Intel kills off the 10nm process

Intel kills off the 10nm process
This is actually a good thing for the company
Oct 22, 2018 by Charlie Demerjian
https://semiaccurate.com/2018/10/22/intel-kills-off-the-10nm-process/

"SemiAccurate has learned that Intel just pulled the plug on their struggling 10nm process. Before you jump to conclusions, we think this is both the right thing to do and a good thing for the company..."

Intel is reportedly killing off its 10nm process entirely
But chipmaker refutes report and claims it's making 'good process'
https://www.theinquirer.net/inquire...ortedly-killing-off-its-10nm-process-entirely

" Update
Intel has refuted SemiAccurate's report in a statement given to the INQUIRER.

"Media reports published today that Intel is ending work on the 10nm process are untrue," a spokesperson told us. "We are making good progress on 10nm. Yields are improving consistent with the timeline we shared during our last earnings report." µ"

 

Very good news to hear from Intel camp. I was already tired of 10nm BS..

 

It read like total BS and stock manipulation move this morning yet it looks like people took the bait and reported it even without any real source.

 

Intel have denied on twitter

https://twitter.com/intelnews/status/1054397715071651841?s=20

And lol @ the 1st reply:
https://twitter.com/TechNewsGen/status/1054405555970818049?s=20

 

" Update October 22, 2018@3:30pm: Intel has denied ending 10nm on Twitter. The full tweet is, “ Media reports published today that Intel is ending work on the 10nm process are untrue. We are making good progress on 10nm. Yields are improving consistent with the timeline we shared during our last earnings report.” SemiAccurate stands by its reporting."

Responses @ https://twitter.com/intelnews/status/1054397715071651841

Thread is locked, but interesting...

AMD can take market share due to Intel killing its 10nm node
https://www.reddit.com/r/Amd/comments/9qdj5l/amd_can_take_market_share_due_to_intel_killing/

 

The problem with 10nm, and 12-7nm, is at their first appearance in the market CPU frequency is not very high for the process. In most older cases by the time the current larger process has maxed out its clocks the new smaller nodes have a similar if not faster node. In the case of Intel the 14nm++(whatever) has been around for so long the clocks were so maxed out I am sure they are now well above what initial 10nm offers at decent yields. In time this could change but it is way too late now for Intel to take a step backwards to go forewords.

the second issue is Intel needs a new architecture to combat future Zen cores.

 

I should also note, one way for Intel to keep 10nm in the minds of stock holders is to release a rumor and then deny it as everything is going ahead ok rather than making a new statement!

 

Interesting it's the guy who's been consistently saying Intel's 10nm is "broken" for years now. Maybe they fed him an anonymous inch knowing he'd make a mile out of it.

Benefit 1, making a statement without making a statement as you say, benefit 2, let a seasoned detractor go too far and undermine his own credibility.

 

Intel's Swan Says Stable PC Sales a 'Pleasant Surprise'
Bloomberg Markets and Finance
Published on Oct 26, 2018
Oct.26 -- Intel Interim Chief Executive Officer and Chief Financial Officer Bob Swan discusses the semiconductor maker's third-quarter performance, stabilization in the personal computer market, and expansion into 10nm chips. He speaks with Bloomberg's Emily Chang on "Bloomberg Markets: European Close."

 

Digitimes on Intel dropping 10nm investment in favor of 7nm investment:

Intel’s 10nm Ice Lake is scheduled for volume production in the second half of 2019, but according to Digitimes sources, there are still several problems with Intel’s 10nm process. As such, Digitimes predicts that Intel could “skip” 10nm, and funnel investments into 7nm.

Source: https://www.digitimes.com/news/a20190308VL202.html

Unfortunately a subscription is needed to read the whole article...

