Intel will announce its Coffee Lake processors on August 21. They will be the last generation of 14nm(++) Core processors before 10nm Cannon Lake and Ice Lake, which is described as using a "10nm+" process:
In an unusual move for Intel, the chip giant has ever so slightly taken the wraps off of one of their future generation Core architectures. Basic information on the Ice Lake architecture has been published over on Intel's codename decoder, officially confirming for the first time the existence of the architecture and that it will be made on Intel's 10nm+ process.
The Ice Lake processor family is a successor to the 8th generation Intel® Core™ processor family. These processors utilize Intel's industry-leading 10 nm+ process technology.
This is an unexpected development as the company has yet to formally detail (let alone launch) the first 10nm Core architecture – Cannon Lake – and it's rare these days for Intel to talk more than a generation ahead in CPU architectures. Equally as interesting is the fact that Intel is calling Ice Lake the successor to their upcoming 8th generation Coffee Lake processors, which codename bingo aside, throws some confusion on where the 14nm Coffee Lake and 10nm Cannon Lake will eventually stand.
[...] Working purely on lithographic nomenclature, Intel has three processes on 14nm: 14, 14+, and 14++. As shown to everyone at Intel's Technology Manufacturing Day a couple of months ago, these will be followed by a trio of 10nm processes: 10nm, 10nm+ (10+), and 10++.
Tick Tock has given way to plus signs everywhere.
Coffee Lake will include the first mainstream 6-core chips from Intel, including the Intel Core i5-8600K and i7-8700K.
Also at Tom's Hardware.
Related Stories
The first "8th generation" Intel Core processors roll out today: a quartet of 15W U-series mobile processors. Prior generation U-series parts have had two cores, four threads; these new chips double that to four cores and eight threads. They also bump up the maximum clock speed to as much as 4.2GHz, though the base clock speed is sharply down at 1.9GHz for the top end part (compared to the 7th generation's 2.8GHz). But beyond those changes, there's little to say about the new chips, because in a lot of ways, the new chips aren't really new.
Although Intel is calling these parts "8th generation," their architecture, both for their CPU and their integrated GPU, is the same as "7th generation" Kaby Lake. In fact, Intel calls the architecture of these chips "Kaby Lake refresh." Kaby Lake was itself a minor update on Skylake, adding an improved GPU (with, for example, hardware-accelerated support for 4K H.265 video) and a clock speed bump. The new chips continue to be built on Intel's "14nm+" manufacturing process, albeit a somewhat refined one.
Source: Ars Technica
takyon: Also at AnandTech. Where's 14nm++ Coffee Lake?
In the past we are used to a new numbered generation to come with a new core microarchitecture design. But this time Intel is improving a core design, calling it a refresh, and only releasing a few processors for the mobile family. We expect that Intel's 8th Generation will eventually contain three core designs of product on three different process design nodes: the launch today is Kaby Lake Refresh on 14+, and in the future we will see Coffee Lake on 14++ become part of the 8th Gen, as well as Cannon Lake on 10nm.
[...] So when is Coffee Lake on 14++ (or Cannon Lake) coming? Intel only stated that other members of the 8th Generation family (which contains Kaby Lake Refresh, Coffee Lake and Cannon Lake) are coming later this year. Desktop will come in the autumn, and additional products for enterprise, workstation and enthusiast notebooks will also happen. As for today's 8th Generation U-series announcement, Intel tells us that we should start seeing laptops using the new CPUs hit the market in September.
(Score: 0) by Anonymous Coward on Wednesday August 16 2017, @04:58AM (2 children)
Is that Idiot for "10+ nm"?
(Score: 0) by Anonymous Coward on Wednesday August 16 2017, @06:26AM
nmmmmmmmm
(Score: 0) by Anonymous Coward on Thursday August 17 2017, @06:57AM
The article mentions that the pluses signify unspecified "enhancements" as compared to the 10 nm process. So I guess that the line width is 10 nm, or perhaps slightly less for certain features.
(Score: 0) by Anonymous Coward on Wednesday August 16 2017, @05:35AM (1 child)
Samsung & Global Foundary, both saying they're ready to skip from 14nm to 7nm & EUV in their next generation?
(Score: 2) by takyon on Wednesday August 16 2017, @10:06AM
They already shit talked everybody else's X nm processes:
Moore's Law: Not Dead? Intel Says its 10nm Chips Will Beat Samsung's [soylentnews.org]
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 2) by RamiK on Wednesday August 16 2017, @05:40AM (2 children)
Not to nitpick, but Ice Lake isn't a CPU architecture. It's a line or a series. Core, Atom and Pentium are families. x86 and ARM64 are architectures...
Or am I missing something?
compiling...
(Score: 0) by Anonymous Coward on Wednesday August 16 2017, @05:54AM (1 child)
Sure, x86 is an architecture, but there are low-level details that are implemented in very many ways, not to mention extensions for specialized functions.
Programmers (that is, compilers) tend to target x86, but it would be possible to target Ice Lake by writing code that is particular to the Ice Lake processor; that is what makes it a [micro]architecture.
(Score: 0) by Anonymous Coward on Wednesday August 16 2017, @10:34AM
You can't go around referring to microarchitectures as architectures. Reason being, "What's special about the Sky Lake architecture?" and "What's special about the Sky Lake microarchitecture?" are different questions: The former refers to new instructions while the latter will link you to https://en.wikipedia.org/wiki/Skylake_(microarchitecture)#Architecture [wikipedia.org] .
