After new CEO Lip-Bu Tan presented his bonafides to the Intel Vision 2025 audience yesterday comes some product news from the event. Well, just about. In truth, there was no major product bomb, but Intel did confirm that it's next-gen Panther Lake CPU was on track to go into production later this year with availability seemingly pencilled in for 2026.
Jim Johnson, Intel's head of client computing, told the assembled Intel Vision 2025 throng that Panther Lake, "combines the power efficiency of Lunar Lake, the performance of Arrow Lake, and it's built to scale 18A and it's on track for production later this year."
Just to break that down a little, Lunar Lake is Intel's low-power chip for laptops, which has on-package memory and relatively limited core counts and configurations, while Arrow Lake is the mobile version of Intel's current desktop CPU architecture and scales up to higher core counts than Lunar Lake. Both are mostly manufactured by TSMC rather than Intel's fabs.
Lunar Lake has particularly impressed with its power efficiency, so a high-performance CPU that can deliver similar battery life to Lunar Lake would be pretty sweet, that's for sure.
As for the "production later this year" bit, the screen behind Johnson on the keynote stage clearly labelled Panther Lake as a 2026 product. So, that likely reflects the lag between production start and actual availability.
In other words, it will take some time from the initiation of manufacturing wafers with Panther Lake CPU dies to package them into functional chips and ship them out to laptop makers before actual systems with Panther Lake are available to buy.
That Panther Lake is slated as a 2026 CPU implies that production will probably start quite late in 2025, therefore pushing actual availability into 2026. This all represents a delay of around a year given Intel originally said 18A was set for production at the end of 2024.
Of course, there was supposed to be a 20A node before 18A, but that was entirely cancelled last year. And if you're wondering why 18A is late, well, there have been reports of bad yields but what's absolutely for sure is that it's an ambitious node that doesn't just dramatically improve Intel's transistor density, it also debuts key new technologies such as backside power, which is arguably as significant as the node's improvements in transistor density.
In that regard, 18A reminiscent of Intel's notorious 10nm node and not in a good way. 10nm was also highly ambitious, ended up arriving at least five years late and was indeed the node that got Intel's fab into the trouble from which they've since struggled to emerge.
Thus, and as we've said on countless occasions in the last few years, the wait continues for Intel's undeniable return to technology leadership. Will Intel 18A really go into production later this year? How competitive will it be with TSMC's current N3 nodes let alone its imminent N2 nodes? Eventually, we'll have some answers.
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
