Off-Topic / Dumping Thread

[+OLD CS1]

2 hours ago, GDMike said:

From an article I read yesterday it looks like they're still in talks. 

Whether they are in talks or not, if these are activist investors, they are likely planning a hostile take-over if they cannot agree on terms.  Texas Instruments and its "loyalist" stockholders are likely trying to iron out a path least likely to destroy their holdings.  If they were in a healthy enough position to stave off an activist take-over, they would have just ignored the advances.  Investment companies are almost always looking to "maximize" their return, which means carving up the company and selling off its bits.

[Gary from OPA]

14 minutes ago, OLD CS1 said:

Whether they are in talks or not, if these are activist investors, they are likely planning a hostile take-over if they cannot agree on terms.  Texas Instruments and its "loyalist" stockholders are likely trying to iron out a path least likely to destroy their holdings.  If they were in a healthy enough position to stave off an activist take-over, they would have just ignored the advances.  Investment companies are almost always looking to "maximize" their return, which means carving up the company and selling off its bits.

The biggest assets TI has is their patents on various technologies and most of their revenue comes from royalties on those patents. Worse case the investors turn it into like ARM which becomes just a licensing system to continue to earn money from the patents and original chip designs

[+OLD CS1]

1 minute ago, Gary from OPA said:

The biggest assets TI has is their patents on various technologies and most of their revenue comes from royalties on those patents. Worse case the investors turn it into like ARM which becomes just a licensing system to continue to earn money from the patents and original chip designs

Does TI no longer fabricate ICs?

[+TheBF]

Off topic for sure but I found it on r/Forth. 

 

The Canon Cat – the computer that tried to banish mice • The Register

[+FarmerPotato]

14 hours ago, Tursi said:

I don't know if it matters. Once this starts it means the liquidators have their sights on you... 

 

I had to read the letter in detail. 
 

I see some hints of fallacies there but taking it at face value: 
 

Elliott and Co saw a beaten-down stock with an apparent disconnect between market demand  and their big investments in new fabs in Texas. Elliott argues that two of the fabs are unneeded. That's their whole idea. 
 

Elliott's thesis: TXN trades below potential because ratio of CapEx/FCF is out of whack at 23%. Buy low, get them to fix this ratio, profit.  That's what asset managers pitch at morning meetings. 
 

 

So Elliott buys a big chunk, 2.5B or about 1%, which is just enough to get a seat at the table.
 

They want Texas Instruments to spend far less,  return the cash to shareholders, as TXN have done for a long time.  
 

They may pound the table for  Texas Instruments to "lever up" which equals more cash out. 
 

That said, I've been a bit underwhelmed by my very positive return on TXN.  
 

Here's Elliott talking: 

 

Quote

TI has made no commitments to the investment community (other than its commitment to spend). No multi-year growth plan has been provided to investors. No multi-year targets have been outlined to assess whether TI is executing against its strategic plan. No articulation has been made regarding how or why TI's revenue-capacity targets are still necessary on the same timetable. TI is asking its investors to support a ~$30 billion investment program but has not provided a long-term revenue growth framework, including anticipated market share gains, which could form the basis for the Company's plan. Most other public companies would have provided a far more robust justification, including having the CEO publicly convey the strategy (which TI's CEO does not do during TI's annual capital-management event or quarterly earnings calls)

I'm not sure that's all true. But there's the "shareholder revolt". 

[Tursi]

1 hour ago, FarmerPotato said:

I had to read the letter in detail. 

Interesting. It still smells like Toys R Us to me. Short term profit, long term destruction. I hope I'm wrong.  

 

If not, maybe we can take up a pool and buy the logo.

 

 

[+OLD CS1]

1 hour ago, FarmerPotato said:

I'm not sure that's all true. But there's the "shareholder revolt". 

Well, I trusted Pelts and Rasulo to turn Disney around, I suppose I could extend the same consideration here, as well.  I remain skeptical in the face of any demands for more money out of the company.  I see this leading to not getting enough money so we should just start selling off stuff.

 

5 minutes ago, Tursi said:

If not, maybe we can take up a pool and buy the logo.

I bet we could buy the home computer rights for a song.

[+FarmerPotato]

16 hours ago, Gary from OPA said:

The biggest assets TI has is their patents on various technologies and most of their revenue comes from royalties on those patents. Worse case the investors turn it into like ARM which becomes just a licensing system to continue to earn money from the patents and original chip designs

No, income from licensing is not it. Texas Instruments's enormous profits are from catalog sales of analog and mixed signal chips.  They've said that TI is happy to hold finished inventory and pile up work in progress (chips on die, for  packaging when needed.) Because since 2020 they want to stock 90 days  inventory of customer demand.  

The largest share of their business is automotive.  77%? Your typical new  automobile has thousands of chips in it. (3000?) 
 

 That market does not want cutting-edge nodes for CPUs with a trillion 3nm transistors (exaggeration, but hold your breath!) Instead they want chips on big dies at relatively big nodes like "legacy" 45 nm. They want chips robust in extreme temperatures with ever-lower power needs and noise floors. Those are the fabs Texas Instruments is building more of. 

