DFI LanParty UT X58-T3eH8 LGA 1366 Motherboard Review | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| by 3oh6 | March 12, 2009 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Voltage Regulation Voltage RegulationWhat you set in the BIOS is definitely the voltage being supplied in reality...correct? Well, that is not always the case. In fact, it is rarely the case. Then there is the discussion about whether or not what the software voltage readings are providing are accurate. The standard way to test for actual supplied voltage is to measure directly from the motherboard with a digital multi-meter (DMM). Recent high end motherboards have made this a dead simple task by providing onboard, easy to reach, read points for various voltages. ASUS started the trend with the Rampage Extreme II and the EVGA X58-SLI also offered a limited selection of voltage read points. Not to be outdone, DFI has provided a class leading eight voltage read points down in south bridge corner. ![]() These read pads are quite easy to access for those working with the motherboard on an open bench setup. In our particular setup here, they are in the ideal spot. With the EVGA X58-SLI we had to solder leads to their read points just to access them but with the DFI, that won't be necessary. We would still like to see test leads that can connect to these points to provide continuous reading like the Rampage Extreme II offers, but we certainly can't complain with the sheer number of voltages DFI has routed to this location for easy access. Before we take a look at how well voltage is regulated on the LanParty X58-T3eH8, we will use the onboard read points to determine exactly how accurate the BIOS settings for voltages are. Below is a chart of what we set in the BIOS, what is read in Windows by software at idle/load, as well as what the DMM picks up from the read points on the board at both idle/load. For the DMM, we will be using our trusty calibrated UEI DM393 True RMS multi-meter that provides three digits after the decimal for accuracy. Below are the results of this testing with our overall overclocked settings (w/vDROOP disabled) we have already used for benching. Load will be taken after 10 minutes of OCCT CPU test load.
It is quite apparent that we have vDROOP disabled in the BIOS as vCORE not only didn't drop, it actually raised the slightest bit under load. The actual voltage really isn't that far off what the vCORE was set to in the BIOS so overall we are quite pleased. With vDROOP enabled, there is about a 0.04v drop from idle to load with the same setup as used above. The next very impressive voltage tested is vDIMM. Here we see a rock solid vDIMM output with the exception that the software reading is just slightly low. What is actually being fed to the memory, however, is as close as it gets to what we set in the BIOS. After suffering through the EVGA X58-SLI vDIMM giant droop, this DFI LanParty X58-T3eH8 certainly has us pleased with the voltage regulation. This will be confirmed shortly with a look at the OCCT chart. The rest of the voltages are pretty much bang on to what is set in the BIOS with minor fluxuations to VTT. At idle VTT is a hair below what we set in the BIOS, and under load it is a hair high. The movement is in quite an acceptable range though. Before we look at the voltage charts for vCORE, vDIMM, VTT, and the IOH/ICH (north bridge), let's have a look at the effects of enabling vDROOP VS disabling it.
I think the line in the images tells the story. Without vDROOP disabled, our configuration shows heavy droop from idle to load. Keep in mind this is the software reading but it still coincides with the DMM readings we were getting at the same time. With vDROOP enabled in the BIOS, we get a slight rise in vCORE in our DMM testing but from idle to load in OCCT we see a completely steady line. The two schools of thought on vDROOP are all over the internet. Some say it is necessary and should be enabled, while others feel vDROOP should be eliminated so that you have lower idle voltages. However you look at it, no conclusive evidence has been linked to enabling or disabling vDROOP. There hasn't been an apparent rash of damaged processors since vDROOP elimination options started showing up in BIOSs, and there has been no public outcry from Intel asking MB manufacturers to remove the option to allow vDROOP to happen. Either way, the option on the X58-T3eH8 works, and works well. Use it if you want, or don't if you choose not to. Here now are the rest of the voltage charts from OCCT that were captured during testing for the above chart under the same conditions. ![]() As we mentioned above, vDIMM is rock solid, all the way across. Even with the DMM and its four digit accuracy, it very rarely flickered off of 1.698v at any point. VTT shows a little variation, as we expected after the chart of voltages from idle to load showed above. There is nothing of too much concern here but keep in mind, VTT does rise slightly at these settings under load. On the whole, the DFI X58-T3eH8 has really excellent voltage regulation, and where it counts the most. Memory clocking should be quite consistent on this motherboard as we have found out and whether you want or don't want vDROOP, the DFI has you covered. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |
| Latest Reviews in Motherboards | |||||||||
|