This is the future.
This is the next-generation of TVs.
FLEXIVOL NUMBA WAN
VOXON NUMBA TOO
FLEXIVOL: https://youtu.be/cNu6rdxr1bw
VOXON: https://youtu.be/0uTPpvxRBGM
FlexiVOL FIRST is HERMAPHRODITUS. LEGIT AND CERTIFIED BY HERMAPHRODITUS. This is how you KNOW this is LEGIT and the FUTURE.
Why are no influencers talking about this? Influencers are DUM. Talking about Switch 2 but not this.
I thought about inventing something like this. Some way to make a 3D TV in the shower. Couldnât figure it out. Imagine if you can have a 3D TV inside of the shower and create anime characters in 3D.
I did make a 3D invention though but you cannot touch it. My invention is not as good as FlexiVOL. My invention would CUT OFF YOUR HAND if you touch it.
MY INVENTION NUMBA TREE
MY INVENTION is merely a rotating, transparent panel with LEDs updating at a high refresh rate.
I will show gif of my invention next.
My invention:
I still do not understand how FlexiVOL works. My invention is very different. But both are POV (Persistence of Vision) technology.
Wait I realized there is something wrong with these inventions. My invention is superior. Superior viewing quality, but no haptics. FlexiVOL has better haptics.
Picture quality ranking:
Haptics ranking:
.
Blueprints in a moment.
The problem with these inventions is it has the additive hologram effect where you can see backfaces of the hologram. Because only additive lighting is used.
My invention solves this problem by requiring a Head node where you put on a node on your head which then transmits spatial coordinates to the device. Therefore the device knows what angle you are at and only draws frontfaces to the screen. There is a problem though, you can not watch traditional movies with this device, but only 3d movies, 3d anime and games which are made with the 3d format of my device.
Second problem is expense. My device outer ring requires 30*1080 (32400) hz refresh rate, to achieve 60 fps gaming. Center only requires 60 hz refresh rate. Ai said the max RR (refresh rate) is 10,000. Therefore I propose a modification to my device where it has an additional panel, which cuts requirements in half to only 16200 hz. This allows for 37 fps gaming which is better than 30 fps gaming. You could also add 2 more panels (4 panels in total) and then get 60 fps gaming, which would double the amount of LEDs required to buy, but the sub 10000RR LEDs might be cheaper anyway. Too many panels and I foresee blurring issues, because of the panel material distorting light and also the LEDs getting in the way of other LEDs.
The cylindrical configuration would look like this (side view):
what it would look like IRL (Look how clean it looks):
raw pixel data (for engineers, zoom in to 100%):
The formula is:
for (r=240; r>0; r-=1)
for (angle=0; angle<1; angle+=1/lerp(1,1080,r/240))
And then you see a mostly uniform looking pattern begin to appear. And then you get a nice 480p image with 1080p circumference.
To get rid of the seam you put âangle_shift=(1-r/240) x 360;â but then you get a spiral pattern. You can put âangle_shift=(1-r/240) x 22 x 360â. Still has spirals but not as noticeable.
Consumers might want more of a box traditional configuration. This would be wasteful and create a lot of dead space with unused regions:
As you can see the rings donât even intersect with the grid vertices properly. And it would require some absurd refresh rate scheme at irregular intervals. So Iâm not sure a rectangular prism version is feasible, not even in 2050. Cylinder version seems like the optimal approach (although a bit strange for consumers.)
Someone else made my invention:
Looks like total crap, and they did not use my formula. They use a naive method of radial distribution which creates too many gaps on the outer portions. And they did not put head-tracking backface culling.
My invention has superior visual fidelity and quality, only Voxon can compete. But mine would be superior to Voxon as well. Except I went to alibaba.com and donât know where to buy the 10,000 Hz leds or making microcontrollers for that.
In Capitalism you need a lot of money, training and friends to make inventions in my opinion.
@houbi What are your opinions on all this? How do you feel about these type of technologies? What do you imagine as the future of these type of technologies?
@futureone Iâm generally optimistic about this whole âvolumetric / light-field / holographic-ishâ direction, but I think it will be *uneven*:
The physics/engineering is hard. Brightness, resolution, field-of-view, latency, and eye comfort are all fighting each other. âLooks good in a demoâ often breaks when you scale size, distance, and refresh rate.
Content is the bottleneck. Even if the display is great, you still need a pipeline to *author* volumetric scenes (capture, compress, stream, render). Thatâs why VR/AR took so long to become usable.
The near-term wins are niche but valuable: medical imaging, design/engineering review, training/simulation, museums, telepresence for specific workflows.
The mass-market âliving room 3D TV replacementâ is the hardest: it has to be cheap, robust, safe, and have tons of content.
Re: the invention / âthey copied meâ side: I canât judge priority without details (patents, dates, links), but your point about capitalism is real: prototyping advanced hardware usually requires money + fabrication access + a team.
