Just be aware that clock speed is not an indication of the total number of IPS or OPS a computer can perform. AMD chips like the K5 and Athlon used to lag behind Intel chips for over a decade with respect to raw clock speed, but could perform more instructions and operations per second than their counterpart; thus, they were more powerful even at the lower clock speed.
So architecture here is going to be far far more important.
The Ryzen architecture is being compared to the 5960X, which has an average overclock around 4.4-4.5 Ghz; some folks can even get 4.7 out of it. 8 Haswell-E cores at 4.5 Ghz with hyper-threading is a vastly superior platform than 4 Skylake/Kaby Lake cores with HE at 4.8-5.0 Ghz.
The performance difference here is also interesting. You can see between 10-20% faster single-threaded performance across two max clocked Skylake vs Haswell-E chips; however, the vast majority of that performance difference is not in the clock but instead in the architectural difference between Skylake and Haswell-E. You can see this by comparing a 5960X directly against the 4790 using identical clocks as mentioned above and noting that there's only about a 6-7% difference due to the higher clocks; yet again, the Haswell-E offers 100% greater workload capacity.
If we're talking about games, then a game that is highly parallelizable, it will run vastly better on 8 cores than 4.