I saw this review pop up pretty frequently when I was first (desperately) searching for info on my MSI-7612 way back when, and see it's still here as I'm looking up my old resources to print for a buddy. Thought I'd weigh in on what this board is actually capable of under the right conditions, as the review didn't do it any kind of decent justice - and frankly, it's obvious the mobo wasn't handled properly, either.

First of all, as just a forewarning for x6 owners looking for info who might come across this, you do NOT want to overclock an x6 on this rig; you will burn out the voltage regulators almost IMMEDIATELY. Google it to find out why. The NF200 chip and the voltage regulators are the (only) weak links in this configuration that I've seen, but if you don't get greedy when pushing your voltage steps, none of that is a real issue when overclocking in general.

An x4, on the other hand, is a marvel to play with. In most cases, you'll likely only be limited by the processor itself, as my examples should demonstrate. You DO, however, NEED to adjust all voltage settings in the bios! VDD with VID, NB after a point, etc; these settings all depend on one another and need to be incrementally adjusted accordingly. You cannot leave any settings (except SB) at stock or auto values and expect a decent overclock - this is especially true once you start overclocking RAM and your graphics card(s). Worth noting is that this is generally true with almost any configuration if you're expecting a "real" overclock (as opposed to an easy 20-30%, which nearly anybody with internet access can do in a few hours) - so I'm not sure why it wasn't addressed in the review here. Anyway, it's a little more difficult to figure out on this mobo, so perhaps that had something to do with it.

Unfortunately, when I first got my board, it took me a solid two weeks of constant, vigilant searching to find ANY useful information on what the **** certain bios voltage settings actually WERE. A friend of mine recently picked up one of these for a new system he's building (beginner), and much to my dismay he had the same issue only a few weeks ago; there's still an overall shortage of knowledge on some very important details. Fortunately, I take good notes. That's my only real beef with this board (and really it's more an AMI bios + AM3 issue): unparalleled lack of feature documentation and support.

In the way of features, most I either already knew or could guess from previous knowledge and experience--since I've been doing this for a couple of decades now--but a few of the bios Cell Menu voltage settings for my Phenom II were truly counter-intuitive, at best. It made what should be a very straightforward and intuitive process... nightmarish. And to this day, the OC community really only has a "grasp" of how certain voltage settings interact with each other; we basically know "what" they do in terms of how they directly affect our overclock potential and system stability, but there's absolutely no information available, anywhere, that tells exactly HOW they do it. I was even unable to extrapolate answers from CPU manufacturer charts and engineering reports/schems.

Trial and error, then, is really your only option on this board. If you aren't prepared to do that, a more user-friendly option is advisable.

If you are willing to put in the effort, however...

Currently, I've been running an x4 PhenomII 955 BE (3.2 stock) @ 5.58Ghz, 24hr Prime95 stable and with only one BSOD in over a year of running a 24/7 OC (my fault, stupid video-related oversight when doing new overclock on my second GTX 460).

FSB max ended up being between 313-314; I run at 310 to maintain an ideal HT-Link speed, and because I don't accept anything other than 100% stability.

This is all ON AIR. If I use an additional external fan to lower ambient temps around my case by ~4 degrees, I can maintain a stable 6.0GHZ OC - and if I wanted to test the limits of the voltage regulators, and/or go with a more aggressive cooling solution, I'm sure I could go farther - but when my system easily outperforms an x4 i7 in the mid-4GHz range, why bother?

While we all know that the 900 series are incredible overclockers, none of that is relevant without a mobo that can support it. For me, this one has far exceeded my expectations. I just finished helping my aforementioned friend tweak his own system about three weeks ago; he got to 4.8GHz on his 945 phenom and didn't want to push it farther (it's his first system, you remember those days), but the mobo could easily handle it if he changes his mind in the near future.

tl;dr -- do your research thoroughly if you're looking into this board. It has its quirks, but it's generally rock solid; I have yet to hear of one that's been fried due to anything other than VRM failure caused by excessive (and I really mean EXCESSIVE) overvoltage far beyond the usual 0.5 over manufacturer's specs - and in most other cases it's been due to stupid user-errors such as trying to run a high overclock with various power-saving no-nos enabled. I've seen a FEW reports of the NF chip burning out, but I strongly suspect that had to do with physical damage to the chip or its heatsink, or just a faulty module (inevitable). The margin of error when working with those kinds of temperatures might as well be nil.

Happy overclocking!



Append Quote Prepend Quote

When I mentioned that MSI's APS (Active Phase Switching) does not function properly in their forum, I was threated by administrators to keep it quiet. They removed all the posts related to in which situations APS does not function which is when you try to undervolt the processor (or basically change anything in that section of the BIOS I believe) for further power savings!

In addition to that, earlier BIOS versions allowed APS to function even when the processor was undervolted, overclocked or overvolted. As a coincidence, the samples they send to testing/reviewing sites had this feature working fine.

As a matter of fact, one can simply gain more power savings by undervolting the processor compared to using APS (and that can be done practically on any board!)

All considered, APS is a marketing gimmick only and you shouldnt fall for that! This was unfortunately a feature which I had in my decision criteria for this board

Message I got at MSI forum from an admin:
----- -------- ------------ ------- --------- ------------ -------
Author: Bas

APS does'n work when you undervolt.
Open another topic about this and find yourself removed from the forum.
If you have a problem with this talk to MSI, there is nothing we users can do.
As you posted bashing towards MSI your topic was locked.
Opening another as you just did is considered you don't obey the rules.

We do not allow such behavior in the forum.
--------- --------- ---------- --------- --------- ------ ------



Append Quote Prepend Quote

Yes, very true. I know that even with similiar hardware, OC result will vary. When testing for stability, what utilities did you use? How long did you let them run before you declared your settings stable?

Thanks!





Append Quote Prepend Quote

I have an MSI NF980-G65 board with Phenom II X4 955 rev C cpu and G.SKILL DDR3 1600 memory. After reading the review of this board, I set up my OC with some reference from the article:

HT Link Reference Clock of 206 MHz
CPU Multiplier of x19.5, resulting in 4022 MHz
CPU Voltage 1.496v
CPU-NB Multiplier x13, resulting in 2681 MHz
CPU-NB Voltage of 1.306v

However, I left the HT Link Multiplier at default. Setting to x12 as mentioned in the article, refuses to boot.

So, my question is, was the HT Link Voltage raised from stock to a higher setting? What was it set for?

Thanks!



Append Quote Prepend Quote

quote cpp25

I have a MSI NF980-G65 Motherboard and it is a nice board but I'm over My head on this one and for the life of Me, I can not figure out how to set the ram timing. I have Crucial Ballistix Tracer 4GB (2 x 2GB DDR3 1600 ram and every time I try to set the timing of 8-8-8-24 at 1600, it tends to freak out or not even post(Because of Me not knowing what I'm doing). Is there any help or a tutorial?

You can try giving the CPU-NB a little bit more voltage. Additionally, you can increase the timings.

Let me know how it works out.



Append Quote Prepend Quote