83. Intel going Fabless, Global Foundries for RDNA2, Next Gen Memory | Dave Eggleston
Episode description
Dave Eggleston has been a Product Engineer, VP, and CEO at multiple Semiconductor Companies over the past 40 years. He joins Tom to discuss the past, present, and future of memory.
[SPONSOR: https://www.cdkoffers.com/]
25% software discount code: brokensilicon
3% discount code for everything: dieshrink
Win10 pro oem key 13$: https://bit.ly/2Wdfghh
Win10 Home oem key 11$: https://bit.ly/3dsbSFi
Win10 pro oem key FOR 2PC 20$: https://bit.ly/36DFHAB
Office 2019 key 29$: https://bit.ly/2WdfBAz
Win10 pro oem+Office 2019 41$: https://bit.ly/3ea83Vo
0:00 Dave Eggleston Background(AMD, Micron, Global Foundries)
17:54 Future of Global Foundries, & TSMC’s potential domination
21:44 Why doesn’t AMD Manufacture Cards at Global Foundries?
31:05 Can and/or should Intel spin off their fabs?
36:14 How has memory evolved over the last 40 years?
45:45 How has storage design changed over time? What’s the hardest part of designing storage?
49:35 Why did it take so long to get PCIE Gen4 SSDs? Why did HMC?
57:20 Why isn't there more focus on performance versus capacity?
1:03:54 Is Intel Optane a failure? Is it being properly utilized?
1:18:07 CXL – will it replace NVMe? What is the future of memory?
1:23:58 The Future of Consumer Storage
1:41:57 What type of memory will next gen consoles like PS6 utilize?
2:01:07 The Future of Chip Design & Processing
https://www.linkedin.com/in/deggleston/
https://www.techpowerup.com/225687/globalfoundries-announces-its-12-nm-fd-soi-silicon-fabrication-node
https://semiwiki.com/wikis/company-wikis/globalfoundries-wiki/
https://www.congress.gov/bill/116th-congress/house-bill/7178
https://en.wikipedia.org/wiki/Hybrid_Memory_Cube
https://www.computerworld.com/article/3082658/intel-lets-slip-roadmap-for-optane-ssds-with-1000x-performance.html
https://www.nextplatform.com/2019/10/23/a-look-inside-the-groq-approach-to-ai-inference/
https://www.anandtech.com/show/15978/micron-spills-on-gddr6x-pam4-signaling-for-higher-rates-coming-to-nvidias-rtx-3090