Intel 14 nm delayed.

[SiliconWars]

Not necc true. If they wanted to prioritize atom they actually have better yield at 14nm vis a vis broadwell.

That is how I understand it as well:

Hard to say for sure but the die sizes might not be all that different. Intel will try to push down the Broadwell die while perhaps maintaining the ~100mm2 Atom (they need to at least double graphics and probably more). Yes it will still favour Atom but not by a huge amount.

Another problem with the early 14nm Atom is the price. Throw defects on top of that and you have a situation where they can't sell Atom above cost. I won't be surprised if Atom is pushed back a little instead of pulled in.

 

[Nothingness]

This raises a question: back at IDF Brian Krzanich firmly stated that 14nm was on track. And now we learn that he wasn't correct. What does that mean about Intel new CEO? Didn't he know? Did he lie?

 

[AtenRa]

One year has 12 months. 1½ years would then be how many months? Feel free to take your shoes off if you need to, but it's just a hair over half of your figure.

edit: 6 quarters is 1½ years

From April 2012 to April 2013 = 12 months or 4 Quarters
From April 2013 to April 2014 = 12 months or 4 Quarters
From April 2014 to October 2014 = 6 months or 2 Quarters

Total months = 30
Total Quarters = 10

OK now ???

 

[mrmt]

This raises a question: back at IDF Brian Krzanich firmly stated that 14nm was on track. And now we learn that he wasn't correct. What does that mean about Intel new CEO? Didn't he know? Did he lie?

What happens when you hit a setback in the middle of a project? Does that mean that you were lying all the time before the setback?

 

[poohbear]

Is broadwell gonna be 6 cores or 8 cores? It cant possibly only be 4 core what with next gen consoles having 8 cores.

 

[biostud]

Is broadwell gonna be 6 cores or 8 cores? It cant possibly only be 4 core what with next gen consoles having 8 cores.

Yes they can. 6/8 cores will be Haswell-E on 2011-3 platform.

 

[Nothingness]

What happens when you hit a setback in the middle of a project? Does that mean that you were lying all the time before the setback?

IDF was only a month ago. BK said in last night conf call that they had to change a few things and got back good data after the changes. Given the time it takes to change things to correct density issues, I don't think the issue wasn't known back at IDF.

 

[SiliconWars]

Of course it was known back then. If you've been following it closely enough it's been known for months that 14nm was in trouble - http://www.semiwiki.com/forum/content/2640-intel-14nm-delayed.html was by no means the start and that was 3 months ago. Not only did they lie at IDF, they lied at the Q2 CC.

14nm move-in was supposed to happen in Q3 but it did not. I got an update this week and was told it would “probably” not happen until Q1.

Sounds familiar to what we heard last night right?

It was known outside of Intel 3 months ago. Krzanich had to reveal it now because the questions just kept mounting.

 

[ShintaiDK]

Is broadwell gonna be 6 cores or 8 cores? It cant possibly only be 4 core what with next gen consoles having 8 cores.

Steamroller is only 2M/4T as well. But again, a fast dualcore is kinda the same speed as the weak console CPUs.

 

[mrmt]

IDF was only a month ago. BK said in last night conf call that they had to change a few things and got back good data after the changes. Given the time it takes to change things to correct density issues, I don't think the issue wasn't known back at IDF.

I think IDC, pm, CTho9305 or TuxDave could chime in and give us an insight if it's possible to make process changes in such a small time frame, because you or me "thinking" something is the same as nothing. In any case there are a lot of things that could have happened, like BK having old information by IDF time or even wrong info.

 

[Nothingness]

I think IDC, pm, CTho9305 or TuxDave could chime in and give us an insight if it's possible to make process changes in such a small time frame, because you or me "thinking" something is the same as nothing.

You're assuming I don't know that kind of things I just wrote that I "thought" because I have no first hand report on how Intel deal with these issues.

In any case there are a lot of things that could have happened, like BK having old information by IDF time or even wrong info.

I agree but that'd still be an issue: he can't afford to have old information about such critical things (in particular given his background); if someone gave wrong info IMHO it's not acceptable. I guess we'll never know...

