Question AMD preps Ryzen 9 3900XT, Ryzen 7 3800XT, Ryzen 5 3600XT for June 16 ....TweakTown

[JPB]

AMD rumored to refresh Matisse CPUs with Ryzen 7 3850X and 3750X

A big juicy rumor has just hit the internet, with it teasing that AMD would be refreshing its current-gen Ryzen 3000 series CPUs with two new entries.



These two new processors are the Ryzen 7 3850X and the Ryzen 7 3750X, which would succeed the Ryzen 7 3800X and Ryzen 7 3700X processors that are currently on the market. The tweet comes from 'HXL' who published the SKUs of the two new processors and made sure to tell the world they are "Not APU".

The tweet itself was a reply his own tweet of a 'Big day coming...' and that the release would be on 6/16 which after that AMD would launch on 7/7. An interesting timeline of release for the Matisse refresh, as 7/7 was the date that the Zen 2-based Ryzen 3000 series launched in 2019.

If you didn't believe just that single tweet, then how about GIGABYTE kinda confirming the Matisse Refresh in their B550 presentation slide, as noted by VideoCardz. The GIGABYTE motherboards roadmap for support shows "Matisse & Refresh",

"Vermeer", and "Renoir" -- all of which aren't released yet.

AMD seems to be responding to the just-launched Intel Core i9-10900K and Core i7-10700K processors, as these new Ryzen 7 3850X and Ryzen 7 3750X processors would directly compete against Intel's new 10th Gen CPU offerings.

AMD preps Ryzen 9 3900XT, Ryzen 7 3800XT, Ryzen 5 3600XT for June 16

We first heard wind of AMD's upcoming Matisse Refresh in some leaked news barely 24 hours ago, but now we have some even juicier stuff to talk about: the Ryzen 9 3900XT, Ryzen 7 3800XT, Ryzen 5 3600XT.

That's right, according to our friends at Wccftech, AMD is about to launch the Matisse Refresh with the Ryzen 9 3900XT, Ryzen 7 3800XT, Ryzen 5 3600XT processors. The new Ryzen 3000 series CPUs would adopt the "XT" branding, something that Radeon fans will remember and love to see back in the ring.

The new rumor has AMD replacing the Ryzen 9 3900X, Ryzen 7 3800X and Ryzen 5 3600X with the new Ryzen 9 3900XT, Ryzen 7 3800XT, Ryzen 5 3600XT processors -- but increasing the CPU clocks, having enhanced overclocking support, and dropping the price on the current-gen CPUs that they replace.

Ryzen 9 3900XT - 12C/24T + faster clocks + enhanced OC support
Ryzen 7 3800XT - 8C/16T + faster clocks + enhanced OC support
Ryzen 5 3600XT - 4C/8T + faster clocks + enhanced OC support

These new CPUs would compete directly against Intel's latest Comet Lake processors, in the just-launched Core i9-10900K, Core i7-10700K, and Core i5-10600K.

 

[Zucker2k]

Do you think Intel will be able to squeeze 10 cores on 14nm? I would assume 125w 'TDP' What kind of PL2 would you predict?
250w??.

Comet Lake S and Rocket Lake share the same TDP, according to the latest leak by SharkBay. PL2 is same 250w, but only for 8 cores. I can't really say what could be the hurdle for a 10 Core Rocket Lake, power or real estate, but if the 8 Core is supposedly at 250w PL2 already, then it's surely power.

 

[Det0x]

My b-die would pop some wood if it turns out to be true.

What does dram calculator read your bdie quality as ?

 

[DisEnchantment]

Some are speculating that while Zen 3 will launch, it will only be for the server side so desktops won't actually get Zen 3 will 2021. I think it makes sense. Zen 2 is already killing Intel so AMD has no reason to launch Zen 3 right now. They might as well keep refining Zen 2 like Intel and the 14nm++++ for another year and focus on what really brings in the most profits, servers and data centres.

