Monday, October 14th 2019

Intel Scraps 10nm for Desktop, Brazen it Out with 14nm Skylake Till 2022?

In a shocking piece of news, Intel has reportedly scrapped plans to launch its 10 nm "Ice Lake" microarchitecture on the client desktop platform. The company will confine its 10 nm microarchitectures, "Ice Lake" and "Tiger Lake" to only the mobile platform, while the desktop platform will see derivatives of "Skylake" hold Intel's fort under the year 2022! Intel gambles that with HyperThreading enabled across the board and increased clock-speeds, it can restore competitiveness with AMD's 7 nm "Zen 2" Ryzen processors with its "Comet Lake" silicon that offers core-counts of up to 10.

"Comet Lake" will be succeeded in 2021 by the 14 nm "Rocket Lake" silicon, which somehow combines a Gen12 iGPU with "Skylake" derived CPU cores, and possibly increased core-counts and clock speeds over "Comet Lake." It's only 2022 that Intel will ship out a truly new microarchitecture on the desktop platform, with "Meteor Lake." This chip will be built on Intel's swanky 7 nm EUV silicon fabrication node, and possibly integrate CPU cores more advanced than even "Willow Cove," possibly "Golden Cove."
The HardwareLuxx article making these explosive revelations attributes the sudden change in Intel's plans to the company not being able to scale clock-speeds of "Ice Lake" high enough to establish product leadership. It feels "Skylake," which has IPC parity with "Zen 2," has enough scalability and clock-speed headroom to stay competitive with AMD at high clock-speeds. The company will augment next-generation uncore (revamped memory controllers, support for PCIe gen 4.0, Gen12 iGPU, etc.), with "Skylake" CPU cores, over time. Other areas where Intel could grow its mainstream desktop silicon is cache rebalancing similar to its HEDT chips, and implementing the Mesh Interconnect to maintain low latencies as core-counts enter two-figures.

Interestingly, 10 nm "Ice Lake" remains on Intel's enterprise roadmap, where the company appears more desperate not to cede market-share to AMD, especially as businesses around the world set their 5G plans rolling, springing a cycle of hardware updates in the data-center. 2020 could see the introduction of Xeon Scalable processors based on 10 nm "Ice Lake" microarchitecture with "Sunny Cove" CPU cores. In 2021, the company will introduce the "Sapphire Rapids" Xeon processor with even more cores and larger I/O connectivity, spearheaded with PCI-Express gen 5.0.

Update Oct 15th: Intel has released a statement, denying these claims, read more here. Source: HardwareLuxx.de
Add your own comment

148 Comments on Intel Scraps 10nm for Desktop, Brazen it Out with 14nm Skylake Till 2022?

#126
candle_86
notb
Honestly, nothing is sure about the naming or (sadly) date.
But we know for sure that Intel is buying a lot of 7nm EUV equipment from ASML. So something will be launched. :)

That's not a huge achievement when you think that Ryzen was the first good AMD design since 2007's K10.
And 6 years since the awful Bulldozer.

If you compare Intel CPUs from 2017-2018 and 2011 (not to mention 2007), the architecture is similar, but the performance and efficiency are in 2 different worlds as well.

No one says it's a bad CPU. But you paid for 8C/16T and you said yourself you never used 16 threads (and you said it like if it was good...?)
Would you be so happy about 2700X benchmark results (Cinebench etc) if they were 10-20% lower? :)

So the question I asked was: wasn't this a suboptimal choice? Maybe 2600X would be better for you?
Did you ask folks in 2007 why they bought a dual core if they only need a single? The point is someone with an i7 from 2012 is still fine while an i5 from 2012 is painful today, same reason a core 2 quad lasted longer than a duo, it had room to grow.
Posted on Reply
#127
ShrimpBrime
notb
Honestly, nothing is sure about the naming or (sadly) date.
But we know for sure that Intel is buying a lot of 7nm EUV equipment from ASML. So something will be launched. :)

That's not a huge achievement when you think that Ryzen was the first good AMD design since 2007's K10.
And 6 years since the awful Bulldozer.

If you compare Intel CPUs from 2017-2018 and 2011 (not to mention 2007), the architecture is similar, but the performance and efficiency are in 2 different worlds as well.

No one says it's a bad CPU. But you paid for 8C/16T and you said yourself you never used 16 threads (and you said it like if it was good...?)
Would you be so happy about 2700X benchmark results (Cinebench etc) if they were 10-20% lower? :)

So the question I asked was: wasn't this a suboptimal choice? Maybe 2600X would be better for you?
Firstly, your thoughts about Ryzen not being a huge achievement have absolutely nothing to do with why I'm 100% satisfied.
You asked why, I told you and there's really nothing more to it.

