Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Skylake Overclocking
What's causing this?
Quote:In this article we've had a closer look at what effect overclocking an Intel Core i5-6600K to 4.5 GHz has regarding gaming performance. Default this CPU clocks at 3.9GHz and therefore bumping the frequencies to 4.5 GHz makes for an overclocking of 15.4%. On average - over all different resolutions - this accounts for a 2.05% performance bump. This means that the difference is actually measurable but since the scaling is not linear (15.4% clock boost doesn't equal 15.4% performance boost) you really don't have to expect miracles from overclocking your CPU.

What's been pretty interesting to observe is tha fact, that Crysis 3 responds best to overclocking our test CPU. Using our 2160p preset we see a 15% jump in performance. In all other games, there is virtually no performance gain measurable. The Crysis 3 result actually puts a bit of a bias in our overall results since it's basically a statistical outlier.

Apart from the performance benefits overclocking your CPU means your system will burn more power. If you keep things like Intels Turbo active, then the CPU will downclock, when there is no load and the CPU will also change the voltage to a lower value. This results in good energy efficiency. For these test we simply bumped the multiplier to 45 and we didn't change any other parameter, leaving the motherboard and CPU in control of changing voltages. In idle are 46W (stock) and 50W (4.5 GHz), which is a rather small difference. When the CPU is running under load at 4.5 GHz we measure 127W from the wall, whereas the system in idle need 124W. This difference is surprisingly small and therefore it doesn't really matter whether you run an i5-6600K at stock clocks or overclocked to 4.5GHz in terms of power consumption.

Apparently overclocking is fun, but when it comes to the plain facts regarding gaming performance, we see that there is almost no gain.
It's because of bad testing on their part. Crysis 3 is a very demanding game in terms of the CPU. In all of their other tests, they were not CPU bound. They were limited by the GPU. So that's why there was not a performance boost. I'm actually shocked that a tech publication did not realize this.
It's been a long time now that the performance falls on the Graphics now and not the CPU.
Now these same people can't find a meaningful difference between the i5-6600K and the i7-6700K.

Forum Jump:

Users browsing this thread: 1 Guest(s)