The GlobalFoundries and AMD Story

A marriage made in hell.

Bitcoin Address - 1HuL9vN6Sgk4LqAS1AS6GexJoKNgoXFLEX
Ethereum Address - 0xB3535135b69EeE166fEc5021De725502911D9fd2

♥ Buy PC Parts from Amazon below.

►Get seriously cheap games at the links below!

--- Video Links Below ---

[Taliyon]

I've learned so much on this channel. The nature of the content is traditionally distorted. Normally, I'd not have the interest to delve into the intricacies. But, the research here shines. I've made more sense of the literal bullshit that spews out of this industry, it's fantastic. I fully appreciate the consistency of presentation. These history lessons really help portray the nature of current events. This is how you draw interest where it would otherwise die. Couldn't ask more of a media channel. The recipe just works... Thank you.

[arkz]

As I was reading the news of GloFo dropping their 7nm process I thought: adoredTV sure is going to make a video about the history of this company. But wait, given his style with lengthy and comprehensive videos it will take at least a week for research and whatnot... 2 days later: new video notification, yeah!.

[SebastianE]

Honestly just ridiculous how you have stagnated at 65k subs for so long now... I just don't understand... You have so much to give to the tech community, but I guess most people just can't handle the truth... If it means anything to you adored, you are by far my favorite tech youtuber and I appreciate every ounce of work you put into these videos.

[techforge-Nate]

Given the history so far, I think Global Floundaries would have been a better name. Best we can manage is a "They tried" award.

[KonkelVonk]

I'm pretending to understand this guy, only here for the awesone accent.

[rdleh]

Jim you're just getting better and better, you are now my most reliable source of tech news. i'm sticking with you all the way bro, greetings from Portugal.

[heckyes]

He's got that kind of voice where just listening to it makes you feel smarter.

[martigrey]

You forgot about Samsung. They will deliver on 7nm as well as 5 and 3nm because they have to. They can't let TSMC win.

[Avvoider]

The only tech channel I keep coming back to. Very interesting content.

[jamil]

"Alright guys, how's it going?"
Aw, yes.

[saintkamus]

This is how Moore’s law ends; not with a bang, but with a whimper. Investors will not fund new fabs, because the cost is too high and the performance gains too low. (Well, it’s how miniaturization ends, Moore’s law has been dead for years)

[sonsofmatriarchy]

Global Foundries working with Samsung to provide 2-4 billion to produce 7mn process seems like a better idea imo.

[veilmontTV]

Your channel is growing far slower than it should. You do incredible videos

[Najvalsa]

GF's finally got future projections on making good money for the first time, and with AMD not ordering anything from them for 7nm since it's not ready, I can understand them not wanting to chase/bleed for, the bleeding edge.

All of their fabs are busy churning, and they're the leader in FD-SOI technology, with an increasing number of clients for it. So increasing, that they're expanding their fab in China. All of which puts them at a top position, in a growing market.

It is disappointing, and worrying, to see them out of the FinFET race. But, it's also understandable. Maybe when EUV gets sorted out, and with a bit of money in their pockets, they'll join the race again.

[unclerubo]

It's worrying to see that the manufacturing of chips is now in the hands of even fewer players. This will of course increase costs for everyone. Bad news indeed.

[snetmotnosrorb]

I think this is a minor disaster. 7LP has no relation to GloFo's earlier work and mishaps. It was IBM's work that GloFo took over when IBM handed over their Microelectronics business, including employees. You should have mentioned that, Jim. You could also explain the relation IBM have with both AMD and GloFo.

7LP is/was a really good process. Everything was on track and looking fine.

They have spent so much time and money on a process that is essentially finished, with good opportunities for improvements and further development, an obvious roadmap and future for GloFo. And now they just axe it. I really wonder what IBM thinks of that.

If there was a time when GloFo should scale down and change direction it would be immediately after the 14nm fiasco, not now when they have invested so much and have a really good offer. I think GloFo is making a huge mistake.

If capacity really is the problem that hinders them to both reach a positive economy of scale and get customers, then why in heavens name did they not take that into account? The whole point of GloFo, their core business, is to be "bleeding edge" in semiconductor manufacturing as well as offer high capacity. It's been known for quite a few years now that new nodes don't come easy anymore.

I'm not convinced that TSMC has the capacity to cater all customers, nor that AMD is good or big enough to be a priority for TSMC. GloFo would actually be both needed and good this time around, and not only for AMD.

[daviddupoise]

TSMC would do well strategically to broaden its manufacturing facilities out of earthquake prone locations. Like here in northern Virginia for example. We have the workforce, we have Micron facility and others.

[ajc-thei]

Great work, as always.

Now, for the AMD statement saying 12/14nm production will stay with GF, likely, this refers to the current Zen/Zen+/Polaris 10&20/Vega 10 offerings. At least on Zen and Zen+, they have promised corporations to have those chips available until 2020 or so, if I recall correctly, meaning drop in replacements. I wonder, with the deployments in Q3 and Q4 of this year being based on EPYC produced at GF, plus the pro series chips for commercial client systems (which don't need absurdly powerful chips, but are an upgrade in many ways to i5s and i3s currently deployed in commercial settings, not to mention the security vulnerabilities that currently have marred Intel's reputation in security), which would likely satisfy their needs and have been tested since introduction for deployment, how much AMD will fulfill their requirements under the WSA, thereby pushing some of the alternate production under the excess capacity provisions, which don't take the brunt of the payouts otherwise? Just an interesting open question, not challenging the analysis here at all.

[moldoveanu]

Thanks for your time doing research for this video, much appreciated, also this CPU market soap opera is always entertaining, thanks for that

[cabbytabby]

you've been on absolute fire lately mate, mint quality stuff. Much appreciated!