The much bigger problem comes when you can make the game unwinnable, but not actually lose right away (well, you've lost, you just don't know it yet). You merrily continue on for hours until much later you realize (if you do realize) that your mistake those many hours ago doomed you, and you need to go all the way back to a save from before then. Killing the player character right away (or at least soon) would be a much more merciful outcome than letting them blindly proceed down a path to nowhere.

I assume the dual intent of such dead ends were to make things seem more "realistic," as real-world errors are often permanent and aren't always immediately apparent, and also to extend the playtime.

Sudden brutal deaths are still a staple of many games and genres (even the new "modernized" King's Quest), but the silently-unwinnable state has happily fallen more or less totally out of fashion.

That feature was a nice reprieve after playing the Sierra-designed games. Though sometimes the abrupt endings in the Sierra games had their point; for example, forgetting to secure your weapon before taking an inmate into jail would result in a game-over scene in Police Quest 1. Not sure how that could be resolved the LucasFilm way.

Stupid cleaning robots....

In what is one of the most Vivendi "we don't know what we want to do" event, just a couple years later they then decided they wanted to sell their main french telecom company (SFR) and at the same time bought Gameloft and agressively bought Ubisoft shares (the two biggest french video games companies).

I remember thinking that the game designers were probably asked to make some of the puzzles frustratingly difficult so that a post-sale revenue stream could be realized.

I also remember playing through these games without the help of the tip hotline. Sharing ideas and discussing strategies with my family. It made for a very fun experience and a very rewarding feeling when you finally figured something out.

It also made for a very frustrating experience when you did something wrong or missed something along the way and had to go back to a very early save point to fix your mistake. I guess you could say that this was poor design but I still have very fond memories of the series.

Of course for most of that era, the PC was "controlled" by IBM who had zero respect for video games.

Puzzles were difficult, natural language processing is a hard problems, and commands/grammers could get weird/complex. The icon based point-n-click cleared this up a little, but Sierra had a business model build around needing to order hint books (or play with large groups of people).

No one wants hint books, because once you use one hint, you tend to just keep using them. With the Internet and things like UHS for adventure games, this model kinda hit a wall.

Telltale, Double Fine, et. al. brought this gene back by treating it more as an interactive story. The puzzles were less crazy and you could get through them without a hint guide.

Broken Age is a stealer example of both versions of this genre. Part I is more like Telltale, except with better puzzles. It's easy to get through, but still challenging. Part 2 goes back to the Sierra model though, with puzzles that make little sense and are insolvable for most casual and moderate gamers without a hint guide.

https://www.humblebundle.com/store/kings-quest-the-complete-...

Where today we speak of "replay value", of games that you finish and then want to play again, no such concept existed back then. Instead, it was assumed that a game that was "beaten" (let alone "mastered") would be put away, and so a game should resist being beaten for as long as possible to make it "more worth the money."

This was partially the quarter-eating arcade-game philosophy leaking into home-console gaming, and partially the knowledge that parents want the $50 they spend on a distraction for their kids to stretch as long as possible.

In the 80s, since games were size-limited (small tapes/cartridges and low memory headroom), this translated to games that were preternaturally challenging to stretch out their play-time. This is the classic "Nintendo Hard" game-design philosophy; but it's also responsible for grind-fest RPGs (another way of taking a small amount of content and padding it out.)

In the 90s, with lower size-limits, we instead saw the temporary emergence and flourishing of the really long game—everything was advertised by talking about how expansive the world was, and how many hours of content there were to get through. We still look back on this era as being responsible for the creation of many classic action-adventure and adventure-RPG games for this reason.

All this stopped when we hit the 3D era and art-asset production became the overriding cost of producing a game. Games weren't limited in "content" in the scenario/script sense, but they were limited in "content" in the unique-things-to-see sense. There was a temporary burst of interest in exploiting this divide directly (some early PSX games fiddled with procedural generation to overcome the problem), but this was limited to a few minor studios; the majority instead moved their games toward being increasingly high-production-value cinematic affairs.

With a fixed amount of stuff to see, getting the player to care about looking at it all more than once became the overriding concern; thus, replayability rather than difficulty became the goal. (Replay-value was always part of the design philosophy of some studios, but now it became part of the basic culture of the industry.) This is when you start to see things like "missions" and "achievements" emerge, that give you many things to do within the same few set-piece areas.