04:15NVIDIA Ampere (RTX 30) Architecture Deep-Dive: RT Cores, GDDR6X vs. GDDR6, & More
From the creator
We're deep-diving into the NVIDIA RTX 3080 and 3090 architecture (Ampere), prior to our RTX 3080 review going live in the next video. This looks at memory, RT cores, and arch changes. Sponsor: Get the Thermaltake Core P3 on Amazon - https://geni.us/Yf93 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 In this video, we're talking about NVIDIA's Ampere architectural changes versus Turing (and, to some extent, Pascal). We'll be covering the block diagrams for SMs and the GPU on the whole, but also talking about the major changes to RT cores and to the Tensor logic. Memory and memory bandwidth are also major performance drivers, and get additional discussion. This video will help you understand how the RTX 3080, 3090, 3070, and other NVIDIA Ampere (30-series) GPUs work at a lower level, and will help you understand why the performance changes are what they are. You'll have to check back for the review on the 16th for benchmarks. TIMESTAMPS 00:00 - NVIDIA Ampere Architecture Overview 01:50 - NVIDIA Process Changes & 8nm 05:05 - Ampere SM vs. Turing SM Differences 12:56 - Ampere L1 Cache Changes 14:55 - Ampere 3rd Gen Tensor Cores 19:14 - Changes with 2nd Gen RT Cores (3080 vs. 2080 RT) 24:14 - Quick Summary 24:44 - GDDR6X vs. GDDR6 & Memory Improvements 28:07 - Conclusion: A Lot More to Cover, But... ** Please like, comment, and su
Published 2020-09-16 · Watch on YouTube ↗
Availability and playback restrictions are controlled by the original channel.





