NYU Stern Professor Asawath Damodaran gives his analysis of the current market

@riccardoj Very useful to get a systematic view of the market. It's as anyone with a logical mind will suspect but his numbers clarify the degree.
  1. Quality - low net-debt, high profitability 2
  2. Change agents - new economic structure (WFH services, disinfectant products, entertainment at home, shopping at home etc). Reddit, what other "change agent" sectors are there?
  3. Bargain basement - Aerospace, airlines, travel, commercial property.
 
@jonwalter Change agent sectors: tech companies that will survive based on the sheer fact people can stream videos, or people will keep using their products amzn, nflx, aapl, cell phone manufacturers or chipmakers broadcom, Qualcomm etc, cell tower REITs, AMT, CCI, SBAC (those tickers might be wrong, fact check that) companies that will continue to design and manufacture our 5g future such as interconnected vehicles and accelerated streaming/connectivity.
 
@thika 5G is not going to be something that buoys tech for years lmao the services are overpriced garbage which telecoms are pushing on people because they don't have any growth stories left

90% of what consumers need can be done with 4G and that remaining 10% is niche stuff which a fraction of the population needs

The physics of 5G is just broken and essentially every company just sidesteps any questions like "why can't my phone sustain 5G coverage for more than five seconds while walking?" Or "Why has my battery life been cut in half?" Or "Why is it physically impossible to get 5G coverage indoors?"

Those aren't the telecoms' faults as working with mmWave and Sub-6 Ghz frequencies is just friggin impossible given the range limitations of high frequencies which is magnified by path loss
 
@resjudicata I don't dispute that the early stages of the technology is flawed, but when's the last time anyone fully embraces the first generation of a completely new technology? When's the last time the first wave of a new technology is actually the paradigm shift? It's usually the 2nd or 3rd iteration. Don't limit your thinking to simply fast cell service and high res apps/games/connectivity.

Interconnected cars. Automated driving. Smart homes that are actually smart, and not just a fancy thermostat. Continuous blood glucose monitoring that seamlessly connects with phones, user interfaces, and doctors offices. Enhanced government surveillance. Machine learning that prepares schedules, meetings, and finishes busy work so that real analysis can be decided with reduced time wasting filler. Parents that can monitor their children/pets/homes/gardens through wearables.

These companies are continuing to research and design 5g, while other companies continue to build out the new platforms. It's coming.
 
@thika He's talking about limits in the laws of physics. Similar to why haven't we gotten Bugattis to go faster 400 kmph even after a decade of development. We can't change the drag coefficient.

Now, path loss for these mm wave is the limitation.

It's no different with skin effect and magnetic materials permeability in electrical components. Hence, we haven't seen a lot more applications of Silicon Carbide or WideBand Gap.

We're limited by the physics.
 
@resjudicata As full time network engineer I can say you are dead on. It's crazy how much hype there is here. Sure I like bandwidth, I even had a shirt "Will work for bandwidth" but 5G isn't some holy grail.
 
Holy shit you're the first person I've seen mention SiC or wide bandgap semis on reddit before. Now THAT is a technology that has the opportunity to change the world once the big three wafer suppliers can ramp more capacity and improve yields. GaN too but to a lesser extent.
 
Aren't they saying SiC is over hyped?

I'm only loosely aware of some applications at work, but to hear the power team talk it sounds like it has real potential. Obviously they're not going to trash their own project, and I'm secretly not actually an EE so I don't know that much about the details.
 
Some people might think that but we don't believe it is because it offers massive benefits over silicon in maximum voltage, switching efficiency, heat dissipation, and pretty much anything else. For markets which are running constantly, the high initial cost of SiC pays for itself through lower electricity usage.

If you look at it from a demand point of view, device manufacturers can't get enough of it. Rohm's internal SiC wafer unit SiCrystal for the first time announced it was doing merchant market sales a few months ago as the few suppliers who can produce it at scale are at capacity. Cree has signed numerous long-term contracts totaling over $1B, much larger than Cree's most recent fiscal year revenue. One of those deals is with STMicro, who ordered $500M in wafers from Cree and purchased Norstel as an internal source before contracting Rohm for another $120M worth just two months ago.

The only problems are cost, yield, and wafer size (currently moving to 8 inch versus silicon on 12+ inch wafers), which are typical for really any new semiconductor technology but especially for SiC. The process of growing SiC boules is nothing like the process for silicon ingots and requires completely separate machinery so the existing silicon guys can't just switch over to SiC easily. So essentially demand is there while supply isn't.
 
@resjudicata I get what you’re saying, but I think it just echos the voices of the past saying a gigabyte is more storage than a consumer needs. Yes, but only for now.

5G is still an increase in bandwidth and lower latency, even if the range is terrible right? Supply sorta eventually creates demand, similar to how building more traffic lanes encourages more people to drive.
 
@clemi I don't really see it happening with in home broadband though. My speeds have only gone up slowly in the last decade, and my typical usage has actually declined, even as some providers are trying to push gigabit speeds.i just don't see at this time what the usage could possibly be.
 
@leaderturk Maybe something like a multiplayer augmented/virtual reality on your phone, where the augmented/virtual world is persistently editable across the in game universe by everyone? That’d be a ton of transfer

I agree, the usages aren’t obvious but I think people will find it
 
@thika If the tech is status quo, then companies like AMT will have to spend a lot of money to get the same amount of revenue.

In other words, more towers for same amount of expected users.

Surely the royalties and license players in 5G are better?
 
@jonwalter AMT buys the towers that T and VZ build, then they lease them back to The and VZ at a profit. It's an excellent business model, with great leadership.

IMHO
 
@thika Its not just the early stages though. The technology in and of itself is the flawed part and like I said, not the fault of the carriers or equipment OEMs. mmWave has absolutely horrid range because it's such high frequency and requires MIMO which is absurdly expensive, and because of its low range, carriers need significantly more base stations/cell towers to deliver acceptable service to anyone. Plus signals that high can't penetrate surfaces so there isn't really a fix to the whole indoor thing.

Sub-6 GHz has more of a chance for mobile as its lower frequency means better range but still, people can do 90% of what they need to do with the existing 600 MHz-3000 MHz frequencies.

If anything, yes 5G's real opportunity is in industrial applications but I don't really see how it can work with vehicles. If it can't maintain coverage when someone's walking then I'm not sure how it'll magically hold coverage going 60 MPH.
 

Similar threads

Back
Top