• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Qualcomm bid for Intel

That seems likely, plus does Qualcomm want to run a foundry? That would be another source of government opposition. The government wants those fabs. Actually make that governments. The EU also wants semiconductor manufacturing spread out more.

The other problem is a deal like that would take a long time to close and Intel is "in flux" right now. They're going to keep working on 18A and new products and looking for foundry customers. Depending on how all that goes the value of the company could change dramatically before they could close out an acquisition deal. That could make coming to an agreement difficult unless Qualcomm is prepared to offer a large premium over Intel's current stock price.
 
Now it looks like Qualcomm is laying off to. Your link takes you to MSN but now jumps back a fore between MSN and WSJ.
 
Qualcomm isn't going to be the only player....
Anti trust is also likely to give the green light to any major US player, save perhaps AMD who I doubt is really thinking about it. I mean they don't offer anything to AMD they don't already do better... but it would be a funny twist of fate. The Gov at this point just wants a US company fabricating chips... Intel has the hardware, but perhaps they don't and never will have the proper management. I believe the Gov would welcome a well run US player assuming the physical hardware... and coming in and being ruthless with the staff, cut it to the bone and fix it.

I still say expect a joint and maybe even 3 way hostile take over before the end of the year. Qualcomm/Microsoft... and maybe a third player that actually wants to run a fab. Or maybe they just split the assets and let Pat keep the Intel name and run run the fab business. I could see a deal where QCom and MS offer a ton of cash to shareholders after buying up the % they can legally buy first... with the deal involving shareholders retaining shares in Intel Fab. The US gov would be more then happy if Intel had nothing else to distract them from making the fabs run on time.
 
OK. But either way, Intel looks like one sick puppy, With a managenent that probably clueless and scared out of its gourd.
 
Why would us defense care if the foundries are owned by intel or qualcomm?
For the same reason that they can’t and won’t use the TSMC foundry in Arizona when it comes online because of the possibility of a Chinese connection.
 
Isnt qualcomm a us company?
Sort of, they are a US company because they have their headquarters in the States, but it’s a figurehead of an office. The CEO and all leadership work is done out of their “Main Office” in Hong Kong.

Qualcomm has been denied a number of US government contracts over the past decade because of an abundance of CCP oversight on their operations.
 
Doubt it for many reasons. Anti-trust would be #1, but then there's the fact that Intel is about to release new chips that would explicitly denounce Apple and Qualcomm's ARM based chips. Those Snapdragon X laptops aren't exactly selling either. Qualcomm isn't worth enough to be able to do a buyout. Sounds like Qualcomm isn't getting their way and they want to buy out the competition before things get really bad. They did approach Intel, not the other way around. Especially when you consider that Qualcomm is about to lose their exclusivity license with Microsoft. Intel has more than double the employees and several times the assets. Intel is criminally undervalued given their assets and Qualcomm knows it. Stock market valuations are stupid. They're hoping that Intel's top brass is greedy enough to take the buyout deal, which to be fair isn't a 0% chance.
 
Sort of, they are a US company because they have their headquarters in the States, but it’s a figurehead of an office. The CEO and all leadership work is done out of their “Main Office” in Hong Kong.

Qualcomm has been denied a number of US government contracts over the past decade because of an abundance of CCP oversight on their operations.
Err... sorry, are we talking about the same company? https://en.m.wikipedia.org/wiki/Qualcomm

The founders and the current ceo dont look like from they're from hong kong either.
 
I may be confusing them with Broadcom….
That makes more sense.

Yep I’m thinking Broadcom, I’ll just see myself out now.
1726898410430.gif
 
Intel is criminally undervalued given their assets and Qualcomm knows it. Stock market valuations are stupid.

Bear in mind, stock market valuations are being done on forward projections, not current assets. Nvidia is being valued right now orders of magnitude higher than the assets suggest because people are projecting that they’re the future of computing, which massively increases their addressable market, and they want to get in front of it. It’s likely they’re overextended, but Nvidia keeps delivering insane growth, which validates that thinking.

Compare that with Intel. What does the market have to be excited about? They’re behind their competitors on all metrics with no pathway to return to market leadership. The CPU space where they traditionally have the lead has been eroding since Ryzen first entered the market. They have struggled to get their newer fab processes online and have not provided any proof they’re positioned to compete with TSMC anytime soon. As a matter of fact, they’re now a TSMC customer despite having in-house fabrication. They’ve launched GPUs now to try to get into Nvidia’s datacenter space now that they’re taking over from Intel’s traditional dominance there with CPUs, but Nvidia has an embarrassing lead that seems almost insurmountable even by GPU veteran AMD. Now we see they had manufacturing defects with the 13000 series. Basically, Intel hasn’t provided anyone a reason to project stronger growth into the future because they’re losing on every metric that matters and have not shown any prospects for breaking that pattern.

