Intel Shows Off Meteor Lake Test Chips

Meteor Lake Chiplets
(Image credit: Stephen Shankman / CNET)
Anton Shilov
Contributing Writer

Anton Shilov is a contributing writer at Tom’s Hardware. Over the past couple of decades, he has covered everything from CPUs and GPUs to supercomputers and from modern process technologies and latest fab tools to high-tech industry trends.

  • Co BIY
    Interesting picture. I though the point of the chiplet concept was to allow the building of the individual parts separately (different specialized wafers, even separate fabs and nodes) and then "package" them together. Here they are all together on a wafer waiting to be cut. I assume that means they are assembled on top of the base wafer and further processed while the base wafer is it is still whole.
    Reply
  • dalek1234
    Co BIY said:
    Interesting picture. I though the point of the chiplet concept was to allow the building of the individual parts separately ...

    If you are right, then that would most likely mean expensive chip to produce due to poor yields, unless Intel can disable defective components without sacrificing performance of the good parts around. This will make it difficult to compete on price.
    Reply
  • Co BIY
    dalek1234 said:
    If you are right, then that would most likely mean expensive chip to produce due to poor yields, unless Intel can disable defective components without sacrificing performance of the good parts around. This will make it difficult to compete on price.

    Now that I have reviewed all the very interesting photos in the CNET link I see that the individual chiplets are "built" separately but then "assembled" on their substrate wafer while it is still whole. After the chiplets are "bonded" to the substrate wafer then the assembled chips are cut.

    Interesting and pretty amazing process. An entire fab is dedicated to that process alone.

    Apparently the Meteor Lake Test chips pictured are package testing chips only. They are not functioning silicon off the Intel 4 process.
    Reply