Intel’s Massive Sapphire Rapids Processor Poses for Camera (Updated)

Sapphire Rapids
(Image credit: YuuKi_AnS)

Update 2/4/2020 6:00pm PT: Added additional pictures of delidded processor. We've added additional analysis below.  

Intel's CEO recently confirmed that the company had started to sample its next-generation Xeon Scalable 'Sapphire Rapids' processors to partners, so it isn't particularly surprising that the first images of the CPU have emerged, proving that the chip is in the hands of Intel's partners. 

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.

  • greatmaharg
    " inconsistent performance from 4 glued-together desktop die." - Intel marketing dept.
    Reply
  • rtoaht
    greatmaharg said:
    " inconsistent performance from 4 glued-together desktop die." - Intel marketing dept.
    ...and they were right. Naples provided inconsistent performance due to the was the memory subsystem was broken down in 4 chiplets. Due to NUMA, whenever one chiplet needs to access the non-local memory (located in other chiplet) via infinity fabric, the performance tanked and it also came with latency penalty. 2nd gen Rome significantly improved (but not fully resolved) this issue with better memory architecture and larger cache. Finally, "glue" is actually a technical term which was previously used by AMD marketing department to deride Intel's old dual core CPUs.
    Reply
  • spongiemaster
    greatmaharg said:
    " inconsistent performance from 4 glued-together desktop die." - Intel marketing dept.
    And Intel was factually correct about the inconsistent performance, though the glue part was unnecessary. AMD acknowledged this when they released the game mode for Threadripper that disabled one of the CCXes.
    Reply
  • Paul Alcorn
    Any guesses on what that FPGA is there to fix? :P
    Reply
  • spongiemaster
    PaulAlcorn said:
    Any guesses on what that FPGA is there to fix? :p
    The Pentium division bug?
    Reply