
Automated System Latency Test Methodology (NVIDIA Latency Analyzer Validation)
This video serves a referential deep-dive for our future total system latency (we're coining "TSL") testing methodology, including NVIDIA Latency Analyzer/Reflex validation and LDAT testing.
Sponsor: Teamgroup T-Force Vulcan 1TB SSD (Amazon: https://geni.us/vIKHLE)
The GN Tear-Down toolkit sold out in 3 days! If you want to back-order the next round, grab one here: https://store.gamersnexus.net/products/gamersnexus-tear-down-toolkit
We won't be running an article for this one, unfortunately, as it's far too long for us to find the time to work into a written piece, and also involves too many visual cues that require movement to really deliver the message well. Sorry!
This is a testing methodology deep-dive that reviews and analyzes NVIDIA's new LDAT tool for latency testing. Specifically, the tool is meant as an easier-to-use replacement for the traditional high-speed camera system -- but because NVIDIA makes it, and because it can be used to test NVIDIA and competing products, we need to perform first-party validation on its trustworthiness. Fortunately, this tool can (1) be completely isolated such that it does not have any means of knowing what GPU is installed in the system, and (2) can be validated against standard high-speed camera approaches. We recently ran a piece where we conducted hundreds of test passes on a 1000FPS camera, and behind-the-scenes, we were also conducting those test passes on LDAT. It qualified. In this piece, we reran recent Google Stadia latency benchmarks (people often call this "input latency," but it's really end-to-end total system latency) in order to verify that LDAT doesn't do anything tricky to cheat data. In our testing, it appears that LDAT produces clean, untweaked data that is just as reliable as high-speed camera methods, but much faster. Further still, our validation testing in this content shows the impressive accuracy of the "old-school" method of frame counting even with all the many variables in between, like slower refresh rates on some cameras (our previous method used a 240FPS camera, and then we updated to a 1000FPS camera for the CPU comparisons).
Google Stadia latency testing: https://www.youtube.com/watch?v=hVTsj66g9bA
TIMESTAMPS
00:00 - Latency Testing is Painfully Slow
01:15 - Traditional Latency Testing & NVIDIA's Open-ness
06:31 - Test Diagrams & Mouse Switch Tear-Down
09:29 - NVIDIA LDAT Isn't Sabotaged
13:24 - Validating NVIDIA's Claims vs. High-Speed Camera
17:43 - Latency Terminology & What This Does
21:56 - Stadia Validation with Metro Exodus
23:43 - Destiny 2 Validation vs. Camera
24:55 - Tomb Raider Validation vs. Camera
28:07 - LDAT is Powerful Enough to Create Bad Data
37:22 - LDAT Conclusion & Usefulness
Like our content? Please consider becoming our Patron to support us: http://www.patreon.com/gamersnexus
** 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:
t: http://www.twitter.com/gamersnexus
f: http://www.facebook.com/gamersnexus
w: http://www.gamersnexus.net/
Editorial, Testing: Patrick Lathan
Editorial, Host: Steve Burke
Video: Keegan Gallick, Andrew Coleman
Other Videos By Gamers Nexus
Other Statistics
Counter-Strike: Source Statistics For Gamers Nexus
There are 10,511,297 views in 25 videos for Counter-Strike: Source. Counter-Strike: Source has approximately 10 hours of watchable video on his channel, roughly 1.15% of the content that Gamers Nexus has uploaded to YouTube.