| Author | Message | ||
     Anonymous Rating: N/A |
Hello People: I'm am a flight sim guy (Falcon 4.0). I want HMD with head tracking that delivers these specs. 1. 60 degree horizontal FOV(Falcon 4 display FOV) 2. 1024*760 resolution (my monitors max resolution) 3. Approx. 3.5 Arc minutes resolution ( 60 deg / 1024 pix * 60 Arc minutes per degree) I can't spend $80,000 for it so I have an idea how to do it for less than $1,000 (WAG). I describe the design at my web page http://homepages.about.com/johncraigrochester/vrvision/index.html Please critique the design. Any suggestions would be greatly appreciated. Craig "Bricklayer" Rochester | ||
| Anonymous Rating: N/A |
Craig Leep Optics does a similar thing but with fiber optic bundles. Your optical mechanical device looks very complex and expensive. Also from the patent it looks like the device needs one leg to be fixed for the mirror to reflect correctly. So this would limit you to one per leg. Also it would limit you to one plane. The optical "elbow" looks like it rotates only 180 degrees. Just some thoughts Kevin | ||
| Anonymous Rating: N/A |
Thanks Kevin, Each joint bends from a relative angle of 45 to 135 degrees for a 90 degree range. Each joint twists 360 deg. axially. The mirror rotates 1/2 the angle change between each tube regardless of either end being fixed. I've mocked up the geometry of three joints linked together. I can fix one end and rotate the other end held normal to my eye. I can get 360 deg. Yaw, 180 deg. pitch, and 180 deg roll. I will have to do some fancy linkage work to keep the rig from binding. I've added a drawing to my page to illustrate. Thanks for the feedback. Craig | ||
| jtridexter Rating: N/A |
I am new to this message board. but I just wanted to say that I have been working on a project with the vfx1. my project involves using NO ISA board. I have been working on a Project that will allow users to fully immerse themselves into any shoot em up games with this body suit connected with the vfx1. and all body movements will be derived into your games. as far as tracking is concerned I have developed a new way to track movements using various Techniques. I am happy to say that the ISA card will no longer be needed instead it will be using an interface box I have created for this project. You have to see this shit I am working on it is really awesome! I will not give out any technical information because it is my own creation and none of my work is protected as of yet and also I should be finished in a couple of months so those really interested I will be selling. Imagine yourself running in your own livingroom and jumping and being able to interact that into whatever game you are playing. using a real full size machinegun in your hands and being able to point and shoot your enemy. you duck for real your ducking in the game for real you run for real your running in the game for real you jump for real your Jumping in your game for real, you get the picture....any questions email me at jtridexter@hotmail.com | ||
| Bricklayer Rating: N/A |
I am progessing in my project. I have included force/motion cueing into the scope. I have also changed my webpage address: http://pages.zdnet.com/johncraigrochester/ I'm still looking for someone to discus optical design for my jointed periscope idea. Later | ||
| Anonymous Rating: N/A |
Craig, I'm impressed. Looks very striaght to the point. Being a plumber once upon a time, its interesting to see the use of the copper. If I had a garage I might attempt to build one myself. Keep up the good work, Kevin | ||
| Bricklayer Rating: N/A |
Hey Kevin, I used the copper tube because my mig welder is down for repairs. I was surprised how strong the copper was. Later, | ||
| Anonymous Rating: N/A |
That looks pretty cool. I take it you would still need to use shutter glasses... I would think fiber optics would certainly make things less complex although for example a fiber optic scope cost something like $700 and I dont even know if that will have a high FOV, but I can see how a mechanical aproach would also allow you to add sensors for rotation of the joints (similar to the ones used in digitizing arms like the ones from immersion), which you can probably feed to a multiplexor/micro-controller having it process and calculate the orientation of the unit, with it connected via USB or serial port. | ||
| Bricklayer Rating: N/A |
I could either use shutter glasses(simplest) or two monitors and an optical path to each eye. I considered fiber optics early on. I didn't find info on FO for visual image transmission. I'm a mech.eng. so I can easily get my head around the mechanism stuff. Yea, the head-tracking is easy and very accurate on a mechanical arm. I've already got the inputs working via a joystick controller chip, three potentiometers communicating with the windows game controller interface. I've got this sending data the flight sim Falcon 4.0 through a Windows "shared file". I may want to improve the resolution (currently 8-bit) by a better A/D circuit. Later, Craig | ||
