Sunday, June 2nd 2024

AMD Zen 5 Storms into Gaming Desktops with Ryzen 9000 "Granite Ridge" Processors

AMD today announced its much awaited Ryzen 9000 series desktop processors. Built in the Socket AM5 package, and drop-in compatible with all current AM5 motherboards with a BIOS update, the processors are based on the new "Zen 5" CPU microarchitecture. The operational part of the processor, the CPU complex dies (CCDs), are built on the 4 nm process, wired to a 6 nm I/O die. AMD didn't get down into the nuts and bolts of the microarchitecture, but briefly mentioned an impressive 16% IPC increase over "Zen 4." Coupled with the fact that the first wave of processors lack 3D V-cache and can sustain higher boost frequencies and TDP, processors in the series should beat the Ryzen 7 7800X3D in gaming performance, which also means that AMD has beaten the 14th Gen Core "Raptor Lake Refresh" processor series by a significant margin.

The 16% IPC increase over "Zen 4" is backed by branch prediction improvements, wider pipelines and vectors, and deeper window sizes across the core design, for more parallelism. The core also features doubling in instruction bandwidth for front-end instructions, FPU to L1, and L1 to L2 data bandwidth, and a redesigned FPU to double AI performance and AVX512 throughput. The company hasn't put out a block design for "Zen 5," and we'll learn more about it in the run-up to the market availability of these chips some time in July 2024.
The "Granite Ridge" processor features a chiplet-based design, just like the Ryzen 7000 "Raphael," Ryzen 5000 "Vermeer," and Ryzen 3000 "Matisse." CPU core counts range from 6-core/12-thread to 16-core/32-thread. The cores are located in CPU complex dies (CCDs). The 6-core and 8-core models feature a single CCD design, while the 12-core and 16-core ones come with a dual-CCD design. Each CCD contains a single CCX (CPU core complex) with 8 "Zen 5" CPU cores. Each core has 1 MB of dedicated L2 cache, and the eight cores share a 32 MB of L3 cache.

The client I/O die (cIOD) appears unchanged from the previous generation. It's built on the 6 nm process, packs a basic iGPU based on the RDNA 2 architecture with 2 compute units; a dual-channel DDR5 memory controller, and a 28-lane PCI-Express Gen 5 root complex.
AMD has planned four processor models for the first wave of Ryzen 9000 series desktop processors. These are led by the Ryzen 9 9950X, a 16-core/32-thread chip with a maximum boost frequency of 5.70 GHz, and a TDP of 170 W. This is followed by the Ryzen 9 9900X, a 12-core/24-thread part that boosts up to 5.60 GHz, with an interesting TDP number of 120 W. If you recall, its predecessor, the Ryzen 9 7900X, had the same 170 W TDP as the 16-core 7950X. So the TDP has generationally lowered.

The Ryzen 7 9700X is the 8-core/16-thread part from the series, with a maximum boost frequency of 5.50 GHz, and an impressive 65 W TDP. The most affordable part from the series will be the Ryzen 5 9600X. This 6-core/12-thread chip boosts up to 5.40 GHz, and has 65 W TDP.

If you're wondering whether the maximum boost frequencies and TDP have gone down generationally, it's because AMD has switched over to the slightly more efficient 4 nm foundry node for the CCDs, besides the "Zen 5" microarchitecture providing the 16% IPC gain.
AMD only put out performance numbers for the top Ryzen 9 9950X processor, which it compared with the Intel Core i9-14900K. In productivity and content creation workloads, the company is claiming performance leadership ranging anywhere between +7% in the UL Procyon benchmark, to +56% in Blender. The real story, though, is gaming performance, where the 9950X beats the i9-14900K by anywhere between +4% to +23%. Mistral LLM is shown running 20% faster on the 9950X compared to the i9-14900K.

