This is how Milo/ Natal Works...

A.) CAMERA IDENTIFICATION

First the Camera (natal) is continually scanning the enviroment displayed to it, untill it finds a human posture it is programed to recognize.

When it find that posture the camera focuses at particular points of that being it detected (Going to try to diagram it)

1.) HEAD -----------------(*)------------------

2.) ELBOES ---------*----------*-------------

3.)Wrist/hands--*-------*-------*----------Abs as well

4.) Up. Legs----------*-------*--------------

5.) Knees-------------*-------*--------------

6.) Feet--------------*----------*------------

Obviously the points would change depending on your Posture. (Camera Tracks the points it makes)

In certain cases, body parts such as Fingers and Toes will also be designated point of interest.

In most cases the camera will set Dozens of minor points outlining your body and depth of facial features (to react to head movement)

B.) COLOR and SOUND RECOGNITION

After calculating what body parts are what, the camera then Scans the areas...

1.) Between and on the arms.

2.) On and above the Abdomen.

3.) Below the headline.

To conclude what color(s) your shirt contains, The Character (Milo in this case) Compliments you on the color of your shirt. (all in a few seconds)

----The Playstation Eye Camera (not to be mistake for the EyeTOY) Can also do these calculations. (Note the 6 year old Eyetoy can also perform alot of these calculations, But not as precise)

Now onto the Milo experience itself.

The Game (Project) itself is programmed to recognize certain Voice Tones and Statements the *Player* may ask, by using Natal's Microphone.

For instance if you ask "How was your Day" the game character would respond with "Good thank you" or something along those lines HOWEVER...as bloged about by some People at E3 this yearWho Asked Milo "Hey are you excited about the new PSP Go Sony is announceing?" Milo understood it was a question by the tone of voice, but did not understand the Words, He simply Nodded at this and stood still awaiting further interaction.

-----Obviously the Playstation Eye Camera has a Mic installed and Games and characters can be programmed to understand and perform exactly as Milo Did/Does.

C.) Now onto the Water interactivity.

At this point in the Demo shown at Microsoft's Conference, Milo asks you to look into the water, Your character automatically moves to the bridge and looks into the water without any help from you. --(One disadvantage of having no remote, though a Motion controller like the Wii-Mote is being Developed by a Third Party for the 360.)--

The game is now shifts focus back onto the camera to recognize Depth of the Players body...AKA if you move your arm back, your ingame character's arm will move back.

In this case the Camera uses the Game's Code to shift focus to your HANDS/Elboes/Head and Waist and tracks the depth/possition changes of your HANDS Depending on thier movement and the reaction that movement has on the Depth and Position of the other tracked points of interest relative to the Camera's Position...IE.

Your standing straight with Elboes Out and Hands in towards your Chest...In this position...

1.) Your head is behind and above your hands

2.) Waist Point is below and behind Hands/wrist.

3.) Elboes are relatively even with the hands and off to the left and right, below the head and above the waist.

Now before i get to the water, heres a little explanation on how Natal detects Depth and Rotation...

---Now that Natal has set points of interest and trace the body in Minor Points it sends out certain waves (Sorry dont know the exact wording here) Directly at ALL points, and then returns to the camera, depending on the wave legth and time it took those waves to return to the camera's detectors, it can dictate the depth and exact distance and rotation of the object(s) in micro-seconds and with extreme accuracy, this data is also stored and saved to the camera (and/or console/device used to store memory) for facial recognition (Milo knowing peoples name for example)...Back to the point!---

So with these set intervals, if you punch with your right arm to move the water, the camera first detects that your Right Hand is moving closer to the camera using Wave Length, while at the same time tracking your elboe moving IN and left...So the players character mimics the tracked motion to reflect this...Hope ya followed that...now onto the PLAYSTATION EYE/EyeTOY.

EYETOY:-  First off, we'll start with the Old EyeToy This device for the PS2, tracked Depth and Rotation through the Size of the object, For example if you had a baloon, at medium size, the EyeToy would track that as its default distance, if you moved the balloon closer the size of it relative to the camera would get larger, so the camera would know the baloon was getting closer, and vice versa...BUT!...if you INFLATE the balloon from its original size, the camera would still think the baloon was getting closer, and vice versa...This means the EyeToy has extreme limitations when it comes to Motion controlled games and its Rotation detection was just about non existant. Now onto to Playstation Eye.

