Friday, April 19th 2024
AMD "Strix Halo" Zen 5 Mobile Processor Pictured: Chiplet-based, Uses 256-bit LPDDR5X
Enthusiasts on the ChipHell forum scored an alleged image of AMD's upcoming "Strix Halo" mobile processor, and set out to create some highly plausible schematic slides. These are speculative. While "Strix Point" is the mobile processor that succeeds the current "Hawk Point" and "Phoenix" processors; "Strix Halo" is in a category of its own—to offer gaming experiences comparable to discrete GPUs in the ultraportable form-factor where powerful discrete GPUs are generally not possible. "Strix Halo" also goes head on against Apple's M3 Max and M3 Pro processors powering the latest crop of MacBook Pros. It has the same advantages as a single-chip solution, as the M3 Max.
The "Strix Halo" silicon is a chiplet-based processor, although very different from "Fire Range". The "Fire Range" processor is essentially a BGA version of the desktop "Granite Ridge" processor—it's the same combination of one or two "Zen 5" CCDs that talk to a client I/O die, and is meant for performance-thru-enthusiast segment notebooks. "Strix Halo," on the other hand, use the same one or two "Zen 5" CCDs, but with a large SoC die featuring an oversized iGPU, and 256-bit LPDDR5X memory controllers not found on the cIOD. This is key to what AMD is trying to achieve—CPU and graphics performance in the league of the M3 Pro and M3 Max at comparable PCB and power footprints.
The iGPU of the "Strix Halo" processor is based on the RDNA 3+ graphics architecture, and features a massive 40 RDNA compute units. These work out to 2,560 stream processors, 80 AI accelerators, 40 Ray accelerators, 160 TMUs, and an unknown number of ROPs (we predict at least 64). The slide predicts an iGPU engine clock as high as 3.00 GHz.
Graphics is an extremely memory sensitive application, and so AMD is using a 256-bit (quad-channel or octa-subchannel) LPDDR5X-8533 memory interface, for an effective cached bandwidth of around 500 GB/s. The memory controllers are cushioned by a 32 MB L4 cache located on the SoC die. The way we understand this cache hierarchy, the CCDs (CPU cores) can treat this as a victim cache, besides the iGPU treating this like an L2 cache (similar to the Infinite Cache found in RDNA 3 discrete GPUs).
The iGPU isn't the only logic-heavy and memory-sensitive device on the SoC die, there's also a NPU. From what we gather, this is the exact same NPU model found in "Strix Point" processors, with a performance of around 45-50 AI TOPS, and is based on the XDNA 2 architecture developed by AMD's Xilinx team.
The SoC I/O of "Strix Halo" isn't as comprehensive as "Fire Range," because the chip has been designed on the idea that the notebook will use its large iGPU. It has PCIe Gen 5, but only a total of 12 Gen 5 lanes—4 toward an M.2 NVMe slot, and 8 to spare for a discrete GPU (if present), although these can be used to connect any PCIe device, including additional M.2 slots. There's also integrated 40 Gbps USB4, and 20 Gbps USB 3.2 Gen 2.
As for the CPU, since "Strix Halo" is using one or two "Zen 5" CCDs, its CPU performance will be similar to "Fire Range." You get up to 16 "Zen 5" CPU cores, with 32 MB of L3 cache per CCD, or 64 MB of total CPU L3 cache. The CCDs are connected to the SoC die either using conventional IFOP (Infinity Fabric over package), just like "Fire Range" and "Granite Ridge," or there's even a possibility that AMD is using Infinity Fanout links like on some of its chiplet-based RDNA 3 discrete GPUs.
Lastly, there are some highly speculative performance predictions for the "Strix Halo" iGPU, which puts it competitive to the GeForce RTX 4060M and RTX 4070M.
Sources:
ChipHell Forums, harukaze5719 (Twitter)
The "Strix Halo" silicon is a chiplet-based processor, although very different from "Fire Range". The "Fire Range" processor is essentially a BGA version of the desktop "Granite Ridge" processor—it's the same combination of one or two "Zen 5" CCDs that talk to a client I/O die, and is meant for performance-thru-enthusiast segment notebooks. "Strix Halo," on the other hand, use the same one or two "Zen 5" CCDs, but with a large SoC die featuring an oversized iGPU, and 256-bit LPDDR5X memory controllers not found on the cIOD. This is key to what AMD is trying to achieve—CPU and graphics performance in the league of the M3 Pro and M3 Max at comparable PCB and power footprints.
