AMD Roadmap Leak: Major Platform, Graphics Changes Coming in Zen 4

This site may earn chapter commissions from the links on this page. Terms of utilise.

AMD-Ryzen-New-Feature

A newly leaked roadmap claims to show the future of AMD CPUs, with Zen iv highlighted as a major inflection point for the company. Equally is always the case, readers should keep in mind that we're discussing rumors, not fact.

Information technology's been quite a while since I did an AMD roadmap assay, so we need to chew through a chip of what's coming downwards the pipage in 2021-2022 before we talk about Zen 4. The top row of codenames refers to AMD's desktop CPU lineup. The second row is AMD'south desktop and mobile APU chips. The third column — Vermeer, Cezanne, Lucienne, Van Gogh, and Chagall — are the CPUs AMD is leading with in 2021.

Lucienne (third row, third column) is a refresh of 2020'southward Renoir mobile platform, only with slightly better clocks and more threads. Van Gogh is a bit of a puzzle. Information technology looks similar it derives from the work AMD has done on the Xbox Serial X and PlayStation v, with a Zen 2 CPU combined with an RDNA2 GPU, but it reportedly targets a v-18W ability envelope.

There's bear witness from AMD that Van Gogh exists, but it occupies an odd spot in AMD's production lineup relative to Cezanne and Lucienne. This slide references LPDDR5, but some contempo leaked bear witness implies the chip could support quad-aqueduct DDR5. Between the two, LPDDR5 sounds far more probable. DDR5 isn't even in-market yet, and a traditional desktop quad-channel retentivity interface on a laptop scrap would burn a lot of power. Finally, there'southward Chagall, AMD'south Threadripper platform.

One important note: This document aligns Vermeer, which launched in 2020, with Cezanne, which launched in 2021. This certificate is a production family roadmap, non a temporal roadmap, which is to say: It shows which products will be included in a given generation or Ryzen family, not exactly when those products volition launch.

Our assumption is that Warhol will debut on desktop in 2021, but that Rembrandt, Barcelo, and Dragon Crest will all exist 2022 products. The 6nm claim is a little surprising and might not be accurate. Co-ordinate to TSMC, the benefit of 6nm over 7nm is that it offers upwards to eighteen percent increased transistor density, with no communicated benefits in performance or ability. AMD and Intel don't typically try to maximize for density, due to the negative impact it can have on clock speed.

Rembrandt delivers RDNA2 to mobile and desktop APUs in 2022, merely the Barcelo refresh ensures Vega survives into another year. This slide implies that AMD volition launch AM5 and DDR5 support on mobile first, and that Zen 3+ will straddle both sockets, with the desktop flavor limited to DDR4 and the mobile version offering DDR5. AMD might be jumping for DDR5 and LPDDR5 support to take advantage of the power savings both standards offering over conventional DDR4.

If these predictions are accurate, information technology ways Rembrandt volition be AMD'south most avant-garde platform for much of 2022. We doubtable AMD would skip releasing desktop APUs based on Rembrandt, given that the mobile fries are supposedly using a different CPU socket. It looks as though the x86 manufacturer has something else in mind.

What Zen 4, Ryzen 7000 May Bring

Zen 4 presumably launches for desktop in late 2022 or early on 2023. It'll be AMD's first desktop platform on 5nm, its first chip to feature PCIe 5.0, and information technology'll be the first fourth dimension AMD has offered a graphics solution on every CPU.

AMD is the company that showtime popularized the idea of a CPU and GPU sharing the same piece of silicon, all the mode dorsum to its acquisition of ATI back in 2006. Information technology'south ironic, therefore, that Intel has done a better task of making baseline graphics adequacy available across its entire product line. Beginning with Zen 4, this changes, and RDNA2 becomes available across the unabridged product stack. That's a bigger change than it might seem.

