P55 Overclocking Showdown - ASUS, Gigabyte, and EVGA at the OC Corral

[Linus]

Being who I am, I can't give any names, but there are other motherboard manufacturers I've received abysmal support from (far worse than MSI) over the years.

I think some manufacturers get a free ride as far as support goes, and I'm not really sure why...

[Eldonko]

I make my decisions based on personal experience. I have done alot of RMAs over the years. If I get unsatisfactory service once I wont buy anything else made by that manufacturer. These days there are too many good products out there to be forced into dealing with any manu you dont want to deal with.

[zsamz_]

Originally Posted by Sagath

The rap of MSI bad company support still runs rampant in the enthusiast circles. I dont care how far they've come, even if its 300% better, that still only means they've gone from 3% satisfaction to 9% .

Until MSI starts doing backflips for their customer base, the enthusiast crowd will stay far, far away.

i agree 100%
if we forget the rma support issues
we need to deal with bios issues
msi is offerin 2 crossfire cables n jackin the price but they need to improve bios before thay ask for top dollar for the g80
i would bought an msi if they were 100$ less not worth wastin my time
240$ gor a g80 is ridiculous when you can get a better board for less

[Raja]

Two reasons MSI were not in the article

1) They were still in beta phase when the initial lineup was being finalised - the GD80 was undergoing a revision for PCI/e

2) By the time they were ready, my good CPU had been damaged by the ASUS ROG socket meltdown. This meant I could not provide comparative results to the other 4 boards. The only way to have done it would have been to start again with another CPU, thus having to test the other 4 boards again. Given 2 CPU's were already damaged, I thought it best to run with the almost complete 4 board test results I already had rather than risk going through the enitre test benches again and coming out with little to show for it.


So nothing against MSI personally. Maybe next time.

regards
Raja

[Soultribunal]

Originally Posted by Raja

Two reasons MSI were not in the article

1) They were still in beta phase when the initial lineup was being finalised - the GD80 was undergoing a revision for PCI/e

2) By the time they were ready, my good CPU had been damaged by the ASUS ROG socket meltdown. This meant I could not provide comparative results to the other 4 boards. The only way to have done it would have been to start again with another CPU, thus having to test the other 4 boards again. Given 2 CPU's were already damaged, I thought it best to run with the almost complete 4 board test results I already had rather than risk going through the enitre test benches again and coming out with little to show for it.


So nothing against MSI personally. Maybe next time.

regards
Raja

Foxxcon Sockets?

That seems to be happening a lot these days.

ST

[Linus]

Originally Posted by Raja

Two reasons MSI were not in the article

1) They were still in beta phase when the initial lineup was being finalised - the GD80 was undergoing a revision for PCI/e

2) By the time they were ready, my good CPU had been damaged by the ASUS ROG socket meltdown. This meant I could not provide comparative results to the other 4 boards. The only way to have done it would have been to start again with another CPU, thus having to test the other 4 boards again. Given 2 CPU's were already damaged, I thought it best to run with the almost complete 4 board test results I already had rather than risk going through the enitre test benches again and coming out with little to show for it.


So nothing against MSI personally. Maybe next time.

regards
Raja

Well that answers my question I suppose, thought to me it's kinda like saying "We didn't include the Radeon 5850 in our GPU round up because by the time we got around to testing it, our GeForce GT300 had wrecked the PCIe slot on our test bench, so in order to include it we'd have to re-run the testing"

It's one thing if the GD80 is the one that wrecked your CPU.

That said, the PCIe revisions seem like a major thing. What were they revising? I have a very early board and I haven't heard anything about it. How do I tell if mine is revised or not?

[Raja]

Originally Posted by Linus

Well that answers my question I suppose, thought to me it's kinda like saying "We didn't include the Radeon 5850 in our GPU round up because by the time we got around to testing it, our GeForce GT300 had wrecked the PCIe slot on our test bench, so in order to include it we'd have to re-run the testing"

It's one thing if the GD80 is the one that wrecked your CPU.

That said, the PCIe revisions seem like a major thing. What were they revising? I have a very early board and I haven't heard anything about it. How do I tell if mine is revised or not?

That analogy fits in the instance where the PCI/e slot not working is a complete one off. Given the fact that the socket failures seems to be plaguing me I think I made the right decision. Wringing out each CPU on every board for those benches took 4 weeks, not 2 hours each (so its hardly as if I'm bumming out of easy work as you see it). You might not agree with that, but hey, I think that I can say this forever and people will still disagree with me, so lets just leave it at that.


MSI were supposed to have revised the PCI/e stuff because of reported issues with PCI/e overclocking on their boards.

later

[Linus]

Originally Posted by Raja

That analogy fits in the instance where the PCI/e slot not working is a complete one off. Given the fact that the socket failures seems to be plaguing me I think I made the right decision. Wringing out each CPU on every board for those benches took 4 weeks, not 2 hours each (so its hardly as if I'm bumming out of easy work as you see it). You might not agree with that, but hey, I think that I can say this forever and people will still disagree with me, so lets just leave it at that.


MSI were supposed to have revised the PCI/e stuff because of reported issues with PCI/e overclocking on their boards.

later

I'm not saying you're bumming out on easy work. Quite the opposite. I know how long it takes to tune in a board to the point where you can publish results on it, and it takes a very very long time.

I'm questioning the priorities.

If we're talking about excluding a T2/T3 vendor (DFI, ASROCK, BIOSTAR, etc.) on account of "it was a hassle because this platform sucks to test" we're having a different conversation right now.

Instead we see 1 board each from a couple of T1 vendors (ASUS, GIGABYTE), and then 2 boards from a T2 (are they even a T2??) for some reason. Meanwhile we're ignoring a host of other T2 manufacturers and a T1.

Obviously I'm not in your lab. For all I know the eVGA boards were the first to arrive and they were tested first, then you tested two more before one blew up your CPU.

Honestly though, I'd have accepted "for this extreme overclocking guide, we're throwing out every board that has a Foxconn socket because we only trust the LOTES socket for our testing". That is an unbiased approach that gives everyone a fair shot at being included and saves your CPU so that you can update the round up later when ASUS/Gigabyte/MSI are able to deliver LOTES boards to you.

Hindsight is 20/20 though, eh?

[Raja]

Exactamundo on the hindsight. If my crystal ball had told me that the Foxconn sockets were going to be the sole cause of my woes I would have done just what you said. MSI were ready to ship their revised GD80 in week 4 after it had all gone down. That's the last I'm going to say on this.

[Linus]

So you don't know how to tell if you have a revised board or not eh?