Hey All,

Yay.. My LCD/Board arrived the other day. Can't wait to start "cutting" ^_^

Just curious does anyone have a recommendation as to what voltage can be put into this unit? Weill a standard 12VDC wall-wart work?

Thanks.

Hey, never heard anything from this post so just retouching.

Sorry for the long wait on replying, I have been at the IEEE VR 2012 conference. Great place, great ideas, great people, not so great for getting time away. Will try to find time to reply tomorrow, expect a massive infodump at the end of the week.

heya PalmerTech,

IEEE VR 2012 sounds pretty awesome. I've had the pleasure of attending various Siggraph conferences, in the past, but I'm sure something more "VR" specific sounds even better.

Needless to say, take you time with the info dump. ^_^

Cheers.

--Mike

Has this thread died?

I plan on having a kit available by June at the latest, just for the people on this forum. Cost would be between $300 and $500, depending on if I include things like a head mount.

Mbogucki, here is a picture. You can cut along the red lines if you want to keep VGA, and also the yellow lines if you do not need it. You can actually trim more than this, but you risk killing the board, and there is no point trimming it a few extra millimeters, IMO.

Do you know what Resolution and FOV your kit might be...? Just when I thought my HMD choice was down to two (Sony, ST1080)...A Big FOV would put you over the top of the other two I think.

I LOVE the idea of a kit..Feels like when computers and other cool stuff would first get to the diehards/hobbyist in the form of a kit.

Hi PalmerTech,

Very Cool. Thank you very much for the reply. Hopefully within the next few days I'll give it a shot... ^_^

I'm looking into seeing if I can get the actual driver chip, (initially as a sample)... to develop a smaller VGA only type board for this particular display.

Thanks again.

--Mike

Depending on how the next few months go, I might be buying two, or I might be eating from bins.....

What will be in this $300-$500 kit?

At the very least:

1) Cut lenses and assembled optical units, one for each eye

2) Display panel/s

3) VGA display control hardware

4) Easy to assemble display enclosure

5) Any required video/power cables

Depending on how things work out:

6) Adjustable head mount

7) Light blocker for eyes

8) Camera for seeing outside world/AR

9) Motion tracker

10) Integrated headphones

Anything I forgot, or input on what should be included? I think most people will want a head mount, for example, but no reason to include it if enough people plan on making their own stuff (Like has happened with the Sony).

Great work Palmer.

Please keep us updated.

Hey Palmer!

I have (silently) followed your progress on the PR (and your career, congrats on that!) and i'm more than impressed! You really got me psyched again for VR!
Theres barely anything left of my own build (just the shell), so i keep watching this forum as a possible customer of yours and wish you all the best!

Greetings,

Sin

Thanks for the info Palmer and looking forward to it. My 2 cents.

- Head mount: The reason everybody is modding the Sony is because those guys didn't design the unit for VR in the first place. I trust your design instincts with regards to a wearable, usable VR unit, and I think you would hit the mark much closer than Sony did for this audience. I think it would be a great idea to design in the head mount.

- Light blocker. I think everyone is going to want this since your unit is firmly in the "total immersion" category.

- Camera: It would be fun to play with, but I don't think it's a priority feature.

- Tracker: I'm 50/50 on the head tracker as it seems that there are multiple options and people might have their favorite solution for their specific needs.

- Headphones: I think the audio should not be a priority since everybody is going to have different solutions for that.


I do think that an important feature will be proper driver support for 2D so that basic desktop navigation can be accomplished and also non 3D-supported applications can be used. But since this is entirely a software issue, it's probably going to require some type of collaborative or 3rd-party effort to solve it.

It was not designed be wearable in any kind of scenario

I thought the welder style straps on the last prototype I tried worked pretty good, especially if the display/lens/housing has a lot of weight (more than 1 pound). If the unit is light, then I think a fabric strap, ski-goggle style, might be fine. Really, whatever works best, but it would be cool if it came included. I guess you can make it easy to mod or have a special bare-bones unit for modders.

