Socket burn, better to go with s1366 vs s1156?

[XJ.]

Well, researching motherboards lately I'm running into reviews where the reviews or community member have had problems with Foxconn LGA 1156 socket contacts burning up and ruining motherboards and CPUs. Of course, 90% of the S1156 boards use Foxconn sockets! Some recent high end boards from Gigabyte (LOTOS sockets) and EVGA (TOYO sockets) - but some of these boards are really expensive ($350 ).

In light of these teething problems for LGA 1156 boards - I'm wondering if I'd just be better of with a tried and true LGA 1366 board (like Asus' P6T v2). It will still cost me more this way, but at least I'd have peace of mind.

Comments and suggestions welcome!
Thx!
-xj

[CrowdAds]

[GWillakers]

XJ,

I like the Bloomfield better myself. I like the additional PCI-e lanes, and the triple channel memory. But if you don't want to OC, the Lynnfield is a good inexpensive alternative. I thought some of those reviews were most unwarranted, being that they were extremely overclocked.

I've quoted my feellings with regard to this matter...

Originally Posted by GWillakers

Every once in a while, One comes upon an article that hits ya right there.
The following is one such article by Rajinder Gill published on Anandtech.

P55 Extreme Overclockers check your sockets
AnandTech: P55 Extreme Overclockers: Check your sockets!

Briefly it states that a Lynnfield solution(LGA1156) with it's reduced pin count is less equipped to handle the increased current demands imposed by "extreme" overclocking scenarios than it's Bloomfield(LGA1366) counterpart.

"we have damaged every motherboard in our possession for the P55 overclocking (extreme) shootout as well as two very expensive i7/870 processors."

The article has merit, in that it identifies a potential weakness with some of these boards. However the tone of the article seems to point fingers of blame where they shouldn't. Namely an innocent manufacturer, Foxconn, its client MB manufacturers, and Intel itself.

"Foxconn's been able to fill the void in the market with what we believe to be a lower quality alternative"

The article implies that Foxconn's socket seems to be inferior to those produced by other vendors. In evidence, Mr. Gill sites the failure of his limited sample of overclocked motherboards bearing the Foxconn socket and he presents pictures of processors extracted from Foxconn sockets which show some pads without contact dimples.

Please note that little to no evidence of failure has been presented to date of Foxconn sockets that have been run within specifications.

Next the article seems to imply that motherboard manufacturers have something to account for when they utilize these fully functional sockets:

"(we) will provide an update once all of the motherboard manufacturers affected have had a chance to properly respond."

Examine closely his disparaging statement towards Intel's new product line:

"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;"

Specifications are set based on the properties of the materials employed. It is up to the manufacturer to identify these and set these down on paper. Overclocking is a deliberate disregard for existing specifications. It is the overclocker who is intolerant, not the design. If the limits of a product are identified and specified correctly, exceeding those limits would probably cause undesired results.

By pointing fingers at Foxconn, motherboard manufacturers, and Intel, Mr. Gill is breaking the cardinal rule of overclocking.... Assume full responsibility for your own actions!

"These problems are the cause of a single component and are repeatable."

No Mr. Gill, these problems are NOT the cause of a single COMPONENT, but rather that of a single ACTION, overclocking. However, you are correct when you state they are repeatable, errors often are.

A lot of people will follow in your footsteps and try to get something for next to nothing. They too will overclock their boards. Maybe not to the extreme, but some. And in doing so, they will ignore the advice of some very learned men/women. People with the proper machinery to analyze the circuits and who really DO KNOW how far these can be safely pushed.

Some of these overclocks will fail. However, the average consumer doesn't get their boards and processors free for evaluation purposes like you. They pay hundreds of dollars. If they fail prematurely due to an ill conceived overclock, will they RMA them back to the vendor? Probably, if it is still under warranty. The vendor, looking to keep a customer satisfied often will accept the board. However the costs will ultimately be passed on and borne by the consumer.

Some consumers worry that a boards overclocking performance is indicative of its longevity. Well maybe. However I am really dismayed when I find owners express a desire to return a fully functional board based on your findings.

Along a similar note:
I got this Freezer. The specs say that if I want to keep ice, that I have to run it no warmer than 32 degrees Fahrenheit. Well I wanted to test it out just to see how good it was. So I decided to run it at 50F. You know what? The ice melted inside of 15 minutes!
It didn't matter whether the ice came from tap water or bottled, the bathroom sink or the kitchen. It melted! This problem was repeatable! I can produce pictures if you wish.
Just to be sure it wasn't something I was doing, I went to the store and got some block ice. While that lasted a bit longer, it too melted.

Now I'm wondering...if I can't keep ice for 15 minutes at 50F, how will I ever be able to keep it for weeks at 20F? Maybe I should return the freezer.

[XJ.]

Thanks GWillakers,

I think that this issue will sort itself out over the next few months. I think some of what is going on is just typical of a new CPU/Socket/Chipset solution - finding out what the limits are and manufacturers getting their boards designs down pat.

All this leads me to think that going with LGA 1366 CPU/motherboards is probably best now, even though I won't be doing any extreme overclocking. Being on the cutting edge has had it's downsides for me in the past.

AnandTech has some updates on this issue FWIW here.

[Übrales]

GWillakers, excellent response. You bring up some very good, well thought out points.