For people not familiar with the nomenclature, review the table headers over at https://en.wikipedia.org/wiki/List_of_Intel_CPU_microarchitectures [wikipedia.org] and https://en.wikipedia.org/wiki/List_of_AMD_CPU_microarchitectures [wikipedia.org] . There's no reason to excuse sloppy news reporting and press releases.
(Score: -1, Troll) by Anonymous Coward on Wednesday August 16 2017, @05:49AM (4 children)
I have always relied on the support of Intel for the advancement of the white race. Especially with the Unternehmen Barbarossa and the i5. Am I wrong?
(Score: 0, Offtopic) by Anonymous Coward on Wednesday August 16 2017, @05:59AM (1 child)
The best thing to happen to non-white people is white people; not only are white people ashamed of their own achievements (such as tolerance, or multiculturalism, or the ending of slavery as a matter of culture), but white people will be the first to throw themselves into the fires to advance non-white people.
(Score: -1, Offtopic) by Anonymous Coward on Wednesday August 16 2017, @06:52AM
White people cheat on their taxes. They are all egotistical douches, like President "Many Sides". They will all end up in prison, just like the little white boy cornered by the Antifa, who then decided he was not actually a white supremacist, [youtube.com] since he had been in prison (I am not going there, at all, but it does bear mention that most of the core of the German Nazi movement were twisted gay as fuck. Read "Gravity's Rainbow"!)
Intel, never trusted them since they teamed with Microsoft and white supremacy.
(Score: 0) by Anonymous Coward on Wednesday August 16 2017, @07:37AM
And so it is.
Though, mind the small print, as it includes: "Jews first, white goyim later"
[In before Eth. BTW, whatever would he be doing now, overwhelmed already by real world competition?]
(Score: 2) by takyon on Wednesday August 16 2017, @10:07AM
Nope. [cnn.com]
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 3, Interesting) by acid andy on Wednesday August 16 2017, @09:09AM (5 children)
I'm sure this is a naive question, but when you have a transistor that is made of just a few atoms or even a single atom, isn't that many times more vulnerable to damage than a larger transistor?
I get that they are shielded by being inside the chip core but how great is the risk that one or more of the atoms could move out of place and cause a permanent error in the chip's calculations? I'm thinking even just a tiny bit of excess heat or current or something like a stray cosmic ray could easily disturb it but maybe that's my limited understanding of the atomic physics of solids.
Is it a normal occurrence for a few transistors to go bad during the life time of a chip? I know RAM can use error correction techniques like ECC but what about the logic gates that actually execute instructions in the CPU? Do they need redundancy and error correction too?
If a cat has kittens, does a rat have rittens, a bat bittens and a mat mittens?
(Score: 2) by takyon on Wednesday August 16 2017, @10:10AM (3 children)
The processes are worked on for years before they hit the market. If 10nm, 7nm, and 5nm were nonviable, we would know it by now. The bigger problem is getting the EUV tools working good enough to make the economics viable.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 0) by Anonymous Coward on Wednesday August 16 2017, @10:35AM
Takes every facet of degradation into account.
Intel has had a number of engineering failures (as has AMD, although they have a lot more general QC failures (IE random chip is faulty), whereas Intel's are either errata (logic errors/races), or engineering failures, like those SATA controllers on that one southbridge a few years back that would erode the SATA traces/transistors until it got too noisy to communicate/went dark.)
Personally, while I am fine with later generation stuff for portables (being generally less reliable, and more sensitive to power improvements.), I had mostly stabilitized on 45->28 nm processors, with no plans to upgrade so long as I can find a web browser and system kernel that will run on them and have nominally up to date security patches. Given that anything post-32nm has management engine issues, firmware signing, backdoors, etc, it is not a huge loss. Especially now that most games have moved to unity or fully drm'd platforms (xbox, ps4) and run like buggy pieces of shit anyways.
(Score: 2) by acid andy on Wednesday August 16 2017, @01:04PM (1 child)
Oh I'm sure it's economically viable. There aren't many atoms in the transistors of the computers most of us are already using, so it clearly mostly works. I just find the physics of it somewhat mind boggling. If you pick up an object or rub it, or drop it, I would expect that a non zero number of molecules would part company from the surface of that object or at least change their alignment. I'm trying to find where my intuition fails me here. If it was a silicon crystal, would a non zero number of molecules still be dislodged each time it was picked up, rubbed or dropped (I actually suspect not)? If so, what about the molecules inside the object, beneath it's surface? Maybe it's extremely rare for them to become misaligned unless an object is subject to unusual stress, heat, etc.
Can any physicists, chemists or material scientists answer this? If not, a little more of the respect I used to have for soylent will ebb away. : (
If a cat has kittens, does a rat have rittens, a bat bittens and a mat mittens?
(Score: 2) by takyon on Wednesday August 16 2017, @01:38PM
https://en.wikipedia.org/wiki/Rock%27s_law [wikipedia.org]
https://en.wikipedia.org/wiki/Extreme_ultraviolet_lithography [wikipedia.org]
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 2) by snufu on Wednesday August 16 2017, @05:26PM
The primary obstacle to scaling transistors to channel lengths below 100 nm is that thermal effects degrade their electrical signal to noise. Permanent irreversible material damage (crystal damage, electromigration) can be an issue, but reversible, transient thermal effects (hot electrons over short lengths) are the primary culprit in the demise of Moore's law.