Below are the temperatures for each of our Kaby Lake CPUs at stock settings – without any overclock applied. We’ve also compared to previously released Broadwell-E and Skylake counterparts (which we’ve also benchmarked against on the subsequent pages too):
For the 7700K we tested out ES and retail CPUs to gauge whether there was much of a different in terms of thermals and overclocking limit. There were marginal differences, if any.
For this part of the review the maximum overclocked was achieved by using the Corsair Hydro Series H100i GTX.
The objective here is to obtain the best result by either altering just the ratio or altering both the ratio and BCLK. For us, just adjusting the ratio and leaving the BCLK on auto (100MHz) realised the best results.
To simulate maximum load we use AIDA64. Unlike games and other benchmarks which only push the system to a certain point, AIDA64 stability test pushes temperatures to the absolute maximum.
Intel Core i5-7600K
We managed to overclock the 7600K from its base clock of 3.8GHz to 5.1GHz by adjusting the ratio/multiplier to 51. This is more than impressive and with such a significant boost to the clock speed this will give us a performance advantage in the benchmarks to come. To reach this point, our chip required a substantial voltage increase – 1.4v.
Intel Core i7-7700K
Turning to the i7 processor, we were able to hit 5GHz but this milestone wasn’t without it’s problems. The thermal threshold peaks out in the high 90Cs and this therefore isn’t an achievement which would allow for 24/7 usage. In light of this, we also nudged the clock speed to 4.9GHz and established the lowest voltage to attain this frequency.
As you can see, 5GHz required 1.42v and we managed to reduce the voltage of the 4.9GHz overclock to 1.38v.
On the subsequent pages we’ll be benchmarking both CPUs at stock and overclocked settings, showing you what you can expect out of the box and what can be assumed with a substantial overclock applied.