That said, this depresses the valuation, and Intel is far from irredeemable under the right management, so that can make them an attractive takeover target, hence what we’re seeing here.
 
Bear in mind, stock market valuations are being done on forward projections, not current assets. Nvidia is being valued right now orders of magnitude higher than the assets suggest because people are projecting that they’re the future of computing, which massively increases their addressable market, and they want to get in front of it. It’s likely they’re overextended, but Nvidia keeps delivering insane growth, which validates that thinking.
Nvidia's been lucky that there's so many stupid people in the world. First it was crypto and the NFT's which failed exactly as everyone thought it would, and then there's "AI" because ChatGPT makes for a very convincing person. We're just waiting for the market to realize that there's nobody interested in buying a product with the label "AI" stamped onto it and then Nvidia's stock valuations are going to crumble fast.
Compare that with Intel. What does the market have to be excited about?
Intel's about release Lion Cove and Lunar Lake chips that will likely disrupt the market. Why you think Qualcomm is "approaching" Intel for a takeover? Everyone knows there's no way that's going to happen. Intel's new GPU is also around the corner. What does Qualcomm have to look forward to? To watch as every cell phone manufacturer slowly makes their own SoC to displace Qualcomm? To watch as their $1.4 billion acquisition of Nuvia becomes a waste of money as they can't scratch the Windows laptop market?
They’re behind their competitors on all metrics with no pathway to return to market leadership.
Your crazy if you think Intel has left that market leadership. Again, Lunar Lake is like 2 days away and Qualcomm pulls this stunt because they know that Lion Cove and Lunar Lake are going to put Intel ahead in terms of power and performance.
The CPU space where they traditionally have the lead has been eroding since Ryzen first entered the market.
Which is bad for Intel but good for the industry since Intel had total market control. Now Intel has like what 60% market control?
They have struggled to get their newer fab processes online and have not provided any proof they’re positioned to compete with TSMC anytime soon. As a matter of fact, they’re now a TSMC customer despite having in-house fabrication.
It's just a matter of time before Intel gets their manufacturing up and running. Them going TSMC was a good decision because they just need more advanced chips now, rather than wait for their facilities to catch up.
They’ve launched GPUs now to try to get into Nvidia’s datacenter space now that they’re taking over from Intel’s traditional dominance there with CPUs, but Nvidia has an embarrassing lead that seems almost insurmountable even by GPU veteran AMD. Now we see they had manufacturing defects with the 13000 series. Basically, Intel hasn’t provided anyone a reason to project stronger growth into the future because they’re losing on every metric that matters and have not shown any prospects for breaking that pattern.
What company hasn't? What growth does Qualcomm have as Huawei is dominating China's market? When slowly Samsung replaces them with their own Exynos SoC's? Even Intel's GPU have potential, which is more than can be said about Qualcomm's Adreno GPU's.
Yeah about that antitrust issue, there is currently major lobbying going on to replace Lina Khan as the head of the FTC with someone more merger friendly.

https://www.bloomberg.com/news/arti...nt-ftc-s-lina-khan-sec-s-gary-gensler-removed

Once that obstacle is removed, I don't think antitrust will be an issue.
That is scary and another reason we need to abolish donors for the people we vote for.
 
It’s kind of nice working at a large American semiconductor company that nobody ever thinks of. We just sick back, take it easy, and let the wafers sell themselves. 😎
 
A Qualcomm - Intel deal will never happen. Just like when AMD was worth pennies, no one touched it. Why? Because an acquisition would trigger a renegotiation of the X86 licensing - and AMD would have the upper hand by far this time.
 
Last edited:
A Qualcomm - Intel deal will never happen. Just like when AMD was worth pennies, no one touched it. Why? Because an acquisition would trigger a renegotiation of the X86 licensing - and AMD would have the upper hand by far this time.
I'm not sure that is the way the agreements back then were written. Pretty sure the life time cross license survives either company.
When national semiconductor acquired Cyrix they retained their X86 license. (which pissed Intel off and they canceled all their National contracts... until they sold them off, combine that with Cyrix deals with IBM foundries National lost their shirts) When they then sold them off to Via... again the patent cross license remained in place. Even now Via owns 15% of Zhaoxin which makes it 100% perfectly legal for them to produce x86 chips.

