Monday, May 15th 2023

AMD Ryzen 8000 "Granite Ridge" Zen 5 Processor to Max Out at 16 Cores
AMD's next-generation Ryzen 8000 "Granite Ridge" desktop processor based on the "Zen 5" microarchitecture, will continue to top out at 16-core/32-thread as the maximum CPU core-count possible, says a report by PC Games Hardware. The processor will retain the chiplet design of the current Ryzen 7000 "Raphael" processor, with two 8-core "Zen 5" CCDs, and one I/O die. It's very likely that AMD will reuse the same 6 nm client I/O die (cIOD) as "Raphael," just the way it used the same 12 nm cIOD between Ryzen 3000 "Matisse" and Ryzen 5000 "Vermeer;" but with updates that could enable higher DDR5 memory speeds. Each of the up to two "Eldora" Zen 5 CCDs has 8 CPU cores, with 1 MB of dedicated L2 cache per core, and 32 MB of shared L3 cache. The CCDs are very likely to be built on the TSMC 3 nm EUV silicon fabrication process.
Perhaps the most interesting aspect of the PCGH leak would have to be the TDP numbers being mentioned, which continue to show higher-performance SKUs with 170 W TDP, and lower tiers with 65 W TDP. With its CPU core-counts not seeing increases, AMD would bank on not just the generational IPC increase of its "Zen 5" cores, but also max out performance within the power envelope of the new node, by dialing up clock speeds. AMD could ride out 2023 with its Ryzen 7000 "Zen 4" processors on the desktop platform, with "Granite Ridge" slated to enter production only by Q1-2024. The company could update its product stack in the meantime, perhaps even bring the 4 nm "Phoenix" monolithic APU silicon to the Socket AM5 desktop platform. Ryzen 8000 is expected to retain full compatibility with existing Socket AM5, and AMD 600-series chipset motherboards.
Sources:
VideoCardz, PC Games Hardware
Perhaps the most interesting aspect of the PCGH leak would have to be the TDP numbers being mentioned, which continue to show higher-performance SKUs with 170 W TDP, and lower tiers with 65 W TDP. With its CPU core-counts not seeing increases, AMD would bank on not just the generational IPC increase of its "Zen 5" cores, but also max out performance within the power envelope of the new node, by dialing up clock speeds. AMD could ride out 2023 with its Ryzen 7000 "Zen 4" processors on the desktop platform, with "Granite Ridge" slated to enter production only by Q1-2024. The company could update its product stack in the meantime, perhaps even bring the 4 nm "Phoenix" monolithic APU silicon to the Socket AM5 desktop platform. Ryzen 8000 is expected to retain full compatibility with existing Socket AM5, and AMD 600-series chipset motherboards.
119 Comments on AMD Ryzen 8000 "Granite Ridge" Zen 5 Processor to Max Out at 16 Cores
I don't have any data so i'll ask: is there much call for more than 16 cores in the consumer space? I couldn't see it being any benefit in gaming, maybe graphic design?
Though those chips are for far more than just gaming.
Infact thanks to the massive L3 cache it's frametimes are often better than competitors. If you state exaggeration you have not right to complain if someone replies to you with an exaggeration. On paper. It's expensive and even Buildzoid who knows a thing or two about RAM OC could not get 8000 stable across all stress tests. I have not seen any such indication. Even Intel's latest server platform features 5.0. If they introduce 6.0 it will be on server side first.
Also PCIe adoption in general lags years behind 1.0 version certification.
4.0 was released on 2017 but came to mainstream desktop via AMD in 2019.
5.0 was released in 2019 and came to mainstream desktop via AMD in 2022.
6.0 was released just last year and will come in 2025 at the earliest. Neither Zen5 or Meteor Lake/Raptor Lake Refresh will use it in 2024. That's a good thing for consumer. There and no GPU's and barely any SSD's to even take advantage of 5.0 at this time. And these SSD's are first gen models that fall well short of maxing out the 5.0 link. Yet you argue for Intel for some reason. I too am on 5800X3D and i dont feel like my temps are high - around 70c max in games. 75W max. Based on HWinfo64 data on last 38 hours uptime with playing games etc.
Nor do i feel this huge latency penalty you keep talking about. In AIDA64 i get less than 60ns in RAM latency test.
They just care about core counts...
I would love to have but, can not really afford the Threadripper series and I don't really need 4 full PCIe 16×/16 lane slots... for triple the money of the top Ryzen 9 CPU + Mobo price.
see: www.techpowerup.com/review/nvidia-geforce-rtx-4090-pci-express-scaling/28.html
XDNA is already prepared for the future W12, office, and the new Direct XII all focused on Ai so this Zen 4 is the one that should be profitable with adjusted prices and RDNA 3
It is the main weakness they try to overcome by using the 3D cache.
They try to minimize RAM access as much as possible but a better way would be to increase the Fabric speed which essentially saw no upgrade from the previous gen.
Frametimes are fine but when the L3 runs out you will eventually get a slowdown, another wasted opportunity would be to at least increase the Cache size which again saw no upgrade. If it means better performance, why not?
It's not like power and costs are the main focus on the top of High End products.
Besides, let them sweat and find a solution to keep both at bay like they did for decades, but i know, why spend so much RnD money when you can still put out low effort products costing 10 dollars and keep the rest on your pocket. Yes because we needed AMD to give a slap on Intel and they did and we supported them but the story took a wrong turn.
Big Hype, low effort products, expensive prices. Sounds like Intel of the past? YES
But it also describes AMD of today. We needed AMD = AMD not AMD = INTEL 2.0
Same with the GPU sector, but there it's even worse AMD = Chinese Replica nVidia... but that's another story...
Both AMD and intel are running their CPU's to their absolute limits, and not spending the time to tune them in at all.
All AMD's best received CPU's have been the ones without the insane balls to the wall attitude at stock.
And they made half of them OEM only after that for some reason.
Look at this for an example of how it feels some days, 63W to 144W with their OC, for what amounted to .03% FPS gains
That's what it felt like running a 5800x with PBO enabled, tons of power for nothing except R23 scores going up.
Give us a solid single CCX design, slap on 3Dcache, but make it power efficient so it doesnt overheat and thermal throttle to sustain that performance without needing high end motherboards and cooling to go with it.
5800x3D smashed things so it's keeping up with the next gen hardware, but simply undervolting it changes it drastically for the better - and they're so over-volted at stock i can run -30 on curve optimiser AND undervolt it with an offset
Was that just comparing the top CPU of each stack, or when the gens launched? I can't recall the order of the individual CPU launches The current design is working quite well for AMD, it's the inter-core latency that is the issue and thats gone on a single CCX chip - and thanks to intel and AMD running mixed cores, we've now got OS's that handle that better.
my 3700x is a 2x4 setup, and it's starting to show signs of suffering in games that like to use 4+ cores, while smashing out older titles just fine
Very few titles at this stage fit that criteria, so overall its aging quite well, but a single 8 core is going to be a viable CPU For a long time at this rate
My suspicion is the CPU market for home will stagnate for many years.
Servers and workstations is where it will be at.
I think AMD will eventually follow suit and start incorporating E-cores into their CPUs too in the future. We'll probably get a clearer idea of what they're working towards when we see what Zen 5c is like.