|
||
 |
Ati says | |||
| Has anyone had any experience on assembling head mounted display's? One idea I had was to use a fiber optic cable to carry images from a larger monitor directly to the eyes, thus avoiding the resolution problems. Another idea that came to mind was using hacked cell phones as eye pieces, as you can buy a halfway decent (if a little old) cell phone for a pretty low price, but HMDs tend to be very expensive. Anyone have any crit's, suggestions on these ideas? |
||||
| Total Topic Karma: 13 | - More by this Author | |||
 |
hx10 says |
|
||||
| I find the idea of interest to me, I think it would be really cool; but I have little experience working with hardware other than reading schematics and embedded firmware modifications; | ||||||
| - 26 January, 2007 | ||||||
|
Ati says |
|
||||
| I too, have a very limited experience with hardware, which is why the fiber-optics idea appeals to me, as it would probably be relatively easy to set up. | ||||||
| - 26 January, 2007 | ||||||
 |
eviljawdy says |
|
||||
| I see a few problems with the fibre optic idea first - their so large and the cones for the retina are very small (if that's what you meant). Your other idea, about using the displays from hacked cell phones, is a pretty good idea - but I see a few problems again (not trying to rain on your parade!). The video controllers within the phones would need to be hacked - so that you can utilise the whole range of the screens capabilities (resolution, pixel colours), so from there you can them write your own libraries to do stuff like putPixel() and then further that. It most certainly seems plausable - once I've sorted out my other projects, this might be worth looking into. Hmmm! |
||||||
| - 05 February, 2007 | ||||||
|
Ati says |
|
||||
| Well, with the fiber optics idea, my idea was to have a lense focus the light from the display into a single thread of cable, then projecting the light from the other side of the cable onto the surface of the retina. Does this make sense? As for the cell phone display, I actually wasn't talking about running the simulation on the cell phone itself, but instead using the cell phone to run a thin client and running the simulation on a computer else where. |
||||||
| - 05 February, 2007 | ||||||
 |
(Guest) Andrew Ayers says |
|
||||
| Actually, your best bet would be to use the viewfinders from an old VHS camcorder - everything is "pre-done" (lens, eyecups, etc), you just need to figure out the pin-outs to the NTSC/Composite input. You can pick up used camcorders from Goodwill or other thrift stores cheap - don't pay more than $25.00. Another option is to modify a VictorMaxx StuntMaster HMD (see my website http://www.phoenixgarage.org/ for more details on this) - these aren't bad, but they don't have great resolution (although still about as good as a cell phone display, albeit without the crispness). How about using TFT displays from newer model handheld televisions? Also, one could hack up a PS1/PS2 portable display screen to hook it up to VGA. Don't try the fiber optic route, unless you have a lot of time and/or money - back in the day (and even today), such HMDs are possible, and they are very lightweight, but cost insane amounts of money, mainly due to the optical systems involved - you can't just use a single fiber, you need a bundle of them (quality glass fiber, too, with optical properties for visible wavelength light), plus all of the optics to gather the light from the output device and then project it out on the other end. HMDs are not easy to build, but a ton of information is out there - look into the "old-school" VR hacking books, like Linda Jacobs "Garage VR", among others. If anybody is interested in further information, visit my site above, and pop me an email. I have tons of information (all the back issues of PCVR magazine, for instance), tons of books and reference material - and I can point you in a good direction. I hope that someday homebrew VR/AR takes off again, and I want to help facilitate this in any way possible... |
||||||
| - 05 February, 2007 | ||||||
|
eviljawdy says |
|
||||
| AR's coming on leaps and bounds, what with ARQuake, PacManhattan and the German Tron-Cycle game (although, those last two aren't as intense a proof of AR, but hey, worth mentioning). Hadn't thought about using old camcorder viewfinders, that's an excellent idea - and chances are you could get the manuals, or some other "hacker" has already released details of pin-configurations. Some small TFT would be good, and if AR was chosen, then you could get one where the backlight was damaged, as you'd have to remove it anyway (well, you don't _have_ to, but to get semi-transparent lenses you would need to). |
||||||
| - 05 February, 2007 | ||||||
|
Ati says |
|
||||
| the viewfinders might work, but the cell phone idea has some appeal for those not willing to go building a full computer-viewfinder interface (with the phone wll it would require is software modification). But it's a neat idea none the less, and it could be very handy for someone with more hardware experience than myself. | ||||||