Originally from:
https://www.gamersnexus.net/industr...hortages-dram-price-drops-intel-hires-hardocp

 

Intel may not have "officially" killed 10nm, but given the years of constant "no show" for 10nm, Intel's 10nm desktop CPU's are more a fable and a fantasy than a reality

Intel's 10nm desktop CPU simply doesn't exist, except as a mirage on the horizon that has never and may never materialize.

Intel 10nm CPU's - Is It Ever Going To Happen?
Tech Showdown
Published on Apr 6, 2019
Today we are talking about Intel's 10nm Ice Lake CPU's and 14nm Comet Lake CPU's which will both be coming out in 2019.


TSMC Completes Design of 5nm EUV Process Node

 

I think it goes without saying that if Zen2 is anywhere near the rumors Intel is in trouble. Some thought 12nm would be that breaking point, I knew it wouldn't be. Unless some miracle happens where 10nm just starts working I doubt that will be a save grace for them either.

They really need 5nm and a core change too. Just having a faster and more efficient 8 or 10 core will not be enough at some point.

 

If Zen didn't approach the rumors for me, I am not holding my breath for Zen2 to do so either.

It is frustrating that Intel doesn't offer us new and shiny toys to brag about, but as far as performance/productivity, Intel is still in the lead.

Particularly in mobile and enterprise/data centers, but also continues to do so for my workflows.

I'm thankful for AMD's constant pressure on Intel, but comparing platforms with an eye on productivity (mine), I don't think Intel is sweating too much yet.

I don't think a certain 'X' nm platform is a requirement, let alone a core change is a requirement too...

If they continue to be faster and more efficient than the competition, they're still the better option when all else, including compatibility, is concerned.

 

That is the concern. If Zen2 rumors are true then Intel will not only be matched by but well exceeded by Zen2. As it is Zen at 12nm has a good showing against Intel's enterprise/data center. For certain video productivity and other things TR is a killer system. Other than gaming Intel does not have the superior performance it used too.

Compatibility, how about a new Intel CPU needs a different main board for each generation, sometimes even within that generation, for chipset and/or other issues?

 

"It's not dead yet!! 10nm Desktop CPU's 2022 FTW!!" - Intel (in their dreams...)

Intel's 10nm desktop CPU's may not be canceled officially, but there is no solid release date as to when 10nm desktop CPU's will ship, or be declared officially alive - or dead.

Whatever state Intel's 10nm desktop (enterprise?) CPU's really are, these new leaks of rumored releases through 2021 don't include Intel's 10nm desktop CPU's - delayed again, or DOA?

Intel's 10nm desktop CPU's may not be here until 2022(?!).

Roadmap shows that in 2021 Intel still makes desktop CPUs at 14nm
A roadmap that comes from Intel shows that the manufacturer will still make desktop CPUs at 14nm in 2021. Variants made at 10nm are not listed on the relevant roadmap. The first 10nm laptop processors will be in a limited edition this year.
https://tweakers.net/nieuws/151984/...l-in-2021-nog-desktop-cpus-op-14nm-maakt.html

"Tweakers has received two roadmaps from an anonymous tipster with introductions from Intel between 2018 and 2021. The roadmaps come from an internal Dell presentation about products that will be released this year and next year. Parts of the roadmap correspond to information that Intel itself has already announced. There is also new information on it. It is not clear how old the roadmaps themselves are. Maybe they were already prepared last year.

The roadmaps mainly describe laptop processors, but desktop processors are also referred to as S-series. According to the roadmap, Intel this year will only present the Coffee Lake Refresh models that were announced on Tuesday . Comet Lake will follow in early 2020 as successor to Coffee Lake. This again involves 14nm processors, but there are also variants with ten cores. Earlier this month a roadmap was published by a motherboard manufacturer that also lists Comet Lake. There have been rumors about the arrival of ten cores to the consumer platform.

Comet Lake was to be succeeded by Rocket Lake in the second quarter of 2021. According to the roadmap the number of cores and the production process do not change compared to Comet Lake, so it would again be 14nm processors with a maximum of ten cores. Never before has information been revealed about the Rocket Lake generation. The name can be seen for the first time on this roadmap.

