The power to express?

[ninzel]

So as I'm playing the Wii and DS I'm seeing some obvious advantages to their controls,namely the ability to have them act as things like pencils and paint brushes much more naturally then standard controllers.
Which made me think of a game where you would raw your solutions in real time 3D. For example say you are in a level or a room, and you need to escape so you draw a set of stairs to a higher level.
Of course the game would have to be able to recognize that abstract(as far as it's concerned) object and then also give it physical properties so it actually becomes useful.
Are current system powerful enough to do even the these things?Or is this a matter of programming?
Edit: If not how would hardware have to be made differently to have this kind of power? This would seem different than the power to simply render pretty pictures ,but it would need the ability to recognize variables based I guess on the known real physical world.

 

[Capeta]

It could be done but you still would need to have all the possible choices preprogrammed. For example let's take a simple text game. If your text library can only recognize 3000 words then whatever word you write will need to be in that 3000 words database for the game to identify the word and use it.

For a 3D game the stairs would already have to be in the database for the game to recognize your stair drawing. There is already a game on DS that does something similar to what you suggest. It's that Pac-Pix game. There is also another DS game that takes your 2D drawings and animates them as a playable character. I forgot what that game is called or if it's even out yet.

 

[inefficient]

Okami says hi

 

[Capeta]

I'm not familiar with Okami but does the game allow you to draw things? Do you use the thumbstick to draw? Or is the game just using stylized rendering?

 

[inefficient]

One of the core gameplay elements is a "Celestial Brush". You draw things.

You hold down R1 to pause the game then paint your strokes with the L-stick. You can choose different types of brushes too. You can draw a sun in the sky when it is night, you can draw a bridge to create a crossing etc.

Just buy the game!

 

[Capeta]

Cool but drawing with a thumbstick is not very intuitive.

 

[archangelmorph]

Cool but drawing with a thumbstick is not very intuitive.

Not particularly no but Okami doesn't require the player to draw any sophistcated entities..

The subject of this thread is really gesture recognition which is the core of control paradigms such ad the DS stylus and the Wiimote..

The concept of gesture recognition is not new and was even done very well on the first Black & White game on the PC (and using the accuracy of the mouse as the input mechanism, it worked very intuitively..)

As far as whether or not a gesture-based game could truely be sophisticated enough to allow the kind of interactions the initial poster is refering to is much more of a question of design (and limitation) and coding over the accuracy/flexibility of the input mechanism.. The Okami example proves that practically any kind of analogue (or even a non-analogue) input device can be used for gesture recognition but the system itself would have many limitations..

Even using the Wiimote, a game trying to give the player the kind of expressive power suggested would have to constrained significantly in order to provide balance and a well-defined mechanic of play, otherwise the gane would be incredibly diffucult to design, build, test and predict..

 

[Lazy8s]

Whether an idea can be implemented or not is a matter of design. What performance affects is the precision with which it's implemented.

 

[inefficient]

Even using the Wiimote, a game trying to give the player the kind of expressive power suggested would have to constrained significantly in order to provide balance and a well-defined mechanic of play, otherwise the game would be incredibly difficult to design, build, test and predict..

Not to mention most normal people are just awful artists. Ever play pictionary?

 

[archangelmorph]

Whether an idea can be implemented or not is a matter of design. What performance affects is the precision with which it's implemented.

But then the design influences to what degree such precision actually matters..

For example, in Okami the player has to make simple gestures like drawing a line to cut something or a circle to draw a sun making the need of high precision input unnecessary to make the mechanic "work"..

However the gesture system in Black and White for example requires the player to make rather sophisticated gestures to draw complex shapes such as love hearts and strange glyph-like rune-shapes.. the use of a gamepad to try to emulate such shapes would only cause the player a great deal of frustration as the lack of precision would make the task much more difficult than it should be..

So it does in fact have a great deal to do with design over the performance of the input mechanism..

 

[ninzel]

Just to be clear what I'm talking about. You draw for example a cube, and on the fly the game engine recognizes it as a an object with physical mass and assigns physical properties to it. Not a preset object that the game already knows,but a random object that the game engine recognizes based on things like height,width,depth etc.

