AMD Ryzen 4000 7nm+ Zen 3 CPUs Spotted

...while quoting them as saying the opposite.

Hey I was just being fair, they did really go out of their way to squash the 10nm desktop is dead reporting. /sarcasm

Intels 10nm head has said it was a mess and admitted they Fd up hard. That they had to basically throw out 2 years of work cause they got stubborn. So if he is to be believed earlier this year Intel basically tossed their 10nm designs, and are redesigning 10nm with a larger gate (so no longer can it be argued to be as good as 7nm)... considering the time frames invovled in CPU design it doesn't seem very realistic that they could get those redesigned true 10nm parts out before the end of 2020 at the earliest if they hit zero issues along the way.

6 months after he admitted they needed to eat their humble pie... a Intel PR person answers a fringe site reporting that they killed 10nm desktop with once sentence that reads as "nu uuuu we still going to make 10nm desktop parts sometimes". Sure you are.

Who knows though. perhaps Intel has a 10nm Core i9-11980wtfc ready Q1 2020..... and despite pretty much 100% loosing the enthusiast market they have kept it quite. (if they had a 10nm desktop part even half working they would drop hints show it off behind loose leaky NDAs and all the unusual PR crap that makes people hold off on buying anything for awhile)
 
Intel isn't ordained to be the #1 CPU company forever.

We heard the same thing 30 years ago about IBM. There huge, they have armies of lawyers and engineers... and a sale force that takes no prisoners. Sometimes just throwing more and more good money after bad doesn't solve the problem. 10nm was a massive investment that appears to be a fail. Which means Intel has next to nothing to move their product forward for close to 2 years. They brought in some new head Engineers to try and fix things but really they waited to long to backtrack. Which means any redesigned 10nm now won't hit market in time to make any impact on the market. Its a minor miracle they got 14nm+++ to a point where they are still even in the game at all on the high end.

Anyway if it happened to IBM it can happen to anyone.

I just said it was unlikely not impossible.
 
  • Like
Reactions: ChadD
like this
Hey I was just being fair

About as fair as MoG's proselytizing ;)

Fair is that 10nm has had setbacks. Really, it's about four years too late -- but that doesn't mean that it won't be successful when released.
 
No offense but I hope they dont release a Threadripper 4 until at least 2021. Gawd I'm so tired of being obsolete 3 months after buying a new platform.
Not obsolete obviously. Just not the biggest e-pen in the room :)
 
Like, 10nm and 7nm products are on their way?

Every production line running weak 10nm products is one that's not running strong 7nm products. I at least half expect them to produce a handful of 10nm SKUs, declare victory, and abandon the line. There were never any other chips to go along with the i3-8121U, for example.
 
Every production line running weak 10nm products is one that's not running strong 7nm products. I at least half expect them to produce a handful of 10nm SKUs, declare victory, and abandon the line. There were never any other chips to go along with the i3-8121U, for example.

They have world-beating ultrabook CPUs available from OEMs today, currently sweeping all of the benchmarks.
 
Every production line running weak 10nm products is one that's not running strong 7nm products. I at least half expect them to produce a handful of 10nm SKUs, declare victory, and abandon the line. There were never any other chips to go along with the i3-8121U, for example.
The fact that there ain't shit speaks volume. 50% yield for that (no iGPU) and nothing much since?

>OH YEAH BUT GUYS THERE IS 10NM PRODUCT COMING TO DESKTOP TRUST US
>wait for Vega navi 10nm
lol sure thing shintel shills.
 
As a APU user I am hoping Renoir is more robust on the iGPU side. I feel the Ryzen x86 cores are being held back by the limits of Vega 11 being to weak now. AMD is on the cusp of delivering 1080P low detail settings for many popular games, for some AAA level games you may have to drop to 720P like in Battlefield 5 ultra low preset gets 50-60 it can spike higher but the holding average is in that 60FPS range. Skyrim plays well at 1080P or best at 1600x900 and Fallout 4 dependign on how much visual you want to sacrifice 1080P is very possible with high FPS if you use ultra low texturing mods or comfortably playable with removing unnecessary stuff and tweaking the games .ini file.generally stable frames.
 
Fair is that 10nm has had setbacks. Really, it's about four years too late -- but that doesn't mean that it won't be successful when released.
Come on, 10 nm is demonstrated, confirmed and admitted outright failure. It's not late, it's DoA.
 