AMD may have adopted chiplets for its desktop CPUs, but its APUs are firmly monolithic, including the eight-core Cezanne/Ryzen 5000 APUs available in mobile. There's good reason for this. Despite the marketing around them, chiplets are not a unilateral positive. AMD pays a penalty in terms of die area, latency, and ability consumption compared to a monolithic chip. In the desktop space, these disadvantages are small-scale — especially considering that AMD can field a 12-cadre and a xvi-cadre desktop CPU at a lower per-chip toll than it would otherwise pay. In laptops, however, AMD has chosen to stick with a more than conventional design. AMD's consoles, similarly, are monolithic architectures.

Integrating a GPU into a chiplet CPU design presents certain challenges. AMD could theoretically build a monolithic GPU chiplet to sit down aslope the I/O die, but the cloth requirements from this kind of organisation would be formidable. It might work for one CPU chiplet continued to one GPU chiplet, but nosotros doubt AMD could wire this solution into Epyc.

Alternately, AMD could possibly build a GPU into every Ryzen chiplet, and distribute workloads across multiple chiplets while treating the entire array of chiplets as a contiguous graphics bill of fare. The GPU cores in each chiplet would presumably connect to the CPU cores via the L3 or potentially an L4 cache. In the diagram below, that would equate to a GPU cluster being tucked betwixt the CCX and the "Infinity Textile" cake — separate from the CCX functionally, but nonetheless connected via the LLC.

Ryzen-5K-Topology

Nosotros expect AMD will redesign Infinity Cloth for the Zen 4 die shrink and peradventure include a new I/O die, assuming one doesn't tip up for Zen three+. Inserting a theoretical GPU die into the CCX topology above doesn't change the number of Infinity Fabric links needed across the entire fleck, even so, since the GPU CUs would be fully integrated into each chiplet.

Every Zen four chiplet would contain, say, 128-256 GPU compute cores (this number is entirely theoretical and could be higher). A 12-core or 16-core Ryzen 7000 would offering commensurately more GPU cores. A 64-cadre Epyc with 256 cores per chiplet would offer a maximum of two,048 GPU cores beyond the entire CPU.

AMD has been reticent to discuss AI in much detail, only executives have told ExtremeTech that they aren't blind to the rapid advances or long-term potential of the manufacture. Dissimilar most of its peers, including Qualcomm, Intel, and Apple, AMD hasn't brought a low-power NPU to market or focused on adopting SIMD instruction sets that specifically better AI performance. If AMD wants to compete in this space — and it says it does — incorporating a CU cluster into every Zen 4 chiplet would provide a guaranteed accelerator unit. Distributing a workload across multiple chiplets could also reduce hot spot formation.

I suspect — over again, if this leak is accurate — that this is why the RDNA2 block for Zen 4 is green on desktop but ruby on mobile. All of AMD's conventional, monolithic designs, for both Vega and RDNA2, are indicated in cherry-red. We see one single, solitary green block. Information technology shows upwards at the aforementioned fourth dimension other rumors imply AMD will launch RDNA2 broiled into every Ryzen chip.

From the beginning, AMD'southward big theme with Ryzen has been re-use, with the aforementioned chiplet design scaling from depression-end desktop to high-cease server. A Zen 4 chiplet with integrated RDNA2 hardware (or CDNA2, if AMD went that direction with Threadripper/Epyc) would provide additional processing horsepower for AI calculations by leveraging AMD's existing IP rather than requiring a from-scratch solution. At that place's no proof for this — it's speculation on my function, and I don't accept within information — simply if the company wants to put GPUs within chiplets, it either has to build a single unified GPU block that connects to every chiplet or it has to build a lilliputian bit of a GPU inside each chiplet. The second seems easier to scale than the former.

AMD will be facing off against fresh contest from Intel's Alder Lake and Raptor Lake processors from belatedly 2021-2023. But autonomously from the new hybrid cores coming with Alder, we don't know much about these fries, other than the fact that Intel expects significant power savings thank you to the adoption of hybrid cores.

Now Read: