AMD Zen 4 Ryzen 9 7950X and Ryzen 5 7600X Review: Retaking The High-End
by Ryan Smith & Gavin Bonshor on September 26, 2022 9:00 AM ESTRyzen 7000 I/O Die: TSMC & Integrated Graphics at Last
We’ll start our look at the Ryzen 7000 silicon with the glue that holds the chip together: the I/O Die (IOD). Whether it’s a 7600X or 7950X – one CPU chiplet or two – every Ryzen 7000 processor gets one of AMD’s new IODs.
As with the Ryzen 3000/5000 series, the IOD is a die separate from the CPU core chiplets that hosts everything that isn’t the CPU cores or their associated L1/L2/L3 caches. By producing a 2 or 3 chiplet-based solution rather than a single monolithic die, AMD is able to put silicon that doesn’t require cutting-edge performance on a cheaper process node, while enjoying the yield advantages that come from smaller chiplets and older, more mature nodes.
And yet with that said, for the Ryzen 7000 series it’s AMD’s IOD that’s getting the biggest upgrade. Previously produced on GlobalFoundries’ 14nm/12nm process node, for Ryzen 7000 the IOD has made the leap to TSMC. The new IOD is made on TSMC’s 6nm process, a generation behind the 5nm node used for the CPU chiplets themselves, but a generation (and then-some) ahead of GlobalFoundries’ smallest node.
With AMD no longer tightly bound to GloFo with wafer share agreement exclusivity requirements, AMD is now free to produce chips like the IOD where they see fit. And with AMD also doing very well as a company in the last few years, they do not need to produce IODs at GloFo to meet their purchase requirements with the fab.
The jump to TSMC 6nm brings several advantages to AMD’s central chiplet. The much smaller process node allows for significantly more transistors – 3.4B on a 122m2 die, 63% more than before – which AMD can invest in new features such as integrated graphics. As well, the more efficient process node brings down the power requirements of the IOD, especially in near-idle situations, though AMD is relying on a lot more than process nodes here.
In terms of functionality, the IOD first and foremost is responsible for the CPU-side implementation of all the new features of the AM5 platform. So PCIe 5.0 root complexes, DDR5 memory support, updated display I/O, and more are implemented here. Relative to the Ryzen 3000/5000 IOD then, there is almost nothing here that AMD hasn’t touched in some function; even the USB controllers and infinity fabric have been altered for this generation of processors.
Ryzen 6000 Mobile’s Legacy: Power Control and Pluton
Outside of the obvious gains from moving to a smaller process node, the Ryzen 7000 IOD has also given AMD a chance to implement everything they’ve learned in the past 3 years about power control and power optimizations from the Ryzen mobile lineup. Since the previous-generation IOD was launched for the Ryzen 3000 family in 2019, AMD has launched the Ryzen 6000 Mobile CPU, a monolithic die design that went all-out on power efficiency improvements in a bid to catch up to Intel. Ryzen 6000 Mobile introduced power features such as SVI3, which have since been ported to the Ryzen 7000 design, and principally show up as parts of the IOD.
This was especially easy for AMD as the IOD is made on the same 6nm process as the Ryzen 6000 Mobile chips. Which means that the new IOD has been able to implement many of the same power gating, power binning, deep sleep modes, clock gating, and power state features that were first introduced for mobile. Which is to say that AMD has put a lot of hard work into keeping the IOD from working hard.
The Ryzen 7000 also inherits many of the basic security features of the Ryzen 6000 Mobile platform. AMD is still using their own Arm-based security processor within the IOD. And the new chip is compliant with Microsoft’s Pluton initiative as well – with all the mixed responses that will undoubtedly come from that.