10nm for the desktop is never coming as they will never hit the clock speeds needed. 7nm will be the wait for desktop chips from Intel.
 
Which are?

Note that Ice Lake is showing > 10% performance over Skylake cores.

4.1 GHz is the max speed on Ice Lake and that is 1 core only 3.6 if all 4 cores are used. They would need to hit 4.6 to 4.5 all cores at the minimum to equal their current chips and that would require Intel to actually be capable of making a large chip on 10nm which they have yet to even try.
 
4.1 GHz is the max speed on Ice Lake and that is 1 core only 3.6 if all 4 cores are used. They would need to hit 4.6 to 4.5 all cores at the minimum to equal their current chips and that would require Intel to actually be capable of making a large chip on 10nm which they have yet to even try.

...in a 15w chassis.

How do you know they haven't tried? Because they're still using 14nm for desktops? Lol.
 
...in a 15w chassis.

How do you know they haven't tried? Because they're still using 14nm for desktops? Lol.

That high clock speed on 10nm required 28 watts it's even slower at 25 watts and at 15 it's quite a bit slower. Because if they could they would have them and we would have rumors of them as they are getting their butt kicked on the desktop and server side by current AMD offerings. Intel would do anything they could to stop the bleeding at the server side.
 
  • Like
Reactions: N4CR
like this
...in a 15w chassis.

How do you know they haven't tried? Because they're still using 14nm for desktops? Lol.

If Intel had a 10nm desktop part ready to go, we'd know about it. They aren't going to let AMD eat their lunch if they had a respectable alternative in the wings. As has already been said, they don't. They've admitted it. 10th gen desktop is another 14nm part (new socket of course :rolleyes:). 2021 is the time frame for something other than 14nm. I didn't even realize this was still being debated...
 
Last edited:
I dropped in a 3600 this summer and couldn't be happier. I'm more than pleased switching to Team Red.
 
That high clock speed on 10nm required 28 watts it's even slower at 25 watts and at 15 it's quite a bit slower.

This is the opposite of the reviews I've seen.

Because if they could they would have them and we would have rumors of them as they are getting their butt kicked on the desktop and server side by current AMD offerings.

From a performance perspective, Intel has up to eighteen cores covered; given that desktop users don't generally need more than four cores, I can see Intel not being interested in switching over to 10nm for that line until volume is up.

The server side is basically the opposite: they could use more cores, but unless they have volume, even hinting about such a SKU would be ill advised given their current production issues for 14nm Xeons.

If Intel had a 10nm desktop part ready to go, we'd know about it. They aren't going to let AMD eat their lunch if they had a respectable alternative in the wings. As has already been said, they don't. They've admitted it. 10th gen desktop is another 14nm part (new socket of course :rolleyes:). 2021 is the time frame for something other than 14nm. I didn't even realize this was still being debated...

They've said that it's on the way. At this point, after admitting fault (to themselves and to the world), Intel's marketing appears to have become reserved.
 
They've said that it's on the way. At this point, after admitting fault (to themselves and to the world), Intel's marketing appears to have become reserved.

But when? The only thing I've seen is maybe by 2021 after Comet and Rocket Lake on 14nm. It's certainly not an imminent release.

https://www.tomshardware.com/news/intel-yes-there-will-be-10nm-desktop-cpus

There's certainly not going to be a "surprise" announcement 10+ cores at 5Ghz on 10nm Intel Mainstream desktop part anytime in 2020.
 
There's certainly not going to be a "surprise" announcement 10+ cores at 5Ghz on 10nm Intel Mainstream desktop part anytime in 2020.

Yeah, that's probably unlikely, simply because they're keeping desktop and some mobile on 14nm for now.
 
If Intel had a 10nm desktop part ready to go, we'd know about it. They aren't going to let AMD eat their lunch if they had a respectable alternative in the wings. As has already been said, they don't. They've admitted it. 10th gen desktop is another 14nm part (new socket of course :rolleyes:). 2021 is the time frame for something other than 14nm. I didn't even realize this was still being debated...


you're better off talking to a brick wall then trying to knock any sense into him, but good luck.


