Will Ryzen 6000 CPUs be based on the N6 fab process (Warhol chip) and release in 2021?

[g67575]

Another title could be - What are AMD's plans for 2021?

It looks like TSMC's N6 EUV process may be key to new AMD CPUs in 2021, and possibly even GPUs. Warhol / Zen 3+ processors based on this node may release this year, according to this article:
https://www.kitguru.net/components/cpu/joao-silva/amd-zen-3-zen-4-and-zen-5-details-begin-to-leak/

It's possible this will be the first AM5 / DDR5 platform, this would make sense because in 2021, AMD needs something to compete with both Rocket Lake and Alder Lake (late Q4 or Q1 2022?). Also, the first modules of DDR5 have already been produced, so I think it's definitely doable.

This is mostly relevant to me as I'm still using DDR3 1600Mhz, and some games run poorly (40-50FPS) on low frequency RAM, such as Watchdogs: Legions. So, if AMD releases a AM5 motherboard with DDR5 compatibility in 2021, this will likely be what I choose to upgrade to.

To me, the least speculative part of this information, is the plan to launch N6 fabrication process based CPUs in 2021. According to Wikichip, N6 "volume production (is) expected before the end of the year" (dated January, 2020). Link here:
https://fuse.wikichip.org/news/3227/tsmc-q4-7nm-dominates-revenue-preps-5nm-ramp-6nm-by-eoy/

For a while though, many thought an enhancement to 7nm called 7NP would be used for CPUs in 2021, with 5nm in 2022. Ultimately N6 (EUV) and 7NP (small improvement vs 7N) are both enhancements of 7nm, but I think it makes much more sense to use the more advanced N6 process, partly because there should be more production capacity available (due to less tech companies using this node).

 

[g67575]

Warhol might still be on AM4. Since they are not increasing core count, sticking with the cheaper DDR4 might make more sense. Hopefully, they will use something better than vanilla 7nm. Using a newer GF node for the IO die would also be nice.

Yeah, I really doubt they would stick with standard TSMC 7FF for 3 CPU generations in a row. With Rocket Lake already slightly ahead of the Ryzen 5000 series in single core performance, They will probably need an improved node to compete with Alder Lake's significant fab. process shrink (10nm).

 

[g67575]

There's also this too "It’s generally expected that Zen 3 will be the last AM4/DDR4 refresh for AMD and that the company will move to new DDR5 platforms in 2021". Link here:
https://www.extremetech.com/computi...-coming-in-2020-talks-challenges-in-notebooks

 

[g67575]

Paul of Paul's Hardware has reported that the next AMD CPUs are expected to be 5nm and this will relieve the pressure on 7nm fabs so there will be greater GPU availability.

Wouldn't shifting CPU production over to the N6 fab. process also reduce the pressure on the in demand N7 process?

I think because of the EUV lithography, N6 is a more specialised (and probably expensive to develop) product, and this would make it less attractive to many tech companies, who don't necessarily need the extra transistor density that N6 should provide.

I've also read that the design rules for 6N are the same / similar to TSMC 7nm, meaning that AMD should be able to (more easily) port their existing designs to 6N.

 

[g67575]

I don't believe all this nonsense about Navi 31/ Navi 41 leaks. It's made up, there's no way they could have info about RDNA 3/4 designs at this point.

Also worth nothing that doubling the Compute Unit (CU) count alone wouldn't double performance of an AMD GPU.

Is it likely 5nm GPU single GPU designs are in the works for next gen? I think so.

TSMC 6N fab process is planned for Intel's graphics cards apparently.

 

[g67575]

Tbh, I can see AMD delaying Warhol until Intel's Alder Lake CPUs (10nm desktop) are available to buy with good availability. So Q1 at the latest. This doesn't answer how they will beat Rocket Lake, though, but maybe the Ryzen 5000 series chips will sell well enough regardless.

Another possibility is, there is no 'Warhol' chip, AMD will attempt to rush to 5nm (EUV) CPUs out to compete with Alder Lake, likely in 2022. The lack of concrete information regarding Warhol, is a bit strange, if it is due to release this year.

Kinda tired of seeing "INSANE... Performance" in every video title from Paul (RGT). Clearly, this guy doesn't do understatement.

 

[g67575]

I think Warhol will be significant if utilizes the 6N Fab process. If not, then I agree that it will probably only be a small performance improvement over the Ryzen 5000 series. If DDR5 is ready, can't see a reason for AMD not to release it along with Warhol.

