Overclocking Results
Taking our UEFI screen shots at the top CPU overclock setting meant saving them for last. That’s also when we read the settings for the chart below. By now you’ve probably noticed that MSI’s P67A-GD80 has no UEFI screen shots. Likewise, it didn’t survive to the point where we would have transcribed its adjustment ranges.
| BIOS Frequency and Voltage settings (for overclocking) | |||||
|---|---|---|---|---|---|
| Row 0 - Cell 0 | ASRock Fatal1ty P67 Professional | ASRock P67 Extreme6 | Asus P8P67 Deluxe | Asus P8P67 EVO | MSI P67A-GD80 |
| CPU Base Clock | 95-110 MHz (0.1 MHz) | 95-110 MHz (0.1 MHz) | 80-300 MHz (0.1MHz) | 80-300 MHz (0.1MHz) | Row 1 - Cell 5 |
| CPU Multiplier | Up to 60x | Up to 60x | Up to 59x | Up to 59x | Row 2 - Cell 5 |
| DRAM Data Rates | 1066-2133 (266.6 MHz) | 1066-2133 (266.6 MHz) | 800-2400 (266.6 MHz) | 800-2400 (266.6 MHz) | Row 3 - Cell 5 |
| CPU Vcore | 0.60-1.52V (5mV) | 0.60-1.70V (5mV) | 0.80-1.99V (5mV) | 0.80-1.99V (5mV) | Row 4 - Cell 5 |
| VTT Voltage | 0.66-1.87 (13mV) | 0.66-1.87 (13mV) | 0.80-1.70V (6.25mV) | 0.80-1.70V (6.25mV) | Row 5 - Cell 5 |
| PCH Voltage | 0.78-1.65V (9mV) | 0.78-1.65V (9mV) | 0.80-1.70V (10mV) | 0.80-1.70V (10mV) | Row 6 - Cell 5 |
| DRAM Voltage | 1.20-1.80V (15mV) | 1.20-1.80V (15mV) | 1.20-2.20V (6.25mV) | 1.20-2.20V (6.25mV) | Row 7 - Cell 5 |
| CAS Latency | 5-15 Cycles | 5-15 Cycles | 3-15 Cycles | 3-15 Cycles | Row 8 - Cell 5 |
| tRCD | 4-15 Cycles | 4-15 Cycles | 4-15 Cycles | 4-15 Cycles | Row 9 - Cell 5 |
| tRP | 4-15 Cycles | 4-15 Cycles | 4-15 Cycles | 4-15 Cycles | Row 10 - Cell 5 |
| tRAS | 10-40 Cycles | 10-40 Cycles | 4-40 Cycles | 4-40 Cycles | Row 11 - Cell 5 |
The P8P67 Deluxe leads in CPU overclocking, though the Fatal1ty P67 Professional is not far behind. MSI wasn't able to participate in this test, unfortunately, as its board met an early demise.
MSI did participate in the base clock test, where all motherboards successfully pushed their P67 platform controller hubs to 107 MHz. So, where did the MSI motherboard fail?
The P67A-GD80 refused to cold boot following its memory stability test at DDR3-2008, instead power cycling in one to two-second intervals. Further analysis indicated that this was due to a bad UEFI setting that stuck. Neither the button on the I/O panel nor the jumper on the motherboard’s surface cleared the problem, even after battery removal.
Another method to get BIOS to revert to safe settings is to force a hardware configuration error, yet swapping out components did not work either. When stuck, the P67A-GD80 wouldn't even register a memory or graphics error with those parts removed. Hopes that MSI’s backup ROM on this Dual BIOS-equipped motherboard might have initialized itself turned to hopelessness. One of our engineering friends said that the reboot was occurring before system logic could register the error, because the error was occurring at the wrong part of the initialization process. In all likelihood, this really was an old-fashioned "stuck BIOS” issue.
MSI’s P67A-GD80 has two ROM ICs, but no selector switch to force a boot from the second ROM. It also lacks a removable ROM socket, so hot-flashing it by using another motherboard was not an option. This might just be one configuration where a removable single BIOS ROM would have been more valuable than two soldered-on BIOS ICs.
