Intel 10nm back on track

What is 10nm?

Reminder, Intel gutted their 10nm to get it out the door. AMD's 7nm might be more like Intel's plans for 10nm but Intel's 10nm is going to look more like 12nm now.

Translation: Let's throw out some bullshit to try and appease our shareholders, who are practically standing outside with torches and pitchforks at this point.

A new manufacturing process providing smaller transistors for better performance, lower power consumption, smaller chips, etc.. It's closer to what everybody else calls 7nm but no part of it actually seems to measure 10/7nm across.

So... more "2 core + non functioning gpu" cpus flooding the market?

this guy gets it.

TSMC's 7nm is superior to Intel's original plans for 10nm.

Intel's "10nm" is going to perform worse than 14nm from two years ago, maybe even worse than that. At release Intel expected the real 10nm would perform worse than 14+, and now they've had to significantly devolve the process to make it feasible for mass productions with reasonable yields. Even if Intel manages to get it out the door a few weeks early they have lost their silicon lead and I doubt they will ever get it back.

So is it bigger or smaller than 10? Like isnt 5 the natty limit?

>I doubt they will ever get it back.
t. amdcuck

But they would still be amd just 7nm.

Attached: 984646561661.png (596x576, 226K)

reminder that intel 10nm is worse than 14nm+++++++++++++++++++++++++++++++++

Attached: intel10nmregression.png (741x722, 127K)

I went to check on WikiChip and it's been updated to add the gate width, which is actually 7nm. It's all the other dimensions that are larger.
Gate Width = 7nm
Gate Pitch = 34nm
Gate Height = 53nm

>Gate Height = 53nm
literally a wall

By the time they reach volume production on "10nm" TSMC will already be sampling 5nm. They haven't even begun looking into EUV tooling which is going to be required for all nodes going forward to achieve reasonable yields without insane mask complexity. TSMC will be sampling EUV silicon within a few months of Intel starting to develop and EUV process, and bear in mind too that TSMC has Nvidia, Apple, Qualcomm, and now AMD to support rapid and expensive development of more advanced nodes, whereas Intel is only producing for themselves. Honestly I would not be surprised if they spun off their fabs within the next decade. In-house fabs are just too expensive to stay competitive.

>back on track
I mean technically yeah, they'd have been producing that shit for 4 years so if they start producing it then they're "back on track."

die shrinks aren't really that relevant for desktop processors. we've reached diminishing returns already in my opinion.
p.s. I guess Jow Forums is working again.

this really irks me.
especially when intelifags argue like it's going to keep AMD beat. turtle vs hare analogy etc

>it's going to keep AMD beat
What's going to keep AMD beat is Intel offering 60% discounts on Xeon processors when data centers start looking at ordering 100+ Epyc systems.

I need to upgrade my CPU and I was thinking about buying a 9700K or 9900K but with this news should I be a waitfag and suffer for another year?

Attached: 1537142049877.png (499x409, 322K)

servethehome.com/intel-is-serving-major-xeon-discounts-to-combat-amd-epyc/

Honestly there's no point in waiting, just buy now.

If Intel and AMD get into a price war, I have a feeling AMD is going to come out the winner.

>by 4-6 weeks
A month to a month and a half. Still more than 6 months behind AMD making 7nm.

that would cut into Intel's revenue so far that I don't think it would make much difference.
right now aren't datacenters 100% Intel anyways?
then it would be 100% Intel at 60% discount, and let's be real AMD would get some of that.
98% Intel 2% AMD.
yeah. unless you're fixed on Intel or impatient, then don't wait. if you're going to wait, it'd be to see AMD release Zen 2. And then I'd advise you to skip that and buy Zen 2+ or 3.
ok dude but economics dictate that it's a huge cut out of their profits. can you tell me what the article is implying because I can't gauge just from the breadcrumb.
No point in buying now if you have an Intel from the past 2-3 years.
SSDs are making the biggest strides in the past few years. get a new SSD.

Why would you knowingly buy insecure products with 24+ hardware security vulnerabiltiies in them?

also daily reminder that Intel knowingly launched the 8th gen Coffee Lake processors when they knew about the hardware security vulnerabilities for months before the launch. In fact, they even rushed the launch date by a couple of months (which is why Kaby Lake 7th gen had a fucking life cycle of 7 fucking months...)

