AMD Ryzen 8000 "Granite Ridge" Zen 5 Processor to Max Out at 16 Cores

[Daven]

Im arguing Q2 2024 at the latest (on 4nm). He is arguing late Q4 2024 at the earliest (on 3nm i believe). AMD'st past cadence has been 12-18 months. Zen4 took 23 months because of the massive number of changes. Zen5 should return to normal 18 months meaning we are about 9 months away from Zen5 release.


AM6 wont happen before DDR6. Meaning possibly Zen7.

Ryzen 1800X Zen 1 March 2, 2017 0 years
Ryzen 3900X Zen 2 July 7, 2019 ~2.33 years
Ryzen 5950X Zen 3 November 5, 2020 ~1.33 years
Ryzen 7950X Zen 4 September 27 2022 ~1.85 years
Ryzen 9950X Zen 5 July to November 2024 guesstimate

AMD can release Zen 5 whenever they like but historically we are looking at just shy of 2 years average between major releases.

 

[MikeGR7]

yes i want AMD to win!

I know you do.
But i would prefer the consumer to win.

It is apparent from your enthousiasm that this is your first high end build so instead of messing with corporate shenanigans like we do, i suggest you go have fun and play Fortnite or whatever makes you happy!

Chill you have an excellent build either way don't let me spoil you.

 

[tpa-pr]

I doubt these new chips will warrant a jump from my 7700X but maybe if my Gigabyte BIOS explodes the chip from too much voltage, i'll consider it

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?

 

[Wolverine2349]

I doubt these new chips will warrant a jump from my 7700X but maybe if my Gigabyte BIOS explodes the chip from too much voltage, i'll consider it

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?

No games come even close to 16 cores. In fact do any games really go beyond 8 cores? If any do its barely at all?

Though those chips are for far more than just gaming.

 

[BArms]

Great news for gamers for sure, most games don't scale beyond 4 cores hardly at all. Civilization VI which is one of the rare games these days that isn't GPU-bound barely benefits at all from more than 2 cores even though most of the time the player is waiting for AI to make moves, later in the game at least.

 

[Minus Infinity]

Leaks have already said Zen5 should have a large IPC increase, it's pretty much a brand new architecture they've claimed.

What I want to know is whether AMD will do something like this: combine a "regular" Zen5 Chiplet with a Zen5C chiplet (so the same two chiplet setup), but then you'd have 8 full power Zen5 cores from the first chiplet and then 16 Zen5c cores from the second chiplet (perhaps the reduced cache on the Zen5C chiplet could be offset by a v-cache addition?) and then you'd have a 24 core 48 thread hybrid Zen5 chip for AM5.

Apparently only for mobile! I don't see why if AI accelerators can go in Strix Point or the like, why they can't go in desktop. I'm also bummed '5c' e cores aren't going into desktop either. I will not buy a 8800(X) with just 8 P cores. I would much rather wait for Arrow Lake 15700(K). An 8850 say with 8P + 4/6 5c cores would be very interesting though.

 

[HBSound]

Hope AMD will keep adding extra PCIe lanes just like with the Ryzen 7000 family!

I agree with you. I would like to see the AMD 5 / 6 able to handle 2 - PCI 4.0 x 16 (X16) at full tilt. No dumbing down of the extra PCI slots. I do not feel two full-size PCI slits are a lot to ask at this level.

 

[AusWolf]

Intel is already at 8000 in it's current platform which clearly demonstrates that the next socket will easily max the DDR5 frequency range.

Not really. Intel is at 5600, AMD is at 5200. Anything faster is considered overclocking, and is not officially supported.

It could be nice if it would translate to actuall lower temperatures and not to an irrelevant annual 10-20 dollars electricity save.

It does. My 7700X barely reaches 60 °C in 65 W eco mode.

Ohh yes! That's the correct mentality!
Introduce crappy performance products, let users beta test it and fix it via patches and workarounds 6 months later!
Works like a charm for the gaming industry let's support it on the Hardware industry too!

Because buying into a new platform has always been a problem-free experience, obviously!

 

[R0H1T]

No doubt, but instead of pushing expiremental technologies on users it's better to wait and do it right.
AMD's chiplets are lovely on specsheets to fill the place with cores as long as latency is hidden under the carpet.
Not that i approve E-cores, but Intel saw that AMD got away with it and got baited into their laggy path.

