The arrival of AMD’s Zen 4 Ryzen 7000 series “Raphael” processors draws near, and recent developments make this a critical release for the company. AMD’s previous-gen Ryzen 5000 processors accomplished what was once thought impossible: The chips unseated Intel’s best in every CPU benchmark, including taking the top of our list of best CPUs for gaming, as the company outclassed Intel’s Rocket Lake in every regard.
But then Alder Lake happened. Intel’s new hybrid x86 architecture, featuring a blend of big and powerful cores mixed in with small efficiency cores, pushed the company into the lead in all facets of raw performance and even helped reduce its glaring deficiencies in the power consumption department. But, perhaps most importantly, Alder Lake started a full-on price war with Intel’s new bare-knuckle approach to pricing, particularly in the mid-range that serves as gamer country.
But AMD isn’t standing still, and its Ryzen 7000 chips are now poised on the starting blocks to take the race for performance leadership to the next level. AMD recently demoed a 16-core Ryzen 7000 processor hitting an amazing 5.5 GHz during a gaming demo and completing a Blender render in 31% less time than Intel’s flagship Core i9-12900K. AMD says the final chips will come with up to >5.5 GHz boost clocks and are loaded with new tech, like a new integrated Radeon RDNA 2 graphics engine, and support AI instructions based on AVX-512. We’ve also learned plenty of new details about the 5nm Zen 4 Ryzen 7000 ‘Raphael’ processors and the new wave of motherboards with the AM5 socket.
We’ve gathered all of the information we know, both from official and unofficial sources, into this article. We’ll update the article as we learn more, but here’s what we know so far.
AMD Zen 4 Ryzen 7000 Series at a Glance
- Up to 16 cores and 32 threads on TSMC 5nm process (N5 used for compute die)
- (up to) >5.5 GHz boost
- 6nm I/O die, DDR5 memory controllers, PCIe 5.0 interface
- DDR5 only (no DDR4 support)
- RDNA 2 integrated GPU
- Zen 4 architecture has 8 to 10% performance gain
- >15% gain in single-threaded work, >35% overall performance gain (multi-threaded workloads), >25% performance-per-watt gains
- AM5 Socket LGA 1718, backward compatible with AM4 coolers
- 600-Series Chipset: X670E Extreme, X670, and B650 Motherboards
- up to 170W TDP, 230W peak power
- up to 25% more memory bandwidth per core
- Support for AVX-512
- 3D V-Cache Zen 4 models will come to market
AMD Zen 4 Ryzen 7000 Release Date Window
Image 1 of 2
Image 1 of 2
AMD has set Fall 2022 as the official launch window for the first Zen 4 products, the Ryzen 7000 series for desktop PCs (codenamed Raphael). For the US, fall begins on September 22 and ends on December 22, meaning we’ll see Ryzen 7000 by the end of the year. AMD has already demoed its 16-core 32-thread Ryzen processor, presumably the flagship processor, and if the company follows tradition, we expect it to launch its highest-end products first.
The Ryzen 7000 chips will mark just the first step of the Zen 4 journey as the company delivers on its CPU roadmap and brings them to the desktop and notebook markets. AMD will also use the Zen 4 architecture for its data center CPU roadmap.
AMD Zen 4 Ryzen 7000 Specifications and Features
Image 1 of 4
Image 1 of 4
This is normally where we’d have a full table of specs for any given family of chips. However, we still have much to learn from AMD about the actual end products — AMD hasn’t announced concrete specifications for the individual Ryzen 7000 SKUs yet. However, we know a lot about the overall specifications, like that the chips top out at 16 cores and 32 threads. AMD has also demoed a chip reaching up to 5.5 GHz on several cores during gaming — and that’s with a standard 2800mm AIO cooler. In fact, the company says the chips feature ‘>5.5 GHz,’ meaning we could see even higher boost clock speeds.
Image 1 of 6
Image 1 of 6
AMD shared a block diagram of the chip, and we took a close-up snip of a bare Ryzen 7000 chip during the company’s Computex keynote. The chip houses two 5nm core chiplets, each sporting eight cores. AMD says these are based on an optimized version of TSMC’s high-performance 5nm process technology (likely N5), and they are placed much closer together than we’ve seen with previous Ryzen core chiplets. In addition, we see what appears to be a shim between the two core chiplets, likely to maintain an even surface atop the two dies. It is also possible that this close orientation is due to some type of advanced packaging interconnect between the two chips.
The new I/O die uses the 6nm process and houses the PCIe 5.0 and DDR5 memory controllers along with a much-needed addition for AMD — the RDNA 2 graphics engine. The new 6nm I/O die also has a low-power architecture based on features pulled in from AMD’s Ryzen 6000 chips, so it has enhanced low power management features and an expanded palette of low-power states.
Surprisingly, the new I/O die appears to be roughly the same size as the previous-gen 12nm I/O die. However, given that the 6nm die is far denser than the 12nm die from GlobalFoundries, meaning it has far more transistors, it’s safe to assume the integrated GPU has consumed a significant portion of the transistor budget (possibly due in part to onboard iGPU cache). The large 6nm I/O die will inevitably add to the cost of the chips, as the 6nm die will be far more expensive than the mature 12nm I/O die that AMD used in the Ryzen 5000 chips.
Although AMD hasn’t divulged memory frequencies, AMD’s test notes include a benchmark with the 16-core chip running DDR5-6000 CL30. It’s unclear if those are stock frequencies or XMP/overclock values (AMD tends to use XMP profiles for its benchmarks). AMD recently touted that it expects to have exceptional DDR5 overclockability, making the memory controllers sound impressive from afar, and the new AMD EXPO (EXtended Profiles for Overclocking) tech looks like an alternative to Intel’s XMP branding. Simply put, AMD will support pre-defined memory profiles with dialed-in memory frequencies, timings, and voltages to enable one-click memory overclocks. A newly-filed patent also points to a possible upcoming automatic memory overclocking feature that would provide more of a dynamic approach that exceeds pre-validated EXPO profile speeds.
The Ryzen 7000 chips support up to 24 lanes of the PCIe 5.0 interface directly from the socket (further details in the motherboard section). AMD is busy enabling the PCIe 5.0 SSD ecosystem with Phison, Micron, and Crucial. Crucial and Micron will have their first PCIe 5.0 SSDs in the market when AM5 motherboards arrive on the market. Additionally, the constellation of third-party SSDs will also use Phison’s E26 PCIe 5.0 SSD controllers, meaning we’ll soon see wide availability of even speedier drives. That will come in handy for Zen 4 Ryzen 7000 systems — AMD claims a 60% performance gain in sequential read workloads with PCIe 5.0 SSDs. Phison backed that up with a recent demo showing its E26 SSD hitting up to 12 GB/s of read throughput (more detailed result in that link). PCIe 5.0’s sequential performance potential will be great for Microsoft’s DirectStorage because it relies heavily upon read throughput to reduce game loading times to roughly a second.
The Ryzen 7000 processors come with expanded instructions for AI acceleration through its support of AVX-512 instructions, like VNNI for neural networks and BFLOAT16 for inference. That oddly places Intel’s Alder and Raptor Lake chips at a disadvantage as they have disabled AVX-512 functionality due to the hybrid architecture.
AMD Zen 4 Ryzen 7000 Integrated Graphics iGPU
The RDNA 2 engine supports up to four display outputs, including DisplayPort 2 and HDMI 2.1 ports. All Ryzen 7000 chips will support some form of graphics, so it doesn’t appear there will be graphics-less options, like Intel’s F-series, for now.
AMD has tried to temper expectations for the integrated graphics engine, pointing out that the RDNA 2 graphics are only designed to ‘light up’ displays, cautioning that we shouldn’t expect any meaningful gaming performance. As such, it’s safe to assume we’re looking at probably 2 CUs per Ryzen 7000 chip.
If it’s any consolation, the iGPU’s close proximity to the DDR5 controllers also resident on the die should provide plenty of bandwidth from the main memory. However, we’ll have to wait to learn exactly how many cores the graphics engine has, but we have seen this iGPU running between 1,000 and 2,000 MHz in a recent benchmark submission. Despite the expected low performance, the integrated RDNA 2 engine will help address one of AMD’s key weaknesses in the OEM market where discrete GPUs are a rarity in most machines. It will also be helpful for troubleshooting if you need a basic display out.