 

[Capeta]

In order for the engine to recognize it, you still need to be within the limits of what the engine could recognize. Actually there is already a game that does what you are describing, it's called "3D modeling and animation app".

 

[ninzel]

In order for the engine to recognize it, you still need to be within the limits of what the engine could recognize. Actually there is already a game that does what you are describing, it's called "3D modeling and animation app".

Of course the game app has to have some advanced knowledge,but I was thinking more of variables I mentioned instead of complete objects.

 

[MoHonRi]

Check out "The Room" demo by good ol Peter Molyneux. Not a game, but a cool demo.

MoH

http://www.gametrailers.com/umwatcher.php?id=27393

(there are some cool object creation things in there, but... once again, it's the user creating iconic representations of objects)

 

[archangelmorph]

Of course the game app has to have some advanced knowledge,but I was thinking more of variables I mentioned instead of complete objects.

To do what your suggesting is extremely difficult..

First of all how would your object synthesis system appropriate values such as weight, volume, mass and density just from a given set of dimension?

Plus there's the fact that drawing the object on screen would mean that the expression of the object is already incomplete since your limited to drawing in 2D.. So how does the synthesis system appropriate the rest of the dimensions?

Also how does the object synthesis system appropriate material information from form..?

Also how does it prediction the expressions of the user and draw destinctions between separate objects where the form is very similar (a window or a plane of wood for example..?)

Face it the limited descriptions you'd give the system are so incomplete that it would be near impossible for such a system to accurately understand exactly what it is your trying to express and synthesis it to the degree your expecting..

The only way to make such a system work (to a degree) would be to heavily constrain the expressive power of the system altogether and provide semantics for a well defined set of forms or form-attributes which could be understood and represented using pre-defined data models..

And like somebody has already mentioned, even this wouldn't work amazingly well considering the vast majority of people don't have even a reasonable level of artistic ability..

Which would mean your engine degerates into exactly what was done in Okami, expression by means of simple shapes (lines, circles, arcs etc..) which are translated into context-driven representations in the game world..

 

[ninzel]

I just though it would be a different type of discussion for this technology forum. To explore how technology could be used in new ways.
I figured coming here and listening to developers would be different than listening to the average forum since devs would have a deeper interest than most gamers. It turns out 95% of the discussions around here are graphics graphics graphics. This place is no different and as superficial as any other form,just more knowledgeable.

 

[archangelmorph]

I just though it would be a different type of discussion for this technology forum. To explore how technology could be used in new ways.
I figured coming here and listening to developers would be different than listening to the average forum since devs would have a deeper interest than most gamers. It turns out 95% of the discussions around here are graphics graphics graphics. This place is no different and as superficial as any other form,just more knowledgeable.

Don't get me wrong i'm not trying to shoot down your idea or imply that attempting such a feat would be completely pointless..

I'm merely offering an impression of the feasibility of the system you suggested based on my knowledge of the subject, whilst identifying the areas of greatest diffuclt and why they are so..

Maybe I should have given a more productive approach and offered possible solutions to such problems but unfortunately such solutions are currently beyond my knowledge of the subject matter..

I guess such a system *could* be designed and could work quite well.. However with every system intended to "model" an idea or solve an ambiguous problem, it must have strict and well-defined constraints imposed on it, in order for it to work..

 

[ninzel]

Don't get me wrong i'm not trying to shoot down your idea or imply that attempting such a feat would be completely pointless..

I'm merely offering an impression of the feasibility of the system you suggested based on my knowledge of the subject, whilst identifying the areas of greatest diffuclt and why they are so..

Maybe I should have given a more productive approach and offered possible solutions to such problems but unfortunately such solutions are currently beyond my knowledge of the subject matter..

I guess such a system *could* be designed and could work quite well.. However with every system intended to "model" an idea or solve an ambiguous problem, it must have strict and well-defined constraints imposed on it, in order for it to work..

Nah I was not knocking your comments. It was a general comment on the fact that this board seems to focus a large portion of it's dicussion on graphics as if that's the only thing technology can do for games.