You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
According to Pat Gelsinger:
- Intel 7 in Israel.
- Intel 4 in Ireland.
- Intel 3 and 20A in Arizona.
- Intel 18A in Europe (Magdeburg and Ireland).
- "Intel Next“ in Ohio.
Using chiplets and outsourcing die (TSMC) will change the way Intel manufacturers chips, absolutely.
Where was that announced? It also seems really weird for the Israel fabs to stay only on 7 given the new shell they've been building there. It seems weird for the 18A and 4 to be Ireland (rather than 4/3). Same deal with AZ too. One things for sure even without foundry disag is indeed causing the wider node lifetimes with shallower ramps like their CFO stated a few months ago. Finally it also seems weird for the furtherest out fab (especially when earlier Pat was talking about how it would get 18A or beyond) to be getting 18A while Ohio gets the furthest out node. Although for all I know he misspoke or something.
According to Pat Gelsinger:
- Intel 7 in Israel.
- Intel 4 in Ireland.
- Intel 3 and 20A in Arizona.
- Intel 18A in Europe (Magdeburg and Ireland).
- "Intel Next“ in Ohio.
Using chiplets and outsourcing die (TSMC) will change the way Intel manufacturers chips, absolutely.
It seems Intel's current plan is to have only one fab located in one county in one continent to handle one particular node, except Intel 18A in Germany and Ireland (both in Europe).
It doesn't seem to match the supply chain resilience that Intel CEO Pat Gelsinger is promoting.
Where was that announced? It also seems really weird for the Israel fabs to stay only on 7 given the new shell they've been building there. It seems weird for the 18A and 4 to be Ireland (rather than 4/3). Same deal with AZ too. One things for sure even without foundry disag is indeed causing the wider node lifetimes with shallower ramps like their CFO stated a few months ago. Finally it also seems weird for the furtherest out fab (especially when earlier Pat was talking about how it would get 18A or beyond) to be getting 18A while Ohio gets the furthest out node. Although for all I know he misspoke or something.
If this process node location list is correct, it may indicate Intel is pushing Ohio fabs opening date further down to deal with the financial and market demand issues. Even with the government subsidies from various countries, Intel still needs to invest a huge amount of capital to bring those fabs online (assume there are enough customer demands). It's a very daunting task.
Some pluses, as I see it:
-Investment: Intel is clearly investing again, with many new nodes to put in many new fabs, a genuine huge positive development.
-Execution: I haven’t heard any issues building Fab 52/62 in AZ. The hard work starts soon though.
Minuses:
-No Intel 4 products, just a date in December. Execution risk is high.
-No Foundry customer commitments. Perhaps related to the above. Execution risk is high.
Some pluses, as I see it:
-Investment: Intel is clearly investing again, with many new nodes to put in many new fabs, a genuine huge positive development.
-Execution: I haven’t heard any issues building Fab 52/62 in AZ. The hard work starts soon though.
Minuses:
-No Intel 4 products, just a date in December. Execution risk is high.
-No Foundry customer commitments. Perhaps related to the above. Execution risk is high.
I don't think execution risk is high if they can give out an official date. And there's down payment for 18A capacity in case you haven't read the news. So that isn't true
Perhaps compared to TSMC that's true, but no other company other than the memory vendors are comparable in scale. Intel manufacturing impresses me by mostly by successfully operating fabs over many years in Ireland and Israel, in addition to two sites in the US. I don't think TSMC has proven they can be successful in manufacturing outside of Taiwan yet.
Perhaps compared to TSMC that's true, but no other company other than the memory vendors are comparable in scale. Intel manufacturing impresses me by mostly by successfully operating fabs over many years in Ireland and Israel, in addition to two sites in the US. I don't think TSMC has proven they can be successful in manufacturing outside of Taiwan yet.
I think that's the main reason that people should not seek the spin-off of IFS from Intel. Intel is IDM, which will expand its capacity and extend it over many years.
I don't think execution risk is high if they can give out an official date. And there's down payment for 18A capacity in case you haven't read the news. So that isn't true
