New to the forums? Click here to read the "How To" Guide.

Developer? Click here to go to the Developer Forums.

some videos of overwatch i made you can watch in vr

hoppingbunny123hoppingbunny123 Posts: 391
Nexus 6
edited March 8 in General

download the videos using dishitas youtube plugin for firefox and watch them in vr using virtual desktop, fisheye, 60 fov.

i'm not affiliated with dishita, the plugin is free or pay, i paid and got the 1080p version of the video. if you know a better way to dl 1080p videos from youtube then share how and a link and then get the videos in 1080p that way instead.

i had 48  games, full games, of overwatch. you can't really play overwatch in vr very good right now, i tried, but i think you can watch them fine and still have fun!

so im sharing my games in case you want to watch some overwatch games in vr from oculus! :) now for some treats from the fridge...

but after posting this is went and watched a 720p game and compared it to a 1080p game and the difference was so much i had to delete the 47 games that were only 720p. only games in 1080p are worth watching in virtual desktop fisheye 60 fov.

and again i had to delete the videos there because in vr the games are too fast.  i had to reduce the speed of the video using movavi video editor;

i reuploaded the videos 2 for now, 1 and 2. it will take a while for 2 to upload and 1 to go to 1080p will take a few hours. you need to use the 1080p version.


  • hoppingbunny123hoppingbunny123 Posts: 391
    Nexus 6
    edited March 8
    i updated my previous post.

    i had a problem with the videos being jerky because of my mouse twitching. so i tried movavi video editor stabilization. i was exhaustive in trying different combinations of settings and didn't find any that fixed the jerkiness thats very noticeable in vr in the rift.

    then i found by using the movavi video editor slow motion tool setting it to 82 the video looks not too fast nor too slow in vr. the audio is slow though so i used the movavi audio tool, audio effect, high pitch, to make the sounds normalized but a bit slower because the speed is now at 82 percent.

    the video above shows the rest of the settings to get the game footage and then using movavi. then to watch the video use virtual desktop fisheye 60 fov in the rift.

    i write this because for vr footage people might want to use a stabilizer and its wonky results are hard to fix, but by using slow motion the jittery camera in the video is fixed.

    also for fps like csgo or call of duty you need to take the speed of the game, give it a measurement of 100 percent speed, and lower the game speed to 82 percent, then people can play the game and not get sick in vr. it could be revolutionary, just play csgo or overwatch, or call of duty, or apex legends or fortnite in vr and not get sick by just lowering the speed the game is played at by 18 percent of the regular 2d tv speed.
  • hoppingbunny123hoppingbunny123 Posts: 391
    Nexus 6
    edited March 8
    movavi video editor stabilization settings;

    - accuracy; 4
    - shaking; 4
    - radius; 6
    - smoothing; 30
    - edge cropping; adaptive

    videos (give it time to go to 1080p on youtube)

    before stabilization;

    after stabilization;

    why i'm showing this is to show that when i'm looking, the video in the stabilization video moves around, but when i'm being watched then the video stops moving around a lot.

    this is a useful thing for things like augmented reality. where you have to interact with the real world.

    if you have a camera on the ar glasses, and the camera sends a video feed of what it sees to the persons view as they look out of the glasses, the person and camera see the same thing.

    then the person moves their view all around as they're the ones whos watching, but the moment the video stops swinging all around then the person is the one whos being watched. then the camera seeing the person react to this thing watching them can id tag that image and treat it special in the software.

    for instance with augmented reality occlusion. the image input into the real world thats virtual like a cat image, is seen, its recorded as a still video motion, so the cat is interacting with the person, in effect the still camera means the cat is the one whos watching the person. then the hand of the person covers the cat this is where occlusion works to show that because the cat is the one whos watching then the cat is occluded when the person puts their hand over the virtual cat.

    still cat, moving hand, in order of sequence the cat follows the moving hand but the value of the cat is still there but is hidden by the now moving camera.

    without image stabilization software then the shaking of the camera to guess if your watching or being watched is unintuitive and hard to see.

    you see the government look into pre crime like from the movie minority report. well if the person doesnt stop moving at what they see sees them then they are uninterested in that thing and won't usually bother with it.

    but if they see someone we would worry about them hurting, then they see them they stop moving then that person is a potential target to them;

  • hoppingbunny123hoppingbunny123 Posts: 391
    Nexus 6
    edited March 8
    referring to the pre crime idea, and how image stabilization fits into that, in my previous post.

    original link;

    i got the clip from the movie minority report from youtube, then ran it through the movavi video stabilization, then re-uploaded it to youtube so you can see the wobbliness of the video to show when he is seeing a potential target, pre crime, the camera goes mostly still;

    but when he's doing some sort of task the camera is wobbly again because hes the viewer not the viewed.

    off topic, but the camera in minority report identifies with what mr cruise sees. its giving the movie personality, something the new marvel movie lacks, starring that female hero. i'm guessing since i didnt run the films trailer through the image stabilization.

    something a film director would want to know, if theyre trying to highlight a scene as a guy or guys looking at a pretty girl whos being watched, the camera wobbles, but the girl looking back the camera goes still.

    more wobbly than still for girls, because guys want to see them, guys expect a guy to look at the camera to see the guys more than a girl would, unless the guy is being watched by girls, then the camera wobbles more so the girls can watch the guy and the guy holds the camera still less often but enough to show the girls he sees them.

