Is this what overclocking has become?

[Ben90]

 

[sm625]

Why would anyone overclock on an MSI board anyway? They have a horrible failure rate.

Why is there not a single Asus motherboard on that list? I know they aint that good...

 

[Rvenger]

This is ridiculous! Shame on MSI, and Intel too!

I want my mobo to cost no more than $40 despite having a BoM of $300...and I demand it to be so over-engineered that I am virtually guaranteed (not that I'm willing to pay for said guarantee) to be able to overclock my CPU to 7GHz.

Not only that, but when I do overclock my $60 CPU to 7GHz I expect the CPU manufacturer to send me a rebate check to cover the elevated electricity costs. After all it is my right to get absolutely everything I want, even if I don't pay for it, simply because I'm 'merican and its my gawd dern given right to do what's I like, whens I like, hows I like.

Freedom!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

You ok IDC?

 

[XiandreX]

I like my board, but it is a bit higher end then that one.

I love my MSI Z77A-G45. Honestly i feel his pain but there should be a reason to pay for motherboards as they tier up. Controlling the amount of overclocks is one way to do it.
When i bought my board there was the G41 and G43 versions below which i avoided for a variety of reasons.

Now deciding the threshold.... thats the tricky part. Kinda like the h77/h87 versus the z77/z87 argument.

 

[blackened23]

The power phases provided still matters regardless of the iVR on Haswell's die, so it's not like because Haswell has an iVR that you can completely ignore motherboard cost. Power delivery still matters, more power phases provides cleaner and more efficient power and provides for an easier/higher overclock. If you look at various Haswell board designs, there is a substantial difference in the MosFET area. If you want to overclock a 4770k with a 1.35V vcore on a 30$ cheese motherboard, well, good luck with that. And be sure to post pics of the MosFET being completely burned out from doing so.

Besides which, this thread has already been proven incorrect - the motherboard in question does allow higher than a 40x multiplier as confirmed by various reviews. And that isn't surprising since the stock multi on the 4770k is 39, virtual larry completely misinterpreted this newegg review.

 

[AnonymouseUser]

Why is there not a single Asus motherboard on that list? I know they aint that good...

That is a database of VRM failures specifically, and they have this to say about ASUS:

VRM Over-Current Protection
Warning: Spoiler! (Click to hide)

Over Current Protection (OCP) is something I have recently been examining. Protection features exist against VRM overheating/overloading depending on motherboard model and brand. I believe it is a crucial feature on motherboards today, because this is the function that will protect your VRMs from a catastrophic failure. This is why I have never seen ASUS boards fail even if people take a lowly 3+1 ASUS boards and try to overclock a Phenom II x6 on it; ASUS boards feature this technology, as it is a part of the PWM controller design.

As for MSI's failures being mostly with AMD motherboards, that's irrelevant IMO. They had more VRM failures than any other AMD motherboard maker at the time, so they were faulty compared to the competition. That doesn't mean I would never buy MSI (I even have an MSI 7790), I just wouldn't buy an MSI motherboard for overclocking.

 

[guskline]

We ALL want that $100 mb that OCs as good or better than the $200 plus mb so we can brag about how smart we are etc. In my humble opinion that rarely works out.

The reponse fro the MSI engineer was justified and frankly correct. Belly aching about it is a classic " I want something for nothing argument".

Lines such as " wanting champagne on a beer budget" or "wanting a Cadillac for a Chevy" price most likely apply. I don't blame the mb owner for asking MSI for help. Just don't go ballistic when you get such a reponse about overclocking.

I still buy most computer components from Newegg. However, I place less and less faith in the reviews, especially where the poster fails to list all of the other components. The actual review that complains about the OC problems does not list the memory,psu etc. used when the mb limit was 40x.

 

[Arkaign]

We ALL want that $100 mb that OCs as good or better than the $200 plus mb so we can brag about how smart we are etc. In my humble opinion that rarely works out.

The reponse fro the MSI engineer was justified and frankly correct. Belly aching about it is a classic " I want something for nothing argument".

Lines such as " wanting champagne on a beer budget" or "wanting a Cadillac for a Chevy" price most likely apply. I don't blame the mb owner for asking MSI for help. Just don't go ballistic when you get such a reponse about overclocking.

I still buy most computer components from Newegg. However, I place less and less faith in the reviews, especially where the poster fails to list all of the other components. The actual review that complains about the OC problems does not list the memory,psu etc. used when the mb limit was 40x.

In my experience that's exactly how things work out these days though.

I've probably used 100-110 Asrock Z68, Z77 and Z87 boards in the past year. Every single one of them was capable of running 2500K, 2600K, 2700K, 3570K, and 3770K chips at 4.4Ghz or better. And not one of those boards was more than $110 at MC, most of the time I was able to get them for $99 or less.

 

[AnonymouseUser]

In my experience that's exactly how things work out these days though.

I've probably used 100-110 Asrock Z68, Z77 and Z87 boards in the past year. Every single one of them was capable of running 2500K, 2600K, 2700K, 3570K, and 3770K chips at 4.4Ghz or better. And not one of those boards was more than $110 at MC, most of the time I was able to get them for $99 or less.

I got an ASUS M5A97 and Phenom II X6 1045t for $145 @ Micro Center, which easily runs @ 3.7GHz (40% OC). I also bought an open-box M5A97 which easily runs an FX-4100 @ 4.4GHz (22% OC). There are cheap motherboards made for overclocking, and then there are cheap motherboards that allow overclocking.