I highly doubt AMDs deal is worse then then deal Cyrix signed. Intel also no doubt put the same protections in for themselves... no doubt laughing about doing so. I'm sure if we had a time machine and went back and let the Intel contract team know those clauses were going to come in handy for the companies that will aquire Intels patents in '25 they would have been a bit dumbfounded. lol :)
 
A Qualcomm - Intel deal will never happen. Just like when AMD was worth pennies, no one touched it.
Didn't Intel grant the x86 license to AMD? So why would Intel not simply extend the license to the acquiring company?
Why? Because an acquisition would trigger a renegotiation of the X86 licensing - and AMD would have the upper hand by far this time.
How? See my question above,. How does AMD have negotiation rights when Intel granted the license to AMD? What is missing here?
 
An acquisition triggers a renegotiation because an acquisition means we are no longer dealing with the original company. New company, new owner - new deal. This is obvious.

Note is is a CROSS-LICENSING agreement. Does the company acquiring Intel really want to say goodbye to X64 extensions (among other shared IP)?
 
  • Like
Reactions: ChadD
like this
I'm not sure that is the way the agreements back then were written. Pretty sure the life time cross license survives either company.
https://www.pcworld.com/article/246...s-no-sense-because-of-amds-little-secret.html

it is the 2009 agreement that is subject of who control Intel-AMD:
https://www.sec.gov/Archives/edgar/data/2488/000119312509236705/dex102.htm

Capture Period” shall mean any time on or prior to the earlier of (a) the fifth anniversary of the Effective Date and (b) a Change of Control of either Party.​

Change of control seem an important element:

Termination Upon Change of Control
. Subject to the terms of, and as further set forth in, Sections 5.2(d) and 5.2(e), this Agreement shall automatically terminate as a whole upon the consummation of a Change of Control of either Party.


An acquisition triggers a renegotiation because an acquisition means we are no longer dealing with the original company. This is obvious.

Not really sure of that, I feel there a reason it is mentioned so much and so clearly in the cross license agreement, because the default is when you buy a company you buy all its contractual obligation (asset and debt) of the sort I think, imagine Disney buying Fox if that mean that their TV-Ip deals are all cancelled because of a change of ownership and they need to redo a deal with India cricket league and FCC regional cable-radio use and all of it, have to redo The Simpsons and other IP deal with their creators... If a bank buy the bank your have a morgage with I do not think it mean a renegotiation must occur.
 
Last edited:
Intel is criminally undervalued given their assets and Qualcomm knows it. Stock market valuations are stupid. They're hoping that Intel's top brass is greedy enough to take the buyout deal, which to be fair isn't a 0% chance.
How much leverage have you used to massively invest in Intel? The chips company that missed the mobile and gpu wave, Microsoft did survive massive miss and turned it around, but it is quite the risk. They have nice asset, they also have a bit of unknown-risky big assets and a debt that without being that worrisome start to be significant for their modern era cashflow, over $20billions net by now I think that you need to add to the stock price in your acquisition.
 
Last edited:
Not really sure of that, I feel there a reason it is mentioned so much and so clearly in the cross license agreement, because the default is when you buy a company you buy all its contractual obligation (asset and debt) of the sort I think, imagine Disney buying Fox if that mean that their TV-Ip deals are all cancelled because of a change of ownership and they need to redo a deal with India cricket league and FCC regional cable-radio use and all of it, have to redo The Simpsons and other IP deal with their creators... If a bank buy the bank your have a morgage with I do not think it mean a renegotiation must occur.
Maybe - depends on how the contracts are written. TV and Entertainment are a completely different industry compared to technology. As far as the AMD-Intel cross-licensing agreement, anyone remotely familiar with how it shares each companies IP with the other should also be familiar with the renegotiation trigger upon any acquisition.
 
An acquisition triggers a renegotiation because an acquisition means we are no longer dealing with the original company. New company, new owner - new deal. This is obvious.

Note is is a CROSS-LICENSING agreement. Does the company acquiring Intel really want to say goodbye to X64 extensions (among other shared IP)?
You are correct. I was operating on some bad info there. :) Man the Cyrix legal team must really have had Intel by the hairs (I guess they sort of did as Intel stole P2 from them lol)... its too bad the company didn't operate at the level of their legal dept. We would probably be talking about them and not AMD or Intel.

