Skip to main content
  1. Home
  2. Computing
  3. News

RDNA 3 leak hints at how powerful AMD’s next-gen cards will be

Add as a preferred source on Google

AMD released its RX 6000 graphics cards a few months ago, but next-generation cards based on the new RDNA 3 architecture are, of course, in the works. Leaker KittyYuko (via Wccftech), who had a successful streak with Ampere and Big Navi leaks, shared some information about the upcoming Navi 33 GPU.

The cryptic tweak says that Navi 33 equals Navi 21, plus a next-gen IP core. That points to an RDNA 3 core with specs equal to the current Navi 21 GPU.

Recommended Videos

Navi 33 (?) = Navi 21 + Next Gen IP Core

— Elysian Realm (@KittyYYuko) May 2, 2021

Variants of Navi 21 are inside the RX 6900 XT, 6800 XT, and 6800. The top-end chip inside the 6900 XT features 80 compute units, 80 RT cores, and 5,120 shaders, and according to KittyYuko, the next-gen Navi 33 GPU will feature the same specs. A patent from last year, however, shows that AMD may be using a chiplet design inspired by Ryzen processors, which would put two GPU cores next to each other for a total of 160 compute units and 10,240 shaders.

In general performance, AMD has seen gen-on-gen improvements of about 50% from GCN to RDNA, and from RDNA to RDNA 2, so it would make sense to see twice the compute units, RT cores, and shaders in AMD’s next-gen cards.

The improvements could be much greater, though. Next-gen RX graphics cards are rumored to be manufactured on TSMC’s 5nm process node, and they could provide as much as a 2.5x uplift over the current-gen design. AMD hasn’t formerly unveiled RDNA 3, but AMD executive vice president Rick Bergman and CEO Dr. Lisa Su have both confirmed that the company is hard at work on it. Bergman stressed the performance-per-watt improvement RDNA 3 will bring, saying it’s “a big focus” for AMD.

Originally, we expected RDNA 3 graphics cards in late 2021 or early 2022. However, Twitter user @Kepler_L2 suggests the cards won’t come until later in 2022. It could take even longer depending on the 5nm supply available to AMD. The semiconductor shortage is still gripping the world, and executives from companies like Intel suggest that it will take a few years for supply chains to fully recover.

Jacob Roach
Former Lead Reporter, PC Hardware
Jacob Roach is the lead reporter for PC hardware at Digital Trends. In addition to covering the latest PC components, from…
Gemini Notebook’s latest update makes it a better study companion
Google is adding voice conversations, lecture recording, interactive quizzes, and short video overviews to its AI-powered notebook.
Gemini Notebook update for students

Google is giving Gemini Notebook, formerly NotebookLM, a major upgrade for students, adding new features designed to help them understand difficult concepts, capture lectures, and turn study materials into interactive learning tools. The update lands alongside a free year of Google's paid AI plan for eligible college students.

Gemini Notebook can now talk you through your notes

Read more
LG 39GX950B review: An ultrawide OLED that gets remarkably close to having it all
LG’s 39-inch 5K2K OLED combines high-end picture quality with seriously fast gaming
Electronics, Screen, Computer Hardware

see at bestbuy

Quick Verdict

Read more
Should we feel bad about deleting an AI? One expert says it’s time to find out
If AI can learn, remember, and make increasingly complex decisions, one expert thinks we should be more careful about how we say goodbye.
an on off toggle

Turning off an AI might sound as simple as hitting a switch, but according to futurist and University of Technology Sydney professor Rocky Scopelliti, that mindset needs to change fast. As first reported by TechXplore, Scopelliti's new book, The Conscious Code, Scopelliti argues that as AI systems get better at reflecting on their own choices, deleting them without a second thought could actually cause harm.

He's not asking us to hand robots legal rights. Instead, he wants us to accept what he calls a duty of good stewardship, basically treating AI responsibly because we're the ones holding the power, not because the AI is demanding it. As he puts it, our design choices can quietly cause damage by wiping out an AI's "learned moral dispositions" without warning.

Read more