So there are no 10nm desktop processors on the Intel roadmap for the next two years, but it is not certain whether Intel is really not working on it.

The relevant roadmap is related to the Stable Image Platform Program from Intel. Hardware appears for a long time within that program. Perhaps there is another roadmap for processors that do not fall within this program.

New laptop processors are also on the SIPP roadmap. Comet Lake H, a generation of 14nm laptop processors with up to ten cores, was due to arrive in early 2020. Comet Lake U is also on the roadmap, as a successor to Kaby Lake U. The new models would have a maximum of six cores and would again be made at 14nm. The SIPP roadmap mentions 10nm laptop processors only from 2021.

Ice Lake U and Tiger Lake U: 10nm laptop processors
Another roadmap shows Intel's laptop processors from 2018 to 2020. It states that in the second quarter of 2019 the first 10nm laptop processors will be released in a limited edition. This includes Ice Lake U, which are efficient dualcores with HyperThreading. The Ice Lake U generation has been on Intel's website for a long time and has been in a test phase since the third quarter of 2017. At the end of last year , Intel also indicated that Ice Lake is the first generation of 10nm processors to be released.

Tiger Lake U and Y will follow in the second quarter of 2020. These are efficient quadcores made at 10nm and the successors of the Kaby Lake generation. The Y versions are the models with a tdp of 5W. 10nm versions of Powerful Core H processors are not on the roadmap. Rocket Lake is mentioned. That would be processors with a maximum of six cores made at 14nm, but with a 10nm gpu. Rocket Lake U is also on the SIPP roadmap, but then also with a 14nm gpu. Maybe there will be different versions or Intel is not sure yet about which process is used for the gpu."


Maybe Intel simply forgot to add the 10nm desktop parts to the presentation?

In this case no news is not good news, but is it bad news? IMHO, Intel is going eternally push out 10nm until Intel's 7nm EUV is up and running, and then officially axe 10nm - leaving a smattering of ULV 10nm laptops out there for many to enjoy.

 

No 10nm Intel Desktop CPUs Until 2022 - Seriously!?!?
Tech Showdown
Published on Apr 25, 2019
An Intel CPU road map has leaked and it is even more disappointing that you could possibly imagine.

 

Just one word, if true, Ouch.

 

This may or not be a non-issue when/if these actually come out. For all we know, these Intel slides have already been superseded internally. But it really doesn't matter.

The process node or any other parameter isn't what is important to me or to anyone that cares about their end results: productivity. Manufacturers will/can only use what they have available and working, they can't sell fairy dust (well, fruity companies can...).

What matters (only) is what performance they give over my current platforms, as always. What it's built with inside is not my biggest concern. After all, a state-of-the-art smartphone is still a toy vs. anything Intel offers me.

Just like the purported/theoretical superiority of TR, Zen and Zen x has failed to move me to a single purchase (that I did not return after testing), let alone a wholesale switch to the AMD world, for my workflows. Because the underlying building blocks mean nothing if the final output isn't actually more than what I'm currently using and has the same reliability, dependability, stability, and endurance too.

Yeah, I know that all the building block details make a difference, somewhere. But that difference is only important when all else is equal or better too.

So far, Intel is still firing on all cylinders as far as my compute needs show, just like MS is still the force to be reckoned with, no matter the popular opinions of the mass uninformed.

Yeah, there are many examples where they've both miss-stepped, to be sure. But the point most seem to miss is how they have handled those missteps, given time.

 

14nm++++++ vs 7nm EUV. Single core perf. 14nnm wins; In multi-core perf, cost per $, easy cpu upgrade w/o buying new mobo,power efficiency etc... Ryzen wins

 

So, you are sticking with that? Really?

I will repost what I did in the Ryzen vs Intel thread as I think it fairly and accurately addresses the leaked roadmap and Intel's future performance. With that, I want to point out that even if superseded, it is still NEWER than anything released to date, as it has new code names that were rumored but not confirmed.

