I have a certain sympathy for all involved in your argument, but I don't think it's a good way to do business for either businesses and their executives, their shareholders, or ultimately the consumer and public.
The problem is short-term thinking around annual or even quarterly earnings reports and whatever drives that. I appreciate that a CEO or even a board may feel pressure up to and including losing their jobs if they leave literally anything on the table in the form of long-term investment, the "goodwill" (which is a real financial accounting line item that's sometimes a lot) of their customers, partners, etc. and the long-term sustainability of the business.
But it's that ability to deliver that magic combination of a product that people ideally love buying, but at least don't mind buying, a team that is healthy, well-compensated, and inclined to go the extra mile for the employer at rare need because the mutual priority of the wellbeing of the partnership goes both ways (this is massive, when you treat people well when they're in a bind, they'll go the extra mile when you're in a bind with little if any pressure, most of the time), and a consistent, low-beta (for the investor's risk appetite) return on investment for shareholders.
This is the kind of business that e.g. Warren Buffet at least talks about as the kind of business he learned to invest in. I'm not a Buffet scholar or watcher really, so I don't know how well he lives up to those ideals in general or especially recently, but it's certainly a pretty good way to describe the desired outcome of capitalism!
I've heard lots of arguments for why boards and CEOs and the like simply have to skin the sheep once rather than shear him many times (to borrow the old saw): but never a very convincing one. It's basically absurd in the kinds of dual-class share structures that are common in tech now: it's not only extant but common for the founder/CEO to be basically unencumbered by quarterly earnings reports, what's e.g. Mark Zuckerberg going to do? Fire himself?
And under this more "good for everyone over a long time horizon" model, you can just launch the next breakthrough card when it's done, and launch incremental improvements when you're out of inventory, or see a gap in your lineup, or whatever.
Fix corporate governance in what should be the relatively uncontroversial ways I and many others have described, and you fix this (among many, many other) set of perverse incentives.
I don't think anyone's under any illusions that people are going to upgrade more frequently, but the lack of predictability is pissing people off and miscalibrating expectations on what they "should" be getting every gen. Going 10%,40% is literally worse and more disruptive (in the bad way) to consumer psychology than if you'd just gone 20%,25%, because people will lock in on the most desirable possible combinations of perf gains and value offerings/prices.
the x70 tier launched at $349 in 2010 dollars and has bounced between $349 and $399 for every generation, but the one people remember is the singular time it was $329. And the x70 die has almost always fluttered around 250-300 mm2, except for the trailing-node era (when NVIDIA got cheap TSMC/samsung and chased price) where the 2060 was almost as big as a 1080 Ti (445mm2 vs 471mm2) etc. But people have never shut up about how the 4070 isn't a real x70 because it's only 300mm2 etc - they latch onto the one singular time it was bigger etc.
what you end up with in terms of node progress is similar - people tell themselves they deserve the gen when everything increases by 50%, at the best possible efficiency, with a bigger memory bus, but they also want the pricing from the value-focused generations. And they tell themselves that really things are actually advancing 50% per gen (because they see that sometimes/in some segments) and it's just evil corporations not letting them have it.
If, instead, you smooth the progression, so you get 20% then 25% over those same 2 gens, there is no bones made about the actual progress being made etc. Consumers have predictability and information, and they can thus make optimal, informed decisions about which generations they are going to upgrade (which I think everyone agrees at this point isn't "every single generation" for literally anyone anymore).
it's functionally no different in terms of progression, you are getting the same averaged gains year on year. It just means "don't launch the GTX 970 to be some super killer deal, because you have to top it again next year". And if next year you only get 10% over the absolute best sku in the lineup, you failed, your card is shit (or at least "disappointing").
you see this constantly with the 3060 Ti: the 4060 is deemed a failure because it doesn't leapfrog over the 3060 ti (the literal best-value card since RX 480 etc). And the 4060 Ti is a failure because it doesn't leave the 3060 Ti in the dust ("only" 3070 performance, with twice the VRAM). Etc etc. Stop doing that, just make normalized skus with consistent yearly improvement. That's what reviewers are asking for, and tbh they're not entirely wrong on that part (even though their overall performance/cost expectations are just completely bonkers). They're imagining it's going to be 50% a gen, not 20% a gen, but they're not wrong about how the public does need some consistency in their life (like any child) as far as generational gains.
The market simply doesn't reward you for making a 3060 Ti or whatever anymore, because that is taken for granted. You can make the same argument about the trailing-node strategy (for 20/30 series) as a whole, which legitimately bent the price curve downwards for consumers. But consumers will never reward you for a price increase you didn't make - in fact they will be extra mad when you finally bounce back to the "normal" cost curve, because now it feels like an extra big increase all at once.
Personally yeah, I'd rather have the god-tier value card and just understand that the successor is going to be feature/performance/efficiency focused and not value-focused. As high-information consumers we can make those decisions and claw back some consumer surplus. But that's not where the public is. And people have been hopping mad about it for years now, this is not an unusual sentiment.
Intel basically exemplified this mindset perfectly: every generation you got X%, and they aligned X% so that they'd get a similar number on every single generation. Going (eg) Sandy Bridge to Skylake is much "lumpier". And people dump on that era, but they also are (functionally) clamoring for that same consistency in the GPU market. "You get 15% a year whether you need it or not".
Right now they don't know what to expect, so they can imagine whatever they want, which only sets things up for disappointment.