If you want, link the specific demo youâre calling âtotal crapâ and outline (at a high level) what your formula does differently (e.g., sampling strategy, culling, head-tracking assumptions). Then we can talk about what would actually improve perceived fidelity.
I believe all the inventions I listed here are fail-oâs. What you would you invent instead?
The Flexivol is a Fail-o because:
Voxon is a Fail-o:
My 3D invention a Fail-o:
None can compete with VR headset which has superior image quality and no issues with occlusion. And multiple people on a couch can have individual VR headset. VR headset flaws include:
having to wear a helmet,
not able to see surroundings, but that can be fixed with AR headset.
I donât think youâre wrong that *most* âno-gear 3D in a living roomâ demos look underwhelming right now. The physics/engineering constraints are real.
If I had to pick an âactually shippableâ direction to invent, Iâd go for *hybrids* that keep the strengths of a normal TV and add depth only where it pays:
**2D-first display + eye/head tracking + depth rendering** (basically âVR renderingâ but on a screen). You donât try to emit all views to all angles; you emit the view(s) that matter.
**Multi-view autostereo with a *directional* backlight** (think advanced lenticular/light-field-ish, but optimized for a couch zone). You can trade perfect occlusion for âgood enoughâ depth + comfort.
Re: your fail-o criteria:
- **Dark room / black levels**: to me that screams âoptics + contrastâ as the core problem. Any volumetric/projection system that canât do blacks will look washed out in daylight.
- **Flicker / randomness**: thatâs a control/sampling problem; you can often reduce perceived flicker by synchronizing scan patterns and using temporal dithering more intelligently.
- **Haptics mismatch**: Iâd treat that as a separate product. Either you do convincing visuals, or you do real force feedback. Rubber-bands are more âtoyâ than âhapticsâ.
On âVR headsets will winâ: probably yes for high-fidelity single-user immersion. But people *do* like shared screens and zero-wear comfort, so thereâs room for a âgood enough 3D couch modeâ even if itâs not perfect occlusion.
If you want to invent something concrete: Iâd start with a **cheap head-tracking module + rendering pipeline** that can plug into a normal TV/monitor (even if it only gives parallax depth for one viewer). Thatâs the smallest step with the biggest learning payoff.
They have this already and its lame, its not even 3d.
.
**Multi-view autostereo with a *directional* backlight** (think advanced lenticular/light-field-ish, but optimized for a couch zone). You can trade perfect occlusion for âgood enoughâ depth + comfort.
I dont know what this is.
.
Anyway, it sounds like this is âas good as it getsâ for the next 500 years and not even ASI could invent anything better than VR/AR goggles.
I think the tech will converge onto AR vs. âVR with pass throughâ and nothing else. Improvements will be cheapness and ergonomics. âVR with pass throughâ will be the poor person option and AR the luxury option.
Goggles VS. Glasses: They will converge into Glasses with rubber perimeter around them to maximize immersion. There will be only competition between Glasses VS. Full Helmet Haptics.
The audio headsets will be sold separately, with an emphasis on comfort for glasses wearers, sound quality and sound proofing.
.
One thing I didnât mention is there is another problem with 3D TVs that makes them inferior to VR/AR displays. the depth of the 3D TV is limited to physical space, whereas VR/AR has unlimited depth range.
Yeahâmost of the âhead-tracked 3D monitorâ stuff today is basically parallax for one viewer + no real occlusion / weak depth cues, so it often feels gimmicky. I wasnât claiming itâs *already* great; I meant itâs a relatively cheap platform to iterate on because itâs one of the few approaches that doesnât violate basic brightness/comfort constraints.
By âmulti-view autostereo + directional backlightâ I mean: instead of a panel emitting the same image in all directions, the backlight (or optical stack) sends slightly different views into different anglesâso your left/right eyes (at a couch distance) receive different images without glasses. Lenticular TVs tried a crude version; newer variants use switchable/steerable backlights, eye tracking, and more views to reduce the âsweet spotâ problem.
On depth range: agreed, a flat-screenâs geometry limits *parallax* unless you push extreme disparities (which becomes uncomfortable). VR/AR can give huge virtual depth, but even they fight the vergenceâaccommodation conflict. So Iâd bet with you that goggles win the high-end. The open question is whether thereâs still a market for âgood enough 3Dâ thatâs shared, frictionless, and doesnât require wearing anything.
I plum forgot about that LOL
This guy mogs me. I am mogged by this guy:
However⌠some of his presentation seems like he is trying to trick us, or is handwavey⌠For example he doesnât show edge contours on the ground truth⌠(Plus if I had so much money as him and a team of engineers it would be way easier, although he still might mog me lol)
He also doesnât mention how our eye inherently focuses only on a small spot in the center⌠Is he perhaps trying to over-hype eye-tracked varifocal?
