cancel
Showing results for 
Search instead for 
Did you mean: 

No more Screen Door Effect (SDE) - new Samsung Odyssey+ introduces perceived 1,233 PPI level res

RuneSR2
Grand Champion
So we've got:

Oculus Rift = 461 PPI
HTC Vive Pro = 615 PPI
Samsung Odyssey+ = 1233 PPI (level)

Actually Samsung Odyssey+ is 616 PPI like the Vive Pro, but some magic happens according to Samsung:


Limitless Viewing Experience

With exclusive, evolutionary, SDE-minimizing Samsung display technology, users can perceive a 1,233ppi-level resolution. A 3.5" advanced Anti-SDE Dual AMOLED display lets you experience incredibly immersive mixed reality, and break down the boundaries between the real and the virtual.

9jb3383mvym0.jpg


Looks interesting though, could be simply awesome for movies and Netflix, we'll see when the first reviews arrive - more info here:

https://www.samsung.com/hk_en/hmd/hmd-plus-xe800zba-hc1/

Some other improvements - design looks much like a copy of the Rift:

a5gb54dsw523.jpg

k51xqnacsbs1.jpg

by0be645p43a.jpg

85tbfy1a79ab.jpg

t1dyuuhmp5lw.jpg

It does actually look quite impressive - still those controllers look awful compared to Touch. Given enough time to evolve, Samsung may be a very serious contender, in fact Samsung could evolve to deserve a new logo!

wp5oennqvbip.jpg

Oculus Rift CV1, Valve Index & PSVR2, Asus Strix OC RTX™ 3090, i9-10900K (5.3Ghz), 32GB 3200MHz, 16TB SSD
"Ask not what VR can do for you, but what you can do for VR"

205 REPLIES 205

hoppingbunny123
Rising Star







the only way this could provide any benefit is if the spaces in the panel not lit by a pixel were lit somehow.

the resolution doesnt change. the pixels still show what they show = resolution. but the spaces between pixels are filled.

split the pixel light over two spaces using something like the fiber optic toy that sends light down plastic fiber to;
- the space over the pixel and
- the space between pixels beside the pixel

https://www.scientificsonline.com/product/fiber-optic-light-wand

its cheap to make the fiber optic strand, just lead it to from the original individual pixel and guide it to two spaces, the fiber lights up the space not lit by a pixel.

the benefit isnt increased resolution but loss of screen door effect.

to separate pixels you will need very fine thin walls between spots the fiber sends light too. and extra light being sent out probably means more light nits have to be used.


better yet is when you add in the fiber cable is to not add in walls where the space was before which would split the light of the pixel up into two, but there should be a fiber cable up to the top of the pixel too so if you added walls then the pixel light would have been split up into three, which would darken the pixel unnecessarily.

just use one fiber cable, its base is a small size, then the fiber cable is enlarged on its end to cover the gaps above and beside the pixel.

you just need to package the fibers together somehow. with really ultra thin walls.

edit. i drew a picture to show how to design the fiber so there is a gap-less picture sheen so zero sde.

f3cbds47fz63.jpg
the fiber fits through the hole on the socket to hold the fiber in place. the fiber is small over the pixel and enlarged on the end to cover to gap between pixels above and beside the pixel.

the problem with using a light transport like the fiber cable is the light will be absorbed which will make the light dim. you need to coat the walls of the fiber cable with a reflective coating to send the light shining one wall onto the other wall, but the top and bottom of the cable is not coated with a reflective surface and is able to send and receive light from the pixel.

paint the walls of the cable white.

https://www.youtube.com/watch?v=-yrZpTHBEss




it occurred to me that the screen could be glass, in telecoms the fiber can be glass. to fabricate them just mold clear glass, and paint the walls white leaving the top and bottom walls un-painted.

the light from the screen back light will always send light to the glass, so the white walls of the glass will always reflect some light being sent out of the pixel. but this light might be dim. theres no guarantee the back light will shine through the pixel.

a solution to always have some light being reflected off the walls of the glass from the pixel is to make a wall have a translucent tint painted gap, and facing the translucent tinted gap is a little light bub or light led, some source of light.

shine the light onto the glass painted white wall across from the translucent tined gap shining light from the light source, and the light source light will be
- collected by the pixel light, and that makes the pixel light bright,
- and that sends light from the pixel onto the white wall which is seen by the eye.

because its a small space, you dont fit an entire red green blue light bulb in there, just white light, a tiny impossibly small light bulb.

here's a picture of what i think it would look like;

drro5lejur2w.jpg

why a translucent tint though and not white light? a translucent tint light source will make the black more visible and the colors have more punch. a white light on the black pixel would make the white shine on top of the black like a reflection.

the good thing about this design is you just scale up the screen to fit in the glass bulb putting light through the translucent tinted gap in the glass.

on a different note, vr god rays;

http://i.imgur.com/KNAEdUn.gifv

the rings seen and the god rays are the same thing, the light is shining on the dark and lighting them up. the dark getting light in the rings shows the rings, and the dark getting light on the lens not the ring part brightens the dark part of the lens and is seen as god rays.

the light is diffused, if you put a magnifying glass in-between the screen which is what sends the light, and the lenses which have a sweet spot in the middle, you could focus the diffused light to get rid of the god rays and move the focused light to the sweet spot to light that up.

here i will show a picture, the diffused light that makes the rings and god rays on the left, the one with the magnifying glass on the right, same light source.

zrr7rvfklsbi.jpg

the center being lit up you have a specific area of the lens to add a transparent tint too to get rid of the halo.

pyroth309
Visionary

J/K J/K. I don't know enough about the tech or how feasible it is to manufacture something like that but it's interesting.

