Nvidia is bending amd and Intel over right now
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Intel Now Operating On A Net Loss, $1.61 Billion In The Red
Nvidia is bending amd and Intel over right now
Paid out to share holders duh; that's the big downside of public companies - in years where they make bank, they pay everything out, and then in bad years they have no money to fund themselves
I work in a publicly owned company as well and our management are always looking for excuses to not pay the shareholders, so that we have cash on hand for a rainy day, but it's tough, investors always expect the highest returns and it's difficult to deliver high returns and retain cash in the bank
Last edited:
Count me in. I'm down with the funky ****.
Last edited:
Soldato
50 years? No forget it.
Let say 5 to 10 years? I think it will look really very tough for both AMD and Intel, I doubt both will survive when they dont have enough money for future R&D and launch future products based on TSMC N2 and beyond process nodes. AMD already burnt though billions on cancelled K12, Navi 41, Navi 42 and Navi 4C projects and Intel burnt through billions on Intel Foundry, cancelled Rialto Bridge, Beast Lake, Royal Core projects which Jim Keller was worked on it and also Intel dissolved Advanced Architecture Development Group.
Apple and Nvidia are the only 2 trillion dollar companies that will succeed and enjoy milked billions of profits weekly for a decade or two. It will be interesting to see what Apple will do with TSMC N2 first in the world in 2025 with next generation iPhone 17 with A19 CPUs and Mac with M5 CPUs. Apple already developed future CPUs based on TSMC A14 process due in 2026 or 2027.
Firm Estimates a 2nm Chip Now Costs $725 Million to Design
At least, for a large chip.
N2 cost $725M in total development to design each chip, PLUS $30,000 per wafer.
Imec Reveals Sub-1nm Transistor Roadmap, 3D-Stacked CMOS 2.0 Plans
Chips will go 3D as they shrink below 1nm.
I calculated what future nodes total development cost to design each chip could look like by add 25% each time.
2024 N2 $725M
2026 A14 $906M
2028 A10 $1.132BN
2030 A7 $1.415BN
2032 A5 $1.768BN
2034 A3 $2.210BN
2036 A2 $2.762BN
I cant imagine both AMD and Intel survive in 2030 each cannot afford mindblow $9.9BN for 7 products Epyc/Xeon CPU, Ryzen/Core CPU, Threadripper/Xeon CPU, Datacenter CPU/GPU, High end Radeon/Xe GPU, Mainstream Radeon/Xe GPU and Mobile CPU/APUs based on TSMC A7 node plus maybe $60,000 per wafer. I think both will not be exist in 2036, they cannot afford over $20BN need to develop and launch 7 or more products based on TSMC A2 node plus $90,000 or $120,000 per wafer.
Last edited:
They are, but aren't they are starting to see losses too? Obviously it's swings and round abouts, and I can't imagine enough to go down the toilet though.
Last edited:
Looks at just Intels property portfolio… 2016-19 bought real estate like no one’s business and kept buying more since while closing working plants. I seem to rember one year Intel spent 16 billion.
Pat is just a wiley old fox promoting the need for another couple of rounds of tax payer funding.
Pat is just a wiley old fox promoting the need for another couple of rounds of tax payer funding.
Nvidia is an add in board maker and without anything to add to they would be finished. Nvidia is an Intel and AMD customer.
good job they got stopped from buying ARM
good job they got stopped from buying ARM
The UK should have done the same.
Upvoted for T2 quote.
Soldato
Maybe peeps on the forum now are too young to remember, but plenty of multi national multi billion £ companies have gone down the pan. In fact the two i am going to mention are Kodak and Rank Xerox.
Kodak had losses of $15 Billion, Rank Xerox had losses of $38 Billion. Both happened many years ago now, so in real terms were actually much much bigger than Intel is now.
The thing is though, both went down the pan for the same reason that Intel is heading in that direction. Bad leadership, a failure to embrace tech changes, management failures and board irresponsibility.
I would be very surprised if Intel was even still a global company in 5 years time, let alone 10 years time.
Kodak had losses of $15 Billion, Rank Xerox had losses of $38 Billion. Both happened many years ago now, so in real terms were actually much much bigger than Intel is now.
The thing is though, both went down the pan for the same reason that Intel is heading in that direction. Bad leadership, a failure to embrace tech changes, management failures and board irresponsibility.
I would be very surprised if Intel was even still a global company in 5 years time, let alone 10 years time.
Intel are too big to fail imo. They are backed by the US government,and certain sand people. Intel will never go under. They might liquitade and change name but they will never die. Just my opinion of course.
If Intel can’t move on from its current CPU architectures then it will slowly shrink as AMD grows. Fact is the world needs CPUs and graphics and only Intel can fill that demand.
I think a leaner more competitive Intel will emerge at some point over the next decade.
I think a leaner more competitive Intel will emerge at some point over the next decade.
That’s where Arc comes in.
Critical thing is Intel has national security implications for the US - the US government can't let them fail and that is probably part of the reason they are in the position they are - people don't tend to care when it is other people's money they are spending.
Last edited:
Permabanned
87% of nvidia's last quarter revenue (30billion usd) was from datacentre. the previous quarter was the same (just 15% less total revenues)
Mr J.H. did not spend much time talking about aib's at computex.
I believe the law of large numbers will kick in. AI spending on hopper/and perhaps blackwell will peak, then fizzle. but that's crystal ball stuff. AI has potential to keep programmers busy...well into 'infinity and beyond'. as to how much and how often nv's DC platforms will require updating is questionable.
Intel are also yak yakking about gaudi and what not. in a sense they're pipped to the post on two fronts. NV is expensive but also delivers the perf. in servers and amd are more affordable and energy/performance efficient. in servers.
as much as I like intel for the little boy's pc i have here. i am a niche market
whereas before of course...well we all know. intel/etal will be pushing their newest laptops very soon.
Last edited:
87% of nvidia's last quarter revenue (30billion usd) was from datacentre. the previous quarter was the same (just 15% less total revenues)
Mr J.H. did not spend much time talking about aib's at computex.
I believe the law of large numbers will kick in. AI spending on hopper/and perhaps blackwell will peak, then fizzle. but that's crystal ball stuff. AI has potential to keep programmers busy...well into 'infinity and beyond'. as to how much and how often nv's DC platforms will require updating is questionable.
Intel are also yak yakking about gaudi and what not. in a sense they're pipped to the post on two fronts. NV is expensive but also delivers the perf. in servers and amd are more affordable and energy/performance efficient. in servers.
as much as I like intel for the little boy's pc i have here. i am a niche market
Without Intel and AMD providing Nvidia an X86 base, Nvidia has very limited access to those revenue streams. That’s possibly why Intels big strategy is to get behind the develop dedicated accelerators to stabilise markets like AI.
Well that's exactly what I said. Intel is a literal national security risk (not that they care) Even if they were to literally collapse, they would come back under another name. Or even the same name. Look at Rangers F.C. in Scotland. They are registered as Sevco, but still appear as Rangers F.C. I don't pretend to understand the shadyness of the business world, but you can't deny some serious crap goes on behind closed doors. Money laundering is my guess.
Soldato
Intel Likely To Sell Altera, Freeze The Construction Of Its $30 Billion German Factory, But Retain The Chip Manufacturing Business, As Per The Latest Tidbits Ahead Of A Critical Board Meeting
For the past few quarters, after each significant stock price dump, Intel investors ask the familiar question: is this the bottom? Yet, each time, the answer has been in the negative.
wccftech.com
Intel likely plan to sell Altera, freeze $30BN German fab contruction and keep Intel Foundry business.