And lo On Tue, 08 Jun 2010 18:17:29 +0100, Darren New <dne### [at] sanrrcom>  
did spake thusly:

> Phil Cook v2 wrote:
>> Take Myst, what the heck are you supposed to do yet it was a massive  
>> sellout; was that despite its open nature or simply because it looked  
>> so good that finding tasks naturally occurred as you looked around?
>
> I think even sandbox type games need careful level design and planning  
> of guidance.  Myst is an excellent example.
>
<snip>
>
> So Myst starts with a set of simple goals once you take more than a  
> dozen steps, obvious limitations on where you can go and what you can  
> do, with hints telling you how to solve the puzzles without them being  
> obvious. In addition, the puzzles are logical within the game space,  
> rather than just arbitrary "you find a locked door, go hunt down the key  
> that won't be far away."  There's a reason for things to be locked, and  
> a reason for the key to be available.

Exactly quest leads to quest with a natural progression that affects the  
world around you. Things make sense, kind of.

> The individual levels of Thief also sometimes had a similar structure.  
> You'd have a main goal or two, that may or may not change as the game  
> progresses, but a big sprawling level with no hint as to which way to  
> go, where the secrets are, or how to achieve the goals you've been set.  
> Other than "steal lots of goodies", there weren't really any utterly  
> arbitrary goals there either.

Again yes 'This is what you're supposed to do, go and do it'. And again  
the world would alter depending on your actions. One annoyance is that you  
couldn't lock doors behind you to stop the AI path-finding :-)
<snip>

> Most of the other "Myst-like" games that came right after myst  
> completley abandoned the "open" format. There was even one called  
> "lighthouse" where you start in a house maybe a quarter mile from the  
> lighthouse in question. I spent 20 minutes trying to figure out how to  
> get out of the first room before I realized you had to listen to the  
> answering machine before the door would unlock (eh?). Then to get the  
> front door to unlock, you had to pick up the umbrella in the stand.   
> Then you couldn't walk to the lighthouse - you had to find the car and  
> drive there. Totally not what an adventure game is supposed to be like.

But it does highlight developer thinking. Imagine the same game and I'd  
devised a trap in that you were sprayed with a substance that drives the  
local wildlife into a killing frenzy. Get sprayed and you're dead. So you  
use the umbrella to stop it getting on you. Except the umbrella is back at  
the house you started at. So either I drop my wonderful trap (not going to  
happen) or I turn it into an annoying precog trap (I'm about to get  
sprayed, hey that umbrella back in the house might help) or I just stop  
the player leaving the house without the umbrella despite the fact there's  
no reason at that point for them to take it.

> So, in short, an open world with no hidden bits to find and explore and  
> challenge is just boring. There have to be places where it takes  
> thinking to get to them. And it's best if the thinking is real-world  
> logic. To say "this door is locked, let's find the key" is not nearly as  
> interesting as "this door is locked, but it's made of wood, and there  
> was an axe in the forest, let's see if I can open the door that way."

I'm not sure about hidden bits. Oblivion had plenty of those and it still  
got dull, InFamous had none and I loved it. I agree with the  
multiple-option 'door is locked', but that costs time and money to  
develop. Best to ignore the fact there's an axe and force you to go get a  
key.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message