 

[Arkaign]

I got an ASUS M5A97 and Phenom II X6 1045t for $145 @ Micro Center, which easily runs @ 3.7GHz (40% OC). I also bought an open-box M5A97 which easily runs an FX-4100 @ 4.4GHz (22% OC). There are cheap motherboards made for overclocking, and then there are cheap motherboards that allow overclocking.

True true. I usually just use B-series boards for non-OC/non-K Intel systems (business builds as opposed to gamer boxes). PhII stuff generally needs moar power, and thus is similar to needing great mobos for C2Q overclocks.

 

[blackened23]

In my experience that's exactly how things work out these days though.

I've probably used 100-110 Asrock Z68, Z77 and Z87 boards in the past year. Every single one of them was capable of running 2500K, 2600K, 2700K, 3570K, and 3770K chips at 4.4Ghz or better. And not one of those boards was more than $110 at MC, most of the time I was able to get them for $99 or less.

It turns out that this motherboard can overclock the 4770k just fine too - Balla posted a link of a review with a verified 4.6ghz overclock on the same motherboard.

Virtual Larry (the OP) misinterpreted the newegg review - the entire issue is that the user who posted the newegg review (linked by Virtual Larry) was, quite simply, an idiot. The point still stands however - while I don't bother with MSI motherboards, 200$ motherboards have better features for higher overclocking. It does make a difference - some people might feel comfortable using a 40$ cheese motherboard for overclocking, but i'll pass. I've seen way too many motherboards with burned out MosFETS to risk a high vcore on an inadequate VRM.

 

[blackened23]

Why is there not a single Asus motherboard on that list? I know they aint that good...

Because they're that good. And this is why you don't use a 100$ cheese motherboard when you're going all-out with LN2/water cooled overclocking.....

Do you hear of a lot of Pro/Deluxe failures related to OCing? Maximus? No? There's a reason. Better components and higher quality.

 

[Arkaign]

Because they're that good. And this is why you don't use a 100$ cheese motherboard when you're going all-out with LN2/water cooled overclocking.....

Do you hear of a lot of Pro/Deluxe failures related to OCing? Maximus? No? There's a reason. Better components and higher quality.

$100 mobo (Asrock Z87 @ MC) is just fine for maxing out Haswell.

For intel K-series quads, you're not going to be pushing the watts that will threaten the stability of a board even of the entry level Asrock Z series.

The higher priced boards are there for features.

And LN2 is waaaaaaaaaay waaaaaaaaaaaaay different than water cooling. A Noctua D14 isn't even all that far off from an H100 when it comes to taking a 4670K or 4770K (or even 3570K/3770K) to their 24/7 limits. LN2 is a whole other ballgame, and super high end boards should show some differences with a de-lidded suicide run. Let's be real though, LN2 is used in .000000001% of systems lol. Even water is ultra rare. I can count on one hand the number of water-cooled systems I've run across in the wild with SB and newer builds. And why not? When a 212 Evo for $20 will take a 4670K to 4.5Ghz, and an H100 will take the same chip to 4.5 / 4.6Ghz at nearly the same temps, it's not much of a payoff.

For AMD FX / Phenom stuff and older Intel stuff the high-end boards really do pay off for overclocking more. For intel 1155/1150 the high priced boards are there to offer more features, or to give you the impression that you're getting more OC capability. Maybe with liquid nitrogen or exotic cooling that would come into play. I'd certainly like to think that my Asus Deluxe board was somehow crucial to operation, but to be honest, the AsRocks that I regularly use are just as good at getting 4.5 to 5Ghz clocks from SB/IB, and ~4.4-4.6 out of the typical Haswell.

 

[bunnyfubbles]

1. Z87 should provide full access to multipliers if he actually has a K series CPU (which the reviewer doesn't specifically state what he has), so if that Z87 board has a limitation there, then that's pretty bad on MSI

2. Even if this particular Z87 board was limited to 40x multiplier, if he did have a K series chip he still should have been able to tweak the bclk straps. A 40x multi with the 125 strap gives us the potential for 5GHz, or with 37 x 125 we're hitting his goal for 4.6GHz

so basically I think the reviewer is misleading somewhere; either he doesn't have a K series chip, or he's not the 5/5 tech level like he claims (although could very well be both)

 

[aigomorla]

The Haswell voltage regulator still needs to be provided a voltage and current.. 1.8V by default, often increased to 1.9 - 2.1V by overclockers. You really didn't think those tiny surface-mount components next to the Haswell die could replace the large capacitors and MOSFETs with their own heatsinks that you find around the CPU socket, did you?

Intel seems to be confident :biggrin:

whats funny is u guys are complaining about a cpu which was slated to gimp overclocking, do just that.

400A on the cpu end.. is like... what???? 1.1V @ 400A = ~ 400W on your CPU?? anyone know of someone who didnt cook there haswell with that current?

I have no idea what intel is thinking.... cuz i dont see the math being correct here...maybe my math is grossly wrong..

 

[Shmee]

If I remember correctly, P=IW means you get 440W, but I don't think you use that formula for this.

 

[VirtualLarry]

http://www.hitechlegion.com/reviews/motherboards/36111-msi-z87-g41?showall=&start=3

4.6GHz on a i7-4770k

Thanks for that link. It appears that my conclusion was wrong. I'm glad to hear that MSI apparently isn't capping BIOS overclocking by multi, at least not yet.

But remember their lack of offset voltage overclocking on their Z77 boards, when the rest of the industry had that feature?

Edit: Interesting, here's another user (in our forums!) that cannot raise over a 40x multi with an IB K quad.
http://forums.anandtech.com/showthread.php?t=2339356