Publicado em 2 de mai. de 2020
This overviews our CPU testing methodology for 2020, including what’s being used on AMD’s R3 3100 & 3300X, Intel’s i9-10900K, and more. Ft. code compile, H.265 Handbrake, games, & more.
Sponsor: Buy Thermal Grizzly Conductonaut on Amazon (https://geni.us/1D8Eeq) or Hydronaut paste (https://geni.us/Fsray)
Benchmarking begins again for our CPU test benches. We’ve completely overhauled our test suite from the past year, and this one will be in use for at least a year going forward. We typically get about 12-14 months of use out of our CPU test suites, but we update periodically along the way. We have one major bench overhaul annually, then we have minor updates (like important Windows changes, GPU driver updates, etc.) throughout the test period. These are easier to manage than the full overhauls, which require a few months of validation and data analysis to ensure we’re producing numbers that are useful. In this methodology piece, we’re going to show you some of the ‘bad data’ that we’ve produced (so that viewers can better understand why we toss some benchmarks), then show how we corrected that going forward.
In addition to all of this, we’ve added a whole host of additional validation reports before and after our test suites finish, which allow us to keep an eye on the bench setup to ensure everything remained as expected. This would include memory timings, which we manually control, but also includes CPU or BIOS changes.
New tests include code compile with Chromium, by very popular demand, Handbrake H.264 to H.265 transcode benchmarks, and returns of several from last year (but updated). We also included some standard deviation charts to help illustrate run-to-run variation and to also illustrate a method of catching bad data. We try to determine the best CPU for Adobe Premiere and video editing, Adobe Photoshop, transcoding, code compile/programming, and gaming in our benchmarks, among other things. Although our reviews will go heavy on the AMD vs. Intel CPU performance scaling, in this one, we’re focusing on the ‘how’ and ‘why’ of benchmarking.
The best way to support our work is through our store: https://store.gamersnexus.net/
Like our content? Please consider becoming our Patron to support us: http://www.patreon.com/gamersnexus
00:00 – Introduction, Bad Data, & Reasoning
03:50 – Memory Selection & Important Considerations
05:20 – Cooler Selection & HEDT Cut-off
06:26 – Intro to New Tests: Code Compile & Chromium
08:18 – Chromium Compile Sample Chart
10:43 – Compile & Blender Thread Utilization
11:39 – Handbrake H.264 to H.265 Transcode Benchmarks
14:00 – Total War Three Kingdoms FPS & Standard Deviation
17:50 – Explaining Test Error & Reality
19:18 – Total War Frametimes & Consistency
20:40 – Total War Campaign Benchmark (FPS)
22:00 – The Division 2 Benchmark & GPU Challenges
25:15 – F1 2019 for Real-World Representation
27:00 – Red Dead Redemption 2 Dx12 vs. Vulkan Bench
29:20 – Adding More Medium Tests for Scaling
30:16 – Data Validation & Error Prevention
32:30 – Adobe Photoshop, Premiere Return
33:52 – Returning Games from 2019 Methodology
35:11 – 7-Zip, V-Ray, Blender 2.81 Cycles
36:18 – Power Testing, Conclusions
** Please like, comment, and subscribe for more! **
Links to Amazon and Newegg are typically monetized on our channel (affiliate links) and may return a commission of sales to us from the retailer. This is unrelated to the product manufacturer. Any advertisements or sponsorships are disclosed within the video (“this video is brought to you by”) and above the fold in the description. We do not ever produce paid content or “sponsored content” (meaning that the content is our idea and is not funded externally aside from whatever ad placement is in the beginning) and we do not ever charge manufacturers for coverage.
Follow us in these locations for more gaming and hardware updates:
Editorial, Test Lead: Steve Burke
Editorial, Testing: Patrick Lathan
Video: Keegan Gallick, Andrew Coleman