Discussion Speculation: Zen 4 (EPYC 4 "Genoa", Ryzen 7000, etc.)

[Vattila]

Except for the details about the improvements in the microarchitecture, we now know pretty well what to expect with Zen 3.

The leaked presentation by AMD Senior Manager Martin Hilgeman shows that EPYC 3 "Milan" will, as promised and expected, reuse the current platform (SP3), and the system architecture and packaging looks to be the same, with the same 9-die chiplet design and the same maximum core and thread-count (no SMT-4, contrary to rumour). The biggest change revealed so far is the enlargement of the compute complex from 4 cores to 8 cores, all sharing a larger L3 cache ("32+ MB", likely to double to 64 MB, I think).

Hilgeman's slides did also show that EPYC 4 "Genoa" is in the definition phase (or was at the time of the presentation in September, at least), and will come with a new platform (SP5), with new memory support (likely DDR5).

What else do you think we will see with Zen 4? PCI-Express 5 support? Increased core-count? 4-way SMT? New packaging (interposer, 2.5D, 3D)? Integrated memory on package (HBM)?

Vote in the poll and share your thoughts!

 

[itsmydamnation]

12:49PM EDT - Zen4 by end of 2022

Hmm Seems Zen4 is going to take ~24 Months instead of the established 15 months cadence.
Either that or they are just being purposefully vague as shown in the slides. Otherwise a bit long considering Papermaster mentioned concurrent teams working on the designs.

yeah as said else where , thats the standard way amd presents there road maps now ,

see looks like Zen3 only gets released end of this year.....

 

[zir_blazer]

12:49PM EDT - Zen4 by end of 2022

Hmm Seems Zen4 is going to take ~24 Months instead of the established 15 months cadence.
Either that or they are just being purposefully vague as shown in the slides. Otherwise a bit long considering Papermaster mentioned concurrent teams working on the designs.

Urgh, this is pretty horrible. I think this is an indirect confirmation than we will have a Zen 3 with stacked Cache L3 (Warhol?) as a gap filler for around a year, since there is no other way around it. So, no consumer's guilt for anyone going Cezanne right now given than the next APUs are still far, far away.

I do believe than Genoa will be given launch priority over any consumer Zen 4 variants.

 

[DisEnchantment]

yeah as said else where , thats the standard way amd presents there road maps now ,

Indeed in the future it is gonna be like this

 

[xilli_fiberbit]

There is new patent from AMD related to 3D V-Cache, with teleportation unit on it (i don't know what it does) :

https://twitter.com/x/status/1429503565752639494

 

[uzzi38]

Indeed in the future it is gonna be like this

View attachment 49242

In the future the "In design" and "On Track" will probably go as well lol

 

[eek2121]

Urgh, this is pretty horrible. I think this is an indirect confirmation than we will have a Zen 3 with stacked Cache L3 (Warhol?) as a gap filler for around a year, since there is no other way around it. So, no consumer's guilt for anyone going Cezanne right now given than the next APUs are still far, far away.

I do believe than Genoa will be given launch priority over any consumer Zen 4 variants.

Zen 3 with V-cache provides the performance uplift of an entire new architecture. AMD showed a few games, but nearly all workloads see considerable uplift.

 

[Mopetar]

Seems odd to slot in a Zen3D product that probably won't launch in any appreciable volume until early next year that will be refreshed in under 12 months going by AMD's schedule.

Assuming a more optimistic launch of Zen 4 around this time next year then AMD is really giving that a short life as a product.

Given that AMD still has a lot of the product stack to fill in with Zen 3, it would be surprising if they launched these as anything other than some special Ryzen 5000 parts. Something similar to the XT parts that Zen 2 used as an update/refresh towards the end of the cycle.

 

[Timorous]

Seems odd to slot in a Zen3D product that probably won't launch in any appreciable volume until early next year that will be refreshed in under 12 months going by AMD's schedule.

Assuming a more optimistic launch of Zen 4 around this time next year then AMD is really giving that a short life as a product.

Given that AMD still has a lot of the product stack to fill in with Zen 3, it would be surprising if they launched these as anything other than some special Ryzen 5000 parts. Something similar to the XT parts that Zen 2 used as an update/refresh towards the end of the cycle.

I disagree.

DDR5 will be expensive for a while. Having a DDR4 platform that can compete with Alder for a lower / comparable platform cost and a DDR5 platform that costs more but is faster overall seems like a good way to handle the transition without needing to support both memory standards with a single platform.