Still holding onto my money for a PR2! Sony and others will be diddling in the realm of that sort of half-assed, Joe Blow learning curve, non committed flavor of VR for a few years yet. The thought of being fully immersed (or at least very close to it) in a Palmer unit is extremely exciting!

* Head Mount: YES PLEASE!
* Head Tracker: YES PLEASE!
* Headphones: Not a priority as most users will have options for that at home. Double so for those who rely on integrated mics and such.
* Light Blocker: ABSOLUTELY! When I put on a PR2, I don't want any part of normal reality interfering with my suspension of disbelief.
* AR Camera: Not really interested. Not that there aren't practical uses but I'm in this for completely synthetic realities.

Thanks for continuing to push this forward Palmer and do let me know if you need graphic/logo/branding help when you get closer to the mark Hell, I'd work for a straight across barter/exchange. I will of course have no issue paying full price either! This is the closest thing to actual VR most of us will have the privilege of experiencing unless research labs and corporations suddenly get a clue

120° FOV is amazing, but won't 800x600 look really blocky stretched out that much? Are you using microdisplays? Any news on a possible lens mod for the HMZ or ST1080? And what about that 270° FOV prototype you have or were planning?

Thanks for the feedback and encouragement! I am hoping to keep it under 1lb, and use a ski-goggle style mount, but with an additional over-the-head strap. If it gets too heavy, then it would be either a ratchet style mount, or something along the lines of a VFX1 shell, but made out of flat pieces of laser cut acrylic that can be slotted together. My goal is to design the HMD as minimalistically as possible, no air space like the HMZ-T1 has. That way, if people want to make some other kind of mount, there should not be any need for dis-assembly.

Another idea I had was to have some recessed portions on the front and top of the case, and put 4 lightweight metal discs in them. Then, have 4 wires going through the inside of the case soldered to the back of the plates, and route those 4 wires to the USB connection for HMD power. That way, it would be easy to make a detachable camera module, or a detachable tracking module, or whatever you want to stick onto the head mount. Just use small magnets on the module, they can act as both electrical and physical connectors.




It does not look as blocky as you would think, especially when I use a diffusion filter to blur the pixels a bit. I have actually been considering two versions, one with very high FOV, and the other with less FOV (Around 80 or 90 degrees) that makes the resolution look a little nicer, and does not cut off the edges of the display, either. Not as immersive, of course, but it might be a tradeoff some people want to make. My goal is to make the optical assembly easy to remove and replace, so perhaps both sets of lenses could be included so you can choose for yourself.

I am not using microdisplays, it is just too hard to get a large FOV/exit pupil out of them. I have some lenses that can get a higher FOV on the HMZ-T1, but they are not as high quality as the original optics. If and when I find a good way to improve it, I will post for sure.

As far as the 270 degree prototype, it works perfectly well. Problem is, it needs 4 offset views rendered for it to work properly! No driver support for that in any commercial games, so it is only useful as a research unit at the moment. I could just stretch out the edges of the image, but that makes movement in the periphery greatly exaggerated, and that makes people feel a bit odd. I have a few ideas, and whatever kit I make will be upgradable to whatever future optics I come up with.

This is all coming together, very excited!

Extreme FOV's are cool but... But are there currently any games supporting them? Most games have FOV 70, and some have variable FOV up to 90, I dont even think Quake had higher FOV settings than 120. Also there is the problem with huds, even with high FOV settings the HUD stays in the corner of the screen.

The sad part is I think it will take the Game industry several years to support high FOV, its a bit of a moment 22 problem. The display/HMD manufactures would produce high FOV HMD's because there is no game support and the game indusitry wont produce any games woth high FOV support because there are no high fov units around :/

Yeah but you can run a 90 degree FOV on a 180+ plus screen... it will just be stretched.
Bugger the HUD's, just kill the immersion really. I've noticed most games nowadays have tried to get away from the HUD concept and torwards more 'realistic' type effects like seeing blood on your viewport and your character moving slower when wounded, instead of just a health percentage etc.

