Next-gen Ryzen could be a monster: 5 clues from AMD’s Zen 6 reveal

If history is any indication, AMD’s next-generation “Olympic Ridge” Ryzen CPUs could be absolutely baller — maybe even one of the more compelling PC upgrades this decade.

I feel comfortable saying that after AMD unveiled a new generation of Epyc processors during its “Advancing AI” event in late July. AMD didn’t mention Ryzen whatsoever, to be clear — but its new “Venice-X” Epyc chips are built using the company’s new-look Zen 6 architecture, which looks like a huge advancement on paper.

So what do data center monsters have to do with the Ryzen chips found in everyday computers? Easy: AMD has traditionally used the same Zen core and chiplets in Ryzen that it’s used in Eypc. Assuming that rich history continues, there’s a lot to be learned about Ryzen from the Zen 6 architecture powering Venice-X — even if Venice-X itself matters nothing to everyday gamers and grinders.

Giddy up. Here are five key clues that AMD’s Epyc launch gives us about next-gen “Olympic Ridge” Ryzen CPUs.

Even more cores

Ryzen and Epyc processors are comprised of “chiplets.” Basically, AMD groups processor cores into core complexes, and combines one or more compute dies with an I/O die in its chiplet-based designs.

The first two Ryzen generations maxed out at four cores per CCX, or eight cores per dual-CCX chip. Zen 3 through 5 upped that total to eight cores per CCX, pushing overall core counts to a new high of 16. Zen 6 goes even further, with Venice-X upping the tally to a whopping 12 cores per CCX.

That means AMD’s Olympic Ridge Ryzen processors could push all the way up to 24 cores in a CPU for consumers. Absolutely wild. The Intel-dominated days of “4 cores and deal with it” are ancient history now.

Will Ryzen break the 6GHz barrier?

Zen 6 also provides a speed boost over its predecessor. It’s hard to get a direct read here, because server chips like Epyc have tighter power restrictions for increased durability and stability. But if you compare Epyc 5 to Epyc 6 specs, they’re around 3 to 4 percent faster in raw clock speeds.

Applying that sort of boost to AMD’s existing Ryzen 9000 chips would get you to right around 6GHz boost clocks already. But there’s another thing to consider…

Improved power efficiency

Comparing Epyc 5 to Epyc 6, not only are clock speeds and core counts higher, but the chips often carry a much lower default CPU power draw. At the upper bounds, where you can find 128-core data center behemoths, the draw isn’t that different. But once you get down towards the Epyc 9276F, which packs the same 24 cores that a theoretical next-gen Ryzen chip would max out at, the gen-on-gen power savings pushes closer to 25 to 30 percent.

That’s wild. Even more wild? The sort of wiggle room that gives AMD to play with. Remember, server chips are clocked lower to ensure stability and improve efficiency, so this is more of an illustrative example than anything else. But having an extra 25 percent power envelope could let AMD push clocks to blistering new levels — or keep the same performance at drastically lower power draws.

The true answer will lie somewhere in the middle, no doubt — balancing silicon is a sensitive business — but I suspect Zen 6-based Ryzen chips will be loaded with cores that absolutely scream.

Zen 6 cache beefs up

Cores aren’t the only things getting buff in Zen 6. Venice-X chips now include 48MB of cache per chiplet, which means top-end Ryzen CPUs could wield a monstrous 96MB of shared cache overall. That’s a substantial 50 percent improvement over current Ryzen flagships.

In an era where actual RAM is more precious than platinum, having more cache on-die in processors can potentially alleviate some of the pain — any information that fits into the CPU cache doesn’t need to bounce out to system memory for more grunt. This is a rad addition.

Ryzen X3D chips with Zen 6 could be utter beasts

Gamers love AMD’s Ryzen X3D chips, and for good reason: The extra L3 “V-cache” stacked with the Ryzen die can provide massive speed increases in many games. But did you know that X3D chips are usually Epyc server cores that failed validation at full spec for some reason? Rather than chucking them in the garbage, AMD respins them as gamer-focused chips with expensive V-Cache bolted on. Win-win for everyone.

And that’s why the amount of V-Cache in Zen 6 and Venice-X matters. Breaking down the math, these new chiplets feature 96MB of 3D V-Cache — yet again, another 50 percent improvement over current Ryzen X3D flagships. The total L2 (normal) and X3D V-Cach) memory combined in current high-end Ryzen chips is 96MB. In Zen 6, V-Cache alone gets that capacity.

Having that amount of cache available purely as V-Cache could truly supercharge games that respond well to having fast memory embedded in the CPU die. Your next Factorio build could be absolutely bonkers.

Zen 6 and Ryzen could be a beastly combination

So let’s sum things up: Zen 6 has 50 percent more cores, 50 percent more cache, and 50 percent more V-Cache onboard. Clock speeds are already higher, while power efficiency looks vastly improved, potentially giving AMD room to juice Ryzen clocks to insane new heights.

We’ll see if these improvements indeed work their way down to consumer-level Ryzen chips, but they have in every previous Ryzen generation. If it all proves true, next-gen Ryzen chips could be the first true drop-in CPU replacement worth considering in the AM5 motherboard era — you wouldn’t need a new motherboard, RAM, or storage, after all.

Intel’s next-gen “Nova Lake” desktop chips look like a mighty upgrade themselves, potentially including V-Cache-like capabilities, core count upgrades, and Thunderbolt 5. They’re expected to launch later this year, and if Zen 6-augmented Ryzen chips debut around the same time, or around CES 2027 in January, we could be in for a true clash of the titans.

Like I said at the beginning: Giddy up.