Intel 14 nm delayed.

[PontiacGTX]

So intel will release intel broadwell after 2nd half of 2014

 

[monstercameron]

so, it seems like even the fab champ intel isnt immune to production woes...I hope product delays dont get brought up in technical arguments anymore...

 

[jdubs03]

Its unfortunate, but Skylake is not impacted by the delay so while this is definitely a short-term issue, longer term it is the same. Hopefully the transition to 10nm (Cannonlake) will go smoother, but alas that's 2 yrs away.

I doubt that this will impact Airmont, though I'm surprised the analysts didn't ask that question. This may just mean the release schedule for Broadwell/Airmont will get closer. Intel cannot afford to push back Airmont, if they did they'd be competing against potentially second gen A57's on 20nm. If that is the case I would be considerably worried about attracting design wins.

 

[OBLAMA2009]

They actually sell better than expected. North American market only shrinked with 0.2% now. With some luck sales will go forward next Q.

i dont see anyone looking at or buying desktop pcs in stores anymore, even i dont really check them out.

 

[ShintaiDK]

i dont see anyone looking at or buying desktop pcs in stores anymore, even i dont really check them out.

Good thing we got numbers instead of guessing then

 

[videogames101]

BREAKING NEWS: Nodes below 20nm will have yield issues.

404 Story not found.

I don't expect any fab company to ramp production of sub-20nm nodes without significant yield issues, and neither should you.

http://news.cnet.com/8301-1001_3-57...adwell-pc-chip-production-to-early-next-year/



Well I'm not surprised. Anyone else?

No one is surprised. The only question is will you be surprised when TSMC and GF have the same, if not worse, delays when they transition to a full node below 20nm. Too bad we won't find out anytime soon.

 

[SiliconWars]

Actually quite a lot of people (many in this thread) WERE surprised even though it was staring them in the face for months.

As for your "expectations", let me guess - you have an engineering qualification or you just pulled that from your rear end because if Intel is having issues, everybody else must as well?

 

[videogames101]

Actually quite a lot of people (many in this thread) WERE surprised even though it was staring them in the face for months.

As for your "expectations", let me guess - you have an engineering qualification or you just pulled that from your rear end because if Intel is having issues, everybody else must as well?

Everybody else has been having issues. Have you not followed process development for the past few years? Or do you think TSMC just decides to delay every new node for the fun of it? Is GF holding back because of the phase of the moon?

I'm an EE student with a focus on IC physical layout design.

 

[Revolution 11]

Actually quite a lot of people (many in this thread) WERE surprised even though it was staring them in the face for months.

As for your "expectations", let me guess - you have an engineering qualification or you just pulled that from your rear end because if Intel is having issues, everybody else must as well?

No, it is more like if Intel is having trouble at the 14nm node despite their overwhelming advantage in institutional knowledge, labor talent, process experience, and funding, then the future nodes don't bode well for TSMC, GloFo, Samsung, and IBM.

Which is unfortunate. Easier nodes are better for everyone, the companies, the consumers, the world even. We all need faster, cheaper, more efficient computers yesterday. But physics is a b****.

 

[SiliconWars]

No, it is more like if Intel is having trouble at the 14nm node despite their overwhelming advantage in institutional knowledge, labor talent, process experience, and funding, then the future nodes don't bode well for TSMC, GloFo, Samsung, and IBM.

They had all that at 90nm as well. AMD didn't have their issues.

 

[ShintaiDK]

This terrible delay. And AMD will still be hit back further with its node. So AMDs 28nm vs Intels 14nm.

What a shame, Intel gets a small delay, AMD essentially gets a full year delay from being fabless. Brilliant.

 

[Revolution 11]

They had all that at 90nm as well. AMD didn't have their issues.

Just because one node turned out one way, does not mean all of them are like that. Leakage was not a concern at the 1 um node, it is a big concern now.

 

[SiliconWars]

Just because one node turned out one way, does not mean all of them are like that. Leakage was not a concern at the 1 um node, it is a big concern now.

And just because Intel is having issues doesn't mean everybody else will.

 

[ShintaiDK]

And just because Intel is having issues doesn't mean everybody else will.

Issues is the norm for GloFO and TSMC. Masters of delays and problems.