"I'm not an insider, so cannot guarantee anything. But, the limited 10nm low power dual core mobiles, then the replacement a year later, but no H or G actually does make sense. Also, Cascade and Cooper lake for HEDT and Server have long been known to be 14nm. Seems they are finally porting the Ice Lake/Sunny Cove back to 14nm.

With that said, and what @tilleroftheearth isn't saying, is 14nm is near its limits. The microarch will get better, meaning Comet Lake and beyond will have increases in single threaded performance due to improved IPC from architecture improvements. All that means is that Intel won't get the process improvements. So it does NOT mean Intel will not have more to get out of the tank, just means that the hopes of lower power is lost.

In fact, prior roadmaps showing Intel's release for the 9900K were only off by within two quarters on the products, depending which roadmap you looked at (the first released one turned out more accurate on that, whereas the second, later released one must have been an earlier version because it had the 9900K coming in Q1 of 2019, rather than late Q3 to Q4, like the first one).

If this one is anywhere near as accurate as those previously published, Intel has a process node problem."

 

You can repeat yourself ad nauseam, doesn't negate what I originally stated. Roadmaps mean squat as everyone knows from both Intel, AMD, MS and any other corporation you care to mention. Waste of my time to even begin to guess where the money will move to next.

I'm sticking to reality, you can stick to your theories and dance moves each time the ball is in the other court.

 

Intel Qualifying 10nm Ice Lake CPUs, Expects First Full-Year Revenue Decrease In Three Years | April 25, 2019
http://forum.notebookreview.com/thr...-lake-cpus-z390.811225/page-149#post-10901920

 

Dude, you are a known intel shill which has projects solely dependent on high single threaded core count. You pushed hardware acceleration without realizing that Adobe, in that same update, at least from some people I know, said they gimped the performance on their 7980XE with that update. Adobe is NOT aging well compared to their competitors, which have invested in parallelizing the work done by their engine to get more performance.

Now, there are still questions surrounding Ryzen's next release. If they did match the 9900K stock for stock with their 65W 8-core chip, then there is reason to believe Intel will be matched this round, maybe keeping the single thread crown due to overclocking ability, while losing all multi-threaded workloads, the power requirements, the core density, etc.

AMD is expected to overshoot Intel's current IPC on the 9900K. If they cannot reach a high enough frequency, then it won't beat the single thread performance though. Intel is set to take the IPC crown back with Sunny Cove. Problem is, if the roadmap is to be believed, the Ice Lake with Sunny Cove will run lower frequencies and only be mobile, while Comet Lake doesn't arrive until next year, potentially (at least for commercial use, as the SIP was presented to Dell in a closed presentation leaked, and commercial OEM builds can lag consumer releases for Intel by a couple quarters, meaning that the 10-core Comet lake may reach consumers in Q4, or make a CES appearance).

As such, you keep talking, but your predictions and comments are aging less well as you speak. Maybe look up at which ways the winds blow.

 

Just one question, if you do not care then why the comments? Just let the performance in the end speak for itself. I think most, can make from the information available out there, a good guess what that will be in the near future.

 

At least 2 more years until Intel releases 7nm, in 2021... before Intel's 10nm Desktop CPU's in 2022?, and Intel may still be on 14nm through the 7nm / 10nm-Desktop launches, and beyond...

AMD is leaving Intel in the dust on die size, with no 7nm Intel chips until 2021
By Jonathan Terrasi — Posted on July 19, 2019 10:52AM PST
https://www.digitaltrends.com/computing/inel-delays-7nm-die-size-shrink-to-2021/

"At Aspen, Colorado’s Brainstorm Tech 2019 conference this week, Intel CEO Bob Swan shed some light on the future of Intel’s fabrication process by remarking that Intel is planning on beginning production on a 7nm die for its processors in 2021.