Waiting for 10nm is stupid though.

Attached: 1507417167484.png (200x200, 26K)

>4-6 weeks
so it'll come out in q3 2021 instead of q4 2021?

If they can keep AMD out of the datacenter by losing money then they win anyways.
>I wish AMD was competitive
Literally my IT department. They ended up buying me a shitty 8 core Xeon system that costed 3500 dollars.

desu I think hardware exploits could be mitigated by use of a completely secure opensource firewall.
they're
as in a smart person would wait longer until they get a more efficient pipeline/latencies or increase IPC ? or a smart person would buy now in spite of there being imperceivable difference in performance.
I don't know if Intel can keep winning with the same old tricks.
We're reaching the physical limitations of process improvement.
Cutting 40% off the product with the highest margins just to keep AMD out of the market for the moment is a sign the times are changing.

>a month
HAHAHAHAHAHAHAHA
hope it gets here by 2021

>they're
lul I got distracted
I agree completely which is why I said to wait for Zen 2 to see if it's promising then buy Zen 2+
you can upgrade your GFX card or SSD if you need to spend money. While Intel did move and AMD is getting better, Intel's move was small and AMD is still catching up to Intel.

if after all this intel 10nm still somehow comes out before zen2 the amd pajeets of this board are going to be committing suicide en masse

"""""10 nm""""" aka 12 nm

Attached: 1505147990486.jpg (329x329, 57K)

>inb4 2020 and still no 10nm

So what happened?

Did they solve something? Did Jim Keller work his magic?

Or is this just a press release because the 9th generation is showing up next week?

but AMD has 7nm coming in just a few months

>Back on track
So it's out for 5 years now?

You know Intel engineers recognized that early 10nm will perform worse than good old refined 14+++++++++++++++, right?

>3 years late
>going on 4
>back on track

TSMC has began commercial production of 7nm like... 3 months ago. Intel estimates 10nm CPUs by end of 2019, a full ~9 months after AMD's 7nm.

The only thing holding back Intel are fabs, they're basically small compared to Chink TSMC fabs since they only produce for themselves and Micron NAND.

The moment they move their i7s to TSMC is the end of AMD's lithographic advantage. Just like Nvidia and AMD having the same node for years yet Nvidia was always miles faster and miles ahead.

>always
heh, zoomer

>our stocks are falling
>do something goy
>release some fake news
|>
|

>>The moment they move their i7s to TSMC
In your wildest dreams. Intel is too prideful.

What's stopping you guys from memeing "just wait for 10+" or "just wait for 10++" or "just wait for 7"

lol this

>producing for themselves
That's false, actually. They are doing some production for other companies, like Apple's modems.

>Nvidia and AMD having the same node for years
Wat. Are you rerarded? Tsmc 16nm is completely different from Glofo's LP 14nm that was used for vega and polaris. On previous nodes amd was better than nvidia for half of a time.

|>
|
|3
|
|

>jule 2019
i mean even intel said 2021 now they jumped up to 2019?
kek for sure

>2019 CPU low consume
>2021 big server chips

This whole fucking ordeal is only about confusing customers as the new processors are released. GaymerNewz will raport these right next to each other:
>Intel is on time in their superb processor fabbing
>The new i7 will blow your dick

Additional bonus for stock holder scamming.

What I don't understand is how the open letter of Intel's interim CEO from friday stated that the current 14nm shortage isn't going to affect revenue targets for Q4 18 and Q1 19 at all. This makes no fucking sense, given that AMD should be vastly outselling Intel on all fronts, if the current sample from mindfactory is any indication.

Attached: amd intel current sales.png (1700x650, 152K)

>relesing consumer chips BEFORE server chips

yeah no
LOL

This is what I keep saying about 'die shrinks'. They haven't actually made the transistors smaller, they just changed the shape. The volume remains fairly constant.

Can someone translate? How close are we to the limit? And does it even really matter?

Basically this. All of hose shintelkikes going on and on about "hurr, shittel 14 nm is better than ayymd 7nm" are doing nothing except projecting hard.