Secondly.... Cinebench is NOT the TELL ALL of computing power lol.
So we are comparing it to Intel? My 2700X unzips much faster than my 8700K. But don't tell anyone I know that bit of information.
Posted on Reply
#128
Super XP
notb
Honestly, nothing is sure about the naming or (sadly) date.
But we know for sure that Intel is buying a lot of 7nm EUV equipment from ASML. So something will be launched. :)

That's not a huge achievement when you think that Ryzen was the first good AMD design since 2007's K10.
And 6 years since the awful Bulldozer.
Intel is in a pickle, they are in big trouble. They know they are in big trouble. Especially when AMD is outperforming Intel by 400% per dollar with ZEN 2 EPYC CPUs.

As for Bulldozer, there's history as to why AMD went that route. Bulldozer was OK, Piledriver squeezed out a lot more performance and efficiencies and gave AMD just enough of a push to remain somewhat competitive till its superior ZEN micro- architecture was ready for release. Sure Bulldozer was OK, but I wouldn't call it awful per say.
Posted on Reply
#129
seronx
AMD is going to get rolled over by Tanner Ridge. Zen is garbage.
Posted on Reply
#130
DeathtoGnomes
seronx
AMD is going to get rolled over by Tanner Ridge. Zen is garbage.
Great contribution!

Please keep up the good work.
Posted on Reply
#131
Super XP
seronx
AMD is going to get rolled over by Tanner Ridge. Zen is garbage.
That's why ZEN are selling like Hot Cakes.
That's why Intel hired Jim Keller (Designer of ZEN) because according to you Zen is garbage lmao.
Competition is great, that's what drives Innovation. By saying Zen is garbage, you are against fair competition and innovation.

Interesting,

DeathtoGnomes
Great contribution!

Please keep up the good work.
He's angry because ZEN is very successful. I would think that's a good thing ya know? Lol
Posted on Reply
#132
seronx
Super XP
That's why ZEN are selling like Hot Cakes.
They aren't.
Super XP
He's angry because ZEN is very successful.
It isn't.

Tremontx(TannerRidge) > Zen
Willowcove(Tigerlake-S) > Zen2/3
Posted on Reply
#133
Super XP
seronx
They aren't.
It isn't.

Tremontx(TannerRidge) > Zen
Willowcove(Tigerlake-S) > Zen2/3
Facts are facts,
To each there own I suppose.
Posted on Reply
#134
DeathtoGnomes
Super XP
That's why ZEN are selling like Hot Cakes.
That's why Intel hired Jim Keller (Designer of ZEN) because according to you Zen is garbage lmao.
Competition is great, that's what drives Innovation. By saying Zen is garbage, you are against fair competition and innovation.

Interesting,


He's angry because ZEN is very successful. I would think that's a good thing ya know? Lol
angry or not, its really about the substance of his post. Why not explain why he thinks Zen is garbage? A Troll cant.
Posted on Reply
#136
AusWolf
I am actually happy for this news. Lack of development from Intel means I can stay with my Kaby Lake i7 longer, unless AMD comes up with something that shakes up the gaming CPU market, which I think is unlikely. Fortunately, in the mainstream segment, per-core performance is far more important than having 8/10/12 cores, which makes the shiny new i9s and such completely irrelevant products for me, even without the hefty price tag (which they have). So instead of wasting money on the core count race, I'll just wait for true innovation and stay happy with my KL i7 until then. Thank you Intel!:toast:
Posted on Reply
#137
seronx
DeathtoGnomes
Why not explain why he thinks Zen is garbage?
Zen is garbage for some reasons:
1. Tactically it failed to compete with Intel cores. --Skylake is better than Zen, then there is the ultra-wide cores(Willowcove/Goldencove) simply breaking Zen2/Zen3's back.
2. Strategically it fails to compete with ARM cores. --Neoverse has more cores, more IPC, and lower power.
3. It isn't innovative or competitive. --Only innovation has been outside of AMD. They tried to do SMT2 before and it failed, they tried to SMT2 now and it also failed.
4. It is followed up by a list of failures from the foundries it was built at. -- TSMC is lack of performance, GlobalFoundries is lack of capacity.
5. The architecture is already dead with a new Family of cores replacing it. -- Another ARM core turned into an x86-SMTx core... *sigh* GARBAGE!
Posted on Reply
#138
Keviny Oliveira
seronx
Zen is garbage for some reasons:
1. Tactically it failed to compete with Intel cores. --Skylake is better than Zen, then there is the ultra-wide cores(Willowcove/Goldencove) simply breaking Zen2/Zen3's back.
lol This not is trust a Ryzen 5 3600 has same performance in multitasking and games of a i7-8700K/i7-9700K and Ryzen 5 3600 consumes less and costs much less too!!!

