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.

Overclocking on the Biostar TZ77XE4 was a mixture of disappointment followed by a brief respite of joy. 

Automatic overclocking on this motherboard is flawed, resulting in shutdowns.  As with some other motherboard manufacturers, the overclock is enabled directly from switch on - thus if it is truly unstable during POST, only a ClearCMOS will clear it (then you have to remember to enable AHCI again).  Other manufacturers enable overclocks at the end of POST, which is preferable.  I only experienced this when I adjusted the Load Line Calibration for the CPU, which essentially is best left on automatic.

During the manual testing, when a voltage was selected and we told the BIOS to make all cores the same multiplier, the operating system would remain at that speed and voltage constantly, not invoking any SpeedStep.  This could result in high idle temperatures for users.

Our standard overclocking methodology is as follows.  For automatic overclocks options, they are selected and tested for stability with PovRay and OCCT to simulate high-end workloads and 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.  We also are using Intel's All-in-one Liquid Cooler with its stock fan.  This is a 120mm radiator liquid cooler, designed to mimic a medium-to-high end air cooler.

Automatic Overclock:

For Automatic Overclocks, the only options available are found in OS Software through the TOverclocker program.  This program gives three options - V6, V12 and Automatic (the naming suggests that Biostar are trying to emulate car engines in their overclock representations).  Selecting these options allows the software to apply the overclock, which was not kept between reboots unless the TOverclocker software was loaded again.  Here are our results:

For V6, the system boosted the BCLK to 102 MHz, giving 3978 MHz during single thread load.  Unfortunately, the board also crashed after applying the overclock, suggesting instability.

For V12, the system boosted the BCLK to 104 MHz, giving 4056 MHz during single thread load.  At this overclock, PovRay reported a peak temperature of 61ºC and OCCT had 63ºC.  The system was very stable.

For the automatic setting, the board immediately reset itself, and loaded the OS.  Nothing happened in the OS, and no overclock was applied.  On loading TOverclocker, it acted as if nothing had happened.

Manual Overclock:

During the manual overclocking, PLL Overvoltage was left disabled and Load Line Calibration was left on automatic.  To change the multiplier, the BIOS option 'Fixed CPU Ratio' was enabled and the CPU Ratio option was adjusted.  Voltages were modified in the BIOS as well, using CPU VCore Fixed, which is gradated in 0.010 volt increments.  Here are our results:

Starting at a 1.100 volt setting, the system was set at 44x on the multiplier.  This gave a load voltage in the OS of 1.116 volts, and peak temperatures observed were 70ºC for PovRay and 75ºC for OCCT.

For 45x, a setting of 1.100 volts in the BIOS was required, which gave a load voltage of 1.116 volts in the OS.  Peak temperatures observed were 73ºC for PovRay and 76ºC for OCCT.

For 46x, a setting of 1.130 volts in the BIOS was required, which gave a load voltage of 1.128 volts in the OS.  Peak temperatures observed were 77ºC for PovRay and 80ºC for OCCT.

For 47x, a setting of 1.180 volts in the BIOS was required, which gave a load voltage of 1.188-1.200 volts in the OS.  Peak temperatures observed were 84ºC for PovRay and 85ºC for OCCT.

For 48x, even at a setting of 1.280 volts in the BIOS, the system was not stable.  At this voltage, the system would reach 97ºC during PovRay and then cause a hard crash.

Biostar TZ77XE4 In The Box, Voltage Readings Test Setup, Power Consumption, POST Time, Overclocks
Comments Locked

28 Comments

View All Comments

  • Patflute - Saturday, July 21, 2012 - link

    ...
  • Olaf van der Spek - Saturday, July 21, 2012 - link

    Are fan headers expensive or what? Is it too much to ask for 2 headers top-left and 2 headers bottom-right? Ideally PWM, but anything is better than nothing.
  • yyrkoon - Sunday, July 22, 2012 - link

    Fan headers require traces on the motherboard PCB. That is more expense. Plus additional PWM costs. Not sure exactly how motherboard manufactures achieve this. A few embedded projects of my own. The MCU may have 1-2 PWM I/O pins. After which any other PWM output I would require, would/could require external PWM controllers. More code for me to write. Then more work for my friend ( electronics engineer ) to figure what all is needed. Plus design the given circuits.

    Once that is figured into the cost. the cost of actual PCB layout will probably be much more. Since good electronics engineers typically cost $100/hr ( or more ) in labor and other additional costs. I am no electronics engineer, but I do know that you have to be very careful with consumer products such as these. Putting a power trace in the wrong spot. Could make a good reliable board, into something less desirable.
  • room200 - Saturday, July 21, 2012 - link

    Asus, Biostar, ECS, etc. all have had nightmare motherboards in the past. I've used them all, and I can't say one is any better than the other. I bought this board for 15 bucks at Microcenter, and have had no problems at all with it.
  • room200 - Saturday, July 21, 2012 - link

    I meant 115 bucks.
  • blppt - Saturday, July 21, 2012 - link

    It isn't going to excite any gamers, but the Biostar P43 mobo with a Q6600 in my parents' computer has been up 200+ days of near constant usage now without a reboot (win7 64), and they don't even have a UPS (it only ever gets rebooted for a critical patch or when the power goes out). And they don't just surf the web---my father has become addicted to Skyrim and Kingdoms of Amalur. So, it may not be the most featured robust over clocking board I've ever seen, but it sure is rock-stable.
  • Kent_Diego - Sunday, July 22, 2012 - link

    The difference between a premium manufacturer and low end is support. By support I mean BIOS updates that fix bugs, add new CPU models, and improve compatibility. I once bought a Foxconn motherboard. It had a few issues and they never released a new BIOS. I have a Biostar motherboard and they kept it supported for two years before the BIOS updates stopped. When I wrote them ab email I got a reply in broken English from an engineer there. Not as good a ASUS but better than ECS.
  • Grok42 - Sunday, July 22, 2012 - link

    I'd love to hear from anyone that is considering building a rig now or in the future and has the need or desire to put a PCI board in it. Not only can I not think of a reason, I haven't found a reason for at least the last 8 years. Since MBs started putting audio and network on-board I haven't used expansion slots at all PCI-E or PCI with the exception of a discrete graphics card.
  • yyrkoon - Sunday, July 22, 2012 - link

    DVR boards would be one reason.
  • blppt - Monday, July 23, 2012 - link

    My X-Fi XtremeGamer is PCI, and there haven't been any notable advancements in gaming audio tech worth replacing it with IMHO.

    That said, if an attractive mobo came along, would I rule it out because of that sound card? Probably not---I'd rather be happy with the mobo and just buy a PCI-E x1 replacement from ASUS or Creative.

    I'd stick with onboard, but the analog outs on most motherboards are terribly noisy IME.

Log in

Don't have an account? Sign up now