PlayStation Eye:-- This uses similar, if not the exact same methods of Depth/Rotation/Facial Recognition as the Natal Camera and NOT the Size Recognition of the EyeToy. We know this due to the 3D dementional demo shown at E3.  (Primarily the Sword/Shield demo near the end of its presentation)  The Sword/Shield demo Showed the Character kneeling as the demonstraitor kneeled, Jumped when the demonstraitor jumped and ECT.  This+ the added benefit of the Two 1-1 wireless motion controllers, exceeds the possibilities of Natal (unless you own both Natal and the Third Party controllers)

PSeye may be more accurate when it releases do to the fact the camera is tracking the motion controller+Body movements at the same time and calculating them together for precise accuracy.

All and all, Both Systems now have extrordinary Motion Control and i am excited to see how they Work.

Hope you enjoyed the read, cause i enjoyed writing it!

Ssliasil.

This is how Milo/ Natal Works...

First the Camera (natal) is continually scanning the enviroment displayed to it, untill it finds a human posture it is programed to recognize.

---Now that Natal has set points of interest and trace the body in Minor Points it sends out certain waves (Sorry dont know the exact wording here) Directly at ALL points, and then returns to the camera, depending on the wave legth and time it took those waves to return to the camera's detectors, it can dictate the depth and exact distance and rotation of the object(s) in micro-seconds and with extreme accuracy, this data is also stored and saved to the camera (and/or console/device used to store memory) for facial recognition (Milo knowing peoples name for example)...

Ssliasil.

ok..first I have to admit I do not know how Natal really works (and MS did not tell anyone that bit so far..), but as a physicist who has worked for years with all the basic technology inside Natal I can try to explain why what you are writing is complete nonsense.

What we could agrree on is that Natal basically consists of

a) a strong infrared emitting diode

b) an infrared sensitive CCD-camera (essentially a lense in front of a 400*320 CCD-chip

c) a color-sensitive CCD-camera (again a lense in front of an RGB-CCD-device)

d) a directional-sensitive microphone of whatever build

e) a high performance processor/LSI electronics board to process all inputs

f) extremely clever software

g) there does not seem to be an ultrasound emitter being involved in Natal.

a) to d) are inexpensive off-the-shelf components, e) to f) is the very expensive stuff

So to your main points:

"First the Camera (natal) is continually scanning"

That is not possible, sorry. A CCD-chip accumulates electrons (generated by light of whatever kind) for a certain amont of time and then the chip is read out as an image into a frame buffer. In the case of Natal, it was mentioned that it is able to do that at 60Hz max. The net result of the infrared part of Natal is that it generates an infrared picture every 1/60th of a second (max) of whatever was illuminated in front of the camera.

"and trace the body in Minor Points it sends out certain waves"

No the infrared diode emits infrared light radially into space that is filled with whatever is in front of the camera (be that ontinuous light or flashes of light)

it can dictate the depth and exact distance and rotation of the object(s) in micro-seconds and with extreme accuracy,

The camera cannot do this for a simple reason. Infrared light travels at 30000000000 cm/sec. If you want information in the cm resolution range, you would have to take pictures every 1/30000000000 second, obviously this is completely impossible (as mentioned, Natal maxes out at 60Hz).

So how do you get spatial information? Apparently, here is where the software comes into play. All we have is picture information evrey 1/60th second (max), so as a rough guess (which culd betotally off) the software compares several frames (or more precise, the difference between consecutive frames) to detect motion of body pieces. The software might also compare brightness/size differences of reckognised elements (limbs) to extract depth information from delta frames.. but here is where my knowledge ends.

Sidenote: Your newspaper might have reported in the science portion that the University of Berlin has built and demoed successfully a device (essentially a cap plastered with electrodes) that works with registring brain waves to steer a human figure on screen. Works nicely, the only problem is the lag of 600-1000ms (which is an order of magnitude faster than anything built before). This would be the ultimate Natal system for the lazy, as it does not require any movement at all

Meh, not going to happen. MS is already planning on making Halo and COD playable on NATAL.

orly ? microsoft is planning on making cod playable with natal ?

1. How ? it's not their game.

2. Pew Pew Pew ?

So what?  Tiger Woods is not Nintendo's game, but it's still compatible with Wii Motion+.