The whole point of the FX is not for high FPS on older games(which you can't even tell the diff after 100 fps). It is about the multiple pixel shader fx and cg support. This chip is made to bring cinematic "FX" to the desktop. When newer games start supporting this, ATI will be far far behind when FX owners are able to play games that look as good as the Final Fantasy movie did. Just look on Nvidia's site to see what this means.
After watching Minority Report, I liked the idea of using gestures to interface with your computer. However having to wave your hands around like that would get tiresome real quick.
The most time consuming part of getting things done on a computer(aside from the software) is having to go back and forth between the mouse and keyboard. Even with keyboard shortcuts, it is unavoidable. I started thinking of other ways to use the same type of gesture interface but with your hands only. No keyboard, no mouse.
Muscle movement memory is very efficient. It only takes a few repetative movements to get used to a static environment. Have you ever stuffed envelopes? You get pretty efficient in no time. The reason a keyboard and mouse is not like this(mostly the mouse) is because its position is always different. Your hand has to find it. A keyboard is much better because once you get used to the layout, your hands pretty much stay in the same place.
So how does a gesture-based interface fit into this? What I envisioned was using only your fingers to do the gestures. To change tools, like from cursor movement to keyboard you could use finger movements or a combination of two fingers moving in a direction as a switch, or even lifting your hand higher. This interface would not require you to touch anything. Your hands could be anywhere and in any position. The hardware would monitor your finger and hand movements. You could be standing and resting your hands on your legs while doing it.
Imagine your hands are resting on a hard surface and you are typing, there would have to be a tactile feedback like the little dots on f and j on most keyboards that tell you where you are at. Maybe a range of motion field gets established in relation to your hand positions at that time. Also the hardware would have to provide this tactile feedback like sleeves on your fingertips or gloves. Once the area is set it would be easy to get a feeling where the keys were. Tactile feedback to determine a key-click would be important. When you need to switch to a pointer, you make a gesture with your fingers or hand(s) and fingers. Or you determine a position above the set keyboard space that is the pointer. Like moving your index finger up 2 inches above the keyboard field and using it as a pointer.
I know it wouldn't be as simple as that. It is just a starting point. I also can imagine if it were done correctly you could basically haul ass moving through windows, multi-tasking etc.
The current issues are we have a set area for our keyboard and mouse. We leave that area, we lose our interface. People move around, we use laptops, we like to keep our interface setup consistent when we switch computers. The mouse is never exactly where we expect it to be and is too far away from the keyboard. The position of a keyboard and mouse on a table in front of us is not always the most ergonomic or comfortorable positions.
Gesture interfaces are better because gestures are easier to remember. They can eliminate having a single area for an interface. They are more configurable. You can keep their configuration consistent for any computer you use. It is more comfortorable being able to put your hands anywhere and still be able to work. You could possibly customize the tactile feedback to suit your taste. Gestures can signify complicated tasks to be performed in an application.
Slashdot Mirror
The whole point of the FX is not for high FPS on older games(which you can't even tell the diff after 100 fps). It is about the multiple pixel shader fx and cg support. This chip is made to bring cinematic "FX" to the desktop. When newer games start supporting this, ATI will be far far behind when FX owners are able to play games that look as good as the Final Fantasy movie did. Just look on Nvidia's site to see what this means.
After watching Minority Report, I liked the idea of using gestures to interface with your computer. However having to wave your hands around like that would get tiresome real quick. The most time consuming part of getting things done on a computer(aside from the software) is having to go back and forth between the mouse and keyboard. Even with keyboard shortcuts, it is unavoidable. I started thinking of other ways to use the same type of gesture interface but with your hands only. No keyboard, no mouse. Muscle movement memory is very efficient. It only takes a few repetative movements to get used to a static environment. Have you ever stuffed envelopes? You get pretty efficient in no time. The reason a keyboard and mouse is not like this(mostly the mouse) is because its position is always different. Your hand has to find it. A keyboard is much better because once you get used to the layout, your hands pretty much stay in the same place. So how does a gesture-based interface fit into this? What I envisioned was using only your fingers to do the gestures. To change tools, like from cursor movement to keyboard you could use finger movements or a combination of two fingers moving in a direction as a switch, or even lifting your hand higher. This interface would not require you to touch anything. Your hands could be anywhere and in any position. The hardware would monitor your finger and hand movements. You could be standing and resting your hands on your legs while doing it. Imagine your hands are resting on a hard surface and you are typing, there would have to be a tactile feedback like the little dots on f and j on most keyboards that tell you where you are at. Maybe a range of motion field gets established in relation to your hand positions at that time. Also the hardware would have to provide this tactile feedback like sleeves on your fingertips or gloves. Once the area is set it would be easy to get a feeling where the keys were. Tactile feedback to determine a key-click would be important. When you need to switch to a pointer, you make a gesture with your fingers or hand(s) and fingers. Or you determine a position above the set keyboard space that is the pointer. Like moving your index finger up 2 inches above the keyboard field and using it as a pointer. I know it wouldn't be as simple as that. It is just a starting point. I also can imagine if it were done correctly you could basically haul ass moving through windows, multi-tasking etc. The current issues are we have a set area for our keyboard and mouse. We leave that area, we lose our interface. People move around, we use laptops, we like to keep our interface setup consistent when we switch computers. The mouse is never exactly where we expect it to be and is too far away from the keyboard. The position of a keyboard and mouse on a table in front of us is not always the most ergonomic or comfortorable positions. Gesture interfaces are better because gestures are easier to remember. They can eliminate having a single area for an interface. They are more configurable. You can keep their configuration consistent for any computer you use. It is more comfortorable being able to put your hands anywhere and still be able to work. You could possibly customize the tactile feedback to suit your taste. Gestures can signify complicated tasks to be performed in an application.