Video Game Physics, Controls and Rules

[Terminal Devastation]

I think I might like this discussion.

Controls: I have a agree with a lot of what Koh said here. The more instantly responsive a control is, the better. They shouldn't be confusing either. When I hit the attack button, the character should attack predictably. Sure it would be neat when the attack is slightly different between two, four and six pixels off the ground mid jump, but it shouldn't suddenly change from a screen clearing bullet spam to a useless taunt to a simple jump slash in that amount of precision.
I'd expect them to be much the same function, maybe a different visual for flavor, maybe slightlydifferent power to allow more skilled players to eek out some more efficiency out of every move, but ultimately I should be able to use it to achieve the exact immediate goal I want it to, and expect it to as well.
Nothing is quite as frustrating as knowing a character has the exact move to make a difficult challenge slightly easier, and most of the challedge is just inputing the commands just the exact way to activate the ability in the first place.

Physics: Really, I'd have to argue its not about what the ideal physics are, but what the ideal use of physics are. Floaty can be good in certain games, as can lead weight. But the games have to be balance to take advantage of whatever physics are used, not force the player to fight them at every turn. In my experience, I only really notice how the physics in a game differ before I get used to them if they are significantly different from what I'm used to, or when they start to interfere with how I'm playing.

For example: Floaty physics are pretty good when you can take advantage of the extra air time, like when you're getting off a few extra airborne shots, or need to course correct mid air a bit more. I quite enjoy them there, but they are quite annoying with precision platforming on small platforms, or low deadly cielings, as I tend to over jump. More weighted physics have the opposite problem, once I jump I pretty much have to commit, and it's a lot harder to perform certain jumps, like trying to jump around to the top of the platform above you. But it does lend itself to more precision jumps, and is quite useful for dodging shots, as you can get to the ground faster after each jump to prepare for the next one.

If the game is nicely balanced, more of the like/dislike of a physics set isn't really whether the physics are bad or not, but personal preference. Some like them more floatly, some don't. When the game ISN'T balanced well for the physics set, it's a frustrating mess, and creates a lot of artificial difficulty.

Again, game physics are something a player should use not fight against.

Rules: Now certainly every game of a decent length should have some sort of save. The issue with saves comes in the relation of saves and difficulty.
I'll admit, some of the challenge of a game is completing several challenges in a row. More saves break that up into smaller and smaller challenge sets, making them slightly easier, and taking something away from the challenge. I think we've seen or experienced what too many saves can do to a harder games difficulty.

To a point.
On the other side, parts of certain challenges become mastered by the player. Now if the mastered section is small, it's not that much trouble to go through time and time again to get to the harder challenge and attempt it again.
But I'm relatively sure that we've all experienced at least one longish level that we defeat over and over again just for another attempt at its difficult part. Eventually, most of the level becomes a frustrating time waster rather than part of the challenge. Sure if the rest of the level was fun it takes some of the frustration out or at least delays the how long it takes before it becomes a pain to slog through one more time again, but eventually we just want to get on to that one section we want to beat. More saves help to prevent this.

My solution for a more ideal method? Set-up checkpoints logical (mid-level, etc) but then set up dormant checkpoints next to especially difficult parts of the level that only activate after reaching them enough times while under some sort of qualifying conditions that prove you've pretty much mastered the previous parts and it's only this one part thats giving any consistant trouble anymore. Like say, no-hitting the entire stage up to that one part 10 times before dying past the hidden checkpoint.

I'd certainly be happy if suddenly a few games did this. Of course, just well placed checkpoints and saves would do this just as well, but sometimes you do want to group up the challenges for one bigger challenge.

As far as I'm concerned about difficulty in general, I care more about how difficulty is achieved than how hard it is. Give me a brutal game without artificial difficulty any day. I might not always set said brutal game to the hardest difficulty (I tend to stick to second highest or one higher than normal to start. Higher if its too easy yet or I 'm wanting the challenge), but a lack of artificial difficulty should make even the easier settings more fun. Artificial difficultly can go take a hike.
I personally consider super precision jumping/positioning artificial difficulty, mainly because they tend to make games closer to DDR or a rhythm game than what the actual games genre is (obviously, if they ARE DDR or a rhythm game this doesn't apply). Only one winning set of commands is just annoying. May only be one way to win in some games, but at least you can jump and miss a few more times if you'd like.