The live streaming presentation of AMD’s “forward PCs together” ended some time ago when AMD CEO Dr. Lisa Su laid the groundwork for the release of the next generation of AMD Ryzen desktop processors. Building on the AMD Ryzen 7000 announcement at Computer 2022, the highly anticipated unveiling showed AMD’s release plans for their first processor family based on the Zen 4 architecture, in which AMD will kick things off with a quartet of enthusiast-centric chips. The chips will be released in just over four weeks, on September 27, and AMD is expected to take the performance crown in virtually every category of PC processors, from gaming to content creation, with ease.
Advancing the benefits of this latest generation of AMD’s desktop CPU is a combination of architectural enhancements that support the Zen 4 architecture, as well as moving CPU core chip manufacturing towards the state-of-the-art 5nm process. TSMC. The combination now gives AMD what they say is a 13% increase in IPC over their Zen 3 architecture – an 11% increase according to Computer – as well as a significant increase in CPU clock speed. The high-end Ryzen 9 7950X has a maximum turbo clock of 5.7 GHz, 800 MHz (16%) more than the equivalent Ryzen 9 5950X. As a result, AMD expects a 29% build increase in single-threaded performance and even more for multi-threaded workloads.
Along with the new Ryzen 7000 series chips, AMD’s AM5 platform will be released, which will be the cornerstone of AMD’s consumer desktop platform until at least 2025. AM5 offers DDR5 and PCIe 5.0 support for AMD chips, as well as a new LGA connector. Four chipsets are planned so far – X670 Extreme, X670, B660 Extreme and B660 – with X670 series cards available for September release and B660 series cards in October.
Ryzen 7000 family: 4 Zen 4 SKU
To casual AnandTech readers and AMD followers, AMD’s product stack for next month’s launch should look very familiar. AMD has done very well starting with an enthusiast desktop version built around four high-end processors and with the Ryzen 7000 generation, AMD is no different.
However, AMD is not disclosing further details on the integrated RDNA2 architecture GPU at this stage. So even though all Ryzen 7000 chips have an iGPU, we still don’t have anything to talk about. However, AMD has made it clear in several places that the iGPU in these desktop chips is a relatively small configuration for basic desktop work and is not designed to be a powerful GPU like AMD’s APUs.
Ryzen 9 7950X: the fastest processor in the world
Topping the package is the Ryzen 9 7950X. The best of AMD’s Zen 4 architecture uses two fully compatible CCDs that produce a total of 16 CPU cores. Thanks in large part to TSMC’s 5nm process, it also has very high clock speeds; The base clock speed alone is 4.5 GHz and the turbo clock speed is up to 5.7 GHz on a single thread. The latter is actually 200 MHz more than AMD’s Computex presentation, where the company only managed 5.5 GHz.
AMD claims this chip contains 80MB of cache. It is split between 64MB of L3 cache – 32MB in each CCD – and 1MB of L2 cache, associated with each core of the Zen 4 CPU. This is the same amount of L3 cache in the Ryzen 5000 / Zen 3 family, but the amount L2 cache per core has doubled compared to the previous generation.
The 7950X will also be AMD’s most energy-hungry mainstream Zen desktop chip. The chip has a TDP of 170W, and as we know from previous AMD announcements, the Power Package Tracking (PPT) limit for these chips is 230W. This is one of the main reasons AMD makes such significant improvements to MT performance. . their average ST performance additions, because previous Ryzen 5000 chips were often limited by TDP when all CPU cores were heavily loaded, so they failed. higher clock speeds than they might otherwise support.
Since AMD is the first CPU vendor with this generation of components, the 7950X has no direct competition, although if you’re considering an Intel Core i9 12900K, AMD will sell you a 7950X upsell instead. According to AMD’s numbers, a full-fledged Zen 4 chip should easily be the most powerful desktop CPU in the world, which we’ll have to confirm in a review, of course, but that’s not hard to believe given AMD’s updates and architecture improvements. . TSMC’s 5nm process. However, it will fight a little more with Intel’s Raptor Lake processors once they are released.
AMD somewhat surprisingly priced its new flagship chip at $ 699. It’s actually $ 100 cheaper than the 5950X’s launch price at the end of 2020, which was $ 799. The chip’s higher price isn’t. been particularly well received, but AMD sold everything it could do for a while. However, the chip shortage has decreased and AMD has more access to both wafers and platforms, and the company will not be without competition for the life of the chip.
Ryzen 9 7900X
These are two other CCD parts, but each CCD only uses 6 cores, for a total of 12 CPU cores and 24 threads. This part also holds the entire 64MB L3 cache that comes with the 7950X, which is identical to how the L3 cache resizes (or rather does not) on the Ryzen 5000 series.
With the TDP set to the same 170 W as the 7950X, the 7900X increases the base clock speed slightly. This part offers a 4.7 GHz base, 200 MHz more than the 7950X. However, the maximum turbo clock speed is 5.6 GHz, which is slightly lower than the 7950X, so if you want the best single-thread performance ever, the 7950X is positioned and configured as the fastest option for ST workloads. and MT.
The Ryzen 9 7900X costs $ 549, which is the same price as AMD’s previous 12-core entry, the 5900X. This is something worth noting for making a true apple-to-apple comparison of generational performance gains at a standard price point.
Ryzen 7 7700X
To complement AMD’s Ryzen 7000 stack, we have the Ryzen 7 7700X. This is the 8-core Zen 4 equivalent of the previous generation AMD 5800X. Like its predecessor, AMD only uses a Zen 4 CCD here, but it’s a fully functional CCD. The base clock speeds are 4.5 GHz while the maximum turbo clock speed is 5.4 GHz.
Since there is only one CCD, this also means that the L3 cache is halved to 32 megabytes. Combined with 8MB of L2 cache on the Zen 4 CPU cores, you will see AMD advertise this chip as a total of 40MB of cache.
Meanwhile, unlike the 12- and 16-core chips, the 7700X has a more traditional TDP of 105 watts. This helps with chip power (and cooling), although it will be interesting to see what it means for turbo clocks with heavy MT workloads. With fewer CPU cores, less silicon is activated, but that means even AMD doesn’t get the additional electrical headroom to boost clock speeds.
The 7700X launches with a price tag of $ 399, which, like the 7950X, is a slight drop in price from the previous generation (where the 5800X costs $ 449).