Graphics is an extremely memory sensitive application, and so AMD is using a 256-bit (quad-channel or octa-subchannel) LPDDR5X-8533 memory interface, for an effective cached bandwidth of around 500 GB/s. The memory controllers are cushioned by a 32 MB L4 cache located on the SoC die. The way we understand this cache hierarchy, the CCDs (CPU cores) can treat this as a victim cache, besides the iGPU treating this like an L2 cache (similar to the Infinite Cache found in RDNA 3 discrete GPUs).
The iGPU isn't the only logic-heavy and memory-sensitive device on the SoC die, there's also a NPU. From what we gather, this is the exact same NPU model found in "Strix Point" processors, with a performance of around 45-50 AI TOPS, and is based on the XDNA 2 architecture developed by AMD's Xilinx team.
As for the CPU, since "Strix Halo" is using one or two "Zen 5" CCDs, its CPU performance will be similar to "Fire Range." You get up to 16 "Zen 5" CPU cores, with 32 MB of L3 cache per CCD, or 64 MB of total CPU L3 cache. The CCDs are connected to the SoC die either using conventional IFOP (Infinity Fabric over package), just like "Fire Range" and "Granite Ridge," or there's even a possibility that AMD is using Infinity Fanout links like on some of its chiplet-based RDNA 3 discrete GPUs.
109 Comments on AMD "Strix Halo" Zen 5 Mobile Processor Pictured: Chiplet-based, Uses 256-bit LPDDR5X
Also:If they want to go above 8 cores it makes sense, just as it does on desktop CPU's. Otherwise they have to make some huge dies with everything in one, that will cost a lot and give lower yields.
I guess going >300 mm² isn't something AMD do on CPU/APU's these days.
This product seems quite niche to me. I suppose you can slightly more compact high-performance laptops by saving some space on the dGPU and the associated components, but given the high idle power consumption AMD’s chiplet-based processors tend to have, battery life may not be great. It may also be used in the most premium desktop-replacement laptops just so they can have the best components, but then those would come with the downside of soldered RAM. I suspect that it will be too power-hungry to compete against the M3/M4 Max.
BTW Lisa isnt going to send you a thank you card for meatshielding them.You skipped a HUGE portion of AMD's driver history. After buying ATi AMD never put sufficient resources into their graphic driver development. From the HD 2000 series through the HD 6000 series, it was standard practice to keep multiple driver son hand, depending on what game you wanted to play. New drivers would break older games more often then not. when the HD 7000s came out, the HD6000s and older were left to wither, cut off half a decade before nvidia's fermi was.
There was the Frame Pacing issues that AMD swore up and down didnt exist, until nvidia proved they did, the 200 and 300 series black screen issues that were never truly resolved, and tons of optimization issues. In 2017 after the launch of polaris AMD committed to fixing their drivers, and by 2019 they were in a much better state.
To most reasonable people, it's understandable why nearly 2 decades of poor quality drivers (ATI wasn't a whole lot better) have informed a large portion of the market against AMD. A few years of good drivers dont immediately fix that. And AMD still hasnt escaped controversy. Dropping GPUs after only 6 years (Fury X), not releasing drivers for the RX 6000s for nearly 3 months because of the RX 7000s (something nvidia has never done), and the more recent minor feather ruffling with TinyBuild dont help their image.
They're worlds away from where they were, and for most users they operate without issue. Perceptions, however, do not change on a dime, they take a lot of time and dedication to change.I can imagine a 13-14" laptop would be able to handle it.
Razer, get on it!Dont forget WAY higher speeds. 9500 mhz is available now and will supposedly be used with zen 5 mobile, and 10700 has been shown off by Samsung.
I doubt any laptop manufacturer would put 4 sodimm slots, but they might, just to sell them with only one stick.
Also, if the article is correct, this is all speculation from some randoms on the internet that have seen an image of a thing and came up with some other images themselves.
The specs are direct upgrades from the Zen 2/Oberon APU of the PS5 and Zen 2/Scarlett APU of the Series X too.
However, I too really wish this goes in some sort of ultra-portable or lightweight gaming laptop.I wish, but anything above 100W TDP in a handheld is going to push it out of its category for sure.
Ironically many of these naysayers are not AMD GPU users and just spreading FUD.
With regard to what you were saying, I too have been switching back and forth between Nvidia and AMD without significant issues from one or the other. I started with 3dfx products, moved to TNT2, geforce 4200Ti and then the Radeon 9800 Pro. Between then and now I’ve owned a dozen cards from both companies in my primary and secondary builds.
*Points at build*