Ryzen 7000 Becomes Whole: Integrated Graphics At Last
Of the additional 1.31B transistors in AMD’s new IOD, the single biggest spend against that budget comes from the long-awaited inclusion of integrated graphics on high-end Ryzen desktop parts. Up to this point, AMD’s high-end consumer Ryzen processors have not included integrated graphics; both the early monolithic parts and later chiplet-style parts have forgone them for cost and integration reasons. But now with AMD’s modern RDNA 2 graphics architecture and TSMC fabrication process, AMD has finally seen the (ray traced) light, and is building a small GPU block into the IOD to offer integrated graphics throughout the Ryzen 7000 product stack.
To be sure, this iGPU is not a very powerful one – and it’s not meant to be. With just 2 CUs (technically a single RDNA 2 dual compute unit), it offers one-sixth the number of CUs on AMD’s top-end Ryzen Mobile APUs, or in the bottom-of-the-stack discrete Radeon RX 6400. So this is not a configuration meant to replace AMD’s traditional APUs in laptop or desktop scenarios. None the less, it is a full implementation of the RDNA 2 architecture – ray tracing and all – in a very low performance configuration.
| AMD GPU Specificaiton Comparison | |||||
| AMD Radeon Graphics (Raphael) |
AMD Radeon 680M (Rembrandt) |
AMD Radeon RX 6400 | |||
| Stream Processors | 128 (2 CUs) |
768 (12 CUs) |
768 (12 CUs) |
||
| Boost Clock | 2200MHz | 2400MHz | 2321MHz | ||
| Peak Throughput (FP32) | 0.56 TFLOPS | 3.7 TFLOPS | 3.6 TFLOPS | ||
| Memory Clock | DDR5-5200 | LPDDR5-6400 | 16 Gbps GDDR6 | ||
| Memory Bus Width | 128-bit | 128-bit | 64-bit | ||
| Infinity Cache | N/A | N/A | 16MB | ||
| Manufacturing Process | TSMC N6 | TSMC N6 | TSMC N6 | ||
| Transistor Count | N/A | N/A | 5.4B | ||
| Architecture | RDNA2 | RDNA2 | RDNA2 | ||
| GPU | Raphael Integrated | Rembrandt Integrated | Navi 24 | ||
The idea behind including an iGPU here is to offer a basic implementation of modern graphics across the entire Ryzen 7000 lineup. Even though it’s not very fast, 2 CUs running at up to 2.2GHz is going to be fast enough for desktop work – and maybe even some very light gaming. This makes the high-end Ryzen chips suitable for desktop work and inclusion in OEM desktop systems without a discrete GPU, a market segment that AMD was previously locked out of and Intel enjoyed free reign. And even for users who are bringing their own GPU, an integrated GPU makes troubleshooting all the easier, as now it doesn’t require a separate video card to merely boot a system.
Besides enabling basic pixel processing, the other major function of the Ryzen 7000 iGPU is to enable all of the video decoding and video output functionality expected from a modern GPU. As this part is derived from the integrated GPU that when into Ryzen 6000 Mobile (Rembrandt), it comes with a very familiar feature set. That includes AV1, HEVC, and H.264 video decoding, as well as HEVC and H.264 video encoding. And, as AMD is keen to point out, these video encode/decode blocks remain accessible even with a discrete GPU in play; so unless a user outright disables the iGPU, every Ryzen 7000 system will have access to a modern suite of video encode and decode features.
Meanwhile on the display controller side of matters, this is the block that’s enabling Ryzen 7000 CPUs to drive up to 4 4K@60Hz displays. The iGPU display controllers can drive HDMI 2.1 up to its maximum 48Gbps data rate, or it can drive a DisplayPort 2.0 output at up to the UHBR10 data rate (a feature not even found on NVIDIA’s forthcoming RTX 40 series cards). And as mentioned previously when talking about motherboards and chipsets, motherboard vendors will have the option of exposing these DP outputs either via USB-C alt mode, or by implementing fixed DisplayPorts. All of which, in turn, can be used as active display outputs even if a discrete video card is installed, via AMD’s new hybrid graphics mode.
All told, AMD is taking a similar tack as Intel here with their desktop CPUs. Most (but not all) Intel desktop parts contain a basic UHD-class integrated Xe-LP GPU, which Intel does for many of the same reasons as AMD now is. AMD has a leg-up on Intel in as much as they offer a newer architecture and features to much, such as AV1 decoding. But we’ll have to see where performance lands; 2 Radeon CUs is low enough that these parts may not even catch up to Intel’s 32 EU integrated graphics.
AMD Infinity Fabric: Faster Flowing, Narrower Going
The last major functional update to come with AMD new IOD is the Infinity Fabric, AMD’s die-to-die interconnect. On the chiplet-based Ryzen desktop parts, IF is used to provide a low-latency connection between the IOD and the CPU core chiplet(s) on the package.
For the Ryzen 7000 series, AMD has done some further engineering work to reduce the power costs of the IF links, and thus improve the overall power efficiency of the link and reduce the data movement power penalty for using a chiplet-based approach. The most significant thing here is that AMD has doubled the IF link frequency, but cut the physical width of the IF link in half. As a result, Ryzen 7000’s IF links offer the same bandwidth as on Ryzen 5000,
According to AMD, this refactoring is because AMD’s engineers found that, contrary to conventional wisdom, they could save power running a narrower IF link at a higher frequency. Which is to say that, even though higher frequencies are normally power prohibitive, in the case of IF it’s just powering the physical links themselves which is the most expensive part. Part of that shift, in turn, comes from the fact that the newer process nodes have allowed AMD to run the IF link at a lower voltage,
Finally, further reducing AMD’s power consumption here, the IF links on the Ryzen 7000 now support multiple power states, versus the binary on/off states of the IF links on Ryzen 5000 and earlier. This gives AMD a nice middle ground to save power in lighter load situations, by being able to use intermediate states rather than running at full-power just to have an IF link active.
And while this leaves AMD with an obvious (if more power-hungry) route to take for more IF bandwidth in the future should they need it, for now the company intends to stay pat. According to AMD’s engineers, the amount of bandwidth supplied by the Ryzen 7000 IF links, though virtually identical to the amount of bandwidth on the Ryzen 5000 links, is still enough to meet AMD’s die-to-die bandwidth needs – even with extra memory bandwidth afforded by the switch to DDR5.
Facebook
Twitter
RSS
205 Comments
View All Comments
emn13 - Monday, September 26, 2022 - link
The geekbench 4 ST results for the 7600x seem very low - is that benchmark result borked, or is there really something weird going on? Replyemn13 - Monday, September 26, 2022 - link
Sorry, I meant the geekbench 4 MT not ST results. The score trails way behind even the 3600xt. ReplySilver5urfer - Monday, September 26, 2022 - link
Good write up.First I would humbly request you to please include older Intel processors in your suite, it will be easier to understand the relative gains for eg the old 9th gen, 10th gen as a reliable place I see things all over on other sites, AT is at-least consistent so would be better if we have a ton of CPUs in one spot. Thanks
Now speaking about this launch.
The IOD is now improved by a huge factor so no more of that IF clock messing with the I/O controller and high voltage on the Zen 3 likes it's all improved so I think the USB fallout issues are fixed on this platform now. Plus the DP2.0 on iGPU is a dead giveaway on RDNA3 with DP2.0 as well.
IMC is also improved looking at it AMD operated with synchronized clocks with DRAM now they can do it without that since IF is now at 2000MHz and the IMC and DRAM are higher at 3000MHz to match the DDR5 data rates. Plus the EXPO is also lower latency, however the MCM design causes the AIDA benchmark to have high latency vs Intel even though Intel is operating at Gear 2 ratio with similar Uncore decoupled. Surprisingly the inter core latencies did not change much, maybe that's one of the key to improving more on AMD side gotta see what they will do for Zen 5.
