Test System
Test System |
---|
CPU: | Intel 4770K 3.5 GHz, 8 MB Cache |
---|
Memory: | 16 GB DDR3 (4x 4 GB) Avexir Blitz 1.1 TechPowerUp! Edition |
---|
Cooling: | CoolerMaster TPC812 |
---|
Motherboard: | ASRock Z97X KILLER Intel Z97 Express, BIOS v0.20F |
---|
Video Card: | MSI GTX780 GAMING 3 GB |
---|
Harddisk: | 2x Crucial 128 GB SATA 6 Gb/s SSD (OS & DATA) |
---|
Power Supply: | Thermaltake TruePower Platinum 1250W |
---|
Case: | Lian-Li T60 Testbench |
---|
Software: | Windows 8 64-bit, Nvidia 335.63 WHQL |
---|
Initial Setup
Booting up and installing Windows 8, everything appeared fine.
I booted into the BIOS and pushed for my standard manual overclock. Back in the OS, I ran a few benchmarks and took some screenshots for you to take a look at.
Power Consumption
We measure CPU power consumption since one of our first tasks is to truly verify system stability. I isolate the power coming through the 8-pin ATX connector using an in-line meter that provides voltage and current readings, and total wattage passed through. While this may not prove to isolate the CPU power draw in all instances, it does serve as a good indicator of board efficiency and effective VRM design. Total system power consumption is no longer reported as this figure can change depending on what VGA is installed. The sole board-only power measurements possible without physically modifying a motherboard are those taken via the 8-pin CPU connector, making it the only figure of value worth reporting. I use wPrime with eight threads selected in the options since it provides a consistently high workload throughout the full length of the test and runs long enough for the VRM and CPU to produce a fair bit of heat. Most average workloads will draw far less than that, although distributed computing applications are quite similar. This is not supposed to test stability since I use several other applications to do so, but merely to provide repeatable power draw numbers anyone can replicate. The meter used is an off-the-shelf Zalman unit that has been on the market for some time. It provides similar results in my test environment when compared with a FLUKE 337 clamp meter.
Load Condition | CPU Voltage | Ring voltage | Idle Power | Load Power |
---|
Stock Clocks | 1.128 V | 1.048 V | 8W | 93W |
---|
Overclocked | 1.285 V | 1.150 V | 10W | 126W |
---|
The ASRock Fatal1ty Z97X KILLER consumed a lot of power in the power consumption tests. I was not impressed. I looked up other ASRock models, saw that the Fatal1ty was apparently priced to perform decently, and was left completely embarrassed for ASRock because of the near 30% increase in power consumption. Yet once my overclocked settings were in place, things were actually pretty impressive. So there is lots of potential here. ASRock's BIOS tuning has clearly not yet been optimized for such CPUs as the i7-4770K I use. Manual tuning pays off in spades here.