My 0.02 € ...
AMD has to launch Zen3. On time.
That is how they can demonstrate to the OEMs and hyperscalers that they are here to stay and that they can keep their roadmaps. Its not just about beating the competition in performance and value but also timelines and commitments.
Lisa, Mark, Forrest etc keep saying it. Just couple of days ago Frank said it.
Cray is waiting for AMD to deliver Perlmutter. OEMs are waiting for AMD to be on time for their next desktops and laptops product refreshes. And so on.
Several new OEMs are on board for Zen3. They better not mess it up.
In fact AMD has to give their everything to not mess up so much of things. Desktops, laptops, supercomputers, consoles, mobiles (Hello Exynos 10 next February)😧

That said... I'm still baffled every Tom Dick and Harry can get a hold of Zen 2+ leaks but nothing... on Zen3.🤔


I'm eagerly waiting for a 4950X to complement my 3900X build machine. CircleCI enabled using your local PC for CI/CD now 😊

 

[jpiniero]

Some are speculating that while Zen 3 will launch, it will only be for the server side so desktops won't actually get Zen 3 will 2021.

The rumor is 12 and 16 core only at launch.

 

[DrMrLordX]

My 0.02 € ...
AMD has to launch Zen3. On time.

Agreed. We know/suspect that Zen3 launches in October, and it's only 4900x/4950x (or whatever they call them). The 12c part needs to be $499 and the 16c part needs to be $749 or less.

 

[chrisjames61]

My 0.02 € ...
AMD has to launch Zen3. On time.
That is how they can demonstrate to the OEMs and hyperscalers that they are here to stay and that they can keep their roadmaps. Its not just about beating the competition in performance and value but also timelines and commitments.
Lisa, Mark, Forrest etc keep saying it. Just couple of days ago Frank said it.
Cray is waiting for AMD to deliver Perlmutter. OEMs are waiting for AMD to be on time for their next desktops and laptops product refreshes. And so on.
Several new OEMs are on board for Zen3. They better not mess it up.
In fact AMD has to give their everything to not mess up so much of things. Desktops, laptops, supercomputers, consoles, mobiles (Hello Exynos 10 next February)😧

That said... I'm still baffled every Tom Dick and Harry can get a hold of Zen 2+ leaks but nothing... on Zen3.🤔


I'm eagerly waiting for a 4950X to complement my 3900X build machine. CircleCI enabled using your local PC for CI/CD now 😊

Ryzen 3 info must be locked up like a drum.

 

[french toast]

Comet Lake S and Rocket Lake share the same TDP, according to the latest leak by SharkBay. PL2 is same 250w, but only for 8 cores. I can't really say what could be the hurdle for a 10 Core Rocket Lake, power or real estate, but if the 8 Core is supposedly at 250w PL2 already, then it's surely power.

Yea I think power and corresponding heat is going to be an issue, I think willow has the frequency potential on anciently mature 14nm as in no hard frequency wall..only power and heat holding it back.

I also think they can't regress on cores vs Comet so 10 core for top end, which I think means mesh = slightly higher memory latency.
No way they have got the real estate for 12 core Sunny/Willow, they wouldn't produce the amounts at the kind of frequencies they will need to make a difference to Zen 3, neither would they make acceptable margin (for them -65%?)seeing as they will probably need to drastically cut prices to keep market share palatable for shareholders.
Can't see how it could compete with Zen 3 from those ES samples (if true).

Anyway back on topic, these zen2+ leaks are real interesting, nice little bump which would make Ryzen a must buy if they don't hike prices.

 

[Atari2600]

Some are speculating that while Zen 3 will launch, it will only be for the server side so desktops won't actually get Zen 3 will 2021. I think it makes sense. Zen 2 is already killing Intel so AMD has no reason to launch Zen 3 right now. They might as well keep refining Zen 2 like Intel and the 14nm++++ for another year and focus on what really brings in the most profits, servers and data centres.

If Zen3 is being confined to EYPC at launch, that will be due to extraordinarily high demand on that side (as the process is just an evolution of where they are right now so supply should be reasonably predictable).

That would be extremely good for AMD's long term health and probably better for us in the long run.

Chipzilla aren't going to go away, so having two strong competitors on sound financial footings is best for everyone.

 

[Kenmitch]

What does dram calculator read your bdie quality as ?
View attachment 21611

My 3200MHz CL14 kit.

 

[Makaveli]

What does dram calculator read your bdie quality as ?
View attachment 21611

 

[Makaveli]

double post delete.

 

[jpiniero]

I also think they can't regress on cores vs Comet so 10 core for top end

It's definitely physically 8.