It's not experimental, they've been at it for what 6+ years now? Also what do you mean by laggy path

Yeah yeah, so if Intel sucks let's encourage the rest of the industry on the same path right?

Don't care about Intel sucking because they've been at it since arguably Sandy Bridge! But without AMD's approach x86 would be in a big hole right now, they're not really competing among themselves now but against the likes of Apple & possibly QC (with Nuvia) & lots of custom(?) ARM solutions. Take AMD out & you have basically nothing to show in the "PC" space for the last decade or so

 

[Tomorrow]

Ryzen 1800X Zen 1 March 2, 2017 0 years
Ryzen 3900X Zen 2 July 7, 2019 ~2.33 years
Ryzen 5950X Zen 3 November 5, 2020 ~1.33 years
Ryzen 7950X Zen 4 September 27 2022 ~1.85 years
Ryzen 9950X Zen 5 July to November 2024 guesstimate

AMD can release Zen 5 whenever they like but historically we are looking at just shy of 2 years average between major releases.

You forgot Zen+ so Zen2 gap is not actually 2.33 years. Also 2700X and 3950X existed and Zen 5 will likely use 8000 series naming ie 8950X unless AMD decides to introduce 8000 series exclusive APU's like they did with 4000 series.

No doubt, but instead of pushing expiremental technologies on users it's better to wait and do it right.
AMD's chiplets are lovely on specsheets to fill the place with cores as long as latency is hidden under the carpet.
Not that i approve E-cores, but Intel saw that AMD got away with it and got baited into their laggy path.

Yes lets keep making huge expensive and power hungry monolithic cores on more and more expensive nodes. That will work out well for every consumer...

Of course no one said it!
Here is some food for thought:
Where is the Frametime Analysis section from TPU's reviews gone?? ---EDIT: I merely mention TPU as an example, it is a trend i noticed in many other sites like Tom's Hardware etc.
Wondered why GN was forced to present the full frametime plot for ONLY 2 games before casually returning to bars for the rest of the games in their 7800X3D video?

7800X3D being a single chiplet model does not have any noticeable latency penalties in gaming compared to dual chiplet models.
Infact thanks to the massive L3 cache it's frametimes are often better than competitors.

Your "lol" at the end gives me hope that you got what is obviously a humorus exageration.

If you state exaggeration you have not right to complain if someone replies to you with an exaggeration.

Intel is already at 8000 in it's current platform which clearly demonstrates that the next socket will easily max the DDR5 frequency range.

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.

The article is obviously disscussing future products and given that Intel's platform with Gen 5.0 is EOL points to the logical conclusion that their next Chipset will add Gen 6.0 support.

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.

The same cannot be said for AMD since their current platform's main advantage is the fact it will stay the same for next CPU release, so Gen 5.0 for them.
Nothing too exciting here, it is just the same as the releases always worked.

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.

Nah, don't really like Intel --> my current system is based on 5800X3D which was an upgrade from my previous 7700X but saddly both were a downgrade from my 12700K i had before, wish i had known better...
At least my averages are still great

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.

 

[csendesmark]

I agree with you. I would like to see the AMD 5 / 6 able to handle 2 - PCI 4.0 x 16 (X16) at full tilt. No dumbing down of the extra PCI slots. I do not feel two full-size PCI slits are a lot to ask at this level.

The problem is, most of the people does not even understand how important is this,
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.

 

[Daven]

You forgot Zen+ so Zen2 gap is not actually 2.33 years. Also 2700X and 3950X existed and Zen 5 will likely use 8000 series naming ie 8950X unless AMD decides to introduce 8000 series exclusive APU's like they did with 4000 series.

Zen+ was a refresh with zero silicon changes. Releasing it early was probably easy to do. AMD could absolutely release a Zen 4+ sooner but no rumors of that so far. Historically MAJOR Zen releases have come out a little before two years on average. Obviously APUs don’t count as they are variations of the same architecture. I am skeptical of the 8000 naming rumor as it breaks the naming scheme of the last three product release cycles.

 

[HBSound]

The problem is, most of the people does not even understand how important is this,
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.

Exactly! A lot are forced in the direction of Threadripper Pro / Epyc just to have active multiple PCI lanes. I am pretty convinced, too when everyone was trying to run dual GPU 3090 and etc. I am convinced they did not understand they were not getting full throttle when using the AMD4/5 Gpus and etc. Pretty sad!!!