| - 05 February, 2007 | ||||||
|
Ati says |
|
||||
| Incidentally, does anyone know of any programming languages that work well on windows mobile? I've been thinking of trying to set something like that up with my smartphone. The screen is big enough for two eyes, and if I could get it sending camera data to my PC and receiving screen data from the same, I'm done. | ||||||
| - 06 February, 2007 | ||||||
|
eviljawdy says |
|
||||
| I've played around with different bits of hardware, and the one's that always throw me are the low-level graphics sorts of stuff. Like screens or similar. Sounds like I may have to pr0d certain people I know, to get their input.., hmmm |
||||||
| - 06 February, 2007 | ||||||
|
eviljawdy says |
|
||||
| I always thought windows mobile ran under something similar to Win32, and simply used C/C++... but could be wrong. Having to say the "evil" words here, but "J-a-v-a" could do it (*shudders* at the fact I said that word, that's a type of coffee and a programming language), as Windows CE/Mobile has a relatively complete Virtual Machine. |
||||||
| - 06 February, 2007 | ||||||
|
Ati says |
|
||||
| I used some java back in the day, and I have a passing familiarity with the syntax. Unfortunately, the smartphone I have won't compile java for some reason or another. Fortunately, I looked into it and there are free C++ compilers for Windows Mobile. Looks like I'm going to have to lean C++. |
||||||
| - 06 February, 2007 | ||||||
|
eviljawdy says |
|
||||
| Not too difficult - if you've done any C at all, then moving to ++ is not all that disimilar. Check out http://www.cplusplus.com/doc/tutorial/ for some good C++ tut's. And if you want graphics, you might want to check out some of the OpenGL modules that have been written for smart phones and embedded systems (OGLES - OpenGL for Embedded Systems), or if you're not _yet_ wanting uber 3D stuff, then try SDL (Simple Dynamic Media Layer) as it's easier to initialize from the off, and can do nice BMP loading with just a few lines of code, with relatively low overheads. |
||||||
| - 06 February, 2007 | ||||||
|
Ati says |
|
||||
| Well, for scripting the environment itself, I've already got several programs for that sort of thing - 3d game programming is something I can handle. Basically, all I need is enough knowledge of C++ to write a pair of programs that grab the screen from a computer and transmit it via a wireless network to my smartphone, and which take the camera data from the smartphone, and send to aforementioned computer. Unfortunately, I haven't used C or any variants before (unless you count Java, which has somewhat similar syntax I think). I had a look at the tutorial you provided, and it looks very useful. I suppose the only thing left is to take a deep breath and go for it. |
||||||
| - 06 February, 2007 | ||||||
|
eviljawdy says |
|
||||
| If you've done Java, then you should have no problem moving to C, it does have very similar syntax - not the same, mind you, but similar. And if you're going to do the engine on a different machine, and then display it on the smartphone - you may as well just take screenshots, and transfer the image data. Would need a lot less bandwidth that way. |
||||||
| - 07 February, 2007 | ||||||
|
Ati says |
|
||||
| Well, that's essentially the way I'm planning it; except that I'd also want to send data from the camera as well (I'm planning to use the camera data to work out hand position, etc. Another thing that came to mind is that I also have access to a last-generation laptop I could use. This would be better in terms of resolution, and I could write the code for it in BASIC, but the only thing is that it might be prohibitively heavy. |
||||||
| - 07 February, 2007 | ||||||
|
eviljawdy says |
|
||||
| Have you seen the ARToolkit? This might suit something you're trying to achieve - if you're planning on using the camera and the screen to do interesting things. http://www.hitl.washington.edu/artoolkit/ It's open source, and has a lot of impressive uses! |
||||||
| - 08 February, 2007 | ||||||
|
Ati says |
|
||||
| Thank you, I've never seen that before. This looks like it could be astoundingly useful. |
||||||
| - 08 February, 2007 | ||||||
|
Ati says |
|
||||
| You know, just thinking about it, it seems to me that it might be possible to interface with the body's optic system directly, instead of mucking around with a screen. The optic nerve is pretty well sheathed in skull, but it doesn't really matter anyway, becase unlike standard nerves it contains something like 1.2 million nerve fibers, which would require an electrode density that is impossible to provide at this time. This leaves two other convenient points of interface: The eye itself, and the visual cortex of the brain. because I'd rather not go mucking around with electric current in the brain, lets have a look at the eye. The best point to interface with the eye would be on the surface of the retina itself. Rather than trying to project light directly onto it (as in retinal projectors), let's have a look at using electric current. The retina is quite small, so what we're essentially going to need is something that looks very like the electrical components for an LCD display, except using the current to alter the state of the retinal cells rather than an liquid crystal substrate. In this case, it becomes a simple matter to use a hacked LCD screen, feeding the current into the skin around the eye, and run the current through the retinal cells to produce images on the eye. Does anyone have any thoughts on this idea? |