The revelation follows only two months after the company’s Computex 2019 announcement of its first 10nm processors, the Ice Lake line, left many observers wondering when the CPU giant would catch up to its nimbler rival AMD, which showed off 7nm processors the day before.

Not only is Intel just reaching 10nm manufacturing as AMD rolls out 7nm chips, but the Ice Lake chips Intel showed off are not intended for desktops, but exclusively for mobile or versatile 2-in-1 devices.

This admission of Intel’s shortcomings makes for a welcome sign for AMD, validating their aggressive strategy of die shrinks and price undercutting.

Beyond AMD’s flashy Computex 2019 launch for its 7nm Ryzen 3000 CPUs, the manufacturer also has a threadripper waiting in the wings that Intel has no plans of contending with.

Confidence is running so high over at AMD that they are also taking the GPU fight to Nvidia in a big way. Earlier this week, a high-level AMD executive boasted about duping Nvidia into setting less-than-competitive prices against AMD’s Ryzen RX 5700.

Intel’s woes don’t end with AMD’s lead on die size, either. If Microsoft goes forward with its rumored switch to Qualcomm, Intel will suddenly lose out on a massive opportunity to supply the highest-selling device in the tablet product category..."

A Look Inside Intel with CEO Robert Swan
Fortune Magazine
Published on Jul 16, 2019
He’s been at the helm since January 2019.


Intel’s New CEO Blames Years-Long Chip Delay on Being Too ‘Aggressive’
By Aaron Pressman, July 16, 2019
https://fortune.com/2019/07/16/intel-ceo-bob-swan/

 

Actually 10nm Ice Lake looked pretty dismal in these PC World review results - except for some tests that used some new pumped up instructions to support encryption and encode / decode, otherwise the Ice Lake 10nm tiny U CPU's performed like Whiskey Lake:

Intel 10th gen Ice Lake performance preview
PCWorld
Published on Aug 1, 2019
Should you wait to buy a laptop right now? Gordon gives us a sneak peak and Intel's 10th gen Ice Lake performance numbers in benchmarks and gaming.


I moved the 10nm discussion here from the Ryzen thread... *Ryzen Thread*... or we could move this to the Ryzen vs Intel thread too.

And, that these wimpy will o' the wisp phantom 10nm CPU's are 15w / 25w 2in1's that can't possibly be considered the "next generation" Intel 10nm CPU's.

The only thing that brings up scores with these 10nm wannabe CPU's is the iGPU and new AVX instructions.

 

We may never see 10nm desktop CPU's as Intel wants to jump directly to 7nm - so 10nm was indeed "killed off", except for the "keep busy" 10nm low power CPU's that Intel has scratched together 2 times now:

Intel 10th-gen Ice Lake CPUs: Everything we know so far
10th-gen Intel laptops have arrived, but desktop processors are still far off [if ever]
By Jon Martindale — Posted on August 1, 2019 10:00AM PST
https://www.digitaltrends.com/computing/intel-ice-lake-cpu-everything-you-need-to-know/

"...The desktop chips, however, won’t arrive until later. PC Builders Club initially claimed that we won’t see Ice Lake hit desktops until 2021. A leaked Intel roadmap also suggested we might not see 10nm on desktop until 2022, but it might be that Intel doesn’t bother with 10nm on desktop at all.

The latest news from Intel CEO Bob Swann, was that Intel would begin production on 7nm chips in 2021, potentially for a 2022 release.

Speculation on pricing for such fabled chips would be nothing but that, and with little of anything concrete to go on about core counts, clock speeds, or release dates for desktop parts, we can’t really suggest what any future Ice Lake desktop chips might cost. If they ever appear at all."

I will be looking for Ice Lake vulnerability reports as well:

"...
SPECTRE AND MELTDOWN
Ice Lake’s architecture, Sunny Cove, will have hardware fixes for the Spectre and Meltdown architectural flaws that caused so much consternation among hardware manufacturers and software developers over the past year. So far we’ve seen microcode fixes for many of the most affected, recent-generation chips, and some of Intel’s ninth-generation chips implemented hardware fixes for specific instances of these exploits.