The company didn't talk about pricing, we'll learn more about it as we near the July 2024 launch.
AMD is launching the AMD X870E and X870 desktop chipsets with these processors, heralding a new wave of Socket AM5 motherboards, which we'll see at Computex. These two chipsets standardize USB4 connectivity on all motherboards, as well as PCI-Express 5.0 x16 on all boards, even for the X870. The two chipsets also enable higher AMD EXPO memory frequencies. Again, the processors are perfectly compatible with AMD 600-series chipset motherboards with a BIOS update, and older Ryzen 7000 and Ryzen 8000 series processors should work on AMD 800-series chipset motherboards. AMD is promising longevity for AM5 going beyond even 2027.
Add your own comment

100 Comments on AMD Zen 5 Storms into Gaming Desktops with Ryzen 9000 "Granite Ridge" Processors

#1
Chaitanya
Will wait for reviews and given AMDs habit of slashing prices couple of months down the line will wait 9-11 months before upgrading my AM4 system.
Posted on Reply
#2
Darmok N Jalad
The one that really jumped out at me was the 9700X at just 65W TDP. That sounds like the real winner of the lineup, IMO.
Posted on Reply
#3
regs
IPC slide actually showing FPS
Posted on Reply
#4
JWNoctis
Darmok N JaladThe one that really jumped out at me was the 9700X at just 65W TDP. That sounds like the real winner of the lineup, IMO.
No hint of a 9x00X3D yet. I'd wait until that comes out. ;)

I wonder if the 4nm process really is that much more efficient, or whether AMD adapted the stock undervolt of their X3D processors to non-3D ones, considering the amount of negative CO people with non-X3D Zen 4 processors seemed to be running with.
Posted on Reply
#5
evernessince
Darmok N JaladThe one that really jumped out at me was the 9700X at just 65W TDP. That sounds like the real winner of the lineup, IMO.
The 7700 is a 65W TDP CPU that's barely slower than the 7700X.

That the 9700X is 65W just means that AMD decided it wasn't worth it to squeeze an extra 1-2% performance.
Posted on Reply
#6
JohH
Well, it's on the low end of my estimation of 15-25%:
JohHQuestionable source but about where I expect Zen 5 to land, in the 15-25% range.
I think this means gamers on non-X3D AM5 parts should just wait for the X3D Zen 5 parts.
And if you're on AM5 X3D there's no need to upgrade. If you're not on AM5 then wait to see what Arrow Lake brings.
Posted on Reply
#7
doc7000
I would say the fact that the 9700X rated boost has gone up by 100MHz and yet the TDP is now 65watts is what I find impressive, makes me wonder if they could have pushed clocks speeds on this with a 105 watt TDP. Seems like they may have made some serious gains when it comes to performance per watt.
Posted on Reply
#8
gffermari
I don't see a point releasing so many different models, simple and 3D V cache equipped.

Having Intel with powerful x600K CPUs, I think it would be a good time for a tier shift.
For example the 9600X to be a 8/16 CPU at 65W, the 9700X a 12/24, both non vcache SKUs.
And then 9800X3D 8/16 with all the bells and whistles and one or two SKUs for the 16/32 part.
Posted on Reply
#9
Hyderz
cool! cant wait to see what these cpu's can churn out
Posted on Reply
#10
Dr. Dro
Darmok N JaladThe one that really jumped out at me was the 9700X at just 65W TDP. That sounds like the real winner of the lineup, IMO.
TDP settings for this processor should be identical to the 7700, so an 88 W PPT. Still great for the performance on tap, though.
gffermariI don't see a point releasing so many different models, simple and 3D V cache equipped.

Having Intel with powerful x600K CPUs, I think it would be a good time for a tier shift.
For example the 9600X to be a 8/16 CPU at 65W, the 9700X a 12/24, both non vcache SKUs.
And then 9800X3D 8/16 with all the bells and whistles and one or two SKUs for the 16/32 part.
SKU spam makes shareholders happy and can sometimes induce customers to small purchasing errors that can potentially shorten the lifespan of a system, indirectly generating more sales. It is still not the time to buy, IMHO - to anyone who asks, wait for Arrow Lake and X3D models to make an informed decision, or wait for X3D if you already have the socket AM5 motherboard.
Posted on Reply
#11
JohH
Dr. DroTDP settings for this processor should be identical to the 7700X, so an 88 W PPT. Still great for the performance on tap, though.
7700X is 105W TDP 142W PPT.
7700 is the 65W 88W PPT part.
Posted on Reply
#12
Dr. Dro
JohH7700X is 105W TDP 142W PPT.
7700 is the 65W 88W PPT part.
True, I stand corrected (both by you and my phone) :laugh:
Posted on Reply
#13
atomsymbol
evernessinceThe 7700 is a 65W TDP CPU that's barely slower than the 7700X.
Assuming default BIOS settings, all non-X CPUs are measurably slower than X CPUs in multi-threaded tasks utilizing 8-16 threads on AMD 8-core CPUs.