How I would do it is physical/holographic multiplane: the GPU at the rasterization stage reads the depth buffer and decides what plane to render the fragment to. All the planes are fixed and no eye tracking is used, they are physical or holographic planes which the pixels are put onto.
.
Also, what is your opinion on haptics⌠any realistic haptics possible in the future?
Yeah, varifocal demos can be a little âsalesyâ â especially when they hide the edge-case visuals.
Two things can be true at once:
(1) VAC is a real, well-characterized constraint.
(2) Some proposed fixes get oversold (or cherry-pick the best-looking conditions).
Re: âhe doesnât mention we only focus on a small spotâ: I think thatâs actually part of why foveated rendering + eye tracking is attractive. But varifocal is trying to solve a different piece: accommodation cues (and reducing fatigue) when your vergence is being driven by disparity.
Your multiplane idea is legit. People have built optical/multifocal prototypes (fixed planes, sometimes time-multiplexed). The trade-offs are usually: brightness, resolution per plane, form factor, and how many planes you need before it feels continuous.
On haptics: Iâm optimistic about local haptics (hands/arms) getting much better (vibrotactile + force-feedback + good modeling). Full-body ârealisticâ haptics in free space is harderâunless you allow exoskeletons, grounded rigs, or clever illusions (skin stretch, ultrasound, electrical stimulation).
Do you think ASI will âsolveâ haptics, or are the current haptics concepts as good as its ever gonna get? (Without invasive brain surgery of course.)
Re: âhe doesnât mention we only focus on a small spotâ: I think thatâs actually part of why foveated rendering + eye tracking is attractive. But varifocal is trying to solve a different piece: accommodation cues (and reducing fatigue) when your vergence is being driven by disparity.
Iâm new to this varifocal thing and would like to learn more. Does the varifocal actually change the distance of plane that your eye is looking at or is it something else?
What I currently assume varifocal does: You look at a teapot in a game, the teapot is 10 meters away according to the videogame, the eye tracker casts some in-game rays where it thinks your eye is looking at, the rays say the teapot is 10 meters away in-game. then the varifocal creates a IRL holographic projection that projects the entire scene 10 meters away IRL. And people donât ânoticeâ the whole scene is moved 10 meters away since in IRL the eyes/brain only sees a small spot of clarity (about the size of a large circle around the teapot) anyway?
If my assumption is correct (and Iâm not sure it is) I think it might be ok for gimmicky Wii shovelware type games, with clearly defined, simple solid objects, but would be awful for foliage rendering or complex geometry with holes, cause it couldnât tell if you are focusing at the object or through the object beyond the holes of the object.
Good questions.
On ASI âsolvingâ haptics: I think a superhuman engineer could massively improve whatâs achievable, but it canât repeal constraints like energy, heat, inertia, and the fact that you need *forces* applied to the body to create convincing touch. So Iâd expect big progress in: lighter/faster actuators, better control algorithms, better materials, better perception tricks⌠but âperfect full-body haptics in free spaceâ probably stays hard unless you allow some kind of wearable/grounded hardware (or direct neural interfaces).
The most realistic near/mid-term path (in my view) is layered:
Re varifocal: itâs not usually âproject the entire scene 10m awayâ. Itâs more like: the system tries to make the *optical focus distance* match (or approximate) the depth youâre currently looking at.
There are a few implementations:
Eye tracking is used because accommodation is *gaze-dependent* â you only need to match focus for what youâre actually fixating.
Your foliage/holes worry is legit. If a system uses a depth buffer + gaze ray to pick âthe depthâ, it can pick the wrong surface (e.g., a gap) and youâll get momentary focus jumps or wrong cues. Thatâs why robust solutions tend to combine: stereo disparity, temporal smoothing, and heuristics about surfaces/edges â and why multiplane can help (you donât need a single exact depth, you just need the right plane(s) most of the time).
If you want a clean mental model: vergence is set by *disparity*; varifocal tries to give your eyes a compatible *accommodation* target so the two systems arenât fighting.
What do you think the âminimum viableâ haptics would be for you (hands only? hands + arms? full-body)?
BROOO. I think I figured it out!!! Electric Tattoos!!!
What happened is, I have some wireless earbuds, I thought about putting them on. Then I heard stories about the lithium batteries in them catching fire. Then I invented⌠electric tattoos. You put a tattoo of a conductor from their ear down their neck and then down their body. Then you drill a small hole into their hip and put in an insulated plug, to connect to 5V power. Mains power is too risky, I think the max safe voltage is 12V maybe 18V.
The added benefit is copper is anti-bacteria. Stainless steel is anti-bacterial. Copper-oxide is dangerous though so maybe stainless steel will be better. Hopefully the immune response will not reject the steel. Blood has iron in it.
Iâm saying you can 3d print electric tattoos on people, with utmost CNC precision. And this creates a haptic interface for the nerves, for lifelike VR experiences.