 

[zir_blazer]

Seems odd to slot in a Zen3D product that probably won't launch in any appreciable volume until early next year that will be refreshed in under 12 months going by AMD's schedule.

Assuming a more optimistic launch of Zen 4 around this time next year then AMD is really giving that a short life as a product.

Given that AMD still has a lot of the product stack to fill in with Zen 3, it would be surprising if they launched these as anything other than some special Ryzen 5000 parts. Something similar to the XT parts that Zen 2 used as an update/refresh towards the end of the cycle.

Alder Lake is supposed to launch before the end of this year, while Zen 4 is more than a year away. Assuming than Alder Lake is good enough, the only emergency product that AMD has available to not lose its momentum is precisely, Cache stacked Zen 3, otherwise AMD risk to be under the shadow of Alder Lake for a whole year. Doesn't seems like a good plan to let that happen.
I suppose that we will have news about that right after Intel launches Alder Lake.

 

[Mopetar]

DDR5 will be expensive for a while. Having a DDR4 platform that can compete with Alder for a lower / comparable platform cost and a DDR5 platform that costs more but is faster overall seems like a good way to handle the transition without needing to support both memory standards with a single platform.

Zen 3 with or without 3D cache will still be good enough for that.

Alder Lake is supposed to launch before the end of this year, while Zen 4 is more than a year away. Assuming than Alder Lake is good enough, the only emergency product that AMD has available to not lose its momentum is precisely, Cache stacked Zen 3, otherwise AMD risk to be under the shadow of Alder Lake for a whole year.

Supposing Intel actually tops AMD across their full product stack, which may not be as easy as you think, how much is it going to beat AMD by?

If it's something like 5% is AMD really living under the shadow of Alder Lake? Weren't they living under an even greater shadow with Zen, Zen+, and Zen 2 in that case? However did they survive those dark times?

 

[jpiniero]

However did they survive those dark times?

Low prices? Something tells me they don't want to cut prices. That's why Zen 3D is needed.

 

[Abwx]

Zen 3 with V-cache provides the performance uplift of an entire new architecture. AMD showed a few games, but nearly all workloads see considerable uplift.

Perhaps for things like 7Zip or Winrar and any other things memory speed/latency dependant but for things FP like Cinema 4D, Vray, 3DS Max and such softs there will surely be little difference.

That could make sense if gamers are targeted but not sure that they are a majority of the buyers.

 

[Ajay]

It all depends on pricing, both for AMD's Ryzen 5xxx series and Intel's Adler Lake CPUs. If Intel has an advantage, AMD has the option of reducing Prices to remain attractive if necessary. Then the V3D skus can occupy the top price points in their in their CPU portfolio. Worst case, Intel has an edge, and AMD suffers a small reduction in gross margins. Of course, we are getting ahead of ourselves, as Adler Lake isn’t on the market and hasn’t been reviewed. Competition will, as usual, be in all our interests. Otherwise, prices will keep spiraling upwards. It’s too bad that AMD is still capacity limited.

 

[moinmoin]

Seems odd to slot in a Zen3D product that probably won't launch in any appreciable volume until early next year that will be refreshed in under 12 months going by AMD's schedule.

Assuming a more optimistic launch of Zen 4 around this time next year then AMD is really giving that a short life as a product.

Given that AMD still has a lot of the product stack to fill in with Zen 3, it would be surprising if they launched these as anything other than some special Ryzen 5000 parts. Something similar to the XT parts that Zen 2 used as an update/refresh towards the end of the cycle.

I think we can consider Zen+, Zen 2 XT and Zen 3D all as plan Bs that AMD would market as a quasi new gen whenever the actual new gen takes longer.

• Zen 2 on 7nm took longer after GloFo abandoned leading edge nodes so Zen+ slot in smoothly.
• Zen 3 despite heavy shortage arrived smoothly so the Zen 2 XT chips were quickly forgotten.
• It's still up in the air how Zen 4 will fare as that also depends on DDR5. In the meantime Zen 3D can slot in without AMD having to get all sweaty about the potential market circumstances.

 

[Ajay]

Low prices? Something tells me they don't want to cut prices. That's why Zen 3D is needed.

Just lower prices on vanilla Ryzen 5000 series.

 

[jpiniero]

Just lower prices on vanilla Ryzen 5000 series.

The supply will dry up. That's what makes the vcache idea great - they need to keep the Zen 3 dies going for Milan but now they can use the leaky ones for Zen 3D and maintain or increase margins.

 

[Ajay]

The supply will dry up. That's what makes the vcache idea great - they need to keep the Zen 3 dies going for Milan but now they can use the leaky ones for Zen 3D and maintain or increase margins.