| Flint Hamblin Rating: N/A |
Hey Craig, I actually worked on a system like this a few years ago. I spent months of spare time on it. The mechanical head tracking is a great idea; very precise and low latency (you might want to look into digital encoders as well). The optics of transmitting an image down a small space like a tube is a bit more complex and not without some hurdles. The field of view when looking through a tube without any optics is very, very small. Try it. Get a 2" diameter tube, 3 feet long, put your eye at one end and hold it up to you monitor. See what I mean... Periscopes and boroscopes use relay lenses spaced at regular intervals that transmit the image at a 1:1 ratio down the length of the optical path. Also, they generally require an objective lens and an eyepiece. To transmit an image down a length of fiber-optic is easier, but it is expensive if you want high resolution. You can't use just any fiber-optic. You must use aligned fiber-optic cables. In an aligned fiber-optic cable, each individual fiber is in the same position on one end as it is on the other. The FOV is only limited by the lens arrangement that directs light onto the front of the fiber-optic, but resolution is dependant on the fiber count. For 1024 x 768 you would need 1 to 2 million fibers (the cables are round so you need more fibers then pixels). High resolution aligned fiber optics are very expensive! (On the HMD page on this site is a HMD by CAE Electronics that uses fiber-optics. Its patented.) The above is not meant to discourage. I still think it is a good idea. After much research, I just found that the optical design became very involved. However, I believe you could still make it for less then many profesional HMD's. Also, please don't use shutterglasses! Use the double monitors. Part of the beauty of this system is that you can overcome some of the visual artifacts that shutterglasses have. I hope some of this is helpful. Flint | ||
| Bricklayer Rating: N/A |
Flint, Thank you very much for the feedback. I have spent a lot of time studying optics. The closest optical system I've found to what I want is a submarine periscope. I've purchased two 60 deg. FOV Erfle eyepieces (surplus shed), and scrounged every lens out of every old camera, projector, video camera...that I've found at goodwill and trash picking. I get a decent image using a 50mm Nikon camera lens as the objective with the Erfle. This combo looks acceptable although the peripheral distortion is noticeable. I'm trying to figure out what the relays should be, what lens combos will minimize distortion, etc. This is not an easy subject and I expect someone in the field has many rules of thumb and boilerplate sets for specific tasks. I'm hoping I'll run across a lens master that'll point me in the right direction(s). I did look a FO and it looked the most elegant way to "couple" a remote image. I even found a design that used "micro-lenses" via a aligned FO bundle and the micro-lens sheet(? term) was concave to minimize distortion. Alas it looked wicked difficult and I assumed expensive. Questions: * What route did you pursue? * What were you biggest obstacles * I've never used an encoder. What are the advantages over a high quality POT. Could I get a 1/3600 resolution? * Any suggestions for eyepiece? Have you ever looked at a Nagler Wide FOV eyepiece? Thank you very much, Craig | ||
| Anonymous Rating: N/A |
http://www.ucapps.de/howtoenc.html I think the problem with POTs is probably Digital to Analog conversion, and filtering. I haven't really worked with POTs so I dont know for sure if it is a problem for them, but I have worked with accelerometers and I had a number of problems with noise that required me to reduce the resolution or I would get flickering/jittering. I think encoders as I have read from that web page, basicly send pulses when something rotates, and I guess the pulses tell you how much it has moved, the obvious problem with that is if you some how miss a pulse. I thought I saw another kind of encoder some time ago, that looked like it had binary rings, one ring would tell you which half of the ring it was on, and another ring would tell you which quarter of that half you were on, and they had more rings, and the more rings the higher the resolution, I'm not sure if that is also considered an encoder or not, but that one seemed the most interesting as there would be no A to D conversion at all... |
