Almost two years after releasing its first Ryzen 5000 desktop processors, AMD is finally ready to follow. Today, the company announced pricing and availability for the first wave of Ryzen 7000 processors based on the Zen 4 architecture, along with more details on the accompanying AM5 platform and the performance increases to which early adopters can expect.
The first four Ryzen 7000 processors will be available on September 27, and AMD is using the same strategy it used to launch the 5000 series (if you’re wondering about the skipped number, 6000 series processors are only available for laptops) . It starts with four high-end and more expensive parts, while lower-end processors for mainstream and budget versions will follow next year.
|CPU||MSRP||cores/threads||Clocks (Base/Boost)||Full Cache (L2+L3)||PDT|
|Ryzen5 7600X||$299||6c/12t||4.7/5.3GHz||38MB (6+32)||105W|
|Ryzen 7 7700X||$399||8c/16t||4.5/5.4GHz||40 MB (8+32)||105W|
|Ryzen 9 7900X||$549||12c/24t||4.7/5.6GHz||76 MB (12+64)||170W|
|Ryzen 9 7950X||$699||16c/32t||4.5/5.7GHz||80 MB (16+64)||170W|
AMD is sticking to the same core count it used for Zen 3. The entry-level model is the 6-core Ryzen 5 7600X, launching for the same $299 the 5600X cost in 2020 ; the 12-core Ryzen 9 7900X is also launching for $549, the same price as the Ryzen 9 5900X. The other two chips are a bit cheaper than their Ryzen 5000 counterparts; the 16-core Ryzen 9 7950X launches for $699, $100 less than the 5950X, while the 8-core Ryzen 7 7700X starts at $399, $50 less than the launch price of the Ryzen 7 5800X (technically , that is a price increase over the $299 Ryzen 7 5700X, but that chip was only released almost a year and a half after the 5800X).
AMD claims that “optimizations” made during Zen 4’s development increased its instruction-per-clock (IPC) increase over Zen 3 to an average of 13%, compared to the 8-10% increase promised by AMD. company earlier this year. . The 7950X’s maximum clock speed has already increased to 5.7 GHz, 800 MHz faster than the Ryzen 5950X’s boost clock. All told, this should make the 7950X average 29% faster than the 5950X for tasks that benefit from single-threaded performance, including gaming.
At its launch event and in a Q&A session for media and analysts afterwards, AMD was hesitant to get too far into the weeds on Zen 4’s architecture and pointedly walked away projections on when we might expect more Zen 4 chips to launch. But don’t expect 3D V-Cache versions of Zen 4 Where low-end and cheaper Zen 4 processors until 2023.
Performance and energy efficiency gains
We’ll learn more about the changes to Zen 4’s architecture by the time the processors launch, but the company has shared some details about where the performance and power efficiency improvements come from.
AMD CTO Mark Papermaster says Zen 4 is a revision of the Zen 3 architecture that focuses primarily on the “front-end” of the architecture to more efficiently fetch and hand off tasks to the engine. improved execution that was central to Zen 3 (Papermaster also says Zen 5 will be a more substantial overhaul, but we don’t expect to hear many details until 2023 or 2024). Most of Zen 4’s 13% CPI boost comes from these optimizations, while branch prediction, a doubled L2 cache, load/storage improvements, and other small engine tweaks execution account for the rest.
Specific tasks such as machine learning and AI workloads can also benefit from the introduction of AVX-512 extensions. This puts Intel in an odd position – the company defined these extensions nearly a decade ago and was the only one pushing them for years. But it has disabled AVX-512 support in its 12th generation processors because CPU efficiency cores don’t support it. These extensions have been a bit controversial because their use can consume a lot of power and because the workloads that benefit from them are specialized and relatively rare (Linux creator Linus Torvalds said that he hopes “AVX-512 dies a painful death”). But it’s kind of funny that AMD’s latest processors now support them while Intel, the company that invented them and pushed to popularize them, sells processors that can’t.
Even with added AVX-512 support, AMD claims that a Zen 4 core and its accompanying L2 cache take up 50% less area than one of Intel’s current-gen P cores (though whether that’s at least partly because you’re comparing TSMC’s 5nm manufacturing process to the older Intel 7 process, and partly because a Golden Cove core has 1.25MB of L2 cache while a Zen 4 has 1 MB flat). AMD also claims that a Zen 4 core is “up to 47% more energy efficient” than a Golden Cove core.
AMD also makes big claims when comparing Zen 4 to the previous generation Zen 3, especially when it comes to performance per watt. Comparing the Ryzen 9 7950X to the Ryzen 9 5950X at the same TDP levels, AMD claims Zen 4 should outperform Zen 3 by around 35% when set to 170W TDP, around 37% when when set to a TDP of 105W, and by a whopping 74% when set to a TDP of 65W.
This kind of efficiency improvement is important because CPUs shipped in pre-built OEM systems often use these lower stock TDP levels rather than the boosted TDP levels that are possible with custom systems and motherboards. more complete. Higher efficiency is also handy for mini-ITX systems, where you might not have the cooling capacity to let the CPU consume tons of power and generate tons of heat.