Really however, I think modern VR will require dedicated games (at least at first) since most PC games are designed from the start for screen, keyboard, mouse. Such things as being able to turn your gun around in your hands, move it around anywhere, etc, add an immense amount of immersion. Maybe it will be someone from mtbs who will write the first killer game for VR that will spark the whole ball rolling!

Well, it will look really weird with super fat characters which will kill the immersion But yeah, the hud can be killed, for example I play hardcore in BF3 because I like it how the crosshair and HUD is not present. Dead space had a cool solution were the 'hud' is part of the world (Holographic projection in front of the character)

edit: Also if you stretch a 90 fov image over a 180 fov display you will loose a big point of high fov, the situation awareness, 90 dregrees is small when you see the entire content on a normal screen, it will be even smaller since you will now only see a small portion of this in focus, rest will be seen through the peripheral vision.

The 180 was just an example, 120 would be more realistic, which could be a viable stretch. It definitely wont look real, but may be workable, and you may still get some immersion due to the high FOV. The situational awareness bit will just make you more paranoid and turn your head more. I'd rather lose the ability to see the edges clearly, and gain greater FOV personally. Of course you won't win any online tournaments with that. There is however always the technical possibility of rendering the game in a lower resolution and displaying it centered in the higher res display with black borders, so you lose FOV but gain edge view if such is important to you.

With stuff like 3D Vision Surround / Eyefinity being popular, I expect more and more games to support higher FOV modes.

Some games already have high fov hacks, for widescreen/multimonitor setups. If they will work with 3d drivers, free/glovepie, haptics drivers and whatever else I have no idea.

2) Display panel/s
PanelS? Got other ideas for the display?

5) Any required video/power cables
Always nice to have included, but power cables might not be any use for people in other countrys.

6) Adjustable head mount
If it is something usable, great, but if it is going to be too expensive to include something good it would be better left out altogether.

7) Light blocker for eyes
Definatly something I would want, but maybe not need to be included.

8) Camera for seeing outside world/AR
Not intrested in this. Others might be, make it optional?

10) Integrated headphones
I would probably rather use my own headphones, as I would want something that could give very good sound positioning with the right software, and that would likely drive the cost up too high.

I can't imagine how VR players and "screen players" could ever compete against each other. It seems like the screen players just have a huge advantage in terms of control responsiveness and energy expenditure. And the more realistic you make the VR players control scheme the more disadvantageous it comes. How could you compete against someone that just gets to sit on their ass and flick their hands while you have to spin and jump around?

In your experience, where is the inflection point between "really large screen" and "being inside". Personally I am excited about the possibility of the immersive experience, so if the cutoff lies somewhere just above 90 then I might not want to make that compromise (resolution be damned). In other words: is there a major perceptual shift between 90 and 120 degrees or is it just an incremental change?


Edit: Also, just out of curiosity and so I don't have to dig through all the posts again - can you tell us what your ideal screen dimensions are for this type of rig and lenses. Or more to the point: if you could order an exact screen size, resolution, and interface to build this type of HMD with, what would those specs be?

Probably not for this revision, just keeping my options open.

I don't think there is a major jump at any particular FOV, it is an incremental improvement for sure. More important is good head tracking, and getting the in-game FOV to match the HMD FOV as closely as possible.

If I could order a hypothetical perfect screen, it would be 128mm wide, resolution as high as possible (Though 1080p might be a good limit, since that is the max that current wireless video transmitters can handle) with an HDMI interface. That 2560x1600 6.1" panel that Toshiba has been showing off is the closest thing to perfect out there, but it is not being produced yet, much less with a usable computer interface.

