News Intel will keep using TSMC's services even when 18A is ramped up: 'It is a good supplier

[dalek1234]

No they are not, I just wanted to jump in on the bashing.

I know, right. Bashing Intel is so much fun these days.

 

[phead128]

I know, right. Bashing Intel is so much fun these days.

These multi-billion dollar corporations could hardly care about you. Never understood why there are fanboys of MNC who would sell you in a heartbeat.

 

[TerryLaze]

Not my fault if you don't like to hear the truth.

The truth is that nvidia has maxed out what amount of money they can make from tsmc alone.

That is the truth, anything beyond that is speculation.

My speculation is that nvidia likes money just like any other company so they will figure out something to make more product than they are now and that means making stuff at intel.

You don’t lose 7 billion in 1 year and say oops tax write off.

Which year?! Or what loss are you talking about.
7 bil isn't even close to what they stated.

 

[acadia11]

The truth is that nvidia has maxed out what amount of money they can make from tsmc alone.

That is the truth, anything beyond that is speculation.

My speculation is that nvidia likes money just like any other company so they will figure out something to make more product than they are now and that means making stuff at intel.

Which year?! Or what loss are you talking about.
7 bil isn't even close to what they stated.

2023 they lost 7 billion in their foundry business , 5.2 billion in 2022 and I shutter to think what the 2024 numbers will finally be. But Intels foundry has been taking a hit since the end of the teens and they had planned out sourcing TSMC. They expect to make it profitable again in what 2 years banking on the success of 18A, and being able to help meet global demand of leading edge production capacity. Which makes this article the more concerning it’s doesn’t forebode well to their strategy. You can’t meet your own needs how are you going to meet someone else’s … granted as pointed out there may be cases where they should outsource for some trailing edge use cases … but leading edge … and they still need to go to TSMC how can they be taken seriously as foundry partner to other chip manufacturers?

2024 $13 billion loss in foundry … the fact is Intel went from being the leading edge provider a little more than decade ago … to playing catch up , and they have never been setup as a serious fab capacity supplier to other chip makers. With the explosion of fab capacity demand due to AI and also with ARM based chips and potentially RISC-V thusly growing demand by fabless chip makers it makes no sense for Intel to be where it is … other than its own ineptitude and stagnation. But they have a lot on their side to turn it around.. I liken it to Apple at their crossroads in the early 2000s.

 

[JayNor]

it makes no sense for Intel to be where it is … other than its own ineptitude and stagnation.

Intel will have the two 18a fabs available for the first time in 2024, or perhaps only one of them. They have Intel-4 and Intel-3 up and running, and their Foveros 3d, and have big construction projects in Ohio under way. This was all planned. Looks like they have a leading technology with the bspd, and apparently have popular Xeon-6 products that are ramping up. They have big customers MSFT, AWS and Ericsson signed up for 18a, as well as some government contracts. Those have been in test wafer evaluation up to this point. Product designs start taping in this year.

 

[acadia11]

Intel will have the two 18a fabs available for the first time in 2024, or perhaps only one of them. They have Intel-4 and Intel-3 up and running, and their Foveros 3d, and have big construction projects in Ohio under way. This was all planned. Looks like they have a leading technology with the bspd, and apparently have popular Xeon-6 products that are ramping up. They have big customers MSFT, AWS and Ericsson signed up for 18a, as well as some government contracts. Those have been in test wafer evaluation up to this point. Product designs start taping in this year.

Where they are … not where they may be … no one is a fortune teller but they have been inept or at least poorly executing/planning for a decade or so … and yes 18A leaves them an opportunity … and remember 20A was supposed to be their comeback to ensure 18A tech was better vetted but was canceled so 18A has to succeed. My point is they had the resources and the market forces in place and there is no reason for their current situation but their own self inflicted wounds. Xeon-6 as far as that as a product is concerned AMD has been eating their lunch in the data center space and it seems it will continue to be the case … as they have more performant solutions in EPYC and better capabilities in the realm of AI data center computing needs. I’ll leave the product side of Intel vs the foundry side discussion as separate endeavors which is also sort of the point. Intel needs to separate these entities to maximize the market opportunity before them.

 

[dalek1234]

These multi-billion dollar corporations could hardly care about you. Never understood why there are fanboys of MNC who would sell you in a heartbeat.

None of what you said makes any sense, as a reply to my comment. Are you on crack? Maybe an Intel bot; that would make sense then.