If Adler Lake has no effect on Ryzen sales, then there is no need for price cuts. Companies typically cut prices to improve or regain market share. If there is no change in market share, AMD is free to do as they wish (stock analyst's aren’t likely to say anything).

 

[jpiniero]

If Adler Lake has no effect on Ryzen sales, then there is no need for price cuts. Companies typically cut prices to improve or regain market share. If there is no change in market share, AMD is free to do as they wish (stock analyst's aren’t likely to say anything).

That's why I put 'maintain or increase margins'. If Alder Lake ends up being a dud in gaming, it'll give AMD that much more latitude in raising prices for Zen 3D. There's only so many wafers they are going to get from TSMC.

 

[Mopetar]

Low prices? Something tells me they don't want to cut prices. That's why Zen 3D is needed.

I don't even know if they need to do that. Of course some of it depends on Intel's pricing, but more recently it's been Intel that cut prices whereas AMD just adopted the price points Intel had historically used. As long as AMD can sell everything they have there's no reason they couldn't charge higher prices for Zen 3D.

I think we can consider Zen+, Zen 2 XT and Zen 3D all as plan Bs that AMD would market as a quasi new gen whenever the actual new gen takes longer.

Is there any reason why Zen 4 would be significantly delayed? Some people are claiming that they already have samples back so unless it needs a lot of tweaking or fixes, there's no real reason they couldn't launch by this time next year.

Even some of the examples don't make a lot of sense because AMD would have had incredibly firm plans about the Zen 3 launch in place by the time the Zen 2 XT parts launched. There was only a 4 month window between those and the first Zen 3 chips being released.

There could be some external factors like motherboard manufacturers needing more time (or AMD just wanting to give them more to ensure a smoother launch) or wanting better DDR5 availability, but I don't think there's anything holding Zen 4 back since the rumors make it sound as though there aren't many serious changes that would require more time fixing bugs or other performance hiccups.

We still haven't seen Zen 3 TR parts yet either. I have an inkling that Zen 3D will be used in tandem with that launch. We probably get one or two Zen 3D desktop chips as well, but I don't see it replacing the existing product stack. The manufacturing capacity probably just isn't there.

 

[jpiniero]

I don't even know if they need to do that. Of course some of it depends on Intel's pricing, but more recently it's been Intel that cut prices whereas AMD just adopted the price points Intel had historically used. As long as AMD can sell everything they have there's no reason they couldn't charge higher prices for Zen 3D.

The vcache is not free. The additional wafer costs money, the stacking process costs money. Higher prices are going to be needed just to maintain margins and I think they want to go beyond that.

 

[yuri69]

AMD announced the stacking goes into the "highest-end models". There is a reason for that - the tech is quite new.

So the "Zen3D" is definitely not a top-to-bottom solution for fighting both Alder and Raptor Lake lineups.

It's a stop gap solution for the highest halo chips for both client and desktop. This won't be a fit for popular chips as the 5600 or 5800 class.

 

[LightningZ71]

5800x-5950x ARE high end DESKTOP models.the low end is all three APUs and the 5600x.

 

[uzzi38]

AMD announced the stacking goes into the "highest-end models". There is a reason for that - the tech is quite new.

So the "Zen3D" is definitely not a top-to-bottom solution for fighting both Alder and Raptor Lake lineups.

It's a stop gap solution for the highest halo chips for both client and desktop. This won't be a fit for popular chips as the 5600 or 5800 class.

What matters more for mindshare - value or high end performance?

Because even now Zen 3 is objectively not good for value money in the 6 and 8 core categories anyway, yet it still totally wins on the mindshare battle on what is THE desktop processor series to buy.

If Zen 3D is enough to keep the gaming performance crown and even leadership in a number of productivity applications, how will the situation be any different to right now?

 

[Thibsie]

Well they said highest-end models not highest-end 5xxx models.
If they consider Zen3 as highest-end, they can still do it even on the 5600x as they still have loads of Zen1/Zen2 products.

With what's been said, they can act however they want, in fact.

 

[Thibsie]

What matters more for mindshare - value or high end performance?

Because even now Zen 3 is objectively not good for value money in the 6 and 8 core categories anyway, yet it still totally wins on the mindshare battle on what is THE desktop processor series to buy.

If Zen 3D is enough to keep the gaming performance crown and even leadership in a number of productivity applications, how will the situation be any different to right now?

Historically, AMD has almost always been better value than Intel. Didn't prevent them to almost bankrupt and being laughed at by idiots (huh, AMD LOL).