EDIT - It's possible AMD could introduce an AM4+ socket, with support for both DDR5 and DDR4. Apparently, DDR5 has the same 288 pin count as DDR4. That would be a pretty safe option in my opinion (and would give customers lots of choice), then design AM5 just for Zen 4 / future CPUs.

This would be similar to AM3, which had support on some motherboards for both DDR3 and DDR2.

In terms of overall performance improvement, I think it's likely to be another 10% or more vs the 5000 series, based on previous Zen generations - Even the 2700X managed about 10% single core performance vs the 1600X (a bit more vs the Ryzen 1600).

 

[g67575]

I don't think this (unofficial) leaked roadmap is quite right:

Spoiler: Roadmap

Cezanne based APUs will become available in Q1 2021 (already available on the Chinese market), in this diagram, it's placed too far forward. It should probably be in line with Vermeer (the Ryzen 5000 series).

This is relevant because it (incorrectly) suggests Rembrandt won't release until 2022, which is apparently based on the N6 process. I find it very hard to believe they would utilize N6 for an APU, but not for the higher clocked standalone CPU variant (Warhol). Also, I doubt the type of DDR memory used would be different to Rembrandt.

I'm partly wondering if this is just AMD trying to get fans excited for 5nm and Zen 4, by playing down that the Ryzen 6000 series will be based on N6 and likely release in 2021.

 

[g67575]

It's interesting that AMD hasn't mentioned the N6 node at all on roadmaps or announcements.

 

[g67575]

Does anyone think the Ryzen 6000 series (Warhol) will be 7nm based instead, and utilize ddr4 memory on an AM4 CPU socket?

Its possible the N6 node rumours are wrong...

 

[g67575]

Yes

Personally, I think AMD would rather there were no leaks regarding TSMC's N6 process and AMD CPUs - ofc, they want to keep selling Ryzen 5000 series processors, at least for as long as they have an advantage over Intel.

I think the N6 process could be quite significant though, as it is the first process designed to utilize EUV in the initial version - TSMC have stated they expect most customers to move from N7 and N7+ (EUV) processes to N6. It's possible there isn't much difference between the EUV fab. processes (they both other around 20% transistor density improvement vs standard 7nm), that it's just a rebranded version of N7+, but with the same design rules as N7.

EDIT - Actually, there is a slight difference, explained here "TSMC's N7+ will be the first node to leverage EUV, using up to four EUVL layers, while N6 expands it up to five layers". Then a large improvement with 5nm, "N5 cranks EUVL up to fourteen (allowing for 14 layers)". Link here:
https://www.techpowerup.com/255097/tsmc-expects-most-7nm-customers-to-move-to-6nm-density

I think 6N is quite likely to offer a similar transistor density to Intel's 10nm used in Alder Lake (which does not utilize EUV lithography - something Intel appears to be behind on despite developing this technique...).

If it's not used in desktop CPUs, it does make you wonder what the N6 node will be used for - TSMC has invested considerable resources in it's development. It's possible it will be mostly used in GPUs (we know Intel is planning this for 'XE' graphics cards), at least until there are more optimized versions of 5nm (EUV), that will further improve the transistor density or power efficiency of GPU dies.

 

[g67575]

I think this shows that Intel is in trouble, if AMD uses 7N+ or N6 (transistor density is almost identical) in 2021:

Spoiler: fabrication process comparisons

Especially if they don't release Alder Lake CPUs (10nm) when AMD releases the Ryzen 6000 series. There's also no sign yet that Intel has been able to improve the transistor density of 10nm much since 2019.

Also, the fact that the Wikipedia 7nm article suggests N6 "2020 production" could mean either that the release of 6N based products is imminent, or it's possible there could be a delay in manufacturing that hasn't been reported yet.

 

[g67575]

Apparently, Mediatek is producing APUs (octa-core CPU + five-core G57 GPU) for Chromebooks based on the TSMC 6N fab. process, which will be available in Q2 2021 (later for the MT8195), link here:
https://corp.mediatek.com/news-even...s-designed-for-next-generation-of-chromebooks

So, it looks like the 6N process is ready for use in processors.

 

[g67575]

Lisa Su gonna milk you dry, boy.

I won't lie, this creates some disturbing imagery...

2017 - Ryzen 1000
2018 - Ryzen 2000
2019 - Ryzen 3000
2020 - Ryzen 5000 - note - we skipped the 4000 desktop CPU series because in our humble opinion this series is just *that* good.

Hhmmm, is there some kind of kind of pattern forming here, just beyond the reaches of my conscious mind?

