To gian reto

Eye tracking is very handy in storing quick refference points, precise targeting, in , layering your tactics, dynamic control in afast pace battle scene. Mouse is not enough and very slow to cope in a face pace environment. With this kind of system you can command multiple troups easily. Because eye movement can move swiftly than dragging a mouse.and also its dorky to click with a blink of your eyes. you confirm your command using a joystick.the eyetracker is use to pinpoint the object/your target. Theres so many points to consider but this is only a few to say.
Eye tracking is not enough to stand on its, it must be acompanied by other sensors/controls.
My point is with the use of eyetracking system as an option for mouse cursor you give the player an interface to react quickly in a fast pace scene.

Well, eye tracking COULD be interesting if combined with other technologies that complement each other... like using eye tracking for enhancing immersion in VR Googles.

I am not sure thought if eye tracking can replace a Mouse or Joypad as main input device for games.

Because input devices are about precision and speed just as much as being intuitive and easy to pick up. Now, eye movement IS faster than moving your hand or finger. Is it more precise? I am not 100% sure. I can focus something quickly, and switch to another focus just as fast of course.

But on the other hand, with our current Human-Computer interface, our eyes are the main "Input device" of the Human part of the interaction. If they are now also used as output device, that might lead to lots of conflicts or decreased speed/precision.

As an example: In a shooter where I need to do a lot of leading, and weapons load quite slowly (lets say a ship-to-ship engagements simulator), I have placed my cursor just before the enemy (ship for example), and as the enemy is going straight, I can just leave the cursor in place in between shots. Because my weapons load slowly, I use the 30 seconds or so to check other parts of my interface for other information (is my own course still correct, or do I collide with a rock soon? How much Ammo do I got left? Are there other enemys in this area?)... the usual way to check this is to move my eyes to the correct UI element and have a quick glance at it.

Now, you could give the user the ability to "lock" the cursor in place while the player looks up information in the UI. You could move the UI Elements right next to the cursor and make the cursor ignore small eye movements (As I still need to shift my focus even if its only 2mm).

This is all going to be pretty awkward compared to just leave the cursor control to a mouse or gamepad/joystick, while freeing the eyes to wander over the UI to check whatever needs to be checked, while muscle memory might even give the player a limited amount of multitasking because with training, the arm/hand might be able to do its job without the player looking.

I am also not sure the korean RTS champions would be much faster with eye tracking... I would say they multitask a lot between eyes and hand, and loosing that might not be worth it.

Then there is also the problem of it being not overly intuitive. We humans are used to use our eyes to look at our surroundings, and our hands (or feet) to interact with it.

Using peripherals that need hands or feet to interact with it thus is quite intuitive.

What might be less intuitive is to use the eyes as a pointing device. Might be able to pull it off with training, but it will feel "wrong". And worse yet, not having the eyes free to check the environment. THAT might break the concepts neck. When your eyes have to pull double duty, while your hands and feet are idle, you will not be working at max efficiency.

I am not sure your player would be able to react quicker in a fast paced scene. It will take training to use the interface, and given the game is optimised for it (no UI elements besides the cursor itself), you MIGHT be able to be faster with training.

On the other hand, in normal games, your efficiency will go down.

Add to that the fact that most players are able to react quite quickly and precisely with a good mouse, and even with a gamepad given some training, and that you probably should try to design the game to suit players (thus fast paced scenes not needing inhuman input speed to get through), not players to suit games (could we all play better if we had 4 arms, many eyes and the brain power to use all that arms and look at multiple places at the same time? Sure).

I think you are looking for a problem for your solution, really.

I believe you in that point. Because people those days doesn`t realize yet how to use it effectively. But I understand its potential and I know very well how to use it effectively. Vr otherwise is great but very risky in producing such game in vr. You must have huge money and time to pull it off, not to mention competion. I say its impractical in marketing if your just starting and had little resources on hand. For me I start small in android phones wich has fronfacing cam then along the way move up to vr.

Isn't it even more risk to develop for a product that doesn't even exist yet?

It might take a long time, and cost a lot of money to create a new market, instead of entering one that already exists...

Do you have a working demo of eyetracking working on an android phone? If so, it would be awesome to see some of it.

I thought front facing cameras on general phones was too slow, with too low resolution, and all have IR filters making it very hard to get contrast and any kind of accuracy.

It might take a long time, and cost a lot of money to create a new market, instead of entering one that already exists...

Do you have a working demo of eyetracking working on an android phone? If so, it would be awesome to see some of it.

I thought front facing cameras on general phones was too slow, with too low resolution, and all have IR filters making it very hard to get contrast and any kind of accuracy.

I second that... create a basic prototype and demo your idea.

I have been pretty sceptical in this thread, and unless proven otherwise I don't think your idea works nearly as well as you seem to think. But give me a good demo, one that shows the strengths of your idea and where I can see for myself that you are on to something, and I will be happy to change my opinion...

I think there are a lot of people that are just as sceptical of new input methods as I am. Same happened with VR and the Occulus. If the Occulus team wouldn't have made huge efforts to produce new prototypes that solved the many problems inherent in the first one, people in general wouldn't be nearly as hyped for buying a device costing 600+$, needing a beefy PC to properly run while lacking any killer app yet as many now are.

The only thing why the VR devices haven't bombed YET is because the teams behind them (Occulus and HTC) did a fine job showing constant improvement and providing tech demos showing that it COULD work, given the right application.

Make sure you have your own demo ready to silence the sceptical people before you start hyping your idea.

Like I said yesterday eye tracking is use to establish an initial/starting point then backing up by a joystick for "precise control", eyetracking can not stand by its own it must be complimented by joysticks/controls.

I think this is a good idea. I think so, because I've also had it :)

I even think it could be useful and more ergonomic for general desktop use. Coarse control of where on the screen you are working, and fine control with the mouse. You'll never "lose" your mouse pointer again, since it always is there within view.

It's on my todo-list to make a demo to try it, because without a demo, it's just theory. It might suck when I try it out for real. Or there is some edge case or limitation I didn't think about.

Go make demos. You will never convince anyone with only saying "I know how to do it". Show it.

Demos are very powerful, and well made demos of key tech have started many great companies.