Since I didn't get to elaborate on something I said in the last thread:
When I said only getting one chance to play through a game with multiple paths was an interesting premise, I didn't mean that it was a good idea to apply it to a regular game (i.e. one that you have to pay for) as several people implied I was suggesting. I just meant it could be used in a similar manner as in [TITLE REDACTED TO PROTECT THE INNOCENT], just in something longer and with more content.
The idea would be to give everyone who plays it a unique experience to share. I think nowadays there's a complete obsession with optimal play, what with every freaking game having a wiki and tier list dedicated to it, and it winds up with everybody playing the game the exact same way, which is boring as fuck. I think it would be nice to look into design concepts that personalize a person's experience with a game, and this is one that could work well in a short piece like a flash game.
That's not quite what happened here, since there's only two possible outcomes, but it was pretty funny regardless :V
PAUL GET IN HERE
There is a quite a bit to be said about having a consequence for failure. Many decisions video games try to present you with often don't carry as much weight as they could because they isn't much of a consequence involved. Half the reason a game like Uplink is any fun at all is because there is a relatively strong consequence for failure, at least when compared to other games.
Using this to create a more personalized gaming experience would make for an interesting game. Sure, not everyone likes the idea of not being able to see everything there is to see in a game, but it is certainly something that I would like to see more of.
You wouldn't necessarily have to be as harsh as only allowing a single playthough of course, something like automatically deleting one's saved game would be more than enough for most given a sufficiently long game, as games like Uplink and Steel Battalion have demonstrated.
Way back in the day when you played an RPG or dungeon crawl, you often could set the random seed prior to starting the game, which would dictate the layout of the dungeon/world. IMO this was a really cool feature that shouldn't have been taken away because sometimes you found one that you really liked or found challenging, and you could tell your friends "play it on seed # 5287 bish". Randomized games probably randomize on a per-level/room basis, but I think it should still offer the option of setting seeds.
I really like this sort of thing and there are games that continue to apply similar methods to other genres, such as AI War: Fleet Command with it's way of generating maps from number seeds.
On the other hand, I believe that such methods of procedural generation are only a starting point, as they are still very random and tend to create a sense of monotony.
Many gamers and developers alike often confuse "procedural" with "random". However, the point of procedurally generating content is to do better than just being "random". In many ways it's about getting a balance between getting what you want while allowing room for unexpected possibilities, overcoming the limitations of doing everything manually.
There is a lot of room to improve in creating methods of procedurally generating content that are more have more complex rules and create more varied and interesting results, and that involves going much farther than putting in a number and getting a level.
I think that the area where game development has a lot of room to improve is in creating games that are focused more on systems rather than scripting, in doing things algorithmically rather than relying on the developer to anticipate every possible thing that might happen.
Right now, game developers are really good at creating games that are either very scripted or very random. Most video games made today reflect this, as they are usually either very linear story-based games or very open and random sandboxes.
It's the ground between the two that is currently very difficult, as it means finding a compromise between getting what you want while allowing for the unexpected. This means that the game has to not only respond to a large number of situations, but do so in a way that is more meaningful and relevant than simply being random. It means that the game cannot simply go to the developer for instructions for that specific situation. This sort of thing can be scary for any programmer.
To use an example I've used before, it's like letting your child out into the world and hoping you've taught it enough to fend for itself. It's not impossible, but very difficult and rather scary to attempt.
I... daresay that the above approaches gross exaggeration. Not only are completionist gamers in their own (not all that large) niche in the gamer population, just as there are immersionist and explorative gamers as well, but, with modern design approaches and funding, we get games that can genuinely deliver a personalised gaming experience without having to resort to random numbers, in that manner vastly surpassing many older games in that regard. I am talking about the likes of New Vegas, Mass Effect and Alpha Protocol, which genuinely have varying content based on a player's input, and plenty of different, mutually-exclusive end-game outcomes that tie such input together into a widely-varied experience.
I don't think that games like New Vegas, Mass Effect, or Alpha Protocol really offer that much in the way of a widely varied experience, at least not to an extent that hasn't already been done before. More to the point, I strongly disagree that "modern design approaches and funding" have helped games deliver a personalized gaming experience. If anything I believe they've made things worse in making many assets more expensive and less disposable.
Of course, this goes into a lot of things I believe that plenty of people disagree with. :V