 

[ArizonaSteve]

Who cares? Whatever comes out on 14nm still be quad core and (if we're lucky) a 100 MHz bump in frequency. Yawn...

 

[mrmt]

I agree but that'd still be an issue: he can't afford to have old information about such critical things (in particular given his background); if someone gave wrong info IMHO it's not acceptable. I guess we'll never know...

I do agree with you, and the other possibilities I described by no means save his face. I just don't like to turn what could reasonably be a management mistake in a character murder fest.

 

[krumme]

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



Well I'm not surprised. Anyone else?

No. Who would expect Intel to continue like the market situation is not changed?

I think their process improvement is where it should be but their predictions about the market was very wrong. So they adapt like they reduced capex. Better say its because of some technicalities that could be prioritized, than admit their strategy and predictions was off.
So this is nothing but good news. Intel is adapting to the new market situation. Like every man they just say it was because of some exterior unforseen element.

 

[Fjodor2001]

BS. Broadwell was "first half of 2014" according to management's previous remarks. It is now Q3. Stop spreading FUD.

http://vr-zone.com/articles/intels-...re-broadwell-and-haswell-are-going/47119.html

And that was before this latest announcement that Broadwell is delayed one quarter. So this means that Broadwell now is expected to be released Q4 2014 or Q1 2015...

 

[mrmt]

And that was before this latest announcement that Broadwell is delayed one quarter. So this means that Broadwell now is expected to be released Q4 2014 or Q1 2015...

A launch with less availability would be out of question?

 

[Phynaz]

they lied at the Q2 CC.

Then you should blow the whistle to the SEC and get people sent to prison for fraud.

 

[mikk]

And that was before this latest a...map change. Overall it is nothing new for us.

 

[Arachnotronic]

http://vr-zone.com/articles/intels-...re-broadwell-and-haswell-are-going/47119.html

And that was before this latest announcement that Broadwell is delayed one quarter. So this means that Broadwell now is expected to be released Q4 2014 or Q1 2015...

Again, total FUD, since the delay is with respect to statements made on the prior earnings call,

As far as our 14 nanometer Core launch in our – just our general product launch, I think what we’ve said so far is, first half of 2014 and we’re not going to – we’re not ready to give any specifics beyond that.

 

[SiliconWars]

Probably the Roadmap change from H1 into H2 was made because of the yield issue, it is just that they informed the public much later. Production start Q4 2013 was always nonsense because of the Roadmap change. Overall it is nothing new for us.

And informed investors much later. As others have said, that's a criminal offence.

 

[liahos1]

And informed investors much later. As others have said, that's a criminal offence.

haha trust me SW its not. This is pretty benign relative to what other companies pull. Look at a 3 year chart of MLNX as an example. you're really trying to blow this into something its not.

 

[Arachnotronic]

And informed investors much later. As others have said, that's a criminal offence.

LOL, the propaganda machine continues. I guess TSMC was lying a few years ago when it said 20nm would be in high volume production in 2013?

 

[mikk]

And informed investors much later. As others have said, that's a criminal offence.

It's a standard behave from other companies like AMD. Best example their Kaveri Roadmaps, totally misleading. Or in the past Bullodzer. Lots of delays and they didn't inform their investors (or much much later). Good luck for you trial.

 

[ShintaiDK]

And informed investors much later. As others have said, that's a criminal offence.

So when do your start your lawsuit against you favourite AMD?

 

[SiliconWars]

Typical sad attempt to deflect the issue on to other companies again.

From another forum -

Something doesn't add up here:

• on September 10th, they said they were 'on track'
• it's just barely mid-October, and now it's been pushed back by a quarter, due to...
• an 'already resolved' defect

If the defect is already resolved now, they must have known about it just a few weeks ago when they claimed to be 'on track'. And they couldn't have 'already resolved' it at that time, or they would have known how much of a delay they were looking at and would be required by law to reveal it. And if it wasn't yet resolved, they wouldn't have known how long resolving it was going to take, in which case they shouldn't have confirmed the shipping guidance.

This is the kind of screw-up that can attract shareholder and/or regulator ire.