The CPU clocks are insane, 5GHz on all 16C32T is a huge thing, plus even the 7600X is hitting 5.4GHz. Massive boost from AMD improving their design, plus the TSMC5N High Performance node is too good. However AMD did axed their temps and power. It's a very good move to not castrate the CPU with power limits and clocks now that's out it gets to spread it's wings. But the downside is, unlike Intel i7 series Ryzen 6 also gets hot meaning the budget buyers need to invest money in AIO vs older Zen 3 being fine on Air. That's a negative effect for AMD when they removed the Power Limits like Intel and let these rip to 250W.
Chipset downlink capping at PCIe4.0x4 was the biggest negative I can think of it, because Intel DMI is now 4.0x8 on ADL and RPL, RKL had it at 3.0x8 CML at 3.0x4. AMD is stuck to 4.0x4 from X570. Many will not even care, but it is a disadvantage when you pay top money for X670E they should have given us the PCIe5.0x4, AMD will give that in 2024 with Zen 5 X770 chipset that's my guess.
The ILM backplate engineering is solid that alone and the LGA1718 AM5 longevity itself is a major PLUS for AMD over LGA1700's bending ILM and EOL by 13th gen. Yes the 12th gen is a better purchase given how the Cooling requirement for i7 and i5 is not this high like R6 and R7 and the cheaper board costs plus 13th gen is coming and AMD's platform is new as well you would be a guinea pig. Depends on what people want and how much they can spend and what they want in longevity.
Performance is top notch for 7600X and 7950X absolute sheer dominance but the pricing is higher when you see the % variance vs Zen 3 and Intel 12th gen parts, and added AIO mandatory because they are hot. The gaming performance is as expected not much to see here and the 5800X3D still is a contender there but to me that chip is worthless as it cannot match any processor in high core count workloads. Although 7600X is a champion 6C12T and it beats 12C24T in many things and the 10C20T 10th gen Intel too. IPC is massive in ST and MT workloads as expected. AMD Zen 4 will decimate ARM, Apple has only one thing lol muh efficiency all that BGA baggage, locked down ecosystem is free.
RPSC3 perf at TPU's Red Dead Redemption is weird as I do not see any gains over Intel, given how much of a beast this AVX512 is on Zen 4 with 2x256Bit without AVX offset that too maybe they are not using AVX512. Plus their AMD Zen 3 gauging is also bad because they do not work well vs Intel 9th gen even, I wish you guys cover Dolphin emu, PCSX2, RPCS3 and Switch Emulators.
I think best option is to wait for next year and buy these parts as they will drop, right now no PCIe5.0 SSD in high capacity. no PCIe5.0 GPU even that Nvidia skimped on it. No use of the new platform unless one is running a super damn old CPU and GPU setups.
Shame that OC is totally dead, Zen 3 was hamfisted with its Curve Optimizer and Memory tuning becoming a head ache due to how AGESA was handled and the 1.4v high voltage and lack of documentation. Zen 4 it's even 1.0-1.2v still no OC because AMD's design basically is now pushed to maximum with it's Core TJMax temps and how it works on the basis of Core temperatures over everything else. There's no room here, AIO is saturated with 90C here. Too high heat density on AMD side similar to Intel 11th and 12th gen. Although Intel can go upto 350W and hit all cores at higher vs AMD 250W max. Well OC was on life support, only Intel is basically keeping it alive at this point after 10th gen it became worse and 12th very hot and high heat and now 13th gotta see if that DLVR regulator helps or not.
