AMD to disclose roadmaps in mid-November — The future of Zen 6, RDNA, CDNA, and UDNA expected

AMD has confirmed that it will host its 2025 Financial Analyst Day on Tuesday, November 11, 2025, in New York City. The company plans to unveil details of its technology and product roadmaps at the event beyond what it already disclosed at its Advancing AI event earlier this summer. 

"The event will feature presentations from AMD's executive leadership team, highlighting the company's strategy and growth opportunities, innovative product and technology roadmaps and long-term financial plan." 

At its Financial Analyst Days (FADs), AMD traditionally lays out multi-year strategies and product roadmaps, often going well beyond what it reveals at product launches or technical deep-dives. Therefore, expect AMD to lay out its CPU, GPU, FPGA, and DPU product roadmaps for several years to come. 

On the CPU front, AMD has already disclosed that its next-generation EPYC processor, codenamed Venice, with up to 256 cores, is based on the Zen 6 microarchitecture and is due in 2026. The company has not officially talked about its Zen 6-based CPUs for client systems, so expect it to reveal more details about them this November. 

AMD has never disclosed whether its 2027 EPYC processor codenamed Verano will be based on the Zen 6 or Zen 7 microarchitecture, so expect it to talk about Verano at its FAD as well.  In fact, considering the nature of the Financial Analyst Day event, expect AMD to talk about Zen 6, Zen 7, and Zen 8 offerings there, but do not expect too many disclosures for products that are years away. 

On the AI and HPC side of matters, AMD already disclosed that its Instinct MI400-series accelerators are based on the CDNA Next architecture (CDNA 5?), so expect the company to talk about its following-generation Instinct MI500-series accelerators, presumably based on the UDNA 6 architecture (as the RDNA 6 is expected to be the point when it merges with CDNA branch of GPUs), as well as rack-scaled solutions beyond Helios that use UALink interconnects for scale-up connectivity. 

In addition, the company may reveal more information about its 'Vulcano' 800 GbE network cards compliant with the UEC 1.0 specification, as well as their successors that will be widely used late this decade and early next for Ultra Ethernet-connected AI and HPC data centers. 

Since AMD's AI and HPC Instinct-series GPUs rely on some of the most advanced packaging technologies available, expect AMD to unveil its packaging roadmap that will span towards the end of the decade at least. 

As for Radeon GPUs for client applications, AMD will probably talk about its RDNA 5 GPUs and UDNA 6 products due over the next few years, as well as its opportunities with game consoles. 

Also, AMD will certainly talk about its long-term financial plan, which includes what kind of growth it expects in its main businesses — PCs, gaming, data centers, programmable, and embedded applications. For analysts and investors, the most important part will be the company's guidance on how it intends to expand its AI and data center share against Nvidia, custom silicon of hyperscale cloud service providers, and the emerging army of AI inference accelerators while sustaining profitability.

Follow Tom's Hardware on Google News to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button.