  • hoppingbunny123hoppingbunny123 Posts: 391
    Nexus 6
    edited March 9
    i ran the marvel trailer i linked to in my previous post through the movavi video stabilizer, and uploaded the results to youtube, here it is;

    what i see is unlike the minority report movie, there is very little wobbliness, when there is wobbliness its usually muted or tame.

    that probably why the movie got bad reviews, the wobbly parts should be wobbly like the minority report movie.

    only a few scenes are really wobbly. which translates to being viewed more than viewing, using my logic i showed in my previous 2 or three posts.

    she should be in a wobbly movie like the hero, not tame;

  • hoppingbunny123hoppingbunny123 Posts: 391
    Nexus 6
    edited March 10
    having read a article on road to vr, about brain interface technology over at valve, i came up with an idea for one.

    using this idea of still = watched, and moving = watching, if you had a brain interface you could replace the mouse.

    you would still use the right hand for wasd and ability choice for e left shift, q r, but the right hand that controls the mouse would be totally controlled by the brain using a brain interface.

    the brain interface sees you go still, then you move = mouse movement. so the things in the camera image have values of still and moving tied to them, this lets you move and the thing with the highest move value gets moved too until your over it then your still.

    using the idea i referred to earlier, with the ar camera mirroring and sending that mirrored picture to the persons ar view of the real world, to understand how to fit in augmented reality things like virtual cats.
    the way the persons still camera photo works with his moving and still values is, just like the ms marvel movie has a still wobbly, then the minority report movie has a large wobbly value, the things in the persons view have wobbly values tied to them that separate what the motion is referring too.

    position (0,0) = wobbly = 2
    position (2, 1) = wobbly = 9

    start wobbly from position (3, 1) = moving, go to position (2, 1) = still = mouse movement

    now add in z coordinate and from position (2, 1, 0), start moving to variable z coordinates that each have a wobbly value and go to the highest wobbly z coordinate value = mouse button movement or left click
  • hoppingbunny123hoppingbunny123 Posts: 391
    Nexus 6
    edited March 16
    here's the math to be able to code something in, referring to my previous post, i will give a true and false math.

    correcting correlating errors \\ function that fixes problems like turning a wrench one way is wrong if you want to tighten it.

    - {
    - a1 numerator green
    - a3 denominator blue
    - }

    - {
    - a2 numerator green
    - a4 denominator blue
    - }

    - {
    - a3 numerator green
    - a5 denominator blue
    - }

    job; attach numerator to denominator with highest denominator value
    - there is two ways to do this, using a path that is a process, or a true or false logic which doesn't need a process to figure out.
    - (true shows the a5 if its a5 or a4), (false shows the process that tests this assumption).

    - a4 = position (x = 0, y = 0) = denominator value = 2
    - a5 = position (x = 2, y = 1) = denominator = 9

    - true [(1 / 1) - 1] / 1:
    - true [(a1 / a5) - a4] / a4
    - (numerator = 1) = start from position (3, 1), 
    - (denominator = 1) = go to position (2, 1)
    - (- 1) = subtract go to position (0,0)
    - (denominator = 1) = numerator value 0 go to denominator position (0,0) 0 times

    - false [1 / (1 - 1), [true [(1 / 1) - 1] / 1]]: // or "spam dos" a term i coined
    - false [a1 / (a5 - a4), [true [(a1 / a5) - a4] / a4]
    - (numerator = 1) = start from position (3, 1), 
    - loop to test if a5 denominator value increases enough over a set amount of grow time so = true
    - (
    - // is there a increment in a5 denominator value?
    - (denominator = 1) = go to position (2, 1) = (denominator value > 9) 
    - - 
    - (denominator = 1) = subtract go to position (0,0) = (denominator value = 2)
    - )
    - if (denominator value > 9)
    - (denominator = 1) = numerator go to denominator position
    - = path false
    - else
    - (denominator = 1) = numerator value 0 go to denominator position (0,0) 0 times
    - = path true
    - end loop after enough steps taken to get from start to end point
    - if path false = (a5 = a5 = false), (a5 = a4 = true), 
    - else = (a5 = a5 = true), (a5 = a4 = false)

    - // similar to walking with arms swinging together when moving the legs to walk.
    - // when unsteady, the arms move the same side as the leg when moving forward and backwards, (a5 = a4)
    - // then regaining stability, the arms swing opposite, outwards when the leg goes in, and inwards when the leg goes out (a5 = a5).

    the false had the wobbly value increment until it was larger, like hacking to give you super powers in video games normal players dont have. this false increment was ignored by the second denominator.

    this math shows theres a lot of noise from the false, because hacks are in the persons mind i presume. false logic put there by others, chemical imbalances, instability, fatigue, etc,

    lots of different forms of hacks going on in the persons mind so that finding the wobbly values in their mind to set things like a brain interface for a mind controlled mouse is difficult without weeding out the math hacks in their minds and instead preferring the true wobbly choice values math.

    so if you have the math help a soldier with a brain interface device, so the device acts on the soldier to make them be brave, the wrong math would have the inverse effect by making them feeble when the ai or mind control device geared to help them goes into effect.

  • hoppingbunny123hoppingbunny123 Posts: 391
    Nexus 6
    i updated my previous posts math a bit. sorry for the previous sloppy math.
Sign In or Register to comment.