As a APU user I am hoping Renoir is more robust on the iGPU side. I feel the Ryzen x86 cores are being held back by the limits of Vega 11 being to weak now. AMD is on the cusp of delivering 1080P low detail settings for many popular games, for some AAA level games you may have to drop to 720P like in Battlefield 5 ultra low preset gets 50-60 it can spike higher but the holding average is in that 60FPS range. Skyrim plays well at 1080P or best at 1600x900 and Fallout 4 dependign on how much visual you want to sacrifice 1080P is very possible with high FPS if you use ultra low texturing mods or comfortably playable with removing unnecessary stuff and tweaking the games .ini file.generally stable frames.

if the clock boosts that were seen on radeon VII can be scaled to 7nm vega 11 then we might see some ok gains. but thinking i might pick one up for a system upstairs anyways just to see what kinda stupid stuff can be done with the igp and whether or not it's heavily effected by infinity fabric clocks to see if maybe you could put the cpu in 2:1 fclock and run stupid high memory clocks for a IGP performance improvement.
 
Last edited:
That alone should be a clue how big of a bust 10nm has been for them.

'Bust'? It's behind, and they've had to do an expensive rework, but how can it be a bust when they're opening up new 10nm lines and readying production?

This is an inexplicable line of thinking.
 
'Bust'? It's behind, and they've had to do an expensive rework, but how can it be a bust when they're opening up new 10nm lines and readying production?

This is an inexplicable line of thinking.

Inexplicable? Intel's been on 14nm since Broadwell (2014). When else have they used the same lithography for 5 (going on 6) years? The jump from 22nm to 14nm was 3 years by comparison.

Obviously, everyone is noting there are problems historically for Intel's 10nm. It's the long term effects of the problems that people disagree on.
 
https://www.phoronix.com/scan.php?page=article&item=sandy-fx-zombieload&num=1

Intel still wins but its not the pure blow out that was reported at the FX launch. Its funny seeing some of the server tests where the FX chips gained performance with mitigation. Intel still wins overall... but there are some examples such as context switching (which would be very much like some of the old heavy multi tasking tests you used to see people include in reviews) where Intel went from being 3-4x faster then AMD to basically the exact opposite.

Michaels conclusion;
"Lastly is a look at the geometric mean for all of the tests carried out successfully on all of the processors/combinations. The Intel Sandybridge/Ivybridge CPUs all basically remained faster than the AMD FX CPUs benchmarked even when mitigated and disabling SMT with the exception of the Core i3 2120. The Core i3 Sandybridge processor's geometric mean went from being ahead of all the FX processors to around the speed of the FX-8320E with the default mitigations and behind if disabling Hyper Threading. The Intel Core i5/i7 CPUs remained ahead but the Core i5 2500K saw a 14% hit to the performance in the benchmarks ran with the default mitigations. The Core i7 2700K and i7 3770K each saw around a 14% hit to the default mitigated performance as well but that extended to 18~19% if disabling Hyper Threading in the name of security. The AMD FX CPUs saw minimal change to performance with their relevant Spectre mitigations."

So sure Intel would still have been in the lead... but FX wouldn't have been the train wreck it was portrayed as either. Instead of it looking really bad it would have been more like Zen one... where sure it was a bit slower then Intel but at the price... and bonus core count who knows perhaps it would have been seen differently.

Thanks. I think I'll hang on to my 8320E home server for a bit longer now!

As to a Ryzen 4000, my personal thought is that it will be a Zen 2+ rather than Zen 3, just like the Ryzen 2000 was Zen +. Coming out next year, it makes sense that this is a refresh/improverment, and if the improvement is anything like the 2000 series was over the 1000 series, we shouldn't be disappointed. Of course, if AMD is moving ahead fast enough to release Zen 3 next year - what a ride!
 
I dunno he seems to be going out of his way to ignore all the signs and hardware sites that all say the same thing, 10nm desktop chips are never coming because 10nm for Intel is still a mess. 2021 is the year Intel will have anything for the desktop we care about and that is assuming 7nm goes far smoother for them.

We might get both, as you see in mobile right now, with Intel releasing new SKUs in 10nm and 14nm (both class leading).

The question that seems to be brought up by The Faithful is whether Intel can produce desktop 10nm CPUs; my point is that that's not a question at all, however, whether they will is another matter entirely. The desktop market isn't where Intel is hurting, and we should reasonably expect them to deploy new architectures where they make the most sense. Hence 10nm on mobile, where the new CPU architecture and improved GPU are actually useful over Intel's previous gen.
 
