Single page Print

Our testing methods
Here are the specifications of our test system:

Processor Intel Core i7-6700K
Motherboard ASRock Z170 Extreme7+
Memory Corsair Vengeance DDR4-3000
Graphics card Sapphire Radeon R9 380X
Storage OCZ Vector 180 480GB SSD
WD Black 1TB HDD
Power supply Aerocool P7-850
CPU cooler Cooler Master MasterLiquid 120
OS Windows 10 Pro

Our thanks to Intel, ASRock, OCZ, Sapphire, WD, Aerocool, and Cooler Master for their contributions to our test system. Our thanks to Corsair for providing the case and memory we're testing today, as well.

Our case-testing cycle consists of the following phases:

  • 10 minutes idling at the Windows 10 desktop
  • 10 minutes running the Prime95 CPU torture test
  • 10 minutes running the Prime95 CPU torture test and the Unigine Heaven GPU torture test
  • 10 minutes idling at the Windows 10 desktop

Our testing methods are generally publicly available and reproducible. If you have questions about our methods, feel free to hit us up in the comments or join us in our forums.

The ambient temperature in my office at the time of my tests was 68° F.

Cooling performance
Here are the results of our cooling tests, plotted over time:

And here are some minimum and maximum temperatures from each testing phase:

For the first time in my case-testing experience, the Core i7-6700K CPU I use for testing got so hot in the Spec-04 TG that it had to throttle during the full-load phase of our cycle. I’m guessing that’s because the only place I could mount the radiator was at the front of the case, so all of the waste heat from the CPU was directed back into the enclosure.

Without an exhaust fan on the top or rear of the case, this waste heat seemed to increase the temperature of all the components inside, and the passive ventilation offered by the Spec-04’s unoccupied top fan mounts wasn’t sufficient to let the waste heat exit the case through convection.

To isolate this issue, I dismounted the radiator from the front panel and let it dangle in open air. In this configuration, CPU temperatures under our Prime95 load dropped roughly 10° C from the maximums I observed with the radiator installed behind the front panel, suggesting that the two layers of mesh at the front of the case could actually be reducing airflow to the front fan mounts.

After completing our regular test cycle, I moved the preinstalled 120-mm intake fan to the rear exhaust position and restarted our full test load of Prime95 and Unigine Heaven. After ten minutes of this load, the CPU reached peak temperatures of 92° C. That’s still quite high for a stock-clocked Core i7-6700K in any modern case, but at least the processor wasn’t throttling.

Overall, I’d say that an exhaust fan of some kind is absolutely needed for the best performance from this enclosure. I’d consider relocating the included intake fan to the rear mount if it’s not otherwise occupied. Even with that help, however, the Spec-04’s cooling performance isn’t up to the standard we’ve come to expect from modern cases.

Noise levels
Before we consider noise levels from the Spec-04, it’s worth noting that the noise floor in my office was lower than usual during my tests, so the Spec-04’s noise measurements will benefit from that variance. Prior to the tests, my smartphone app read about 27 dBA, where other tests averaged about 33 dbA. I’ll chalk the difference up to colder weather and fewer people moving about in my apartment building.

When the system is powered on, it makes itself known through a light and undisturbing hum from the fans inside at idle. Under load, though, the Spec-04 almost reaches out and grabs you to let you know it's under stress. At some points around the case, noise levels increased by 10-11 dBA while the system was under load. When we account for the advantage that the difference in noise floor affords the Spec-04, it's neck-and-neck with the Aerocool P7-C0 as one of the louder cases I've ever tested, despite the fact that the P7-C0 Pro had three more 120-mm fans whirring away inside.