I think you make a very valid point here. That's the problem, consumers do not understand the development process very well at all. This problem is visible with games using the examples you illustrate but it goes well beyond games to all software and beyond graphics to be generalized to "software features".

To create an application be it a game, or a desktop app or something else requires a certain amount of creativity both towards solving problems, but also towards the presentation layer, and that includes things like user friendliness/human interface guidelines and aesthetics/eye-candy/effects/etc. Creative people can and do come up with all kinds of interesting and amazing ideas in all of these areas, and it can be very fun, challenging and fulfilling to implement one's ideas as well, but in the process one often discovers that for any number of reasons that a given idea becomes impractical in the end when all is said and done.

It could be that something looks or sounds good at first but when you actually implement it, it isn't that great or doesn't really add value like you thought it would, or it may even create a negative experience. In other cases, a given idea for say some special effect might end up looking amazing but requiring an extremely disproportionate amount of CPU/GPU/RAM/etc. resources to pull it off. Even after optimization such a feature might not give a benefit proportionate to the resources it consumes and may thus not be viable in the end, however at the early stages of development of something, one just does not have this level of information available. As the product comes together and you have to try to make all features work on the targeted platform(s) you can realize that the app is doing way too many things compared to what the actual hardware is capable of. You want your customers to all have a good experience, so you may need to scale back on all of the visual or other features you may have wished to provide in a perfect world.

The thing is, developers don't have a crystal ball to know all of this stuff in advance though. So they either have to be open about it and share what they're doing, then adapt it over time and perhaps pull back on some cool things they may have liked to do as a natural course of action, or they have to be extremely cautious about sharing or explaining any of what they're doing publicly in advance at all, so as to not let people get their expectations up unrealistically.

I've been a developer for a few decades now and I can say one thing for certain. Lets face it, no matter what disclaimers any developer prefixes any kind of software announcement/demonstration/reveal with, as soon as people see or hear something exciting and want it, if you take it away later or can't do it for any reason whatsoever, some people will be very upset as if you promised it to them written in blood.

This is precicely why many companies tend to just never ever publicly disclose or even acknowledge products or product features that they are working on. People can always speculate or desire a company to be working on something, but if they don't have any information at all to go on, then their imaginations and expectations can't take over.

Companies like Valve, Blizzard, id Software went the "no comment" and "When it's ready(TM)" route a long time ago and stuck to this way of doing things and overall it has worked out quite well for them both economically and via popularity. Sure, there are detractors out there too but there are always going to be detractors of anything big, that's just mathematics.

The more information companies reveal in advance about their products and features, the more they seed people's expectations both of getting those features and of seeing the product released sooner than later. When people see things they desire things, and when they desire things the best time to get it is "right now", and any amount of wait is grueling. Don't announce things, don't make release date/time promises and people may want you to, but that's the only thing they have to complain about - is not knowing anything.

So while I want to know about what games are coming out, what features they will have, and when they are coming out like most other people do too, with my developer cap on I also appreciate the "no comment/When it is done." way of doing things and I personally think companies are in a safer place to take this approach even if myself and everyone else are dying to know what they're going to do next.

So in conclusion, I think the best thing that any company can do about this problem is to simply not publicly reveal anything they are doing at all period until they know with absolute certainty that something is 100%, because otherwise there will always be a vocal group of people out there who are upset by it who just can't comprehend the concept that nothing is set in stone despite best intentions.

I watched an interview with the No Man's Sky dev recently where they talk about this lightly and he admitted that they were very hesitant to talk about many of their ideas/plans for the game whether they were implemented yet or not because you never know what you can truly deliver until you actually try to do it. In a perfect world it should be ok to say what you'd like to do then find out later that it isn't viable and let people know it isn't happening. But in the real world, you will be burned at the stake and eaten alive on gaming sites on the Internet for being that level of "honest". Honesty and transparency may be virtues, but they don't mean much if you're laying on the ground charred with smoke coming off of you.

Silence is golden.