We might get both, as you see in mobile right now, with Intel releasing new SKUs in 10nm and 14nm (both class leading).

The question that seems to be brought up by The Faithful is whether Intel can produce desktop 10nm CPUs; my point is that that's not a question at all, however, whether they will is another matter entirely. The desktop market isn't where Intel is hurting, and we should reasonably expect them to deploy new architectures where they make the most sense. Hence 10nm on mobile, where the new CPU architecture and improved GPU are actually useful over Intel's previous gen.

I don't think they CAN get 10nm desktop to the point where it is a real upgrade to 14nm++++++++++++ I think that's where the disagreement lays. Hence why you see a lot of ULV processors and somewhat lower core/clockspeed processors.
 
  • Like
Reactions: N4CR
like this
Thanks. I think I'll hang on to my 8320E home server for a bit longer now!

It didn't get any slower... or faster. If it serves its purpose, why consider a change?

and if the improvement is anything like the 2000 series was over the 1000 series, we shouldn't be disappointed.

Zen 2 is essentially Zen++, in that it's still a refinement of Zen. The reason a Zen 3 is more likely is that AMD must now iterate their architecture to get more performance.
 
I don't think they CAN get 10nm desktop to the point where it is a real upgrade to 14nm++++++++++++ I think that's where the disagreement lays. Hence why you see a lot of ULV processors and somewhat lower core/clockspeed processors.

Given that their 14nm CPUs meet or exceed AMD + TSMC at 7nm where it counts for volume desktop and mobile sales, I can easily see that Intel isn't in a hurry to push out 10nm parts except for ultrabooks where the power savings is a strong selling point.

Where 10nm can help the most is for the Xeons, with higher core counts at lower TDPs, with significantly higher IPC to boot. That's where it makes the most sense for them to focus.
 
I have a feeling Intel has switched priorities. They are developing 3D stacking chips, dedicated GPUs and AI processors. Giving up some traditional desktop, workstation and even some server market share to gain ground in deep learning and ultra compact computing devices. If Intel manages to outperform nVidia GPUs with Nervana, it will end up selling more Xeons anyway because the AI systems will contain them as well.

It may appear AMD has Intel on the ropes, but there is a bigger picture to look at. Think about it. Governments will pay countless millions to have the fastest image recognition systems, for example. That market is far more valuable then what vapid YouTube "influencers" recommend for Call of FortWatch.
 
Giving up some traditional desktop, workstation and even some server market share to gain ground in deep learning and ultra compact computing devices.

They are giving up some tiny fraction of marketshare to AMD, but this isn't by plan: Intel planned to have eight-core 10nm CPUs on the desktop four years ago. Now they're rejiggering architectures while they juggle fab upgrades and yes, having to prioritize what they produce where.
 
Thanks. I think I'll hang on to my 8320E home server for a bit longer now!

As to a Ryzen 4000, my personal thought is that it will be a Zen 2+ rather than Zen 3, just like the Ryzen 2000 was Zen +. Coming out next year, it makes sense that this is a refresh/improverment, and if the improvement is anything like the 2000 series was over the 1000 series, we shouldn't be disappointed. Of course, if AMD is moving ahead fast enough to release Zen 3 next year - what a ride!

Zen 3 is coming next year... and no its not Zen2+.

https://realmoney.thestreet.com/investing/technology/amd-inks-new-server-cpu-deals-15170073

As per Forrest Norrod AMDs head of all things server;

When asked about what kind of performance gain Milan's CPU core microarchitecture, which is known as Zen 3, will deliver relative to the Zen 2 microarchitecture that Rome relies on in terms of instructions processed per CPU clock cycle (IPC), Norrod observed that -- unlike Zen 2, which was more of an evolution of the Zen microarchitecture that powers first-gen Epyc CPUs -- Zen 3 will be based on a completely new architecture.

Norrod did qualify his remarks by pointing out that Zen 2 delivered a bigger IPC gain than what's normal for an evolutionary upgrade -- AMD has said it's about 15% on average -- since it implemented some ideas that AMD originally had for Zen but had to leave on the cutting board. However, he also asserted that Zen 3 will deliver performance gains "right in line with what you would expect from an entirely new architecture."

AMD hasn't been full of any BS for a few years now... so ya 2020 is looking like a bad year for Intel. Zen2 is not what AMD sees as their big gun... 3 is.
 
  • Like
Reactions: N4CR
like this
Back
Top