 

 

 

[+FarmerPotato]

Way back in the 90s, their TMS320 series became  50% of profits as they dominated the market (which they pioneered: beating Intel.) Since then, we've become accustomed to TI as a cash gusher. 

Some hard cap-ex decisions they've made are mentioned in the Elliott letter.  
 

I researched two of them: the  1998 decision to not invest the fixed costs of doing the future of CMOS nodes and process, after their pretty amazing BiCMOS. Second, not trying to compete to fab microprocessors at home. 
 

This was when a cutting edge research fab price reached $1 billion. (Or soon after.) They partnered with TSMC like so many others. They're still making huge profits without announcing a new CMOS or bipolar technology node.  But many other innovations go into new products. 
 

The CPU decision came 

around 2006, IIRC. Texas Instruments  realized they couldn't keep up the race needed to make  new CPUs at home.  So those are made at TSMC for a long time now. Simultaneously, they staffed up to make ARM CPUs.  Embedding them into everything from $1 controllers to big-price-tag medical and space rated applications. 

 

(Anyone know if those fabs have reached the $10 billion price tag yet?)

 

It's become so costly, the world production of the  just-ready-for-prime-time equipment is... single digits of EUV lithography machines. I think Apple just bought all of them for 2024. (Where do they put them? at TSMC?) 
 

Only TSMC and Intel build and operate  cutting-edge fabs. Samsung trails behind, but they make mind-bogglingly dense types of memory. TSMC uses the cooperation of partners all over the world, which generates an unbelievable amount of innovation. Also they work 24/7.! Intel has gone it alone until their "foundry" announcement (around 2021? )
 

Rooting for TXN as a fan, the choice to stop equipping for cutting-edge  process research  shocked me. But they were the right financial decisions.  Continuing those investments would have sucked up billions in cash; partnering with the world's leading fabs has paid off. (that means just TSMC now, maybe was GF once. TI just bought out a GF fab, I think that was Utah.) 

 

Now (past ~5 years) Texas Instruments just wants to multiply its domestic fabs for "legacy" nodes that can generate massive cash returns if fully utilized.  This includes all 300 mm wafers from 150  (300 is the current standard at equipment makers, 450 seems to have been a non-starter.)

 

The math on 300mm is: double the radius, quadruple the area per unit wafer. It turns out yield is 150% more per dollar.

150mm or  6" is like one slice  of pizza:  🍕. 
 

300mm or 12" is like a personal pizza:  🍕🍕 🍕🍕

 

 

TXN has been building, I lost count, maybe 6 new 300mm fabs in Texas or Utah. It seems that as soon as the first was about done, they announced another. (Indeed, they survey the land to accommodate doubling of a fab.)   It's an about-face,  as TXN has typically re-invested single-digit % of cash flow. Since the big supply disruption or 2020, it's a full speed ahead 23%. 
 

When  this got underway, I listened to two quarterly calls. They were really thin on details about this cap expenditure.  Personally, I think it's exciting.   BUILD BABY!  but that's just emotion.  And that's the gist  of the Elliott complaint: "TXN should have been happy to build 4, not 6 fabs" I guess. 

 

I'm going to check the recent quarterly calls. 
 

 

[Gary from OPA]

This person seems to have recreated our ti99 sound chip using a max v cpld.

 

 

[GDMike]

Remember that day

[Gary from OPA]

2 hours ago, GDMike said:

Remember that day

i think i still have my weird af670@torfree.net email when back then they started assigning emails starting with aa000 and moving up as new members joined the system.

[+OLD CS1]

3 hours ago, GDMike said:

Remember that day

Before then, I had an @genie.geis.com email address, based upon my user name (a little more usable than that numeric-dotted mess that was CompuServe.)

[jrhodes]

Oldest email I had was on hotmail in 1997. Second oldest was on lavabit, back in 2008. Stayed with them until the shutdown. Now I got multiple emails addresses here and there for different stuff.

[+Vorticon]

I just realized that my main home desktop PC in 13 years old! It was a top of the line Alienware Aurora R3 when I bought it new, and the only upgrade I've done to it is replace the two 1TB raid drives with a single 2Tb SSD, along with maxing out onboard RAM. Frankly I have not yet felt an acute need to replace it, although I think it will be out of spec for the upcoming Win 12, which might force my hand. It certainly won't run more modern games like the latest Flight Simulator, but honestly most of my gaming is with DOS games anyway. Other than that it still runs all my needed utility programs and emulators, and is fully decked out with a full array of ports and readers as well as a CDR which is very handy. Not to mention that the case looks really cool with its embedded LED lighting. I'll hang on to it for as long as I can, and I'm sure others here can relate.

I can't say that my laptops have had the same kind of longevity sadly, averaging about 4-5 years before breaking down.