AMD Zen 4 Ryzen 7000 Benchmarks and Zen 4 IPC
We tend to see benchmark results posted to third-party benchmark databases as processors work their way to market. Still, we’ve only seen one instance of a Zen 4 Ryzen 7000 processor listing that doesn’t come from AMD — two Ryzen 7000 submissions to the MilkyWay@Home project on the BOINC platform.
The submission doesn’t tell us much about performance, but it does expose the 100-000000666-21_N codename that likely represents the Ryzen 7 7800X that will replace the Ryzen 7 5800X. The other codename, 100-000000665-21_N, lines up with a 16-core model that is likely the Ryzen 9 7950X that will replace the Ryzen 9 5950X.
For now, most of the Zen 4 Ryzen 7000 benchmarks come from AMD, and as with all vendor-provided benchmarks, you should approach these results with caution. These chips are pre-production models, so performance is subject to change, and the test conditions could be favorable to AMD’s chips.
Image 1 of 5
Image 1 of 5
During its Computex 2022 keynote, AMD CEO Lisa Su demoed a 16-core pre-production Ryzen 7000 chip running the Ghostwire: Tokyo game. As you can see from the third image, the chip topped out at an incredible 5.52 GHz, and AMD has since clarified that this boost occurred on multiple cores during the test. The 5.5 GHz peak matches the current desktop PC frequency leader, the 5.5 GHz Intel Core i9-12900KS. Naturally, that comes with caveats: AMD only guarantees that its chips can reach the peak frequency on a single core. However, this is a significant increase over the existing Ryzen family.
AMD also demoed it’s 16-core 32-thread Ryzen 7000 chip against the 16-core 24-thread Core i9-12900K in a Blender render (we included the test notes in the above album). The Ryzen 7000 processor completed the render of a Ryzen 7000 chip in 204 seconds, which is 31% less than the 12900K’s time of 297 seconds.
Blender supports AVX-512, which could contribute to AMD’s lead over Intel in this benchmark, which would be odd: Intel pioneered AVX-512 but disabled the instructions with the Alder Lake chips because of the complexities of scheduling work to the correct cores in the x86 hybrid architecture. Now AMD has it in its arsenal.
Additionally, although we know that the 5nm process should be more power-efficient than the 7nm process, it is possible that the higher 230W provided by the AM5 socket could help improve all-core performance, specifically during an AVX-powered workload. (The 142W PPT limit hampered performance with the 12- and 16-core Ryzen 9 5900X and 5950X during all-core workloads.) It will be interesting to see comparisons of multi-threaded performance in a broader spate of benchmarks. We do have to remember that Raptor Lake will come with four more e-cores and higher clock rates than the 12900K, so we expect a close competition between the chips in heavily-threaded work.
AMD also measured Ryzen 7000’s +15% single-threaded performance improvement by putting an unnamed pre-production 16-core Zen 4 Ryzen 7000 processor with DDR5-6000 memory up against the 16-core Ryzen 9 5950X with DDR4-3600 in a Cinebench R23 single-threaded test. Unfortunately, AMD didn’t share any specific benchmark scores, but this does give us a basic idea of how the chips will fare against Intel’s Alder Lake in this specific benchmark.
According to our benchmarks, Intel’s Alder Lake chips currently hold the lead in the Cinebench R23 single-threaded benchmark. They also hold the overall lead in single-threaded performance against AMD’s Ryzen 5000 chips. Below we’ve boiled this down into a head-to-head with the flagship Alder Lake Core i9-12900K against the Ryzen 9 5950X.
|Tom’s Hardware||Cinebench R23 Single-Thread (% – CB Marks)||Overall Single-Thread Geomean|
|Core i9-12900K DDR5 — 5.2 GHz||100% (1,968)||100%|
|Ryzen 9 5950X DDR4 — 4.9 GHz||83.9% (1,652)||84.9%|
According to our tests, the Core i9-12900K is roughly 16% faster than the Ryzen 9 5950X in the Cinbench R23 benchmark, and AMD claims its 16-core Ryzen 7000 model is 15% faster than the 5950X. That means the Zen 4 chips will likely pull to parity with Intel’s Alder Lake in this benchmark.
Additionally, you can see that the Cinebench R23 result tracks well with our more expansive overall measurement of single-threaded performance that we use for our rankings in our CPU Benchmark hierarchy. This measurement encompasses performance in three single-threaded tests, and its similarity to the Cenbench scores suggests that Zen 4 could basically match Alder Lake in overall single-threaded performance.