Jews choose would money over pride any day

When TSMC releases 7nm Intel will be 5 years behind in transistor technology

Attached: 1536694146733.png (1200x800, 164K)

YES, WE WON INTELBROS

Attached: 1536612142306.png (504x400, 91K)

so the usual kike tricks form a corrupt to thd core piece of shit of a company, more at 11

when will intel 7nm come out?
in year 3000?

Attached: intell btfo.jpg (638x599, 123K)

Press release, Keller can't unfuck fabs.

He went in and schooled them to fix their shit. Result is working 12nm called 10nm.

>IC: How are you involved in how new processes are developed?

>JK: Well in terms of how do they put metal layers on wafers, no. In terms of how do we define the next generation technology node optimized for performance frequency density power? You bet! In the first 2 weeks we sat down and walked through the metal stack for Intel's 14, 10 and 7nm and I have lots of opinions about that. So you know, I have a simple theory about this kind of stuff: as I tell people my opinions, one of two things happen. They like what I say, or they correct me when I'm wrong. So yeah, I'm going to be quite involved in that.

> In the first 2 weeks we sat down and walked through the metal stack for Intel's 14, 10 and 7nm and I have lots of opinions about that.
> I have lots of opinions about that.
When you can't say "shit is FUBAR".

Just wait for 10nm+++++

Well done Intel, now you're only 202-204 weeks behind

>delayed by 4 years
>actually it's only going to be delayed by 3.9 years
>journalists: INTEL IS AHEAD OF SCHEDULE

l>
l
l3

Keep it simple.

Some really disgusting kikery is going on

Oy vey, stop it with your hate speech goy.

Well, that's not completely true. While it might not increase the performance as much as it used to do, "halving" the node size still gives decent performance uplifts.
But in this case it's especially important since amd is transitioning from a low power node to a high performance node, which should move the voltage wall of the ryzen processors up by roughly one gigahertz. That's what we're all really waiting for. The extra performance/power gains of 7nm are just a bonus.

It wouldn't surprise me if Incel release a broken 10nm lithography desu

Still no hi-end parts till 2021 so it's literally nothing. Blatant stock manipulation from jews.

wccftech.com/amd-zen-2-ryzen-8-core-16-thread-cpu-leak/

> 10-15% higher IPC
> 4 GHz base, 4.5. GHz on turbo
> on an early engineering sample that's still unstable

If that leak is true, it sounds pretty good. Any estimates what overclocks we could get out of this?

How about you develop some reading comprehensions?
The leak ends at AMD CPU Roadmap. Everything after that is speculation from literal nonames.
What's known is that eng sample clocks at 4 GHz and 4.5 GHz turbo unstable, can't even finish benchmarks. You can expect like 4.2 GHz base clock on highest end finished products, not sure about turbo/xfr. The part about 10-15% IPC increase is bullshit. It may be 0% IPC increase or 50% IPC increase, nobody has info on it.

I'm not holding my breath

Attached: akko4721.jpg (361x361, 17K)

Considering my 2700x barely goes over 4ghz-4.35 if say 4.5ghz is easily done.
I wonder what 7nm+ will be like and zen 4

Even if literally all of that is true it only means Zen 7nm will be equivalent to coffee lake lmao Intel wins again

>4-6 weeks
Who cares.
It's already 3 years late.

You mean its like original Zen ES, which had 2.8Ghz base and 3.2 boost and couldn't finish benchmarks as well? ES are testing stuff for motherboard manufacturers they won't be signally their real capabilities.

>trying to undestand jewish tricks

They shifted suppy to Xeons which have much higher margins.

You forgot that unlike intel AMD operates under TDP. Turbo might as well go over 5 Ghz, but not base clock. Reminder that 8700k's base clock is 3.7 GHz.

I mean if you bundle speculation with the leak it doesn't become a leak, it's still speculation.

Only to cut down those xeons into fuckup lake XEs and sell them with a 200% less profit?

Not anymore, Intel has been offering 60% levels of discounts on their Xeons to anyone that phones them telling them they plan to buy EBYN. And what yields do Xeons have compared to EPYC?
As I understand the schedule called for a ramp down in 14nm production to make way for 10nm, and moving stuff like Apple's modems and NAND to 14nm, not decisions or contracts that you can just backpedal with

Weren't apple modems cancelled already?