With the varied skills of all gamers these days it is very hard to judge just how hard a "hard" mode should be, and thus, how easy the "easy" mode should be.

One of the better ways I've seen the combat this problem is that companies are now introducing a "dynamic difficulty" for want of a better phrase. This idea is in its infancy but in theory is a great idea. Games that have gone down this road that I can think of are Max Payne, if you are doing well then there are actually less health pickups around the maps to increase the difficulty, then if you die horribly the pickup amounts are increased making the game easier.

Now take that idea and make it bigger, enemies that use different AI depending on how well the player is doing, ie. They do/dont use cover, do/dont call for backup, do/dont charge blindly at the player etc.

But then what do you do if your a gamer who likes a serious challenge (ie. a very hard game)? Well you could get the player to pick an easy/normal/hard at the start and depending thier decision, limits can be imposed on the dynamic difficulty so that things never get *too* hard or *too* easy.

I think more developers may start using these ideas in games to come, as it is by far the most flexible. 

dynamic difficulty.

Take that concept too far and you end up with elastic band AI. I believe the man who invented it was dragged out into the street and shot in the face. 

if you think about it, criticising something because it's too hard or easy isn't really valid since it *does* depend on who's playing. (unless it's obviously too hard, like there's no health at all and the player is almost 100% certain to take tons of damage)

the only thing that should be criticized is if it's consistent. imagine playing a map where you gunning down grunts left and right and all of a sudden you get ambushed by multiple big monsters...

i dunno, does that make sense? *shrug* 

Take that concept too far and you end up with elastic band AI. I believe the man who invented it was dragged out into the street and shot in the face.

I totally agree.

And lets keep it simple.
A Game should have preset skill levels, at least 4 or 5.

Easy.
Medium.
Hard.
Very Hard.

Its the age-old method.
IMHO, it works perfectly.

end of discussion. 

People play on anything but the hardest? PUSSIES!

Kidding, but seriously.. I've always enjoyed the idea of multiple entry points into an environment, and depending where you enter, the difficulty varies - only you find out how much it varies - but also gives the map some replayability.

Always wanted to incorporate that into my maps when I attempted them. 

Yet I had some reports that it was too hard.

Go figure.

I'd rather make something that everyone runs through at least once though, rather than something that pisses 90% of the people playing it off & makes them quit before seeing it all.

Making things harder (and I can't remember if I did this in A3 or not) does not have to be just about adding more monsters - tweak ammo values, make players conserve, demand that they don't waste... 

Do games need to be focused on competition/challenge in order to be good? After playing thousands of maps/games, do you, upon seeing a new one, say to yourself "Woohoo, a new challenge!".

Maybe you do; I make no judgements. I still enjoy a good challenge once in awhile. But I'm finding that my satisfaction from a game is becoming less and less despendent on overcoming the obstacles within. It might be interesting to take a look at the demographics (genres, target audience etc) of games, and compare which ones use skill settings and which don't.

On a topical note, nintendo has been saying that they dislike how hard games are becoming, especially wrt to interfaces, and how they're trying to elegantize controls. 

I've always enjoyed the idea of multiple entry points into an environment, and depending where you enter, the difficulty varies.

This is something I'd like to try at some point - a two-map unit or somesuch. Now that I think about it, I may have had something like this in mind for my abandoned Q2 project. (Which I may pick up again after Ontranto; by that time, I'll probably be thoroughly sick of Q1 medieval!)

Let's consider that as an example. The idea was/is that you have to find a way into the base, and there are several:

1) The front gate (easy to reach, hard or worse; mind you, this is the only way out)
2) Air ducts (middlin' resistance, if you can open the dang things)
3) Climbing down a crevasse and sneaking in through a plant room or something (hard to get to, but easy at first)

At least, that's how I understand what Phait said. 

multiple entry points,
but one thing i'm dying to see in a quake map -- or any map for that matter, is different paths depending on difficulty...

Say, on easy, certain hallways connect to certain rooms, and on medium, some of the rooms on easy mode are completly inaccessible, in place there is another room which is even harder, and same for hard...

not sure how this would work out though... and players wouldn't be able to see the whole map without playing it multiple times on different skill levels... but it could still be cool... may even up the replay value... 

can't you give plain func_walls skill spawnflags? If so it would be very easy to do something like that. 

but imagine the work involved with effectively creating three seperate maps all curled up into one.