|
|
|
#1 |
|
Senior Member
Last Online: 04-10-2004 10:37 PM
Join Date: Sep 2003
Posts: 693
|
I just ran across this.
I know this is in a pretty primitive stage right now, but it has me incredibly excited for some reason. If it manages to gain a substantial foothold and the technology gets streamlined and all, this could very well shape the future of games.(Or it could go the way of the virtual reality headset) Imagine, if you will, an adventure game equipped with such technology. It could make the 'conversations' with the NPCs an altogether different affair. The challenge to the player would be to think within the gameworld to decide what to ask the NPC as no discrete choices will be available, the player's emotions coupled with the voice recognition technology (yeah) will drive the conversations. The NPCs will have sophisticated AI with their own set of emotions modeled using either the classic Finite State Machine model, or we could get adventurous (no pun intended) and make them Fuzzy i.e Non-Deterministic. The NPC will have either hard-coded or procedural responses to queries by the player depending on numerous factors, a few of them being the current state of the NPC (the NPC's mood), which is in turn determined by the player's previous queries. The NPC will have significant memory, it will remember the events in the game world (caused by the player) and its responses to the player and other NPCs will be dependent on the knowledge gathered There would, of course, be a considerable paradigm shift in puzzle design if this were to take place. Um...yeah so what do you guys think of this? Am I out of my mind ?
__________________
|
|
|
|
|
#8 |
|
Senior Member
Last Online: 04-21-2010 01:44 PM
Join Date: Mar 2004
Posts: 2,771
|
Heh, you're optimistic
No but seriously, I don't see this technology adapted to gaming in the next twenty orso years. This hardware/game-thing was made purely for scientific purposes (see it as a prototype), to see if this could be done. It's merely the continuation of a very old project. I believe they managed to make a skiing game some 10 years back in which you control the characters X axis (lef and right movement) while going downhill. The only (typical) drawback was that the machine had the size of an overweight snapshot-booth. The principle was that, if they could detect controllable thoughts (like thinking of a difficult math question, or spelling a difficult word), they could control a computer (game) with it. The MAJOR drawback was, however, and STILL IS, that the device is still only capable of 'detecting' either A or B. This technology works differently but also similair. There are boxes on the screen that are flickering at different frequencies, by focusing on either one of them, your game character will balance left or right. I don't know about you but I don't like flickering things in my games It's also an indication that this technology has very limited possibilities in gaming, but alot of possibilities for people who can't move at all. So if you'd want to control a game focussing on spots on the screen using your brain, it'll probably be a long time before we can detect that using a headset, It's more likely we'll someday use our eyes for controlling that aspect. In other words: I think this is the wrong technology for the applications you're implying. (What do NPCs have to do with anything at all, btw?)Personally, I find this alot more exciting: Nasa develops "mind-reading" system In short: they've found a way to detect what a person is about to say. This works with small sensors near the vocal chords or something like that. Anyway, the fun part is, it appears we can use it to "speak without speaking" (WITHOUT ANY LIP MOVEMENT) . One of NASA's goals with this is to make it available for every cell-phone, they actually see this being used in crowded public places. So far the software is only capable of comprehending six words and 10 numbers, but it picks it up 92% of the time, making it (IMO) alot more reliable than any vocal voice-recognition already. Ok, it is, in itself a form of voice-recognition, but capping the 'signal' somewhere in between the brain and the vocal chords seems an amazing accomplishment to me. If the technology is really this reliable, I don't why it couldn't be put in adventure games in the NEAR future! It wouldn't be all that hard I guess. It would be cool to see someone implementing like some form of keyword-recognition, for example: In an AG; if you need to know where an artist is performing tonight, you just say (or not say ) the following:"Hello, Do you know where G-blahblah is performing tonight?" Then the game would pickup the words "know", "where", the name of the artist and "tonight" among others, and based on that information, it should know what you want to know. Since most recent AGs seem to have vast amounts of dialogue, and most of which either completely irrelevant to game or exagerated into dullness, I don't see why they shouldn't make a line of dialogue for all possible story-related questions, right? Ahhh, just dreaming.. Last edited by jjacob; 03-25-2004 at 06:10 AM.. |
|
|