Power Consumption & Overclocking
Power Consumption
I don’t typically dedicate a whole page to power consumption but there’s a pretty substantial story lurking behind the numbers you see below and how they directly relate to TDP claims from both Intel and AMD. Without getting too technical, the way these two companies go about measuring TDP is fundamentally different from one another.
What you need to know is that TDP values are a universally poor way to determine actual power consumption for end users since they are simply thermal design guidelines that are given to system integrators. As I say in every review, TDP is not actual power consumption so don’t take it as such.
As both Intel and AMD recommend, the best way to measure true power deltas between processors is via a simple (yet calibrated) power meter plugged into the wall outlet. That’s exactly what we do but add in a controlled 120V power input to eliminate voltage irregularities from impacting the results.
Intel's Coffee Lake processors are fabricated with on a refined 14nm++ process node which allows them to attain similar TDP levels as their predecessors despite offering more cores / threads. It seems like that's exactly what happened with these CPUs.
The i5-8400 consumes about as much power as the i5-7500 but it also offers substantially more performance in some instances. The same can be said about the comparison between it and AMD's Ryzen 5 1500X but in that comparison the Intel part comes out well ahead in the performance per watt category.
That i7-8700K also pulls ahead of the 1700-series in perf per watt statistics but not quite as convincingly. It does however offer up some very competitive numbers against the i7-7700K and i7-7740X.
Overclocking Results
Intel’s Coffee Lake processors incorporate a number of additional advances specifically meant for enhancing their overclocking capabilities but many of those were already available in Kaby Lake CPUs. For example, per core overclocking and real time memory latency control were rolled into later BIOS updates for Z270 motherboards but broader PLL trim options are new. However, features like enhanced memory capabilities could allow all of these newer processors to hit record high DRAM frequencies.
I had some initial worries that Intel’s already-high frequencies would directly lead to limited overclocking headroom, especially when syncing all cores to the same speed. However, my overclocking journey with the i7-8700K was pretty straightforward and all that was needed for a stable 5GHz on all cores was a multipler change and a bump in CPU voltages to 1.40V in the BIOS. ASUS’ own internal Dual Intelligent Processors 5 program actually allows you to do this with the click of a button but I found it did increase the voltages a bit more than necessary. Basically, the process followed closely in the footsteps of Kaby Lake.
Going beyond the 5GHz mark wasn’t too much of a problem either and even though I was using a Noctua NH-U12S with the single fan’s speeds set at a constant 60%, temperatures never went above 80°C. I had set a hard limit of 1.45V and with that, the best I was able to achieve on all cores with complete 24/7 stability was 5.145GHz. That’s a pretty respectable result when you consider this particular i7-8700K hovered around 4.325GHz
(note a previous version of this review incorrectly showed 4.1GHz here) with all its cores loaded.
Meanwhile memory overclocking seemed to be limited by the kit being used rather than any controller bottleneck. I was easily able to hit 3900MHz, though timings needed to be loosened a bit. With that added to the already-impressive CPU frequency, every program received a substantial performance improvement. I’ll be exploring how this stacks up to some older processors (read: Sandy Bridge) next week. Until then, I think the results speak for themselves; while the lack of overclocking on Intel’s i5-8400 is disappointing, the unlocked Coffee Lake processor shouldn’t have any trouble consistently hitting clock speeds of 5GHz and beyond.