 

[SiliconWars]

Issues is the norm for GloFO and TSMC. Masters of delays and problems.

Is that so? Why don't you tell us what issues they've had and at what nodes, I could do with a history lesson.

 

[ShintaiDK]

Is that so? Why don't you tell us what issues they've had and at what nodes, I could do with a history lesson.

Their ever promising roadmaps and never able to even remotely close to deliver on time?

The biggest joke of the industry, GLoFo are obviously both shipping 22nn and 20nm. And shipped 28nm for a looong time.

Oh wait, its not. FAIL FAIL FAIL!

 

[SiliconWars]

So basically, that's one delay at 28nm (which incidentally has been shipping for almost a year).

 

[ShintaiDK]

So basically, that's one delay at 28nm (which incidentally has been shipping for almost a year).

One? LOL!!!!

 

[SiliconWars]

One? LOL!!!!

Wow, even by your standards I thought you'd be able to figure this one out.

That's a long term roadmap. If one node gets delayed, everything after it gets delayed. If you had a long term Intel roadmap you'd see how the delays were creeping in to that as well. You would see how 10nm *will* be delayed because 14nm is. That's just how it goes.

So yeah, one delay at 28nm.

 

[AtenRa]

Every engineering project has problems, some are bigger and more complicated than others. You may overcome them within the time frame of your project or it may take longer than anticipated.

As nanometer production is getting more complicated(smaller and smaller), the bigger, more complex and unexpected the problems will be. This is the reason why each node doubles in cost, because they must have them ready at the same interval each time. So they pour more resources to the problems to get the node production ready on time.

 

[guskline]

This report didn't shock me AND if your READ the article, Intel has already fixed the problem. Production will ramp up in the first quarter of 2014. To somehow read into this article that Intel is slipping or in trouble is frankly, SILLY.

 

[SiliconWars]

This report didn't shock me AND if your READ the article, Intel has already fixed the problem. Production will ramp up in the first quarter of 2014. To somehow read into this article that Intel is slipping or in trouble is frankly, SILLY.

Seems I'm one of the few who actually did read the article. They *think* it's fixed, in other words they don't know for sure.

As for thinking Intel is slipping is SILLY? Do you really believe they would come out and say it now if it was only a minor slip? I am willing to state right now that it will be LATE Q4 2014 before we see Broadwell products.

 

[Phynaz]

Wow, even by your standards I thought you'd be able to figure this one out.

That's a long term roadmap. If one node gets delayed, everything after it gets delayed. If you had a long term Intel roadmap you'd see how the delays were creeping in to that as well. You would see how 10nm *will* be delayed because 14nm is. That's just how it goes.

So yeah, one delay at 28nm.

Manufacturing node development doesn't happen sequentially. A delay in one node doesn't have to impact the following node.

You are aware that TSMC completely cancelled a node, right? Or, more factually, renamed it after it was two years late. As for GloFo, you can get whiplash trying keep up with their roadmap slides.

 

[AtenRa]

This report didn't shock me AND if your READ the article, Intel has already fixed the problem. Production will ramp up in the first quarter of 2014. To somehow read into this article that Intel is slipping or in trouble is frankly, SILLY.

They are certainly not in trouble but they are sure slipping off the two year tick-tock timeframe with each node.

First 45nm product release on Q4 2007 (October 28 2007) Core 2 Extreme QX9650 (Penryn)
First 32nm product release on Q1 2010 (January 4 2010) Core i5 and Core i3 (Clarkdale) ------> ~27 months
First 22nm Product release on Q2 2012 (April 23 2012) Core i5 and i7 (IvyBridge) ------> ~28 months
First 14nm product release on Q4 2014 (October 2014 ???) Broadwel (first time Mobile product gets the new process before Desktop) ------> ~30 months

 

[Khato]

Seems I'm one of the few who actually did read the article. They *think* it's fixed, in other words they don't know for sure.

It's fixed. Feel free to try and play word games with what has been publicly stated though, since apparently it wasn't adequately declarative. Doesn't change the fact that the primary issue has been resolved. My guess is that they waited until now to say anything for the simple fact that announcing that there's been a problem without stating that it's been resolved is far more detrimental than staying quiet in the interim.