MSI Z77A-GD65 Gaming In The Box

Products aimed at the gaming crowd are draped in PR and have to be visually striking.  Along with this should be a sense that the gamer is being looked after, and that they are getting more for their money, in terms of features and support.  So when you offer access to specific forums and support admins when something goes wrong, that is great – but also there is an element of camaraderie for competitions between other brands.  Part of the way to gain this mindset is to give more stuff to the user, either free or with what they purchase.  As part of a gaming product, I would want extras themed in the overall style, but not too cheesy. 

In the Z77A-GD65 Gaming, we have:

User Guide
Installation Guide
Driver CD
MSI Gaming Door Sign
Four SATA Cables
Rear IO Shield
Flexi-SLI Connector
Voltage Read-point attachments
M-Connectors

Even though we are dealing with a sub $200 motherboard, the kit in the GD65 Gaming is incredibly light. By virtue of not having an extra USB 3.0 port on board we do not get an extra USB 3.0 panel included, like on the Gigabyte Z77X-UD5H.  Back in the review of the UD5H I criticized it for not having a rigid SLI connector, and I could say the same for the MSI here, but various MSI GPUs have heatpipes that block rigid connectors, so a flexi-connector makes sense.

MSI Z77A-GD65 Gaming Overclocking

Note: Ivy Bridge does not overclock like Sandy Bridge.  For a detailed report on the effect of voltage on Ivy Bridge (and thus temperatures and power draw), please read Undervolting and Overclocking on Ivy Bridge.

Experience with MSI Z77A-GD65 Gaming

The MSI overclocking experience is always a little odd.  As a general rule they have one automatic setting (which either does or does not work) and the manual settings are either a gamble in the OS or a mire in the BIOS.  The GD65-Gaming fits into this rule, and I have a feeling that all MSI motherboards will fit into this rule until the BIOS is organized or more automatic overclocking options are offered.  ASUS offers 3-4, Gigabyte offer 3-4, ASRock offers 6-10, MSI offers one – that is a big discrepancy.

Our overclocking experience with the Gaming was a little crazy.  For various reasons we are using a new i7-3770K CPU, and the performance of this CPU seems to be very poor, failing to hit even 4.5 GHz on the Z77A-GD65 Gaming.  This could be either CPU or BIOS issues. Nevertheless, our overclocking experiences are detailed below.

Methodology:

Our standard overclocking methodology is as follows.  We select the automatic overclock options and test for stability with PovRay and OCCT to simulate high-end workloads.  These stability tests aim to catch any immediate causes for memory or CPU errors.

For manual overclocks, based on the information gathered from previous testing, starts off at a nominal voltage and CPU multiplier, and the multiplier is increased until the stability tests are failed.  The CPU voltage is increased gradually until the stability tests are passed, and the process repeated until the motherboard reduces the multiplier automatically (due to safety protocol) or the CPU temperature reaches a stupidly high level (100ºC+). Our test bed is not in a case, which should push overclocks higher with fresher (cooler) air.

Automatic Overclock:

Automatic overclock options come from OC Genie, which is enabled either via the Control Center (in OS), in the BIOS (OC Genie mode), or phyiscally with the button.  Unfortunately each one provides different results.

The OC Genie via Control Center method applied a 4.2 GHz overclock on all cores as well as jumping the memory to XMP but back a strap (2133 instead of 2400).  In this mode we recorded a load voltage of 1.256 volts in the OS, a peak temperature of 81C during OCCT and a score of 1530.96 in PovRay.

The OC Genie Mode via the BIOS was fairly basic, applying only MultiCore Turbo (39x on all cores), and did not touch the memory.  This gave a load voltage of 1.256 volts, a peak temperature of 75C during OCCT and a score of 1421.96 in PovRay.

The last method, physically depressing the button, reacted similarly to the Control Center method, giving a 4.2 GHz overclock on all cores and a memory bump.  Load voltage was registered as 1.248 volts, giving a peak temperature of 83C and a score of 1535.10 in PovRay.

Manual Overclock:

Here's where things get a little interesting.  For various reasons, we are using an untested, different 3770K CPU to previous reviews.  Normally I would start this testing at 4.4 GHz and 1.1 volts, something the three 3770K CPUs that have crossed my path (only one for reviewing) over the past 12 months have all done.  In a real world situation explaining how varied CPUs are, this CPU, in the GD65 Gaming, behaved very different.  In fact, we had to start at 4.0 GHz at 1.1 volts, then perform our normal testing.  The results are as follows.

For 4.0 GHz (40x100), we needed 1.100 volts in the BIOS which gave 1.088 volts at load.  OCCT gave 62C and PovRay scored 1465.55.

For 4.1 GHz (41x100), we needed 1.125 volts in the BIOS which gave 1.112 volts at load.  OCCT gave 66C and PovRay scored 1497.88.

For 4.2 GHz (42x100), we needed 1.175 volts in the BIOS which gave 1.160 volts at load.  OCCT gave 68C and PovRay scored 1527.51.