seronx
2. Strategically it fails to compete with ARM cores. --Neoverse has more cores, more IPC, and lower power.
lol Zen2 consumes much less than a iCore processor and ARM not competes with processors x86.
seronx
3. It isn't innovative or competitive. --Only innovation has been outside of AMD. They tried to do SMT2 before and it failed, they tried to SMT2 now and it also failed.
lol this is only your hater for AMD, like will fail if AMD has not tried yet?
seronx
4. It is followed up by a list of failures from the foundries it was built at. -- TSMC is lack of performance, GlobalFoundries is lack of capacity.
But worse is Intel than even the ability to manufacture 10nm processors in 2016 it has, it's 2019 and she's given up on 10nm because it knows it's a failure, AMD has at least proved that it will always try to innovate as much as possible. may differ from Intel.

seronx
5. The architecture is already dead with a new Family of cores replacing it. -- Another ARM core turned into an x86-SMTx core... *sigh* GARBAGE!
lol you use x86_64 instructional processor that was created by AMD not Intel or "ARM", LMAO.
Posted on Reply
#139
notb
Keviny Oliveira
lol you use x86_64 instructional processor that was created by AMD not Intel or "ARM", LMAO.
By all means, AMD was not the first company to introduce 64-bit processors.
AMD was simply the first to launch x86 64-bit specification for consumer CPUs. This specification became the standard.
Posted on Reply
#140
candle_86
seronx
Zen is garbage for some reasons:
1. Tactically it failed to compete with Intel cores. --Skylake is better than Zen, then there is the ultra-wide cores(Willowcove/Goldencove) simply breaking Zen2/Zen3's back.
2. Strategically it fails to compete with ARM cores. --Neoverse has more cores, more IPC, and lower power.
3. It isn't innovative or competitive. --Only innovation has been outside of AMD. They tried to do SMT2 before and it failed, they tried to SMT2 now and it also failed.
4. It is followed up by a list of failures from the foundries it was built at. -- TSMC is lack of performance, GlobalFoundries is lack of capacity.
5. The architecture is already dead with a new Family of cores replacing it. -- Another ARM core turned into an x86-SMTx core... *sigh* GARBAGE!
1. Clock for clock Zen was faster in slot of things compared to Skylake, the reviews showed that, it lost in games which for 95% of the buying public is irrelevant.

2. Arm and x86 cover two different areas, if arm was a better chip I'm sure don't and Microsoft would be doing the next generation consoles on it, but they aren't because arm can't reach the same performance as x86 without going large and power hungry.

3. Innovation is what infinity fabric actually is, it's even being adopted by Intel.

4. To label tsmc and global foundries failures is the most moronic statement I have ever read. Your statement implys that everyone but samsung and apple that make microchips are clueless.

5. It is far from dead, it's now faster per clock than anything Intel has outside of games. Workstation and server loads are faster on zen2 than on Intel's parts.

Your entire post reminds me of the garbage I used to read when people defended the Pentium D or Pentium 4.

notb
By all means, AMD was not the first company to introduce 64-bit processors.
AMD was simply the first to launch x86 64-bit specification for consumer CPUs. This specification became the standard.
When it comes to x86 there is a reason they call it AMD64 even on Intel chips, it's because Intel hadn't started on a 64bit x86 chip, they put all of their resources into the failed itanium design.
Posted on Reply
#141
notb
candle_86
2. Arm and x86 cover two different areas, if arm was a better chip I'm sure don't and Microsoft would be doing the next generation consoles on it, but they aren't because arm can't reach the same performance as x86 without going large and power hungry.
ARM and x86 use different instruction sets, so compatibility is a challenge. But behind the interface they are just processors. They perform arithmetic operations.