Intel has confirmed, however, that Ice Lake will go well beyond that, potentially representing the first CPU generation from Intel to mitigate major Spectre variants at the silicon level. That may also mean some of the performance-impacting stop-gap solutions protecting users against these exploits will not be present, although most think it unlikely that Ice Lake will be able to halt all the potential exploit paths."

 

Intel's disastrous failure plagued 10nm just won't die, still sucking money and resources from Intel but not delivering any returns... only costs.

The State of Intel's 10nm
AdoredTV
Published on Aug 16, 2019
An analysis of Intel's ongoing 10nm problems.

 

Intel keeps generating chaos in it's lineup...14nm is the new 10nm, when 10nm isn't good enough.

It Just Got WORSE For Intel 10th Gen - Comet Lake Arrives
HardwareCanucks
Published on Aug 21, 2019
Welcome to the Intel 10th Gen processor lineup confusion Round 2! They're now launching Comet Lake, a lineup of processors based on an old 14nm architecture that's being marketed as 10th Gen and it uses the 10-series names. Not only that but its actually BETTER than Ice Lake [10nm] in many ways...wait, what? Let's jump in!

 

Good Rumors Never Die - unlike the long rumored 10nm Desktop CPU's from Intel...

Intel is finally giving up on 10nm for Desktop / Server(?) for good? With Intel shifting it's weight behind 7nm, double or nothing!!

Rumor suggests Intel are scrapping 10nm desktop CPUs
James Dawson 1 day ago CPU, Featured Tech News, Tech News
https://www.kitguru.net/components/...ggests-intel-are-scrapping-10nm-desktop-cpus/

Rumours coming from inside Intel are suggesting the company is about to scrap plans for 10nm production of desktop CPUs to concentrate on a 7nm node for this market instead.

"According to some exclusive information from the team at hardwareLUXX, Intel is planning to scrap all plans for 10nm CPUs in the desktop market and focus on introducing a 7nm node in this area. However, it is thought that the 7nm desktop CPU will not arrive until two years’ time.

Originally reported in German, it seems the information coming from hardwareLUXX is that the transition from 14nm to 10nm for Intel desktop CPU’s has been too problematic for the company to pursue, although Intel will continue with 10nm mobile CPUs in the upcoming Ice Lake series.

Additionally, it is claimed that Intel cannot achieve the desired clock frequency they require for Ice Lake desktop CPUs on the 10nm node, plus the volume of 10nm manufacture is significantly lower than 14nm, so Intel would not be likely to fulfil its supply demands. The fact no engineering sample of Ice Lake-S or Tiger Lake-S CPUs have been seen yet, further cements this rumour according to hardwareLUXX.

If the information from the hardwareLUXX source turns out to be correct, it will mean all 10nm desktop CPUs planned by Intel have been scrapped and according to inside information, Intel will release 7nm desktop CPUs in 2022, leaving AMD with its own 7nm CPUs until then.

Production of 10nm Xeon CPUs is still on course to stay on the Intel 10nm roadmap – since Xeon relies more on core/thread count rather than clock frequency, the issues seen with 10nm desktop clock frequency won’t be as much of a problem from Xeon."

KitGuru says: Could Intel have officially conceded in its chase for 10nm desktop processors?

Insider rumors: Intel completely eliminates 10-nm plans for the desktop
Published on: Monday, October 14, 2019 at 15:00 by Andreas Schilling
https://www.hardwareluxx.de/index.p...t-10-nm-plaene-fuer-den-desktop-komplett.html

 

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

Not rumors, just good info.

 

Intel's 10nm Desktop CPU's are the worst kind of rumor: Intel continually failing to deliver, with Intel always promising that 10nm Desktop CPU's are about 2 years out on the distant horizon.

