#I5 4690k novabench tests windows#

The Pentium G3258 on the other hand was way down the table with 1,579 points overall compared to 2,328 for the Core i5. It was actually fairly close to toppling the Core i7-3770K, despite the latter sporting Hyper-Threading and four extra virtual cores as a result and did manage to better the Core i7-2600K, being over 200 points clear of the Core i5-2500K overall. Overall, in the Media Benchmarks it was a good chunk faster at stock speed. The extra 100MHz stock and Turbo-Boosted frequencies saw noticeable gains over the Core i5-4670K across the board. In fact, we regularly saw similar temperatures around 4.8GHz and 1.3V so it's clear Intel's new thermal paste seems to be doing a good job. Needless to say, 1.4V at 4.9GHz and barely touching 90☌ is a lot cooler than you'd see a Core i5-4670K running at. In the end, the exact same voltage as we used with the Core i7 proved to be the lowest stable setting, although with decent water cooling, it may be possible to get slightly higher than we achieved here with our Corsair H80i. We gradually ramped up the voltage in an attempt to get 4.9GHz rock steady but even at a toasty 1.4V, which saw the load temperature breach 90☌ in Prime95's smallFFT's test, it still wasn't totally stable. However, Prime95 lasted just a few seconds before we were met with a BSOD. We headed back into the EFI to try for 4.9GHz and were surprised to see it boot okay.

#I5 4690k novabench tests windows#

Using the same voltage and 48x multiplier, we booted into Windows first time and 100 per cent stable too. The press sample Core i7-4790K managed 4.8GHz using a vcore of 1.33V so we used that as a benchmark for the retail Core i5-4690K. Overclocking We should note that we haven't received a press/engineering sample from Intel so this is a retail CPU so the overclocking potential should be similar to what you can expect if you went out and bought one now.