What you had in mind are results of conscious choices. ARM is mostly made with mobile devices in mind, so the cores are optimized for this scenario.
But it is possible to make "big" ARM CPUs and we'll see them fairly soon
Similarly, it is possible to make "mobile" (small, frugal) x86 CPUs (like Intel Atom lineup).
3. Innovation is what infinity fabric actually is, it's even being adopted by Intel.
Adopted? What? :eek:
Your entire post reminds me of the garbage I used to read when people defended the Pentium D or Pentium 4.
Your entire post is a perfect example of what fanboyism can do with - I assume - fairly normal person. You've been here for a long time, so I bet you used to write a lot more balanced posts about CPUs before Zen came out.
Why so much hate? What's your problem?
When it comes to x86 there is a reason they call it AMD64 even on Intel chips, it's because Intel hadn't started on a 64bit x86 chip, they put all of their resources into the failed itanium design.
Well I just said that AMD was first with a consumer standard. But that doesn't mean they invented 64-bit :)

And Itanium was far from "failed". It worked perfectly fine, but Intel switched to the common standard that already got a lot of traction.
Otherwise we would have multiple instruction standard for x86 and huge compatibility issues.
Posted on Reply
#142
candle_86
notb
ARM and x86 use different instruction sets, so compatibility is a challenge. But behind the interface they are just processors. They perform arithmetic operations.

What you had in mind are results of conscious choices. ARM is mostly made with mobile devices in mind, so the cores are optimized for this scenario.
But it is possible to make "big" ARM CPUs and we'll see them fairly soon
Similarly, it is possible to make "mobile" (small, frugal) x86 CPUs (like Intel Atom lineup).

Adopted? What? :eek:

Your entire post is a perfect example of what fanboyism can do with - I assume - fairly normal person. You've been here for a long time, so I bet you used to write a lot more balanced posts about CPUs before Zen came out.
Why so much hate? What's your problem?

Well I just said that AMD was first with a consumer standard. But that doesn't mean they invented 64-bit :)

And Itanium was far from "failed". It worked perfectly fine, but Intel switched to the common standard that already got a lot of traction.
Otherwise we would have multiple instruction standard for x86 and huge compatibility issues.
Actually the one off balance is you, Intel has stagnated since Sandy bridge much as they did during Pentium 4
Posted on Reply
#143
notb
candle_86
Actually the one off balance is you, Intel has stagnated since Sandy bridge much as they did during Pentium 4
In the previous comment I said something about x86 vs ARM (where you were very imprecise) and IF adoption + AMD64 (where you were wrong).

And all you can come up with is "Intel stagnated since Sandy Bridge"...
It seems you aren't really interested in discussing anything.
Posted on Reply
#144
candle_86
notb
In the previous comment I said something about x86 vs ARM (where you were very imprecise) and IF adoption + AMD64 (where you were wrong).

And all you can come up with is "Intel stagnated since Sandy Bridge"...
It seems you aren't really interested in discussing anything.
No I simply don't waste slot of effort on Intel zombies
Posted on Reply
#145
SaLaDiN666
candle_86
Actually the one off balance is you, Intel has stagnated since Sandy bridge much as they did during Pentium 4
Ivy Bridge delivered up to 15% over Sandy Bridge, average perfomance gain was something about 9-10%.

Haswell delivered up to 8%, average perfomance gain was something around 5%.

Skylake was something similar like Haswell.

So pretty much, you have no clue what you are talking about.
Posted on Reply
#146
candle_86
SaLaDiN666
Ivy Bridge delivered up to 15% over Sandy Bridge, average perfomance gain was something about 9-10%.

Haswell delivered up to 8%, average perfomance gain was something around 5%.

Skylake was something similar like Haswell.

So pretty much, you have no clue what you are talking about.
You mean similar gains we saw when Intel went willimate to Northwood, then to Prescott, and then to Prescott 2m. Maybe you shoulda been around in 2000-2006.

Seems my point is valid
Posted on Reply
#147
SaLaDiN666
candle_86
You mean similar gains we saw when Intel went willimate to Northwood, then to Prescott, and then to Prescott 2m. Maybe you shoulda been around in 2000-2006.

Seems my point is valid
Your point is completely invalid because those gains were mainly caused by increased frequency, not by architectural changes. Pentium 4 was a failure and had it achieved such gains, it would have been actually faster clock to clock than the AMD counterparts. Not to mention, there was performance regression between the gens you mentioned when compared clock to clock depending on workloads. Prescott was actually often slower than Northwood.

My gains, which I mentioned, are compared clock to clock.
Posted on Reply
#148
candle_86
SaLaDiN666
Your point is completely invalid because those gains were mainly caused by increased frequency, not by architectural changes. Pentium 4 was a failure and had it achieved such gains, it would have been actually faster clock to clock than the AMD counterparts. Not to mention, there was performance regression between the gens you mentioned when compared clock to clock depending on workloads. Prescott was actually often slower than Northwood.

My gains, which I mentioned, are compared clock to clock.
so are mine, Northwood P4 vs Willimate was a 15% improvement, and northwood C was another 8-10% faster making it faster than athlon XP, and PRescott was faster than northwood in SSE aware software, you just wernt around i suppose
Posted on Reply
Add your own comment