EDIT: For what its worth now that I read the actual legal bits... it sounds like AMD could agree to an Intel sale. I can see the issue for Intel though why would they?
I can also see some possible loop holes here. If Intel does split in two... the Fabs could retain the core x86 patents, and license them for life to the other half Intel design that gets sold. I mean that could end up in court for years... but I can see some legal shenanigan ways to avoid the renegotiation. Its also possible that a player like MS gets invovled in a sale they could say fine we don't care, throw out the x86 ISA. Retain the design teams put them to work on ARM. Depending on the player it may be in AMDs best interest to agree to a cross licence. In the case MS is involved in a Intel sale (which I still suspect they are strongly consiering) AMD will have to come to at least reasonable terms or MS could make x86 a second class citizen pretty fast.
 
Last edited:
I can also see some possible loop holes here. If Intel does split in two... the Fabs could retain the core x86 patents, and license them for life to the other half on Intel that gets sold. I mean that could end up in court for years... but I can see some legal shenanigan ways to avoid the renegotiation.
Other way around ? Logic design side would keep, as TSMC and other can make fabs x86 cpu core no problem, it is more the cpu design team that seem to need it.

t sounds like AMD could agree to an Intel sale. I can see the issue for Intel though why would they?
For the buyer (thus Intel if they want to sell), is it really that AMD need to agree to an intel sale or more need to agree for the new buyer can make x86-AMD64 cpu and would negotiate with it for that right, they could be more greedy now than back in the days (as becoming possibly the only one making them would be quite valuable).
 
  • Like
Reactions: ChadD
like this
You are correct. I was operating on some bad info there. :) Man the Cyrix legal team must really have had Intel by the hairs (I guess they sort of did as Intel stole P2 from them lol)... its too bad the company didn't operate at the level of their legal dept. We would probably be talking about them and not AMD or Intel.

EDIT: For what its worth now that I read the actual legal bits... it sounds like AMD could agree to an Intel sale. I can see the issue for Intel though why would they?
I can also see some possible loop holes here. If Intel does split in two... the Fabs could retain the core x86 patents, and license them for life to the other half on Intel that gets sold. I mean that could end up in court for years... but I can see some legal shenanigan ways to avoid the renegotiation.
Cyrix had cross-licensing but just couldn't design a CPU to keep up with the Pentium. AMD was barely better but survived.

Intel can sell to whomever it wants, it's the IP that it shares with AMD run so deep now that Intel will be gutted on the CPU end. Fabs are not design. IP is design and engineering principles. Fabs are not part of the Cross-licensing. Fabs only produce what is designed elsewhere. Be it Intel or maybe Qualcomm. In any case, the designs being pushed to the fabs need to be legally produced. That includes licensing.
 
  • Like
Reactions: ChadD
like this
Cyrix had cross-licensing but just couldn't design a CPU to keep up with the Pentium. AMD was barely better but survived.

Intel can sell to whomever it wants, it's the IP that it shares with AMD run so deep now that Intel will be gutted on the CPU end. Fabs are not design. IP is design and engineering principles. Fabs are not part of the Cross-licensing. Fabs only produce what is designed elsewhere. Be it Intel or maybe Qualcomm. In any case, the designs being pushed to the fabs need to be legally produced. That includes licensing.
I don't want to go off too far with Cyrix... but Fabrication held them back. It might be a bit telling with what is happening as I think about it though. The big mistake Cyrix made was agreeing to a settlement on Pentium Pro/PII. Intel clearly lifted the P Pro/II register-renaming techniques directly from Cyrix. Who had to turn their design to prove it wasn't Intels in the previous 486 case Intel lost. The Cyrix design was lean and mean. Intel killed Cyrix by forcing them in that agreement to only produce Cryix chips in Fabs that had a licence with Intel. At the time that seemed reasonable to them and they needed cash. The cash wasn't enough and when National Semi bought them and tried to move their fabrication to their more advance nodes the Intel agreement scrwed them and forced them to conitune using IBMs aging fabs. If Cyrix had been able to hang on for a couple years they would have gotten a big pay day when the courts eventually ruled on that one, the court would probably have been pissed to boot... as Intel clearly used the courts to gain the information. Intel setteled unexpectely fast with reason.