 

[Wirko]

I agree with you. I would like to see the AMD 5 / 6 able to handle 2 - PCI 4.0 x 16 (X16) at full tilt. No dumbing down of the extra PCI slots. I do not feel two full-size PCI slits are a lot to ask at this level.

Or better bifurcation support, in order to split the 16 gpu lanes into 4x4 and/or those two other x4 ports for SSDs into four x2. That's where PCIe gen 5 would show its strength on the desktop. Of course also gen 5 gpus need to become commonplace, and gen 5 SSDs need to become half less hot.

 

[csendesmark]

Exactly! A lot are forced in the direction of Threadripper Pro / Epyc just to have active multiple PCI lanes. I am pretty convinced, too when everyone was trying to run dual GPU 3090 and etc. I am convinced they did not understand they were not getting full throttle when using the AMD4/5 Gpus and etc. Pretty sad!!!

I have a RX 7900 XT 20GB PCIe gen4 but only 8× lanes, no drawbacks, so I would say dual 3090 on 2×8 lanes should be fine too

 

[HBSound]

I have a RX 7900 XT 20GB PCIe gen4 but only 8× lanes, no drawbacks, so I would say dual 3090 on 2×8 lanes should be fine too

Just to be transparent. You are possibly right. I do not have the experience to debate. From my point of view - my thoughts and comments are based on what I have and the specs.

 

[csendesmark]

Just to be transparent. You are possibly right. I do not have the experience to debate. From my point of view - my thoughts and comments are based on what I have and the specs.

TPU often checking this out,
see: https://www.techpowerup.com/review/nvidia-geforce-rtx-4090-pci-express-scaling/28.html

 

[david salsero]

On the slide yes, in the text of the article no.
PD. In Windows 11 it has AI functions integrated for a few months.

I don't want to wait for Zen 5, as forums are already impatient for OEMs to introduce the famous 2.2-pound ultrabooks. Right now a safe purchase is: AMD ZEN 4 7040 Phoenix = LPDDR5 + RDNA 3 + USB 4.0 + HDMI 2.1 + AI Artificial intelligence with XDNA architecture developed by Xilinx and all at 4nm, the only processor that surpasses the M2 until the M3 reaches 3nm but at what price ........
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

 

[MikeGR7]

7800X3D being a single chiplet model does not have any noticeable latency penalties in gaming compared to dual chiplet models.
Infact thanks to the massive L3 cache it's frametimes are often better than competitors.

The thing is that while better than dual ccd models, the main latency bottleneck on AMD's architecture is the Infinity Fabric.
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.

Yes lets keep making huge expensive and power hungry monolithic cores on more and more expensive nodes.

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.

Yet you argue for Intel for some reason.

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...

 

[Vader]

I'm more excited for x770 (?) Chipset. Hopefully they're able to connect it to the cpu using PCIe 5.0 lanes, which ryzen 7000 already has, but the chipset didn't support it yet for x670. This would double the bandwith vs x670 and make that motherboard class more exciting for prosumer activities.

 

[Mussels]

They need to work on their default power settings and make these chips not overvolted to the moon - regular CPU voltages, not SoC problem related.
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

 

[TumbleGeorge]

Give us a solid single CCX design

Or just complex CPU without chiplets and additional latency.

 

[Mussels]

Ryzen 1800X Zen 1 March 2, 2017 0 years
Ryzen 3900X Zen 2 July 7, 2019 ~2.33 years
Ryzen 5950X Zen 3 November 5, 2020 ~1.33 years
Ryzen 7950X Zen 4 September 27 2022 ~1.85 years
Ryzen 9950X Zen 5 July to November 2024 guesstimate

AMD can release Zen 5 whenever they like but historically we are looking at just shy of 2 years average between major releases.

I love it when people quote facts.


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

Or just complex CPU without chiplets and additional latency.

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

 

[mb194dc]

Virtually no use case requires the kind of power available even from the last two or three generations.

My suspicion is the CPU market for home will stagnate for many years.

Servers and workstations is where it will be at.

 

[Pepamami]

Make one CCD of e-cores and one CCD of p-cores. This is the way!

E-core was invented by Intel for marketing purposes, so they can slap 16cores tag on their CPUs, E-cores are bad for consumer in every way, with chiplets its even worse.