The research and development of a technology company is a cornerstone on which its foundation can be based, but it must adapt to new changing times such as this.
Swwan was aware of this and his new team had decided on a series of specific tactics he revealed in an interview with the Times.
10 nm lithographic process, Lakefield and more
Knowing that AMD and TSMC will take the lead, the company's first real rate when Swwan officially arrived was the milestone in the 10 nm lithographic process.
Not that he would lose the leadership of the semiconductors, but further delays have given a bad name to investors. Therefore, he contributed significantly to the final construction of the node and increased its production to the limit, when at the end of the year we should live the war between Zen 2 and Ice Lake-S
The second step was to research and expand the technology and 3D and chip technologies, where Lakefield and Foveros appeared, an integrated design firm that left AMD and TSMC on paper far, at least for now, and where it would have to be designed and adapted to high performance processors, because focusing only on small machines or devices.
The ascendancy of Jim Keller, the father of Zen architecture
After his journey with Apple, AMD and Tesla, Jim Keller returned to Intel to help and bring his experience in chip design. After a year and a half he has been appointed to the position of Senior Vice President of Intel TSCG and the General Manager of Engineering.
This appointment has led Intel into a new evolution for the future, when Jim Keller changed the CPU development process past. At the company's highest level, CPU designers rarely interacted with processors, where many large teams were formed and this was inefficient and unchanged.
Keller urges Intel to establish smaller teams, which can allow the creation of specific components or chip locations so that collaboration from both employees is unnecessary and focused on specific tasks.
Jim said the initial implementation of the program is paid off, when this upgrade of the R&D process reduced the time needed to develop and test chips from weeks to just days.
This, he says, makes the complete construction of the chip look three times faster than the past few months. We won't even notice these changes 14 nm ++ not even at 10 nm, so it is expected that the speed of Intel's roadblocks in the coming years is a flaw of this systematic move submitted by Keller, where once completed, the blue giant returns to the battle in time and profit providing both TSMC as Samsung.
In any case, it will be the next year in which we see all these changes, where Intel begins with the benefit of being a judge or making its products.