||||||
| - 11 February, 2007 | ||||||
 |
p0ss says |
|
||||
| Ati, that last idea, i am not letting anywhere near my eye. I have been following retinal implants, and optic nerve interface systems, i still think that headsets are the best bet, a tip for the wise, check out icuiti. I wish i could comment further on this thread but i currently have a patent pending on a system involving hmd's and i can't disclose too much. |
||||||
| - 11 February, 2007 | ||||||
|
p0ss says |
|
||||
| i tried to include links, but they didn't work, so here: http://www.icuiti.com/ http://www.stereo3d.com/hmd.htm http://www.eye-chip.com/ http://www.2-sight.com/index.html |
||||||
| - 11 February, 2007 | ||||||
|
Ati says |
|
||||
| Well, the last idea wasn't so much a practical idea at the moment, as a possible project for some time in the future. Certainly not to be undertaken without considerable consultation with a doctor. The Icuiti thing looks interesting- fairly cheap for the resolution. The other thing that occurred to me is that might might be possible to retrain the brain to interpret tactile impulses as visual impulses (There's a device like this currently in existance for the blind that uses the tongue). In this case, it would be a matter of transmitting visual data into a nerve on the skin, while showing the eyes the same signal, and training the brain for a while to reinterpret the tactile data as visual. I'm curius about this patent of your. Be sure to send me a summary of it when it clears. |
||||||
| - 11 February, 2007 | ||||||
|
eviljawdy says |
|
||||
| My understanding of patents, is that once you have one pending, it doesn't matter if someone else tries to file one, as you'd already have yours date-stamped - but i could be wrong! Anyways, the stuff about retinal stimulation that I've seen already has been "interesting" to say the least. But this sorta stuff is kind of getting beyond "homemade" an into the industrial/theoretical ranks... far beyond my plans/expectations/skills/anything-else-that-lowers-me  |
||||||
| - 11 February, 2007 | ||||||
|
p0ss says |
|
||||
| Eviljawdy, even once i have the patent, i still wont disclose trade secrets or component information, both of which would help on this thread :/ | ||||||
| - 11 February, 2007 | ||||||
|
Ati says |
|
||||
| Your right EvilJawdy, once you've got a patent pending, another one can't be filed because you've got the earliest time stamp. But if the patent doesn't clear for some reason, you've got an open playing field if someone else wants to submit a similar patent before you do. Likewise, patents aren't universal protection, and I think our friend P0ss here is worried about uncrupulous people taking his ideas and using modified versions of them, whitch the patent system doesn't protect against. |
||||||
| - 11 February, 2007 | ||||||
|
Ati says |
|
||||
| And yes, if your going to do retinal projection or similar, it would probably have to be as part of a major corporate venture with a significant amount of money and resources behind it. | ||||||
| - 11 February, 2007 | ||||||
|
eviljawdy says |
|
||||
| Aaah, I see - is there anyway you could give help/advice without disclosing too much or any sensitive information? Or is that the whole point about your patent, that this subject IS the application? | ||||||
| - 12 February, 2007 | ||||||
|
Ati says |
|
||||
| Well, I guess well have to wait until it clears, then he can send us the public release of the patent. | ||||||
| - 12 February, 2007 | ||||||
|
Ati says |
|
||||
| A though just occured to me. I wonder if there is some way you could pattern a tactile signal running through a skin nerve to look like its a visual signal... Perhaps you could convince the tactile center of the brain to relaize that this wasn't its field and re-route it to the visual cortex. just a thought. |
||||||
| - 21 February, 2007 | ||||||
 |
Nadeem says |
|
||||
| I doubt that would work. I remember reading somewhere that the input neurons for the visual cortex are directly connected to the optic nerve. Tactile sensations never get there. |
||||||
| - 21 February, 2007 | ||||||
|
Ati says |
|
||||
| Yeah, that's true. I do know that its possible for other kinds of data to be re-routed to the visual cortex though; I think there was some product for the blind they were developing using the tongue, and the recently blind said the impulses eventually just started being treated as visual. |
||||||
| - 21 February, 2007 | ||||||
|
|
