Doom's Vulkan renderer might not put the most stress on a single core, but when a game runs this fast, interesting things can still happen. We tested Doom with all of its settings maxed at 2560x1440 using the GeForce GTX 1080 Ti Founders Edition graphics card.
Vulkan is usually the great equalizer for Doom performance, so there's not a ton of difference between the slowest and fastest CPUs in this test. Having more cores on tap is still good for a small boost in average frame rates for our many-threaded chips, though. Surprisingly, the Core i7-7820X delivers the worst 99th-percentile frame times in this test by a wide margin. In absolute terms, a 14-ms 99th-percentile frame time isn't the worst thing in the world, but it's a bit lackluster in this company.
Our "time-spent-beyond-X" graphs can be a bit tricky to interpret, so bear with us for just a moment before you go rocketing off to our productivity results or the conclusion. We set a number of crucial thresholds (or bins) in our data-processing tools—50 ms, 33.3 ms, 16.7 ms, 8.3 ms, and 6.94 ms—and determine how long the graphics card spent on frames that took longer than those times to render. Any time over the limit ends up aggregated in the graphs above. Those thresholds correspond to instantaneous frame rates of 20 FPS, 30 FPS, 60 FPS, 120 FPS, and 144 FPS, and "time spent beyond X" means time spent beneath those respective frame rates. We usually talk about these results as a proportion of the one-minute test runs we use to collect our data. If that's still too much to bear, just understand that more time spent in these graphs means worse performance.
If even a handful of milliseconds make it into our 50-ms bucket, we know that the system is struggling to run a game smoothly, and it's likely that the end user will notice severe roughness in their gameplay experience. Too much time spent on frames that take more than 33.3 ms to render means that a system running with traditional v-sync on will start running into equally ugly hitches and stutters. Ideally, we want to see a system spend as little time as possible past 16.7 ms rendering frames, and too much time spent past 8.3 ms or 6.94 ms is starting to become an important consideration for gamers with high-refresh-rate monitors and powerful graphics cards.
While all of our chips spend a little time holding up the graphics card beyond 16.7 ms, those holdups should be practically invisible. For the fast-running Doom, the more interesting results can be found past 8.3 ms and 6.94 ms. The Threadripper 1950X and the Ryzen 7 1800X lead the pack here, while the i9-7900X, the i7-6950X, and the Threadripper 1920X trade blows. The i7-7820X spends over a second under 90 FPS overall, thanks to a strange (and repeatable) pattern of stuttering that's absent from the Core i9-7900X. These stutters mean a gamer could have a potentially less smooth experience on the i7-7820X than any other chip here.
The same story continues past 6.94 ms, although our contenders are more evenly matched here. The Threadrippers and the Core i9-7900X all spend about four seconds of our one-minute test run on tough frames that take more than 6.94 ms to render, but the i7-7820X's curious performance means it spends another second yet on that hard work—and it's therefore the least smooth chip here. Just goes to show that even in this largely GPU-bound test, the CPU can still matter.
|Razer Kiyo and Seiren X set the stage for streaming excellence||7|
|MSI Cubi 3 Silent and Silent S can be seen but not heard||6|
|Massdrop's Vast 35" VA display lives up to its name||17|
|Spitballing the performance of Nvidia's purported GTX 1070 Ti||15|
|Friday deals: a huge monitor, racing gear, audio, and more||17|
|G.Skill 3800 MT/s SO-DIMMs put lightning in tiny bottles||7|
|Cooler Master bedazzles the MasterLiquid Lite ML120L and ML240L||3|
|Razer Electra V2 offers affordable immersion||6|
|Samsung 360 Round camera captures the world from all angles||11|