All in all a good CPU but has some downsides to it. Not much worth for existing 2020 class HW folks at all. Better wait when DDR5 matures even further and more PCIe5.0 becomes prevalent. Reply
Threska - Monday, September 26, 2022 - link
Maybe people will start delidding.https://youtu.be/y_jaS_FZcjI Reply
Silver5urfer - Tuesday, September 27, 2022 - link
That Delid is a direct die, it will 100% ruin the AM5 socket for longevity and the whole CPU too. That guy runs HWBot, ofc he will make a video on his bs delid kits. Nobody should run any CPU completely blowing the IHS off. You will have a ton of issues with that. Water leak, CPU silicon die crack due to Thermodynamics and the pressure differences over the time, Liquid Metal leak. Total bust of Warranty on any parts once that LM drops on your machine game over for $5000 worth rig there.AMD should have done some more improvements and reduced the max TJ Max to say 90 at-least but it's what it is unfortunately (for high temps and cooling requirements) and fortunately (to have super high performance) Reply
Threska - Tuesday, September 27, 2022 - link
There are some in the comments both wondering if lapping would achieve the same and the thicker lid was giving some room for future additions like 3D cache, etc. Replyabufrejoval - Wednesday, September 28, 2022 - link
I'm not sure that PCIe 4.0 "DMI" downlink capping is a hard cap per se by the SoC, but really the result of negotiations with the ASmedia chipset, which can't do better. I'd assume once someone comes up with a PCI 5.0 chipset/switch, there is no reason it won't do PCIe 5.0. It's just a bunch of 4 lanes, that happen to be connected to ASmedia PCIe 4.0 chips on all currrent mainboards.Likewise I don't see why you couldn't add the second chipset/switch to the "NVMe" port of the SoC or any of the bifurcated slots: what you see is motherboard design choices not Ryzen 7000 limitations. That just has 24 PCIe 5.0 lanes to offer in many bundle variants. It's the mainboard that straps all that flexibility to slots and ports.
I don't see that you have to invest into AIO coolers, *unless* you want/need top clocks on all cores. E.g. if your workloads are mixed, e.g. a few threads that profit from top clocks for interactive workloads (including games) and others that are more batch oriented like large compiles or renders, you may get maximum personal value even from an air cooler that only handles 150 Watts.
Because the interactive stuff will rev to 5.crazy clocks on say 4-8 cores, while for the batch stuff you may not wait in front of the screen anyway (or do other stuff while it's chugging in the background). So if it spends 2 extra hours on a job that might take 8 hours on AIO, that may be acceptable if it saves you from putting fluids into your computer.
In a way AMD is now giving you a clear choice: The performance you can obtain from the high-end variants is mostly limited by the amount of cooling you want to provide. And as a side effect it also steers the power consumption: you provide 150 Watts worth of cooling, it won't consume more except for short bursts.
In that regard it's much like a 5800U laptop, that you configure between say 15/28/35 Watts of TDP for distinct working points in terms of power vs. cooling/noise (and battery endurance).
Hopefully AMD will provide integration tools on both Windows and Linux to check/measure/adjust the various power settings at run-time, so you can adjust your machine to your own noise/heat/performance bias, depending on the job it's running. Reply
Dug - Monday, September 26, 2022 - link
"While these comments make sense, ultimately very few users apply memory profiles (either XMP or other) as they require interaction with the BIOS"This is getting so old. Your assumption is incorrect which should be obvious by the millions of articles and youtube videos on building computers. Not to mention your entire article is not even directed to "general public" but to enthusiasts. Otherwise why write out this entire article? Just say you put a cpu in a motherboard and it works. Say it's fast. Article done.
Why not test with Curve Optimizer? Reply
Oxford Guy - Tuesday, September 27, 2022 - link
This text appears again and again for the same reason Galileo was placed under house arrest. Replysocket420 - Monday, September 26, 2022 - link
Could someone, preferably Ryan or Gavin, please elaborate on what this sentence - "the new chip is compliant with Microsoft’s Pluton initiative as well" - actually means? This is the only review I could find that mentions Pluton in conjunction with desktop Zen 4 at all, but merely saying it's "compliant" is a weird way of wording it. Is Pluton on-die and enabled by default in Ryzen 7000 desktop CPUs? Reply