On another note, someone posted another review.

Let's
be real, you didn't come here to find out about the crazy comfort
changes they made to the Odyssey+. You came here to find out if the
anti-SDE filter really works. I'll say right now, it's a night and day
difference between the Odyssey+ and every other current gen headset.
Which headsets have I owned or extensively tested? HTC Vive (since
launch), Oculus Rift (1 year), Dell MR (~1 month), and Odyssey (~1
month). In the Odyssey+ have to just stand and stare at the worst
possible spot (large, bright areas; blue sky, etc) and even then I can
only maybe begin to see some tiny screen door effect [SDE]. Here are a few pictures I took
(take into account this is a smartphone picture of a VR headset, center
of my picture is most clear, zooming in isn't representative of
anything, and none of this substitutes for really seeing it).





The only headset that comes close to having the least SDE as possible
is the 5K+, but the Odyssey+ even tops that in terms of pure SDE. Which
should you get? 100% without a doubt anyone who can afford it and has
the GPU horespower to drive it get the Pimax 5K+. The SDE is super
minimal, not quite as low as the Odyssey+ but it comes with the game
changing 170 ° horizontal FOV, unlike the Odyssey's+ same as all other
headset's 100 ° FOV. Not to forget the 5K+ has much more clarity as
well due to the higher resolution screens.





If you don't have the budget for the Pimax, then this is by far the
only current gen headset worth your money. I'm incredibly impressed by
Samsung, what seemed to be an iterative update to the original Odyssey
in every regard, ended up being iterative for everything except the SDE,
the SDE is almost eliminated. I'll definitely be keeping the Odyssey+
for now, at least until I get my Pimax. For anyone interested in a
better perspective on my opinions, here are my Pimax impressions.





EDIT: A few things I thought I'd clarify. Samsung has done all this
without massive blurring, there might only be a slight difference when
compared. The new material feels a lot nicer on the skin than the last
model. The nose flaps are gone, you can play without the fear of
constant suffocation. Colors are still amazing, not muted at all. Not
including bluetooth in the original was a massive oversight which I'm
glad they corrected. I plugged the headset in, turned on the
controllers, and everything was already paired and working. Sadly the
headphone mechanism doesn't extend any further than it did in the
original, I know it didn't reach many peoples ears correctly. It still
doesn't quite perfectly sit on the ear center, but comes very close for
me that it's not really a problem.



For me there is one unfortunate annoyance which has been here since
the original model. It is simply a fault of the head-strap design. While
comfortable in how it manages weight distribution, it fails to really
sit on your face. It never feels solidly attached, the bottom always
feels just a bit off my skin. This causes the headset to move a bit side
to side when turning or looking around at a slightly quicker pace. Oh
and the cable is still the same extremely disappointing length, make
sure you pickup extensions.




I still see blur on a lot of the pics but he is another that says the blurring is minimal. Could just look blurry with the phone camera zoomed in idk.




RuneSR2
Grand Champion
It really does look amazing - it looks quite sharp and I don't notice any SDE - here are the pics taken with a phone camera showing images through the Odyssey+ lenses (which will cause some blurring). So this is the second reviewer highly praising the Odyssey+

guuw2q7jgbzc.jpg

sdrgeietvr5j.jpg

1r6553to8vjg.jpg

ogltzxzdc7ct.jpg

Oculus Rift CV1, Valve Index & PSVR2, Asus Strix OC RTX™ 3090, i9-10900K (5.3Ghz), 32GB 3200MHz, 16TB SSD
"Ask not what VR can do for you, but what you can do for VR"

pyroth309
Visionary
I'll find out for sure around November 5th-7thish lol. And it's the 3rd reviewer, but 2nd one with pics.

I will say I've been conflicted and going back and forth about canceling my order and just going after the Pimax 5k Plus. I've watched a lot of through the lens videos and it's pretty incredible. Yea it didn't meet its advertised specs for the kickstarter but I didn't back it in the kickstarter so doesn't really matter to me personally. But, I feel like I want to wait another GPU generation before I go all in on a brute force headset like the pimax so I can really push the realism levels. I also want to let valve come out with their knuckles and 2.0 tracking and all that.

So yea, I'll have an Odyssey in a week or so. Will definitely add my $0.02 when I get it.