Agreed. We know/suspect that Zen3 launches in October, and it's only 4900x/4950x (or whatever they call them). The 12c part needs to be $499 and the 16c part needs to be $749 or less.

AMD definitely has leeway on pricing. There has been zero about Intel releasing another HEDT product any time soon, and they still haven't shown they are willing to properly supply DIY for the mainstream products. Now if Vermeer is slower than Rocket in gaming, they can adjust prices when Rocket launches. If it's slower than Comet, that's a problem.

 

[AnandThenMan]

My 0.02 € ...
AMD has to launch Zen3. On time.
That is how they can demonstrate to the OEMs and hyperscalers that they are here to stay and that they can keep their roadmaps. Its not just about beating the competition in performance and value but also timelines and commitments.
Lisa, Mark, Forrest etc keep saying it. Just couple of days ago Frank said it.
Cray is waiting for AMD to deliver Perlmutter. OEMs are waiting for AMD to be on time for their next desktops and laptops product refreshes. And so on.
Several new OEMs are on board for Zen3. They better not mess it up.
In fact AMD has to give their everything to not mess up so much of things. Desktops, laptops, supercomputers, consoles, mobiles (Hello Exynos 10 next February)😧

That said... I'm still baffled every Tom Dick and Harry can get a hold of Zen 2+ leaks but nothing... on Zen3.🤔


I'm eagerly waiting for a 4950X to complement my 3900X build machine. CircleCI enabled using your local PC for CI/CD now 😊

In summary AMD better have flawless execution or game over? Intel on the other hand can bumble around on 7nm+10²⁴ and ...... all good.

dammit I screwed up the joke, 14nm+10²⁴ not 7. Feels like Intel has been on 14nm for decades at this point.

 

[dacostafilipe]

Zen 2 is already killing Intel so AMD has no reason to launch Zen 3 right now. They might as well keep refining Zen 2 like Intel and the 14nm++++ for another year and focus on what really brings in the most profits, servers and data centres.

Going by what AMD people said in the past, they are not thinking like this. They are aware that Intel could be back in the near future and that only way to beat them is to push hard like they do now. I mean, Intel will come back, until then AMD needs to shrink the gap or even outpace the future Intel monster

IMO, those XT models only show how confident AMD is at the moment. Even with a small number of them, maybe because they are not easy to bin, they will be their Halo products until Zen3 and provide a healthy margin on top.

Oh, and it speaks a lot for how flexible AMD can be. That's something shareholders love.

 

[Gideon]

Going by what AMD people said in the past, they are not thinking like this. They are aware that Intel could be back in the near future and that only way to beat them is to push hard like they do now. I mean, Intel will come back, until then AMD needs to shrink the gap or even outpace the future Intel monster

IMO, those XT models only show how confident AMD is at the moment. Even with a small number of them, maybe because they are not easy to bin, they will be their Halo products until Zen3 and provide a healthy margin on top.

Oh, and it speaks a lot for how flexible AMD can be. That's something shareholders love.

Besides all of the above, they have also stated publicly multiple times that zen 3 will come "late this year". Postponing it significantly would not be taken without some flack (also from stock-owners).

 

[piokos]

Some are speculating that while Zen 3 will launch, it will only be for the server side so desktops won't actually get Zen 3 will 2021. I think it makes sense.

I think it doesn't make sense. Which absolutely doesn't mean it won't happen.

Zen 2 is already killing Intel so AMD has no reason to launch Zen 3 right now.

Zen2 (or rather: TSMC's node) is an obvious leader right now, but this might change with Tiger Lake. Why would AMD want to wait for Intel to catch up?

They might as well keep refining Zen 2 like Intel and the 14nm++++ for another year

Not doing a X+++ was meant to be the key differentiating factor that AMD fans point out all the time. Will they accept something similar from AMD?

and focus on what really brings in the most profits, servers and data centres.

You may be overestimating the profitability of datacenter segment.
AMD undercuts Intel's consumer lineup by maybe 20% (order of), while their server CPUs are pretty much 3x cheaper.
The general understanding is that server chips are pretty much given away at break-even prices (or maybe even under) to quickly expand market share (which clearly works).

At this point we don't know what Ice Lake-SP will bring, but best case scenario is that it matches Zen2 on performance per socket - still at a higher price.
In other words: AMD could keep selling Zen2 for another year (just like they skipped Zen+).