I mention all that as I can see the potential for some Karmic Justice if AMD was to say we'll keep the cross licence with the provision that "Intel" x86 will only ever be fabricated by Intel fabs. A few buyers like say Microsoft might actually be ok with that... it would help AMD long term no more Intel doing Tic Toc TMSC->Intel->TMSC. I think really when it comes down to it if the right company is looking to aquire Intel... it might well be in AMDs best interest to agree. If all of a sudden the only x86 company in existence is AMD. Its not actually a win for AMD... it simply accelerates the switch to ARM for everything. Especially IF the company doing the buying is say Microsoft themselves. I still say MS might be in on Intel at some point even if it means the end of x86. MS only really needs to add a GPU design team to complete their more to in house silicon for all things. Qualcomm on their own? Ya I don't see it either... I doubt Qcom cares about x86 however. I suspect its other compute patents they are really interested in.

I think I'm saying Intel might be worth 60-70 Billion for the design side of things even if x86 dies. The fab get spun off anyway reducing the potential cost of aquisision. I mean AMD paid almost 9 billion in todays money for ATI at point point.... are all the intel chipset/graphics/non x86 compute patents + their deisgn teams worth 10x that? I think maybe. If the calculation also includes an outside chance at getting AMD to agree to a reasonable new deal on x86_64 perhaps. Its also possible Intel limps along for a few years like AMD did.
 
Intel fabs just are not competitive. Period. Maybe next year or two, maybe not. That is one of the he biggest gambles as far as Intel goes. The fab spin off is probably the smartest move intel has made in five or more years. It almost makes the rest of intel "sellable" - except for the aforementioned cross-licensing with AMD.
 
  • Like
Reactions: ChadD
like this
Other way around ? Logic design side would keep, as TSMC and other can make fabs x86 cpu core no problem, it is more the cpu design team that seem to need it.

For the buyer (thus Intel if they want to sell), is it really that AMD need to agree to an intel sale or more need to agree for the new buyer can make x86-AMD64 cpu and would negotiate with it for that right, they could be more greedy now than back in the days (as becoming possibly the only one making them would be quite valuable).
Logic would like to keep sure... but they might have to leave the patents with "Intel". Intel being the fab business going forward. Companies often have their subsidiaries hold patents and license them to the main company.
I don't if its possible in this case... just saying there might be a creative legal way around this. Sort of like what Via is doing in China... Via owns 15% of the Chinese company making x86-64 chips. Just enough to make it legal. Its possible and Intel merger/sale/acquisition could be a messy thing with Intel remaining a company with a fab, and a x86 patent holding.... and retaining a small % of ownership of the previous Intel design business. I suspect it could be made legal is my point.

AMDs x86-64 position might also not be as strong as some believe. Yes their is a cross license between AMD and Intel. Still Via x86-64 exists... and 64 was developed by AMD long after Cyrix and their x86 cross patent was sold off. AMD might have a small say in what happens with Intel... but considering the state of the industry I don't think it would be wise for them to let Intel die instead either.
 
Intel fabs just are not competitive. Period. Maybe next year or two, maybe not. That is one of the he biggest gambles as far as Intel goes. The fab spin off is probably the smartest move intel has made in five or more years. It almost makes the rest of intel "sellable" - except for the aforementioned cross-licensing with AMD.
The mistake was made when they decided to announce they were open for Fab business. No one is going to trust the direct competition to be a good fab partner. Someone should have smacked Pat in the head when he announced it. AMD isn't going to hire Intel fabs, Nvidia isn't going to hire Intel fabs, no one is... Intel directly competes with everyone. The only buisness that was on the table for them was the odd ball stuff like Amazon and Microsoft silicon. Stuff MS really only started seriously chasing this last year. (and contracts not substainal enough to please investors)
 
Outside Amazon-Microsoft maybe you can have television (Sony of the world), automobile (even Tesla for example would it consider intel a big competitor for most of what goes in their car, certainly the Toyota-BMW-Honda would be open), robots, industrial, (US military for example, seem to have big contract lined up for intel foundry service, probably as a designer too but maybe for some existing chips), a bit of all what mediatek do, Matrox type, mobile only, network not directly in Intel market, memory-solid/state that could be interested in good enough nodes without seeing Intel too much as a direct competitor at the good price. We can imagine a future if advanced node get in volume that chips on them get everywhere

The doing amd-nvidia-apple (well maybe they could end up feeling Intel so far to consider it at some point, would they mind if Intel was used to do their old tech by now Switch 2 or "3050" card in 2024...) do sound too much of an IP issue for sure without a clean split.
 
Last edited:
Back
Top