4 ms·

The issue with TSMC seems to be there is demand for their leading node but not enough demand for their 7nm. Also the cost of build the next bleeding edge node i

The issue with TSMC seems to be there is demand for their leading node but not enough demand for their 7nm. Also the cost of build the next bleeding edge node is getting higher and higher and at some point even though Physics of building lets say 0.5 node feasible but the economics wont be there. That means their rivals can catch up with them in 5 year span.. while they remain stagnant.

zarzavat 4y ago
TSMC only has one rival: Samsung. Intel is not really a rival, no matter how good their process is, they cannot win big contracts with Apple, Nvidia, AMD.
- paulmd 4y ago
  > they cannot win big contracts with ... Nvidia https://www.tomshardware.com/news/intel-ifs-lands-3nm-to-make-3nm-chips-for-major-customer https://www.tomshardware.com/news/intel-ifs-lands-3nm-to-mak... > "I am very happy that we were able to add a leading cloud, edge, and datacenter solutions provider as a leading-edge customer for Intel 3," said Pat Gelsinger, Intel CEO. "Including prior customers such as MediaTek, we now have a lifetime deal value of greater than $4 billion for IFS." Hmm, I wonder what "leading cloud, edge, and datacenter solutions provider" that would be? /giant thinking emoji What's NVIDIA got coming in 2024? Any tegras, or Mellanox, or heck that's also their next consumer graphics window. If it's not NVIDIA that's a pretty short list. Maybe Tesla? Broadcom? Are those companies working at leading-edge? Anyway: > Intel's IFS has been steadily gaining momentum. The foundry unit has already signed Qualcomm and Amazon Web Services (AWS) as initial customers and won a contract from the U.S. Department of Defense. In addition, Intel says it is actively working with many of the largest fabless chip designers on specific engagements, which is quite an achievement for a contract chipmaker that has been in the business for less than two years. > "We also have an active pipeline of engagements with seven out of the 10 largest foundry customers coupled with consistent pipeline growth to include 43 potential customers and ecosystem partner test chips," said Gelsinger. "Additionally, we continue to make progress on Intel 18A, and have already shared the engineering release of PDK 0.5 (process design kit) with our lead customers and expect to have the final production release in the next few weeks." Regardless of what you think, this time is different. It's do-or-die for Intel, their design teams can't be relied upon to keep the foundries busy anymore, and the design teams can't rely on the foundries to execute flawlessly every node either. It's in everyone's best interests to disambiguate these, you can't rely on P(FAB_SUCCESS) * P(DESIGN_SUCCESS) when every other fabless company or foundry has a success probability of merely P(DESIGN_SUCCESS) or P(FAB_SUCCESS) individually. You can't stake the fabs on the design groups executing well and keeping them busy, and you can't stake the design teams on the fabs not fucking up and bottlenecking your release pipeline. The fabs can still be a huge moneymaker (see: TSMC) but it's time to go arms-length between the business groups. And yeah it's gonna be a huge shift for everyone involved - that's part of why Sapphire Rapids stalled out for 12 steppings or whatever. Intel 4 is actually ready to go and the design teams can't get their shit together in a world where they can't go down the hall and have the fab team fix their shit, that's why you're seeing some of these products spin out to ridiculous numbers of steppings. So who's gonna fill all that fab capacity that's sitting there idle? Not Intel products, that's for sure. And that's the call Gelsinger is making, these are now two arms-length business groups who use a lot of each others' services, but they do business with other companies for those services too. Samsung does the same model FYI, they sell custom foundry to their own competitors (famously, Samsung vs TSMC SOCs in iphones). So does LG and many others. For example sometimes Apple products or other brands will end up with better displays than the actual LG or Samsung products - cause they're distinct business groups and LG Electronics just happens to buy a lot of LG Display panels.

zarzavat 4y ago

TSMC only has one rival: Samsung. Intel is not really a rival, no matter how good their process is, they cannot win big contracts with Apple, Nvidia, AMD.

paulmd 4y ago
> they cannot win big contracts with ... Nvidia https://www.tomshardware.com/news/intel-ifs-lands-3nm-to-make-3nm-chips-for-major-customer https://www.tomshardware.com/news/intel-ifs-lands-3nm-to-mak... > "I am very happy that we were able to add a leading cloud, edge, and datacenter solutions provider as a leading-edge customer for Intel 3," said Pat Gelsinger, Intel CEO. "Including prior customers such as MediaTek, we now have a lifetime deal value of greater than $4 billion for IFS." Hmm, I wonder what "leading cloud, edge, and datacenter solutions provider" that would be? /giant thinking emoji What's NVIDIA got coming in 2024? Any tegras, or Mellanox, or heck that's also their next consumer graphics window. If it's not NVIDIA that's a pretty short list. Maybe Tesla? Broadcom? Are those companies working at leading-edge? Anyway: > Intel's IFS has been steadily gaining momentum. The foundry unit has already signed Qualcomm and Amazon Web Services (AWS) as initial customers and won a contract from the U.S. Department of Defense. In addition, Intel says it is actively working with many of the largest fabless chip designers on specific engagements, which is quite an achievement for a contract chipmaker that has been in the business for less than two years. > "We also have an active pipeline of engagements with seven out of the 10 largest foundry customers coupled with consistent pipeline growth to include 43 potential customers and ecosystem partner test chips," said Gelsinger. "Additionally, we continue to make progress on Intel 18A, and have already shared the engineering release of PDK 0.5 (process design kit) with our lead customers and expect to have the final production release in the next few weeks." Regardless of what you think, this time is different. It's do-or-die for Intel, their design teams can't be relied upon to keep the foundries busy anymore, and the design teams can't rely on the foundries to execute flawlessly every node either. It's in everyone's best interests to disambiguate these, you can't rely on P(FAB_SUCCESS) * P(DESIGN_SUCCESS) when every other fabless company or foundry has a success probability of merely P(DESIGN_SUCCESS) or P(FAB_SUCCESS) individually. You can't stake the fabs on the design groups executing well and keeping them busy, and you can't stake the design teams on the fabs not fucking up and bottlenecking your release pipeline. The fabs can still be a huge moneymaker (see: TSMC) but it's time to go arms-length between the business groups. And yeah it's gonna be a huge shift for everyone involved - that's part of why Sapphire Rapids stalled out for 12 steppings or whatever. Intel 4 is actually ready to go and the design teams can't get their shit together in a world where they can't go down the hall and have the fab team fix their shit, that's why you're seeing some of these products spin out to ridiculous numbers of steppings. So who's gonna fill all that fab capacity that's sitting there idle? Not Intel products, that's for sure. And that's the call Gelsinger is making, these are now two arms-length business groups who use a lot of each others' services, but they do business with other companies for those services too. Samsung does the same model FYI, they sell custom foundry to their own competitors (famously, Samsung vs TSMC SOCs in iphones). So does LG and many others. For example sometimes Apple products or other brands will end up with better displays than the actual LG or Samsung products - cause they're distinct business groups and LG Electronics just happens to buy a lot of LG Display panels.