 

[Gary from OPA]

There only one left.. if you ever needed a full 6 farad supercap, we not talking micros here a full 6.25 of zapping power across the room for just under $3k the time is now to grab it. (With rubber gloves)

 

https://www.digikey.com/en/products/detail/eaton---electronics-division/XVM-259R2635-R/10441148

Edited June 8 by Gary from OPA

[jrhodes]

I think Bob the dog may be my spirit animal

[+OLD CS1]

Microsoft takes a step back to reassess Recall.

 

https://thehackernews.com/2024/06/microsoft-revamps-controversial-ai.html

[tater1337]

21 hours ago, OLD CS1 said:

Microsoft takes a step back to reassess Recall.

 

https://thehackernews.com/2024/06/microsoft-revamps-controversial-ai.html

nope, this is no a step back, the is just damage control

the software will still be installed on your machine, and four updates from now it'll get turned on

after that noise subsides then someone will notice that even when turned off, that it is still recording data

and after THAT news gets old, someone will notice that it is sending data to servers online

[+OLD CS1]

1 hour ago, tater1337 said:

nope, this is no a step back, the is just damage control

That is the "reassessment" bit.

1 hour ago, tater1337 said:

the software will still be installed on your machine, and four updates from now it'll get turned on

I posited this, already.

1 hour ago, tater1337 said:

after that noise subsides then someone will notice that even when turned off, that it is still recording data

Like other "features" people do not want, Microsoft will just turn it on for you to make your life better.

1 hour ago, tater1337 said:

and after THAT news gets old, someone will notice that it is sending data to servers online

And, by the way, if you do not agree to these terms, you must immediately cease use of your Windows computer.  What about your data?  Get wrecked.

 

For now, this requires a computer with an "AI processor."  Until Microsoft decides Windows 12 has requirements which make it impossible to run on your old computer or any without an AI processor.

 

I am going to throw this bit of flamebait out there: while I am considering Linux on my desktop for the first time in my long professional career, I am compelled to point out that 10 or 20 years ago, Linux could have been a force against Apple and Microsoft had it not been splintered so much.  It seems the only people who could agree on a direction for the operating system were those with similar corporate interests as Microsoft and Apple.  What we get is angel investors for apps and programs which run on the operating systems which use us as the device.

 

This is the Judean People's Front and the People's Front of Judea and the Judean Popular People's Front.  Splitters.

[+mizapf]

I believe the "splintering" thing is a bit overrated. By the library concept, you usually get all required libs for the applications on installation time. I usually don't choose the application to work on my openSUSE installation in particular.

 

There is another issue, though: It's just with the library concept. Since most applications are not statically linked, you rely on certain versions of libraries on the target systems. This is why the containerization was so successful in the last years (e.g. Docker), although I actually don't like it. Th problem is that you get the same library in all containers in different versions without a good way to patch them.

[Tursi]

28 minutes ago, mizapf said:

I believe the "splintering" thing is a bit overrated. By the library concept, you usually get all required libs for the applications on installation time. I usually don't choose the application to work on my openSUSE installation in particular.

 

There is another issue, though: It's just with the library concept. Since most applications are not statically linked, you rely on certain versions of libraries on the target systems. This is why the containerization was so successful in the last years (e.g. Docker), although I actually don't like it. Th problem is that you get the same library in all containers in different versions without a good way to patch them.

I agree with that. Containers are a reasonable solution to a problem that didn't need to exist. 

 

It's like open author authors looked at Windows and went "DLL Hell? Well, watch this!" Remember when they used to make fun of it?

 

To add my take, one of the strongest reasons Windows dominates is backwards compatibility. You can still run your software on the new version without changing or updating it the vast majority of the time. This concept seems lost in the Linux community. "Just works" is kind of important to most people. That said, Windows 10 and 11 are showing strong signs of backsliding.

 

Like OLD CS1, I am considering Linux on my desktop for my first time, too. I obviously know how to maintain it. My laptop already switched a couple of years ago, just to see if I can stand it (so far yes). But I don't do heavy work when I travel. When I maintained Linux software for my work, I was hitting those obstacles almost daily. But I probably won't hit them as often as a mere user of average software.

 

[+Vorticon]

4 hours ago, Tursi said:

Like OLD CS1, I am considering Linux on my desktop for my first time, too. I obviously know how to maintain it. My laptop already switched a couple of years ago, just to see if I can stand it (so far yes). But I don't do heavy work when I travel. When I maintained Linux software for my work, I was hitting those obstacles almost daily. But I probably won't hit them as often as a mere user of average software.

I've done the same with an older laptop (ironically a first gen Microsoft Surface Book) which has had Linux Mint installed for 3 years now. Quite usable, but the familiar issue of of Linux not being supported by my academic organization such VPN remote access to my office computer or to the medical EMR remains a huge problem for mainstay use, at least in my case. Not to mention that hobby programs like Classic 99 (to name only one) remain Windows bound even though most, but not all, can be run, albeit imperfectly, with WINE. Until that changes, I will unfortunately remain inexorably anchored to the Microsoft ecosystem...

[jrhodes]

[+FarmerPotato]

From the Almanac of Improbable Utterances: 

 

"Shoot! I left the floppy disks in the car! In this scorching Texas heat!"