Wondering what the headphones would be like if you included them? Would they be suitable for binaural positional audio? Does the quality of the headphones even affect that? The reason I ask is that binaural recordings always seem to be behind or inside my head. Never in front. Is it me, the same as some people can't see 3d, or is it the poor quality headphones I have been using?

So it sounds like there could be a phone screen out there that would come close to the ideal, right? Are those only available via OEM direct to the manufacturers? The other problem is of course the interface - one part electrical circuitry and one part software driver. So I wonder if there any phone screens that are
a) available for individual purchase
b) designed to be paired with a graphics chipset that can also be purchased off the shelf
c) has andoid/linux drivers for it (or has enough spec info so that a driver could be written)

I know this is much more complicated than the all-in-one solution, but it would be great if some type of group effort could assemble all the parts to make an even better HMD.

This is sounding really good, I am certainly interested in getting one. I'd probably rather the really huge FOV one, even if it looked blurry or software support was lacking. There are already going to be decent options for moderate FOV like the Sony or SMD. I'm not sure how high the FOV was on the last prototype I tried, but it certainly gave a more immersive experience than anything else I've tried (even the Sony). It was great, at least in terms of immersion.

Yes cybereality, I think big FOV is what this HMD is all about. I don't think any other HMD will have that anytime soon if ever.

Just because you used it and liked it, I would like to know also what FOV you were using cybereality.

@3dvison: Honestly I don't know. It was for sure over 90 degrees but I'm not sure by how much. Maybe Palmer will chime in...

This is an important question and the most relevant one I think with respect to what people are trying to get out of the PR2.

Since most of us haven't had the privilege of experiencing anything over 90d FOV, what is the actual immersive tipping point? Is there a calculable moment where you effectively trick the visual brain into feeling immersed? Does a hamstrung resolution destroy that immersion factor or does pixel diffusion make that a moot point?

For myself, I will be using the PR2 to play games but not just any games. Only games with appropriate pacing (slow and or heavy on environment/atmosphere) and only games that allow me to hack the fov or remove/move HUD elements (there's a number of them ready for that). There's a few classics I'm dying to go back to and hopefully feel like I'm standing there, rather than looking at a little window.

Is that hoping for too much? Am I being unrealistic with my expectations of the PR2? Can I load up Skyrim and feel like I'm standing in a dark dungeon or an open, expansive cathedral?

As a guess, when you increase FOV, immersion becomes more common, but the true tipping point would be when it starts to approach covering your entire vision.
I remember virtuality which I think was about 80-90 degrees (i'm a bit leary as to the specs posted on Vrealities or somewhere that claim it was less) and that had a very noticable black border. It was like watching a widescreen movie however. That system could and did still immerse you, but often you still realised you were in a VR due to the black edges. The best immersion for me came after I got into the game and was just playing it like a game, and then I stopped noticing the black edges and actually had some sensations that I was IN the game. Its the difference between a screen in front of you showing 3D, and scuba diving in a pool of 3D.

As the black edges reduce, the sense of immersion increases.
To answer the specific question, if the HMD has a FOV of 120 or greater then I believe you will have quite a few moments where you feel you are inside the game. However, I think you'll find that higher FOV is required for full time immersion.

@ Brantlew, I agree somewhat re VR competing with non VR gamers, but I can think of a few situations where VR could actually give you an advantage. You can just stick your eye around the corner to see, stick your gun aruond, and shoot, where non VR gamers would have to expose their entire body (if they don't have lean support). You can stick your gun through a hole in an obstacle and use the obstacle as a shield. In room clearing, you can stick it around the corner of the door and just spray....

Thanks WiredEarp. Having played the old Virtuality system I know exactly what you're saying.

Palmer on the other hand has had a very rare opportunity to play with cutting edge technologies not available to the public sector. He should be able to tell us, somewhat subjectively, about how the PR2 stacks up against the other units he's handle with respect to overall immersion. Is it an on again, off again sense of immersion that sort of works when it works or is it an always-on feeling of immersion?