Intel’s Raptor Lake will come with the same Golden Cove architecture for its performance cores (P-cores) as we saw with Alder Lake, but we expect Intel to dial up the clock rates to boost performance. As such, we can expect quite a battle for single-threaded superiority between Ryzen 7000 and Raptor Lake.
- 2017: Zen 1 — +52% IPC
- 2019: Zen 2 — +16% IPC
- 2020: Zen 3 — +19% IPC
- 2022: Zen 4 — +8 to 10% IPC
The TSMC 5nm process hit 5.52 GHz during AMD’s gaming demo, which was incredibly impressive, but AMD has clarified that we’ll only see an 8 to 10% improvement in IPC over Zen 3. That’s less than we’re accustomed to seeing with new AMD architectures, but improved power delivery can help deliver much larger gains in threaded workloads. It also isn’t unheard of for AMD to tout higher performance numbers as it reaches final silicon (like with Zen 1’s IPC measurements).
AMD is eager to show that those relatively tame IPC improvements aren’t all Zen 4 Ryzen 7000 brings to the table. At its Financial Analyst Day, AMD also shared a slide showing a greater than 25% performance-per-watt and greater than 35% gain in overall performance in a multi-threaded Cinebench benchmark.
This benchmark used a 16-core 32-thread Ryzen 7000 desktop PC processor against the 16-core Zen 3 Ryzen 9 5950X. The slide is a bit misleading as it uses a non-zero axis that visually amplifies the gain, so keep that in mind. However, these are impressive generational performance gains — regardless of whether they originate from IPC, frequency, or improved power delivery and multi-core boosts.
The Zen 4 processors will also support up to 25% more memory bandwidth per core, a marked increase that comes from both the step up to DDR5 and likely from widened pathways in the chip to deliver additional bandwidth to the cores. That will provide quite the uplift for the bandwidth-hungry AVX-512 extensions that AMD added for Zen 4.
We’ll have to wait a bit longer to get a clearer view of performance from third-party benchmarking. As always, we won’t know what you’ll see in real life until we snap the chips into the socket on our testbed.
AMD Zen 4 Ryzen 7000 Power Consumption
AMD originally stated that Socket AM5 would have a 170W Package Power Tracking (PPT) limit, meaning that would be the peak amount of power the socket could provide to any given processor. However, AMD later clarified that the original number it shared is in error, and the peak power consumption (PPT) for the AM5 socket is actually 230W. That’s a significant increase over the previous-gen Ryzen 5000’s 142W limit.
This equates to a 170W TDP for some processors designed for the AM5 socket, like Ryzen 7000, which is a significant increase over the current 105W TDP limit with the Ryzen 5000 processors. Overall, the increase represents a 65W TDP and an 88W PPT increase over AMD’s current flagships.
This increased power delivery will help the Ryzen processors in heavily-threaded workloads, like the Blender benchmark the company demoed during Computex that showed Ryzen 7000 thrashing Intel’s Alder Lake Core i9-12900K. The increased 170W TDP also means it’s entirely possible that we could see souped-up 12- and 16-core Ryzen 7000 chips with a 170W TDP for extreme users, while 105W 12- and 16-core models slot in for more mainstream uses.
Increasing the TDP and PPT will help AMD deliver more performance, particularly for its higher core-count models, during heavy multi-threaded workloads. In many cases, AMD’s previous limit of 142W with the previous-gen AM4 socket held back performance, so the additional 88W of power will be particularly helpful with the newer 12- and 16-core models. In addition, AMD has specified that it will use the standard TDP and PPT calculations for chips that drop into the AM5 socket — you can simply multiply the TDP by 1.35 to calculate the maximum power consumption of the chip (PPT).
AMD Zen 4 Ryzen 7000 Architecture
AMD has shared precious few details about the Zen 4 microarchitecture, but we do know that AMD has also doubled the L2 cache per core to 1MB for Zen 4, giving the CPU cores a bigger slab of near memory for workloads.