Many games even in year 2024 are utilizing just a few CPU cores.
Posted on Reply
#14
SL2
JWNoctisI wonder if the 4nm process really is that much more efficient,
Exactly, they improved the IPC by doing some basic undervolting in BIOS

/s
Posted on Reply
#15
atomsymbol
SL2Exactly, they improved the IPC by doing some basic undervolting in BIOS
There is no causal relationship between IPC (Instructions Per Clock) and [undervolting in BIOS or CPU wattage].
Posted on Reply
#16
JWNoctis
JWNoctisNo hint of a 9x00X3D yet. I'd wait until that comes out. ;)

I wonder if the 4nm process really is that much more efficient, or whether AMD adapted the stock undervolt of their X3D processors to non-3D ones, considering the amount of negative CO people with non-X3D Zen 4 processors seemed to be running with.
SL2Exactly, they improved the IPC by doing some basic undervolting in BIOS

/s
For the record, I was referring to the reduced TDP target. Ryzen 7 7700X is an 105W part while Ryzen 7 9700X is a 65W part, actual consumption notwithstanding.
atomsymbolThere is no causal relationship between IPC (Instructions Per Clock) and [undervolting in BIOS or CPU wattage].
Technically there's a very slight relationship, if the higher clock from undervolt resulted in comparatively longer wait states. If anything, it should be slightly lower. ;)
Posted on Reply
#17
SL2
atomsymbolThere is no causal relationship between IPC (Instructions Per Clock) and [undervolting in BIOS or CPU wattage].
Wow, maybe you should figure out what /s means
JWNoctisFor the record, I was referring to the reduced TDP target. Ryzen 7 7700X is an 105W part while Ryzen 7 9700X is a 65W part, actual consumption notwithstanding.
Model numbers doesn't really tell us much anyway.

If you're right, it might be a way to make a 9800X3D look more competitive outside games.
Posted on Reply
#18
atomsymbol
JWNoctisTechnically there's a very slight relationship, if the higher clock from undervolt resulted in comparatively longer wait states. If anything, it should be slightly lower. ;)
It is very unlikely that there is some relationship there. In information theory, SIGNAL is "the thing" that is distinguishable from NOISE. Thus, if you are unable to distinguish those IPC changes from generic noise that by definition is present in all measurements of IPC, then there is no causal relationship there.
Posted on Reply
#19
azrael
I'm so curious to find out whether or not the 9000 series still supports ECC memory or if you need a socket AM5 EPYC for that now.
Posted on Reply
#20
SL2
TPUAMD is promising longevity for AM5 going beyond even 2027.
I find it funny that no one comments on this. It was kind of exected, but not a given.
Posted on Reply
#21
Onasi
Well, this was an announcement that pretty much lands where most sane people assumed it will land in terms of performance improvements. We’ll see more in the reviews, but it seems overall to be a standard gen on gen jump. I agree with @Dr. Dro that whatever the result, the best play, if you are in a market for a new CPU, would be to wait and see what Intel comes out with and the inevitable X3D chips.
Posted on Reply
#22
napata
JohH7700X is 105W TDP 142W PPT.
7700 is the 65W 88W PPT part.
Has AMD ever said why you need to mutiply their TDP by 1.35x to get the actual power limit? It just seems so weird where you have to multiply it with a random number.
Posted on Reply
#23
atomsymbol
SL2Wow, maybe you should figure out what /s means
Sarcasm in a previous message has no effect on [a true statement in a subsequent post that is citing the previous message without /s]. If I wanted to react to the whole previous message, then the /s would be included in the citation.
Posted on Reply
#25
SL2
atomsymbolSarcasm in a previous message has no effect on [a true statement in a subsequent post that is citing the previous message without /s]. If I wanted to react to the whole previous message, then the /s would be included in my post.
Maybe you didn't miss the sarcasm, but you did indeed miss the context. This has already been explained.
Posted on Reply
Add your own comment
Dec 21st, 2024 21:55 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts