> I can already see Nvidia rubbing their hands together in expectation of the massive influx of customers to their cloud gaming platform.

Roblox is not popular because of its graphics. Younger gamers care more about having fun than having an immersive experience.

I love it when I get my Robloxhead daughter to test drive some of the games I play on my 5090 box. "Ooooh these graphics are unreal" "Can we stop for just a moment and admire this grass" :-D

I think we're talking about 2 different things. I'm not sure where Roblox fits into what I said.

The problem I describe is companies pushing towards the "rent" model vs. "buy to own". Nvidia was just an example by virtue of their size. Microsoft could be another, they're also eying the game streaming market. Once enough buyers become renters, the buying market shrinks and becomes untenable for the rest, pushing more people to rent.

GPUs are so expensive now that many gamers were eying GeForce Now as a viable long term solution for gaming. Just recently there was a discussion on HN about GeForce Now where a lot of comments were "I can pay for 10 years of GeForce Now with the price of a 5090, and that's before counting electricity". All upsides, right?

In parallel Nvidia is probably seeing more money in the datacenter market so would rather focus the available production capacity there. Once enough gamers move away from local compute, the demand is unlikely to come back so future generations of GPUs would get more and more expensive to cater for an ever shrinking market. This is the vicious cycle. Expensive GPU + cheap cloud gaming > shrinking GPU market and higher GPU prices > more of step 1.

Roblox is one example of a game, there are many popular games that aren't graphics intensive or don't rely on eye candy. But what about all the other games that require beefy GPU to run? Gamers will want to play them, and Nvidia like most other companies sees more value in recurring revenue than in one time sales. A GPU you own won't bring Nvidia money later, a subscription keeps doing that.

The price hikes come only after there's no real alternative to renting. Look at the video streaming industry.

> then that is more a claim about modern self-pwned gamer behavior than megacorp conspiracy.

Riddle me this: does anyone pursue a self-pwn intentionally?

"Conspiracy theory" is just dehumanizer talk for falling prey to business as usual.

The paradox of the paradox of tolerance.

Yeah, in the future we will mourn the loss of the art of creating a new JS framework every week.

It's like learning to read English after speaking fluently for a few years. You may only need the letter sounds and then you can guess the rest. Learning Chinese works that way. You learn some basic characters and then you can guess the rest. (Learning to write without a computer is definitely more of a challenge though.)

> How can we have a constructive conversation about this, come to a consensus that things are not okay now, and move forward?

How can it be a constructive conversation if you have presupposed the outcome?

I don't necessarily disagree, but researchers are not required to be good communicators. An academic can lead their field and be a terrible lecturer. A specialist can let a generalist help explain concepts for them.

They should still review the final result though. There is no excuse for not doing that.

I disagree here. A good researcher has to be a good communicator. I am not saying that it is necessarily the case that you don't understand the topic if you cannot explain it well enough to someone new, but it is essential to communicate to have a good exchange of ideas with others, and consequently, become a better researcher. This is one of the skills you learn in a PhD program.

That is how it should be, yes. Do PhDs always meet that standard though? No.

Society depends on long-term 'happiness'. Short-term 'happiness' often makes society worse.

what you are actually saying is that a certain class of people "know better" than what another class thinks they want.

If you look at financial markets and finance theory, there is no validity to the idea that people are long term blind and short term mistaken. markets discount the future, they are the best estimates of the future rather than somebody with no skin in the game magically "knowing better"

> they are the best estimates of the future

its plausible for companies to be worth less than their assets. While it might be the best estimate its still not necessarily the best one. aka the market can stay crazy longer than you can stay solvent. Markets measure confidence as much as they measure value.

What about externalities? What about policy makers going after short-term gains?

Ironically, quite a few Linux-compatible games only work for me in compatibility mode.

Yeah that sounds like wisdom, not intelligence.