Also with Zen, we had Zen+ and then Zen 2. I always thought it was weird AMD didn't announce a Zen 2+ and is instead going straight to Zen 3. If there is a Zen 2 refresh coming, it would explain why AMD and the board partners originally didn't want to support Zen 3 on the B450 chipset.

Zen2+ is pretty much happening right now - under the "XT" SoC naming. Slightly improved node, higher clocks.

 

[geokilla]

Besides all of the above, they have also stated publicly multiple times that zen 3 will come "late this year". Postponing it significantly would not be taken without some flack (also from stock-owners).

COVID-19 excuse can be used and I don't think many people will blame them. COVID-19 is killing the world right now.

 

[DrMrLordX]

AMD definitely has leeway on pricing.

Not as much as you might think. AMD has already developed a healthy margin for their products (46% as of their last earnings report). If they hold to their current pricing in the face of lackluster competition, there emerges the potential for market share expansion. Which would you rather have? 5-10% higher margin or significant gains in OEM marketshare?

Mind you, I am aware that DiY pricing has less to do with prices paid by OEMs. AMD may "double deal" by offering OEMs discounts while increasing prices to the DiY market. AMD has made significant inroads in the DiY market already.

Zen2 (or rather: TSMC's node) is an obvious leader right now, but this might change with Tiger Lake.

Unlikely. Intel isn't using Tiger Lake in enough product segments for it to matter that much.

 

[Ajay]

Agreed. We know/suspect that Zen3 launches in October, and it's only 4900x/4950x (or whatever they call them). The 12c part needs to be $499 and the 16c part needs to be $749 or less.

Why? Zen3 will blow CML out of the water, why should AMD lower it's prices???

 

[Ajay]

Man, this is just lame marketing stuff:

 

[DrMrLordX]

Why? Zen3 will blow CML out of the water, why should AMD lower it's prices???

Uh

3900x was $499, 3950x was $749. 3950x carried that premium because it was the first time a 16c had happened on desktop. Now they may not lower it, but if they are going to leave open the prospect of a higher core count on a desktop socket . . .

 

[Panino Manino]

3900XT全核能上4.6？有这么猛吗？ - 电脑讨论 - Chiphell - 分享与交流用户体验

3900XT全核能上4.6？有这么猛吗？,不知道是真的还是假的，感兴趣的可以去看看https://www.cpu-monkey.com/en/,电脑讨论,讨论区-技术与经验的讨论 ,Chiphell - 分享与交流用户体验

www.chiphell.com

"laughs"
Is these benchmarks are really true and the latency improved a bit, makes me feel bad for Intel.

 

[Hitman928]

You may be overestimating the profitability of datacenter segment.
AMD undercuts Intel's consumer lineup by maybe 20% (order of), while their server CPUs are pretty much 3x cheaper.
The general understanding is that server chips are pretty much given away at break-even prices (or maybe even under) to quickly expand market share (which clearly works).

This is not true. In one of the recent earnings calls, AMD stated that margin for Epyc is well over 50%.

 

[piokos]

This is not true. In one of the recent earnings calls, AMD stated that margin for Epyc is well over 50%.

That's gross margin. Possible, given 46% total gross margin. Can you link the source?

AMD's profit margin was 9%.
And it was a very good quarter - they actually expect the rest of the year to be worse.

By comparison, Intel's profit margin for PC CPUs is 35% and for data center chips it's around 40%. So in their case Xeons are more profitable, but again: their reference pricing is way above AMD's.

 

[jpiniero]

Not as much as you might think. AMD has already developed a healthy margin for their products (46% as of their last earnings report). If they hold to their current pricing in the face of lackluster competition, there emerges the potential for market share expansion. Which would you rather have? 5-10% higher margin or significant gains in OEM marketshare?

OEMs would be fine with selling Matisse Refresh for gaming desktops.

Uh

3900x was $499, 3950x was $749. 3950x carried that premium because it was the first time a 16c had happened on desktop. Now they may not lower it, but if they are going to leave open the prospect of a higher core count on a desktop socket . . .

I imagine the 3600X/3800X/3900X will be discontinued when the XT parts launch, and the 3950X when Vermeer launches.