However, with Intel’s chips, we’ve seen larger L2 caches primarily benefit data center workloads. Larger L2 caches generally reduce L3 cache accesses (theoretically by ~40% in this case), which reduces contention on the fabric, thus enabling better scalability and performance in all-core workloads — as opposed to enabling big boosts to single-threaded work. That means there’s a chance Zen 4’s increased L2 capacity will pay off more handsomely for the EPYC Genoa server chips than it will for most desktop PC applications.
We’ll update this section with more architectural details as they emerge.
AMD Zen 4 Ryzen 7000 Processors and AM5 Socket
Image 1 of 4
Image 1 of 4
The top of the Ryzen 7000 package looks busy with a full complement of capacitors spread out across the PCB. This eliminates the need for capacitors that face into the socket, like the large arrays of capacitors we see spread among the LGA pads on Intel’s processors. (Ryzen 7000 pad image from @ExecuFix – not official from AMD.)
Ryzen’s capacitor arrangement necessitates the great-looking heatspreader, but it also likely eliminates any chance of AMD adding a third die to the chip. AMD has said that Zen 4 chips will top out at 16 cores and 32 threads, just like the previous-gen Ryzen 5000 series. AMD has told us that AM5 will be a similarly long-lived socket as we saw with AM4, so it’s possible we could see higher core counts in this socket in the future with newer generations of Ryzen.
Image 1 of 6
Image 1 of 6
A motherboard vendor shared a video of a Ryzen 7000 processor being slotted into the new AM5 socket but then removed the video. Luckily, we grabbed some screenshots before they took the video down. This new socket marks a big departure for AMD — the company is moving from its long-lived Pin Grid Array (PGA) AM4 sockets to a Land Grid Array (LGA) AM5 layout. Despite the entirely different LGA1718 socket interface (1718 pins), the AM5 socket will still support AM4 coolers.
An image of the underside of Ryzen 7000’s integrated heat spreader (IHS), shared to a Facebook group by an unknown poster, has also emerged. We actually learn quite a bit about the chip from the image, such as that AMD will continue to use solder thermal interface material (TIM) with its Zen 4 Ryzen 7000 processors. The IHS also appears quite thick, which helps with thermal dissipation, thus easing cooling requirements.
We can also see the glue at each of the mounting points on the eight ‘arms,’ which is a departure from AMD’s seal-all-around approach with the Ryzen 5000 chips. The two compute dies ride one edge of the heat spreader. As you can see, there isn’t room for a third die inside the package unless AMD were to alter the die placements significantly. Finally, we can clearly see the cutouts that make room for the surface mount devices (SMDs) on the top of the PCB (these are mostly capacitors).
AMD Zen 4 Ryzen 7000 600-Series with AM5 Socket: X670E Extreme, X670, and B650 Motherboards
Image 1 of 5
Image 1 of 5
AMD’s socket AM4 has served for five years across five CPU generations, four architectures, four process nodes, 125+ processors, and 500+ motherboard designs, so it’s time for a new socket, AM5, which also means a refreshed series of motherboards.
The Raphael processors will drop into a new AM5 socket that supports the PCIe 5.0 and DDR5 interfaces, matching Alder Lake on the connectivity front. The Socket AM5 motherboards will offer up to 24 lanes of PCIe 5.0, the most PCIe 5.0 lanes direct from the socket in the industry.
The chipset also supports up to 14 SuperSpeed USB ports up to 20Gbps and Type-C, along with support for Wi-Fi 6E with DBS and BlueTooth LE 5.2. AMD hasn’t specified, but the Wi-Fi 6E support comes as a discrete chip from the company’s initiative with MediaTek. The Ryzen 7000 chips also feature the SVI3 power infrastructure that supports more power phases from the motherboard and enables faster voltage response.
The X670E ‘Extreme’ chipset will support PCIe 5.0 for two graphics slots and one M.2 NVMe SSD port. This chipset is designed for motherboards that aim for extreme overclockability, carving out a new tier above AMD’s standard lineup.
The X670 chipset powers the ‘standard’ high end motherboards and will come in multiple flavors with varying PCIe support. The M.2 port will support the PCIe 5.0 interface, but the first graphics slot can support either a peak of PCIe 4.0 or PCIe 5.0, which will vary by the motherboard. This offers a lower-cost sub-tier of PCIe 4.0 X670 motherboards.