For 4.3 GHz (43x100), we needed 1.225 volts in the BIOS which gave 1.216 volts at load.  OCCT gave 70C and PovRay scored 1575.73.

For 4.4 GHz (44x100), we needed 1.300 volts in the BIOS which gave 1.288 volts at load.  OCCT gave 81C and PovRay scored 1603.28.

4.5 GHz was not stable - at 1.325 volts the CPU was reaching 95C during PovRay, getting very hot very quickly. I was unwilling to push it any further.

MSI Z77A-GD65 Gaming Software Test Setup, Power Consumption, POST Time
Comments Locked

37 Comments

View All Comments

  • hansmuff - Friday, April 19, 2013 - link

    Good comment on the polling rate, marketing at its best. I do notice a difference between a 125 Hz rate and a 250 Hz rate in terms of smoothness of the desktop cursor when you move it, but any higher and I personally can't make out a difference.

    It should also be noted that polling with a higher frequency uses additional CPU cycles, since as the name implies USB is a polling interface.
  • sherlockwing - Thursday, April 18, 2013 - link

    My favorite part of the MSI Z77 board is the 90 degree USB3.0 header, it really helps hiding that huge usb3.0 header.
  • boogerlad - Thursday, April 18, 2013 - link

    Is this motherboard a joke? Covering the fins is very stupid, and fins that are shaped merely for aesthetics is even worse. Are aesthetics more important than performance and practicality these days?
  • mwildtech - Thursday, April 18, 2013 - link

    Bitch Bitch Bitch - Beat it!
  • HisDivineOrder - Thursday, April 18, 2013 - link

    Dare you deny the dragon?
  • Crono - Thursday, April 18, 2013 - link

    I dare not. I actually was planning on doing a custom paint case, red and black, with the Dragon Army emblem from the new Ender's Game coming out in fall. The motherboard would fit well, though really even with a side panel no one looks that closely to see some of the theme details on a motherboard.

    "Your ass is dragon!"
  • Sunstorm - Thursday, April 18, 2013 - link

    Hi Ian, I wanted to point out a piece of misinformation that MSI seem to be allowing various review sites to propagate. The board is not capable of 2X SLI (yes, standard two way) when the third PCI-E 3 slot is in use. I was told by MSI tech support that this is due to how the pcie lanes are provided for the third slot, which differs from the majority of other Z77 boards. This occurs even if the third slot is used for a soundcard ie you are not trying to do tri-SLI, SLI is simply disabled.

    Multiple reviews of the GD65 (and therefore presumably this gamer version) do not highlight this issue and presumably review sites aren't testing this. MSI seems content to allow this misinformation to be spread far and wide, which led to me buying the GD65 and finding it was not adequate for my needs several months after purchase. Thankfully the retailer accepted a return.

    Please update the review to reflect this, if reviewers are able to test this and publish it then maybe a few people will avoid the hassle I had
  • IanCutress - Friday, April 19, 2013 - link

    Hi SunStorm,

    I am communicating with MSI directly on this to see what is up. Personally I feel there may be something wrong with the SLI certification flag. It probably confirms that the x8/x8 arrangement is SLI capable, but the x4 from the chipset is not, but there may be a bug that states if the x4 is populated, disable SLI completely. I am hoping to get a straight answer for you.

    Please note I would have tested it if I had access to three of the same NVIDIA GPU, but alas I do not :(
  • IanCutress - Friday, April 19, 2013 - link

    Hi SunStorm. I got an answer, and it boils down to this.

    "In order to run SLI you need either a x16 slot running at @16 or @x8."

    So when the third slot is populated, it brings all slots down to x8/x4/x4. Because the second slot is x4, it fails SLI certification.

    If you have examples of SLI working in x4 mode *anywhere*, please email me (ian@anandtech.com) as this is rather interesting. If this is truly the case, it's an NVIDIA restriction, not an MSI one.

    Ideally if you want 2xSLI and an x4 sound card/RAID card, you need a board that does x8/x8 on PCIe slots and then takes the final x4 from the chipset, such as the Z77 OC Formula or Z77 Extreme6.

    Ian
  • Sunstorm - Friday, April 19, 2013 - link

    Thank you Ian, the fact that you took the time to take a look in to this is much appreciated. I have only found one forum post anywhere which highlighted this issue.

    Since most reviewers are stating that the MSI GD65 (and presumably related boards) is capable of tri-SLI I would imagine other people buying these boards may end up bitten down the line, even if they do their research by looking at reviews. I think in most people's case, given the price level, the fact that it is not tri-SLI capable is not a concern; however, the fact that ANY card in the third slot disables SLI is pretty important for people to be aware of, particularly as it is a behaviour not seen on most Z77 boards. As far as I am aware, the majority of Z77 boards take the lanes for the third slot from the chipset (as you describe), I've now switched to an ASUS v-pro for this reason.

    The only instances where I recall I may have seen SLI operating at 4x or lower are the tests of the impact of PCI-E bandwidth on graphics card performance. I'm not sure how those are conducted though.

Log in

Don't have an account? Sign up now