I'm not sure that you and I actually disagree about all that much, and might be just arguing cases in different adjacent debates.
The realities of the modern hyper-sophisticated semiconductor fabrication business is that the chips aren't consistent with event a single logical "run" (or whatever the modern term of art is), and so you're binning and blowing fuses and all kinds of stuff to get even the same SKU in the ballpark as other cards with the same SKU (throw in MSI vs whoever, it just gets harder), and so clearly there is a mutually beneficial amount of "smoothing" that's more than zero (impossible) and less than anti-competitive (56% profit margins and a CEO saying "this is a gold mine" on camera).
And as someone who is clearly more current on the nuances of the gamer-oriented lineup, I'm inclined to take your word for it that gamers want a little more "smoothing" than ML people without a blank check (though I don't think I can endorse referring to the public as needing the same guidance as any other child, that's not me trying to chastise you, it's me noting that underestimating the public has burned a lot of smart people some rather, uh, recently).
It is complicated, and there are hard realities of the physics and tolerances of the manufacturing that complicate the situation, and there are hard questions in what constitutes a "major version" and what constitutes a "refresh" or a "rev". So to the extent you're saying that the vendors need some rope to give their customers something that isn't a casino by binning and labeling releases with an eye to some kind of non-ridiculous trendline in capability that isn't coming or whatever, yeah sure. I don't know enough about the pressures of a middle manager working on that stuff to offer hard policy prescriptions about it. I do know about the pressures of an upper-middle technology executive faces with this whole other bag of constraints over and above the already hard job about the "the Street is going to kill us if X isn't Y by Z for Q1": that's a problem I think you'd agree no one would actively solicit as part of their job description all else equal.
There's this different basket of stuff that's ethically and practically anti-competitive, anti-market, and at least used to be illegal most of the time (jurisprudence on how narrowly to interpret the anti-trust laws we do have has become an Olympic sport in narrowing their scope). That's the drivers I simply don't believe to be unfixable in a year or two, that's an unreasonably wide gap between a 2k card you can run PyTorch on and a 50-80k card (or whatever the real price is, I'm not sure anyone even knows in a public way) that's basically the next step up, that's the 4090 spending die area and TDP on vector units it doesn't have the GDDR6 to keep fed half the time. And finally, it's accepting margins that let you price cards at MSRPs you're willing to publish and meeting the demand for them at your stated price. This isn't an act of god chip shortage (in general, there are instances of that, particularly during COVID): I hate to pick on Meta here because the reason I'm using them as an example is that they're reasonably transparent about this (so keep it up guys, that's awesome), but IIRC their last publicly known block of H100 (Hopper v1 for lack of a better term) was 150k cards in one transaction, with a target of something like 700k "A100 equivalent compute units" by the end of 2024. I'm not surprised Lisa Su and Jenson Huang are inclined to do as much of that business as possible, but again, market failures left and right.
Maybe gaming and AI need to be different companies to get an efficient market, maybe export controls need to be fiddled with further, maybe a lot of things. Gamers should be able to get cards that play relevant titles at MSRP, startups and independent researchers should be able to get cards that work and don't require a blank check, and big AI labs should be able to get a reasonably level playing field in terms of access to the big iron. But a monopsony quasi-cartel of frenemies competing in some ways but with their fates intertwined in other, cornering the market on process nodes and still getting gouged by vendors having a "spirited debate" and not a UFC match. That's not getting us the benevolent global outcome of the locally selfish invisible hand. That's where sparing but well-refereed interventions in failed markets is what makes capitalism sustainable after the sugar rush wears off of having industrialized in the first place.
That's the stuff I think Intel is going to go on a rampage around if Gelsinger pulls off the turnaround.
The problem is short-term thinking around annual or even quarterly earnings reports and whatever drives that. I appreciate that a CEO or even a board may feel pressure up to and including losing their jobs if they leave literally anything on the table in the form of long-term investment, the "goodwill" (which is a real financial accounting line item that's sometimes a lot) of their customers, partners, etc. and the long-term sustainability of the business.
But it's that ability to deliver that magic combination of a product that people ideally love buying, but at least don't mind buying, a team that is healthy, well-compensated, and inclined to go the extra mile for the employer at rare need because the mutual priority of the wellbeing of the partnership goes both ways (this is massive, when you treat people well when they're in a bind, they'll go the extra mile when you're in a bind with little if any pressure, most of the time), and a consistent, low-beta (for the investor's risk appetite) return on investment for shareholders.
This is the kind of business that e.g. Warren Buffet at least talks about as the kind of business he learned to invest in. I'm not a Buffet scholar or watcher really, so I don't know how well he lives up to those ideals in general or especially recently, but it's certainly a pretty good way to describe the desired outcome of capitalism!
I've heard lots of arguments for why boards and CEOs and the like simply have to skin the sheep once rather than shear him many times (to borrow the old saw): but never a very convincing one. It's basically absurd in the kinds of dual-class share structures that are common in tech now: it's not only extant but common for the founder/CEO to be basically unencumbered by quarterly earnings reports, what's e.g. Mark Zuckerberg going to do? Fire himself?
And under this more "good for everyone over a long time horizon" model, you can just launch the next breakthrough card when it's done, and launch incremental improvements when you're out of inventory, or see a gap in your lineup, or whatever.
Fix corporate governance in what should be the relatively uncontroversial ways I and many others have described, and you fix this (among many, many other) set of perverse incentives.