The B650 chipset will support PCIe 5.0 for a single NVMe port, but only PCIe 4.0 for the graphics slot. This chipset also supports overclocking, but as per usual, you won’t find as robust of power accommodations as you will with the more expensive boards.
Socket AM5 motherboards support up to four display outs via HDMI 2.1 Fixed Rate Link (FRL) and Displayport 1.4 High Bit Rate 3 (HBR3) outputs, powered by the RDNA 2 graphics engine onboard the 6nm I/O die in the Ryzen 7000 processors.
Image 1 of 2
Image 1 of 2
We’ve now seen five of the upcoming flagship X670E motherboards, with offerings from MSI, ASRock, ASUS, Gigabyte, and Biostar.
AMD hasn’t clarified its dual-chipset alignment yet, but recently leaked pictures of the MSI X670E motherboards and ASUS Prime X670-P Wi-Fi motherboards have confirmed many of the details we previously uncovered. According to our sources, AMD’s mainstream B650 platform will come with a single chipset chip that connects to the Ryzen 7000 CPU via a PCIe 4.0 x4 connection. However, documents we’ve seen say that a PCIe 5.0 connection is available on some AM5 processors.
Meanwhile, the enthusiast X670 platform employs two of these ASMedia chips (our sources confirm the chips are identical, not a north/southbridge-type arrangement), effectively doubling these connectivity options. Furthermore, these chipsets are daisy-chained together. This stands in contrast to AMD’s approach with the current 500-series motherboards, which use different chips for the X- and B-series motherboards. The new approach will obviously provide cost and design flexibility advantages.
Another report about the 600-series chipset (codenamed Promontory 21 – PROM21) has backed up our findings and provided more insights into the power- and cost-saving features of the 600-series chipset design.
Finally, AMD has confirmed that the AM5 socket will only support DDR5 memory. The company says that DDR5 provides the extra performance to justify the cost, but we’ll have to watch pricing closely. As we’ve reported, DDR5 continues to be more expensive than DDR4, largely because DDR5 marks the first generation of mainstream memory with onboard power management ICs (PMICs) and VRMs. Unfortunately, those have been in constant shortage due to the pandemic, but luckily, DDR5 pricing has fallen as PMIC and VRM supply improves. Unfortunately, DDR5 is still more expensive than comparable DDR4 kits.
However, DDR5’s more complex power circuitry and design mean that these modules will continue to command a premium over DDR4. DDR5 also has in-built ECC mechanisms for data at rest, which requires additional dies to provide the same memory capacity as DDR4. This means DDR5 will remain more expensive than DDR4, regardless of supply.
AMD Zen 4 Ryzen 7000 Pricing
AMD hasn’t shared specifications for the Ryzen 7000 product stack yet, so naturally, we don’t know how pricing will land. However, it is noteworthy that TSMC’s 5nm process is rumored to be much more expensive than the 7nm process was at this stage of production. The 6nm I/O die is also expected to add cost compared to the 12nm I/O die that AMD used with the Ryzen 5000 series.
The price of the chip you buy isn’t always all that matters, though: The X670 and B650 AM5 platforms support only DDR5 memory, which has pricing implications for platforms built around AMD’s upcoming Zen 4 processors. Though the pricing differences will become smaller over time, DDR5 will remain more expensive than DDR4, regardless of supply. That means Intel’s Raptor Lake will likely have a platform pricing advantage with readily-available DDR4 platforms, which could pay off in the mid-range and low end of the product stack. AMD has a counter with less-expensive PCIe 4.0-only X670 motherboards, but we’ll have to see how that pans out when those boards come to market.
All these factors mean you might have to pony up some extra cash compared to competing Intel Raptor Lake platforms, at least with the inaugural Zen 4 ‘Raphael’ Ryzen 7000 chips for Socket AM5. As a result, much like we saw with AMD’s high-priced debut for the Ryzen 5000 processors (AMD just finally released lower-cost Zen 3 chips a year and a half later), you can expect to pay a premium for AMD’s first Ryzen 7000 platforms when they arrive later this year.
We won’t have to wait long to see how pricing stacks up — the 5nm Zen 4 Raphael Ryzen 7000 chips and the accompanying 600-series chipsets are due on the market in Fall 2022. We’re sure to learn much more as we get closer to launch, so check back for updates, which we’ll add to this article regularly.