposely designed the i5 to NOT be overclocker friendly, at least high overclocks.

If you plan to overclock past 3.6, I would surely go with the s1366 or AM3.

I have a feeling those that have not had problems yet, will. And your going to be $#%^ out of luck trying to rma your burnt chip or mobo.

Where is your proof of this claim?? I read the article about chips getting fried on more that 1.45 volts past 4.2ghz. I haven't seen anyone frying their chip below 1.4.

 

[cory1234]

What test did you read? Because in Tom's test a board fried at 1.36v and another fried at 1.4v multiple times and took out the cpu.

http://www.tomshardware.com/reviews/budget-p55-motherboard,2436-15.html

though a certain Core i7 motherboard failed repeatedly from high VRM stress and eventually killed a CPU. But these are the new, more efficient motherboards, so we certainly should not see that mess again…right? Unfortunately, the same brand of board that failed three times before failed again, was replaced, and failed a second time at the same transistor while using a “gentler” 1.40V. But this time ASRock wasn’t alone.

The third try is supposed to be charmed, but it wasn’t for MSI. The P55-CD53 blew out one power phase at a mere 1.36V under full CPU load.

http://www.anandtech.com/mb/showdoc.aspx?i=3661

anandtech:

At first glance, one might be inclined to think LGA-1156 based processors are intolerant of high-end overclocking, almost as if by design. This is correct to some extent; a quick glance at Intel’s white papers for socket 1156 CPU’s reveals that there are around 175 pads for VCC compared to over 250 for socket 1366 CPU’s. This means socket 1156 has around 66% of the current capacity of socket 1366, the caveat being that when overclocked, processors from both platforms draw similar levels of current.

Unfortunately when overclocked, the s1156 and s1366 cpu's have the same current draw, although the s1156 only has 66% of the current capacity of the s1366.

If I'm not mistaken the anandtech article said only in EXTREME overclocking scenarios above 4.3ghz.
Did you read that article on tomshardware? They booted the MSI board @1.44v.
So all those boards on tomshardware failed at ~1.45v which is in the 4.2ghz+ range.
Like i said before...as long as your under 1.4 volts you should be fine..or keep it under 1.3 for safety. (you only need 1.2v to hit 3.6ghz)

I have not read any actual people claiming their i5 750 was fried and I've seen people hit over 5ghz on mid range motherboards.

 

[cory1234]

Dude apparently you have selective reading comprehension.

The MSI blew at 1.36v in Tom's article. Let me quote it again for you since you seem to be a little slow.

The third try is supposed to be charmed, but it wasn’t for MSI. The P55-CD53 blew out one power phase at a mere 1.36V under full CPU load.

The Asrock failed 4 times at a mere 1.4v in Tom's hardware article.

Unfortunately, the same brand of board that failed three times before failed again, was replaced, and failed a second time at the same transistor while using a “gentler” 1.40V. But this time ASRock wasn’t alone.

I quoted from Tom's article. What fantasy land are you in?

since a 0.371V offset voltage was required to get the system to boot at 1.44V

 

[notty22]

Intel purposely designed the i5 to NOT be overclocker friendly, at least high overclocks.

If you plan to overclock past 3.6, I would surely go with the s1366 or AM3.

I have a feeling those that have not had problems yet, will. And your going to be $#%^ out of luck trying to rma your burnt chip or mobo.

Your taking actual facts, that some boards have failed at a given voltage, and making a false presumption. These failures are a negative anomaly that is being addressed on those certain foxcon sockets. Meaning they are failing not because of a Intel Plan, or lack of engineering planned or unplanned tolerances. It is obviously something to be aware of, for now. Are there any known boards that will definitely NOT have this potenital socket.

 

[notty22]

Actually I have excellent reading comprehension. You realize your quoting a articl that said it MIGHT BE INCLINED, a Asuumption then backed by actual egineering data that MAY contribute to the problem. There is no quote from a Intel engineer, or any engineer. This is just a conclusion based on engineering data. Its been widely touted that the x58 boards are server level class hardware. there is a difference between server hardware specs and micro personal computer hardware . Its also why they mentioned its a combination of load pressure and the socket. Where if both specs are not perfect, this problem arises. Out of spec, out of spec, compounds to create a problem. Its why some are failing and others don't. This problem will be corrected. The server level boards are over engineered for extremes. I know the hobbyist always ignores these facts. WD has different lines of HD's etc. If everything is as it should be, those boards/sockets won't fail.

edit: we should all be grateful for this problem to be identified, it is DEFINITELY a problem that needs to be adressed and a buyer right now should consider his odds of the product he may acquire right now. Regardless of his intended usage now or down the road. Maybe someone in the household builds this computer , and 3 years from now his son decides to O/C and like lightning finding the golfer, havoc arises.

 

[endorphines]

zipzoomflyhigh is just being a dick, he doesn't realize that you can argue without insulting, selective behavior comprehension i think.
If i remember correctly that was the board's fault that a fault of the i5/i7 platform, and something that the manufactures are addressing, and a subsiquent pcb revisions / bios releases will implement changes.
those problems are all a while ago now, my thought is that if you do a bios flash to the newest revision, which you should anyways, and you're not goin nuts with your overclocking, which is always risky anyways, that you're not going to run into issues.

 

[weilin]

Are there any known boards that will definitely NOT have this potenital socket.

Yes there is actually, I believe DFI and EVGA have both annouced that they have stopped using Foxconn sockets. A quick glance at newegg pictures show EVGA uses a LOTES socket. However, DFI boards (at least in the pictures) still have foxconn. I would wait a week or two for newegg to turn over inventory and then get a EVGA or DFI board. They seem the safest bet as of right now.

 

[knuppel]

Good tip weilin, thanks.

 

[eleventh]

Yes there is actually, I believe DFI and EVGA have both annouced that they have stopped using Foxconn sockets. A quick glance at newegg pictures show EVGA uses a LOTES socket. However, DFI boards (at least in the pictures) still have foxconn. I would wait a week or two for newegg to turn over inventory and then get a EVGA or DFI board. They seem the safest bet as of right now.

I bought my P55 SLI from Newegg for this very reason. The socket? Foxconn.

 

[RJR]

You may want to have a look at where Anand got there information, XS forum thread called: UD6 Burn (or something like that, can't miss it) even has the same pictures as the Anand article and a lot more info.

As for TH's review, anyone who is halfway serious about there OCing would at least get a middle of the road board and then they won't have any problems unless they REALLY push it. I've taken my UD4P with an i7 860 to 4.6 GHz and run it 24/7 at 4.0 GHz without a problem with the foxconn socket. After reading through the XS post I even removed my processor to make sure it was OK and it was.