- A rumor from reputable trade press firm, Digitimes, claims that Intel's next generation 14-nanometer PC chip codenamed "Broadwell" is delayed
- Such a delay throws off Intel's entire product pipeline and suggest delays for 14-nanometer mobile prodcessors
- Without this process lead, Intel's key structural advantage in mobile/computing begins to erode
I have to hand some pretty serious props to SemiWiki's Daniel Nenni. While many (including yours truly) gave Mr. Nenni quite a bit of flack for claiming that Intel's (NASDAQ:INTC) 14-nanometer process was delayed (even with Intel publicly claiming that 14-nanometers were on track as late as September 2013), it turned out that he was right on the money, as he had heard from his sources that the equipment move-in for the 14-nanometer node was delayed, implying that the process itself wasn't ready.
At the 2013 Investor Meeting, Intel's management claimed that the first 14-nanometer processor, code-named "Broadwell", was on-track for volume production during 1Q 2014 and a 2H 2014 system availability. Intel also claimed that its next generation Atom processor, code-named "Cherry Trail", would be available in late 2014.
Unfortunately, Digitimes claims that Broadwell has been delayed yet again to the 4th quarter of 2014, which further implies that Intel's first 14-nanometer mobile SoC ("Cherry Trail") will be delayed to early 2015, if true, as Intel's low power SoC processes tend to lag the mainstream CPU ones.
I don't know if this is true, but if it is, then the repercussions could be pretty serious and it is very likely that Intel will see its "process lead" that it has claimed against TSMC (NYSE:TSM) and the rest of the ARM (NASDAQ:ARMH) ecosystemm erode. This is in light of the fact that TSMC is in high volume production of its 20-nanometer process today and claims to be on-track for production of its 16nm FinFET process for early next year.
Please, Not Another Disappointment!
I don't understand why Intel seems unable to leverage its process lead to really "win" this market, and I further don't get why Intel is having so much trouble with 14-nanometer while TSMC/Samsung claim to be on-track for 2015 production (right around when Intel will have it).
It had a great opportunity, a several year window, of FinFETs to itself. Its product development teams completely missed the mark in smartphones and tablets during 2013 (when this should have been a year during which nobody could have matched the performance/power/area of its mobile processors), 2014 doesn't look much better (since the 22-nanometer products that could be competitive won't roll out until the end of the year), and now it looks as though even the vaunted 14-nanometer process will be delayed into 2015 for mobile parts, implying that Broxton - the "absolute leadership" part from Intel - could be delayed into late 2015.
Now, if TSMC doesn't really ramp 20-nanometer into the market until late 2014/early 2015, then this gives Intel the entirety of 2015 on 14-nanometer to fight TSMC's 20-nanometer process. However, if TSMC moves more quickly than expected, then it'll be a match up between Intel's 14-nanometer process against TSMC's 16-nanometer process.
While Intel claims that it has a significant performance/density advantage with its 14-nanometer process against TSMC's 16-nanometer, there has been a lot of mudslinging and distortion behind these claims, so it's tough to know what to believe. At any rate, process is just one (major) enabler and the actual SoC design is equally, if not more, important than the underlying process. Intel had the better process at 22-nanometer, but its product definition teams didn't do a great job forecasting exactly what the market needed that it all turned out to be for naught.
I still remain long the shares in the hopes that the company really gets its mobile act together by the end of this year and in 2015 because if the company is successful, there's a LOT of operating leverage to be had and money to make owning the stock. However, with each passing day, it seems that the thesis that I've done so much work on is turning out to be more fantasy than anything based in reality.
With the ARM ecosystem benefiting from exploding leading edge volumes, enabling the likes of Samsung and TSMC on the manufacturing side of things, Intel's process lead is at serious risk, as is its long-term viability as the world's leading semiconductor vendor by revenue.
Here's hoping for something great from Intel at Mobile World Congress.
Disclosure: I am long INTC. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.