I had the pleasure of QAing Dune II on the PC while I worked at Virgin Games in 92. The game really sucked me in and was fun to play, despite the bugs.
It's universal.
We may have different tastes, but I can't see how some games went through development, QA, management, etc. and no one said "This isn't much fun to play." Or even recognized that people weren't excited and smiling at the game!
How screwed up of an organization does one have to have for there to be that little feedback shared and listened to?
Further, defining fun as a KPI presents challenges of sociology, psychology, and philosophy: People are happy to exclaim "this is fun!" And then quit playing within a few minutes. Absent a very, very careful understanding of the belief structures that drive the combination of purchase and interest that makes someone a "fan", the metric degrades to raw "engagement", which leads to meetings where it is determined that the engagement metric should be boosted through the introduction of gambling-style addictive loops.
In other words: lots of people like playing good games. Almost nobody knows precisely why they like a game, and therefore they do not know how to make a game equivalent to what they like. What the industry does instead is to clone whatever worked last time and try to give it a twist or add a little more scope. Whatever formula worked for another game, everyone is going to try adding to theirs, whether or not it's sensible. Specific game teams manage to see farther and make exceptional work by dint of being able to frame the discussion appropriately, so that it brings in everyone's perspective. Without that, every meeting is just a discussion of deliverables. But in many, many cases, the team has to make the game twice to be able to sink their teeth into the problem. Hence the common thing of game sequels being much better than the first entry.
Everyone knew it was about to happen, but this was a multi-year product where everyone had been pointing out the issues more than early enough but management simply wouldn't hear it. Concerns about unrealistic scope, poor design, etc. were met with demands that we believe in the game and put our backs into it.
Even then, I did find myself enjoying the game and every now and then lost track of what I was supposed to be testing (as a software engineer testing my changes, not QA) and simply lose track of time playing the game.
The other variation I've experience is everything starts out great with a clear design and reasonable time line. Then a couple of years later, when you are almost done, some people start worrying that what you're making might be a bit too 'niche' and start adding some 'minor' changes to make the it appeal to a larger demographic.
You can tell a book isn't fun but you may not know how to make it better.
Or the game may just go in a different direction that you originally planned. I've read about authors complaining that their characters didn't develop how they initially wanted and you may think "How is that possible?" but the same can happen with games (or software or any product you design).
- Indie devs who need to skill up/budget up/work on it more
- big corp sweatshops that don't care because they're doing the micro-transaction whale hunting/gambling addiction angle.
Same with 'bad' movies. The people releasing the bad movies aren't oblivious to the fact that the movies turned out bad and not what they hoped it would be, but they're out of time, out of money and contractually obligated to get the thing out the door.
And a lot of good games suck until very close to the end where suddenly everything clicks and they become amazing. It's not really a very predictable process & at the end of the day you have to spend a lot of resources (time & money, some of it you're on the line for to publishers or investors) before you find out.
It can very easily be better financially (or at least less bad) for a studio to finish up a game they realized (possibly too late) isn't as good as they hoped than to just abandon it - you have contractual obligations, you get paid on milestones, and everyone still has to get a salary even if you decide to abandon the game.
Do studios not prototype games?
Maybe it's rose tinted glasses from a simpler time, with shorter development cycles, but almost all these famous game retrospectives seem to begin with "once we had the initial systems in playable form, before assets or polishing, it turned out the game was a lot of fun."
Not every game is fun in prototype form (think for example of SimCity or Civilization - you have to have a certain level of complexity before it's interesting) and other times something that seems fun for a few minutes in a prototype doesn't necessarily translate to a fun X hours game. Or you can find a fun core (matching gems in a match 3 game and causing a lot of cascades is fun!) but is not something that will on its own make a game commercially viable in 2022.
And sometimes you run out of time before finding a very promising prototype and have to get started anyway or you'll have people idlying (while getting paid & costing you money you may not have before you sign a publishing deal) so you just take a leap of faith on whatever seemed best among what you did find...
The cases in your quote are really the rare exceptions where everything went great but it doesn't go that way every single time.
And if you're a small company/startup without a lot of money you can't chance getting stuck without funding, so play with the cards you were dealt and try what you can...