I'm fine with 6 CPU cores personally, although a lower clocked 8 core CPU would probably be worth it in the long term, if it only cost a bit more. AMD does not seem keen to release a Ryzen 3700X successor, except for the impossible to buy, awkwardly named Ryzen 5800 (oem).

 

[g67575]

well, its not surprising, the 5600x has higher ST performance than current Intel CPUs. The prices are likely to fall when rocket lake releases - that is, if Intel can produce sufficient stock.

 

[g67575]

There's some info about AMD's server 'Genoa' platform, that could release in 2021. Apparently, it has a new SP5 socket and potentially DDR5 RAM support, link here:
https://www.wepc.com/news/amd-inadv...epyc-zen-3-milan-and-zen-4-genoa-server-cpus/

This reminds me of the similar rumours about Intel 'Sapphire Rapids', DDR5 server processors in 2021.

There's also some 7nm+ Zen 3 server CPUs ('Milan') that entered production in Q3 2020, link here:
https://www.tomshardware.com/news/a...noa-architecture-microarchitecture,40561.html

 

[g67575]

Isn't anyone thinking about the Ryzen 6000 (or Alderlake) processors yet? Maybe it's too soon?

 

[g67575]

I'd be more interested in Alderlake if the enhanced SuperFin 10nm process actually improved the transistor density, from the original Intel 10nm in 2019. Still, if it if does release this year, I may be tempted to build a new system, if they can offer a cheap B series chipset and CPU combo.

Rocket lake is gonna be a very short lived CPU generation, if Alderlake releases this year - considering Rocket lake needs a new motherboard + chipset (unless upgrading on an expensive H470 or Z490), it's looking like a bad option in terms of value.

I'm unsure if the Warhol generation actually exists - If it doesn't release this year, I can't see how it would fit into AMD's lineup. I think this codenames are often made up tbh, or based on early speculation or designs that aren't taken up.

My best guess would be, that Warhol may release alongside Genoa (4th gen), if so, they could share a lot of features. It looks doubtful though, AMD is going to release Milan (3rd gen, Zen 3) server CPUs soon, they previewed them at CES in January 2021, slide here:

Spoiler: Slide

A simple refresh of Zen 3 is probably what AMD will release this year, maybe based on the N6 fab. process.

 

[g67575]

almost certain I read the next AMD cpus would be on a new socket as they has taken am4 as far as they could

Maybe you could try to find where you read that, it sounds interesting.

I think they will probably stick with AM4, but with a 'new' fab. process like TSMC 6N. That would give them ~20% transistor density improvement vs 7nm Ryzen. Or, it could use the very similar 7nm+ (EUV) instead, which is what the delayed 3rd gen. server products (Milan EPYC) will apparently utilize. That will should give them a large enough perf. improvement to equal or exceed Rocket Lake processors (which offer a 19% IPC improvement vs Skylake gen 1 CPUs).

Probably not DDR5 on Ryzen 6000 unfortunately. AMD does not need to compete with Alder Lake DDR5 systems until they release in 2022, I think 1 year after Rocket Lake releases next month.

 

[g67575]

Whatever chip is used for Ryzen 6000, be it Zen 3 (refreshed), 'Zen 3+', or Zen 4, AMD has no intention of making any statements about it yet. they'd rather talk about 2022 and beyond. Afterall, the 5600x and 5800x are doing very well, AMD is making more net profit than they have in a long time (ever?).

To me, this lack of AMD statements / official roadmaps for 2021 doesn't seem like a positive thing. They mention 2022 I think, because the technology isn't ready yet, Milan (Zen 3 + DDR4 based, planned for 2020) is due to be released quite soon.

In fact, it looks like AMD wants to focus on releasing mobile products and improving chip production capacity in the 1st half of 2021.

A 5.3ghz (boost) 8 core Rocket lake is only 2-8% faster than a 4.8ghz (boost) 12 core Ryzen 5000 CPU in games according to a CES Intel slide:

Spoiler: Slide

I notice a (likely) deliberate absence of XT products in the current Ryzen 5000 line-up, which were a part of the Ryzen 3000 series. I think AMD will simply release (about 10%) overclocked versions of the 5600X, 5800X and 5900X to compete with Rocket Lake, maybe with extra L3 cache if necessary. So, that would put them at 5.0ghz or above.

Clock for clock, the Ryzen 5000 series appears to still be (slightly) ahead of Rocket Lake in terms of IPC, so I think overclocked versions would make sense.