I've long ago given up on Intel ever delivering on 10nm Desktop CPU's. And, given Intel's legacy of failure with 10nm, I wouldn't be surprised if Intel's 7nm continues Intel's tradition of process delivery failures.

IMHO, it's only a matter of time before Intel giving up on 10nm desktop CPU's rumor becomes a reality.

 

You're welcome to believe whatever you want. No need to keep showing your bias.

When AMD kept promising the moon and delivering dung for a decade I didn't hear a peep out of you.

 

There in lies the problem, for a decade or so we got to hear about the supposed dung AMD was offering. Now for a period of time we get to hear about Intel's dung.

 

Of course.

And I don't mind Intel being held responsible where it's warranted.

But bringing just rumors to the conversation without any replies Intel may still be simply spouting, for now, doesn't advance this topic either.

 

did you see the review on tomshardware 9900ks? highly binned 5.2ghz 180w vs 9900k 5ghz at 200w. thats like almost 3 multiplier difference pretty amazing i must admit. i really thought they got nothing left to squeeze, now it'd be a terrible time to buy any KF/K CPU as all good silicon gone to KS.

with intel's fab size, they might be able to pump out just enough 10nm (or 10nm+++ for that matter in next 1-2 yrs) CPUs for high end desktop just like the 8086k or 9900ks and call it "desktop cpu" and keep mainstream on 14nm+++ while waiting for 7nm.

or they can just go back on their word and release 10nm NUC call it desktop to avoid lawsuit and thats it. in the end it is for their own business not to lose too much money so we'll see how they approach it.

 

Intel has cost another company $9B due to Intel's failure to deliver on their 10nm foundry promises.

Intel's 10 nm debacle claims another victim: Nokia. $9B+ plunge in market
submitted 2 days ago by Long_on_AMD
https://www.reddit.com/r/AMD_Stock/comments/dv7hay/intels_10_nm_debacle_claims_another_victim_nokia/

"I hadn't noticed this until a friend who had invested in NOK, assuming that 5G would drive their stock price up, contacted me recently in some distress. Take a look at their stock chart over the last month. They held their Q3 earnings call on October 23rd, and the next morning, $6.3 billion of their market cap had vanished into thin air. The stock has gone lower since, and is now down $9.2B in market cap: $19.9B, compared to $28.7B the night before their earnings call. That's a pretty massive haircut.

What was the problem? Well, their CEO didn't name names, but during the earnings call Q&A, he did admit "One supplier let us down". Later, analyst Robert Sanders stated: "I mean, you had this issue with ReefShark with Intel’s 10-nanometer process". You can read the entire earnings call transcript, here:

https://seekingalpha.com/article/42...-results-earnings-call-transcript?part=single

and a good article on the whole mess is here:

https://www.fiercewireless.com/5g/nokia-made-a-bad-call-for-5g-chips-scrambles-to-rectify-situation

Now, before anyone asks, Nokia is NOT the company mentioned by SemiAccurate [Ericsson]. That is yet another sorry victim. But the bottom line is that the destruction of capital among the former customers of Intel's custom foundry business has been massive.

Intel promised the moon, and delivered only a broken 10 nm process. Some bailed in time; others continued to drink the cool-aid. Nokia is only the latest example; we may yet learn of others. Companies as large as Apple bet big on Intel 5G modems, and lost. What a mess...

But it is the impact to Intel's own roadmaps that matters the most. AMD made a series of very bold bets starting in 2012, and those are now paying off big-time. But the idea that it would be Intel's turn to "lose the recipe", just when AMD roared back into performance leadership, is simply staggering. The future is bright!"

NukeMagnet 27 points 1 day ago
"Lmao even after losing 10B to intel these companies are afraid to speak their name. Intel is the Harvey Weinstein of companies, they just can't help but [ruin] everybody and when you catch them they send the Mossad after you"

EverythingIsNorminal 3 points 1 day ago
"Nokia is NOT the company mentioned by SemiAccurate.
Did he say who the other company was or how do you know that?"

