During Intel's (INTC 1.74%) April 26 earnings conference call, the company's management finally admitted that progress in developing its latest 10nm chip manufacturing technology was going slower than planned.
According to Intel, 10nm is supposed to offer both a significant area reduction and power efficiency improvements over its currently shipping 14nm technology.
Image source: Intel.
In particular, the company claims that the yield rate of chips produced on that technology -- that is, the percentage of the chips produced that Intel can actually sell -- isn't where it needs to be, and it'll take more time than the company had expected to get it to that point.
Although this delay is a clear negative for Intel because it means that the company's new and dramatically improved products that were slated to be built using this technology have now been pushed out yet again, Intel CEO Brian Krzanich explained on the call how Intel can get away with this delay without an adverse impact on its business.
Milking 14nm tech
"We have leadership products on the road map that continue to take advantage of 14nm, with Whiskey Lake for clients and Cascade Lake for the data center coming later this year," Krzanich said.
Krzanich, in defense of the company's continued use of 14nm while it tries to get the issues with 10nm sorted out, said that "14nm process optimizations and architectural improvements have resulted in performance gains of more than 70% since the first 14nm products were launched."
The executive added additional thoughts later on during the question-and-answer session, saying that its 14nm technology "continues to have legs" and that the company thinks that it can "continue to make improvements, both within that process technology and architecturally."
Is a 14nm+++ in the cards?
What I think is particularly interesting is that if Intel doesn't go into volume production on 10nm products until the second half of 2019 (something that Krzanich conceded on the earnings call was a distinct possibility), it could face issues in having new products to sell during the first half of 2019.
For example, Intel mentioned that for the client market, it will be rolling out Whiskey Lake later this year, but Whiskey Lake is a product that's targeted at the mainstream notebook computer market. A large market to be sure, but it's not clear if Intel will be able to introduce new 14nm products for the high-performance notebook and mainstream desktop computer markets for product launches in the first half of 2019.
The question, then, is whether Intel intends to further enhance its 14nm technology to support a new set of products for desktops and high-performance notebooks. There's no doubt in my mind that Intel could introduce, say, a 14nm+++ technology that boosts performance over the company's currently shipping 14nm++ technology, but whether Intel planned well in advance to develop such a technology as well as products based on it for launch early next year is an open question.
This is a question that we're likely to learn the answer to over the course of the year as information about 2019 product plans leaks out from the supply chain.