Anonymous
Not applicable
@hoppingbunny123


Seems a bit complex - wouldn't it be cheaper to take one pixel and convert it into 4 pixels by girdling everything into each other? 1pix to 4 fake pixels but keep the light from bleeding into the next grid of pixels by some black boundy. In theory though - without that extra data - you still have the same problem with it being not as sharp - but at least you wouldn't run into the wash out effect from those images would have plus the added pixel increase. It's like a cheaper way to do upscaling. Sure, you still have SDE lines - but between the pixel count and the space between pixels - it should still look a lot better than current gen one headsets. No needed light and cheap to make.


b92vp2ic929f.png
The nice thing about this - is the fact it should be applicable to any VR headset with the correct current pixel spacing. Of course it be near impossible for a normal use to do it - but future VR headsets could double their pixel count without having to increase demand for it or have extra cost in a upscaler.

p895p96exkm3.png

If I base it off this image - it looks like 1 fake pixels is actually shared by 2 real pixels and that causes the wash out effect as they share the same pixel.





Personally - I think SDE is solved already - the problem isn't so much the space between the pixel as it is in the amount of information per real pixel. Real pixels cost a lot and that is where I hope AI and Eye tracking comes into play. If FOVR/ET can help lower the real pixel performance down, but still allow high end image where we are looking - then stuff like this can feel almost real life by the time we can hit real 4k by 4k with a 16k by 16k upscale technology. Even real 2k by 2k still would be 9k by 9k.

hoppingbunny123
Rising Star
a pixel is made up of three different colored lights, 1 red, 1 green, 1 blue.
so i can see you taking that 1 pixel and adding to what it is sending light too, but theres a problem when you try to get two different colors from the pixel.

c3jdr5x5h3jx.jpg

the rgb combine to make one color, so you need two sets of rgb to make two colors, and two different colors = 2 pixels.

on a different note, i tested my idea using a lamp shade, a blank piece of white paper, a large emergency flashlight and a ultraviolet flashlight used to see stains in the dark.

- i taped the blank white printing paper to the lcd screen, there was half the lcd screen not covered by the paper.
- i held the lamp and moved it around the lcd screen watching to see if i needed to cover only the white part of the lamp shade on the inside of the lamp with a dark color black.

i covered the white plastic part in the inside of the lamp with black socks and then let the light shine out of the top of the lamp so the monitor could reflect that light from the black socks, and it did, it washed out the black lcd color regardless of how i darkened the inside reflecting part of the lamp.

this meant reflection from the inside of the lamp washed out the black on the monitor and so the top part of the lamp needed to be out of view of the screen to not wash out the black, so i do need to put a tint to stop any of the white light from washing out the color of the pixel.
- then i drew two lines on the paper and these were measurement markers to measure the light i shone on the paper from my uv flashlight, the inner measurement was the ending part the flashlight lit up.
then i held my lamp, and moved it towards the white paper and it moved the lit up part of the paper towards the outer marking, without adversely changing the uv color.
- then i did the same thing but used a white light and when i moved the lamp towards the lit paper it moved the lit up part towards the outer measurement., without changing the color white from the flashlight to a darker tone.

so my idea works. you might need a dark color tint or dark black or blue.

edit. and the tint has to be translucent not transparent. if its translucent the light is distributed across the tint evenly, or mostly evenly, but if its transparent you would probably get a shine on the black from the light bulb and that ruins the effect.

SpaceStranger
Explorer
I know it's WMR but I hope Oculus is taking notes on this for Rift 2.. no hope for new research for the Quest at this point on this though so screen door still shipping with the Quest I guess lol.  Already at a disadvantage.

Anonymous
Not applicable
It seems that Oculus is starting to worry about the advantage that WMR is taking (see link). The cancellation of the Oculus Rift 2 project and the resignation of Brendan Iribe will surely cause delays, but the truth is that if the people of Oculus do not move quickly, they will be left behind. Samsung Odyssey is already going for its second version of Odyssey, which is clearly superior to the Rift and at almost the same price (the original version is at $ 349). The positive part is that when there is competition, the beneficiary is the user.

https://techcrunch.com/2018/10/31/after-canceling-rift-2-overhaul-oculus-plans-a-modest-update-to-fl...

pyroth309
Visionary


It seems that Oculus is starting to worry about the advantage that WMR is taking (see link). The cancellation of the Oculus Rift 2 project and the resignation of Brendan Iribe will surely cause delays, but the truth is that if the people of Oculus do not move quickly, they will be left behind. Samsung Odyssey is already going for its second version of Odyssey, which is clearly superior to the Rift and at almost the same price (the original version is at $ 349). The positive part is that when there is competition, the beneficiary is the user.

https://techcrunch.com/2018/10/31/after-canceling-rift-2-overhaul-oculus-plans-a-modest-update-to-fl...



Well on one hand it's good that they recognize the flaw in having their "flagship" be made outdated by even their mobile tech offerings but that's kind of disappointing that it seems they're abandoning the cutting edge/a real CV2 for a long while. It's a bit late in the game for me though. I'm still rolling an Odyssey for now. I'm looking forward to the portability of the Odyssey+ as much as I am the SDE reduction. I'll compare this new Rift refresh to the Pimax when I'm ready for another ugprade. 

pyroth309
Visionary
https://youtu.be/VkxDacHXZio?t=1685 ;    

They talk about the SDE here at 28:05.