Eric Howlett of Leep claimed that around 70degrees was a tipping point.
His studies were done with cameras not rendered scenes, but the resolution was low.
I don't know of any real psychological studies on the subject, but it would surprise me if they hadn't been done.

I'd suspect that in rendered scenes there would be more contributors than just FOV.

Let me ask this, with the PR2 or any HMD with a large fov. Is there a point, if the FOV is large enough, that it will begin to look as if, the floor/ground and ceiling/sky are under your feet and over your head ? Not right under your feet and over your head, but as if they are heading in the direction of your feet and head ? Or is there always a gap in that plane between you and the screen no matter how big the FOV is ?

It depends on the design. I would only include headphones if they were integral to the design and stability of the HMD, like with the VFX1. I am going to avoid that if I can, because people all have different sound budgets and preferences. As far as binaural audio goes, it depends on a lot of things. You want the quality to be as high as possible, and you want them to be in-ear for best results (I use HeadFi RE0s for in-ear, AD-700s for everything else). On top of that, a binaural recording should ideally be made using molds that are cast from your own ears. If your hearing system varies too much from the recording, the effect is greatly diminished, which might explain your results.




I don't think there is an exact tipping point, nor is FOV the only factor. If you have good software and good tracking, then the resolution and FOV are definitely sufficient to feel a sense of presence! Getting commercial games good enough is a little tougher, since you cannot actually move and tilt your head in all the dimensions possible in real life. Pixel diffusion is a mixed bag. You lose a little sharpness, but get the huge advantage of not being able to see the pixels at all. One limitation of my HMD is that everything is at the same focal plane, but that is a problem that all displays have, be they 2D, 3D, HMD, projector, LCD, whatever.



The Virtuality systems were only 70 degrees diagonally, surprisingly! Like you say, the black border is an immersion killer. With my high FOV lenses, you cannot see the edges of the display at all, the entire lens is filled with image. The FOV of the unit I sent to Cyberreality was about 110 degrees, my new lenses will beat that, with less distortion to boot. I could go all the way to 270 degree FOV, but like I said earlier, no way for normal games to render that!

As far as maintaining immersion/presence, that is tricky. 120 degrees is definitely enough to maintain a solid sense of presence, but hiccups like tracking jitter or having the HMD shift on your head are going to hold you back.




Yes, there is a point where that happens. In fact, if you press the lenses of my HMD right up against your eyelashes, then the lenses (Which are fully filled by image) are actually reaching over your brow and cheek, so your vertical FOV is just as high as it is in real life! As long as the field of view is correctly rendered to match the HMD, then it will feel very much like the ground and sky are in their place.

Anything I missed? I am trying to see if there are ways I can "cheat" my way to a higher FOV that do not require special rendering, perhaps using plastic panels lit up with LEDs that match the colors onscreen.

Thanks PalmerTech, the high FOV of your PR systems sound wonderful.



Bit like Ambilight eh! I had another idea for doing this passively - how about using mirrors or thin prisms around the edge that reflect/map the image edges. That way, if you have green and red pixels at the edges of the screen, the surroundings will also show green red to fill up your peripheral vision.

No way to do it effectively with mirror, unfortunately, I have tried. Using prisms or waveguides for a passive setup is possible, but there are issue with that, too. It weighs more (Especially prisms), and the brightness is far too low. A few pixels on the edge are just not very good for lighting a whole panel! I am going to try a little more work with waveguides made for frontlighting, but I am not sure it will work well enough.

I could also try just heavily blurring the side displays of my 270 degree unit, but movement is still going to be way too accelerated, so that is probably not a solution. Might just have to stick with 120 degree FOV, but with a light blocker, it is still very, very immersive.

With the Sony and ST1080 being around 45-50degrees, I would kill to have a 120degrees HMD at a good price.
If going with 120 degrees gets your HMD to market faster, thats what I hope you do. 120 degrees over 45-50degrees will be a huge step for VR use.