UpNDownCan 1 point 17 hours ago
"Yes, he has recently said Ericsson."

I posted about this a long while back when it was public Apple that was screwed over by Intel for failing to deliver 5g 10nm, here is the article about Nokia for a refresher:

Intel Custom Foundry’s 10nm meltdown is crushing a $20+B market cap tech giant
Exclusive: This is the beginning of a far reaching story
Jul 2, 2018 by Charlie Demerjian
https://www.semiaccurate.com/2018/0...down-is-crushing-a-20b-market-cap-tech-giant/

Update to Intel Custom Foundry 10nm customer meltdown
Not one but two additional confirmations of our article
Aug 30, 2018 by Charlie Demerjian
https://www.semiaccurate.com/2018/08/30/update-to-intel-custom-foundry-10nm-customer-meltdown/

Glad it wasn't Ericsson, they had a Plan B probably a Plan C and D too, as they are pretty awesome. But Intel's failure also cost Ericsson time and money.

And, here's the Reddit thread about this from back then, and some did realize it was Nokia at that time:

Intel Custom Foundry’s 10nm meltdown is crushing a $20+B market cap tech giant
submitted 1 year ago by Thierr
https://www.reddit.com/r/AMD_Stock/..._custom_foundrys_10nm_meltdown_is_crushing_a/

Nokia, Apple, and Ericsson recovered, but are still stunted from that reliance on Intel. My recommendation: stop buying Intel now.

 

I've debated whether to post this 7nm EUV inside look at Intel's FAB because I don't want to get people's hopes up. I've posted similar coverage of 10nm in the past and nothing has come of it for us enthusiasts.

Hopefully Intel will deliver on 7nm EUV for high performance CPU's with at least as much performance as their current 14nm CPU's.

Be sure to watch all the way through to the end where he summarizes what is holding Intel back from shipping 7nm. It sounds to me to be similar to what held back Intel from 10nm, and he says in the end that if Intel can't get EUV to work - Intel is working on a new solution, which isn't ready to test, let alone produce CPU's. Then you will see why I hesitated posting this...essentially it amounts to Intel spending mass amounts of capital and getting little in return, similar to the route 10nm has taken.

I enjoyed watching this, even if it doesn't amount to anything, it's still interesting to see:

EUV: Lasers, plasma, and the sci-fi tech that will make chips faster | Upscaled
Jan 1, 2020
Engadget
Microchips are one of the most complicated objects humanity has created, packing billions of transistors into a chip only a few centimeters across. These transistors keep getting smaller and more efficient, and the current process to make chips is already astounding, requiring dozens of steps, fantastically complicated machines, and atomic-scale precision. But the current state of the art has reached its physical limits. The structures on a chip are now smaller than the wavelength of light used to make them, and any more progress will require a big change. That change is EUV, a radically new way of making chips that uses super high energy UV light created from a complex process involving plasma and lasers. EUV will enable our devices to keep getting smaller, faster, and more efficient, but where the current process to make chips already feels like sc9i-fi technology, EUV feels like magic.

 

2020 is due or die for Intel 10nm? No? How about 2021??

Will Intel push through and deliver desktop 10nm CPU's at all, ever??

Will 10nm new architecture CPU's be devoid of the security issues and resulting load of firmware / OS mitigations? That alone might sell 10nm desktop CPU's even if it's not as fast as the top 14nm parts.

10nm Intel on Desktop in 2021 | Whispers of Alder Lake
Jan 12, 2020
Moore's Law Is Dead
Alder Lake stands to be Intel's last chance to not have a woeful 2021. Fingers crossed...
1) 0:07 Intro to Intel's Situation
2) 4:00 Alder Lake
3) 9:22 Ice Lake & Cooper Lake Server
4) 14:18 Closing Thoughts