@@hanneskarlbom6644 I'm being honest idk how to implement it, like should the whole screen be random noise and have that effect? That's the only good idea I have rn
@@v0idbyt3 You can try something like rain or fog? Motion blur might work too. There are also those blind games where you use a scanner/LIDAR or sound to see, it might give some inspiration. The effect is basically that we see a contrasting movement, but when nothing is moving there's no contrast to be seen. Something like Super Hot(The game), time/the environmental effects only move when you don't or do move or look around allowing you to see it, you could also add some extra flavour of being able to move slowly to sneak but the environment/time won't move so you won't be sure where it is. You can also make it so the ¨monster¨ can still move when invisible by having the effect only apply to its texture like a chameleon. You could put in the effort to make it FP or just make something like Pacman. Hope this helps.
This is a really good example of text that computers aren’t able to analyze as easily as current computer vision technologies prosess video one frame at a time.
@@tecnicrazeComputers can calculate value deltas, would not be that effective. They analyze things frame by frame, but can keep the context of previous frames in memory for comparison
take 2 frames, compare the difference, boom instant picture of the thing to food to an algorithm. only useful if its obscure so people dont bother writing somenhing for it
I think that may be due to compression artifacts that are spread out by a set number of pixels, therefore becoming less prevalent at higher resolutions. That means it might not be there if you played the project instead of watching the video.
As a child I discovered the Pulfrich Effect by accident. As my parents were driving on long trips, I had a pair of sunglasses. Looking out the passenger seat window at the passing scenery, if I covered one eye with one of the lenses, I saw what I could only explain as covering my right* eye everything looked smaller, yet closer, while covering my left eye everything looked giant and further away -- that is except for cars which were driving alongside us. * I may have the eyes flipped as to which caused things to look near or far, as this was something I discovered several decades ago. Years later, some TV shows used this effect to create 3D. They had 3-D glasses you could get from 7-11 before the show which were effectively the same thing, tinting only one eye. The show was staged so that foreground movement moved left to right while background movement moved right to left. Static elements of the set or actors were centered in the depth field. This gave the sense that there was actual depth in a scene by making parts of the stage appear closer to the camera or further away. It's a really neat effect and it's great that you've used it. If you really want to explore it, you can have the static moving faster or slower depending on the depth in the scene, and have the static for objects which extend outward going the opposite direction. To really take advantage of this, you can apply that shift on things like the spinning donut, so that the velocity of the static changes at different depth of the surfaces and control the overall depth by changing it as the object moves closer and further in the field.
Thank you for the comment! I'll have to try doing that next time I'm on a long drive. I'll play around with the shader and see if I can add more depth to the effect. I feel like a side scrolling game would be really good with this. The player could be on a moving train, the background could move in one direction, and a train in the foreground could move in the other
There's a Doctor Who / Eastenders crossover charity special from 1993 that made use of the effect using judicious camera movement and VFX animation. It's up on UA-cam - "Doctor Who: Dimensions in Time". That's how I learned of the effect several years ago... very entertaining!
The effect being taken advantage of here is called persistence of vision. Your eyes aren't like your phone camera, taking individual frames. Instead, your eyes are more like a radio sending a continuous analog signal to your brain. But there's billions of radios, one for each of your rods and cones. This allows for a certain degree of "smearing" in our vision which allows us to see patterns over time that a frame by frame image won't capture
I believe some animals actually see only the change (or see it much better). It would make sense that some animals stand still not to be detected. Movement in general is easy to see
This is the basis behind animation too. In this case, we can differ the part that moves as the foreground, and the static part as the background (or vice versa). When paused, everything is static so there's nothing to differentiate. If you look deep into the B&W object, that'll become the background.
i believe it is actually debated whether our perception is "continuous" or if it happens in "discrete frames". it's been a while since i read any research on that though, so i don't remember the arguments for it other than the "wagon wheel effect" (the whole "a wheel in motion appearing to slow down or move backwards). it's not a great argument though as i believe it is mostly explained by lighting conditions
Awesome work, Jack, and thanks for linking to my videos in your description. What you've done here is way more sophisticated than my simple XOR-based code and gives a really interesting effect. Looks like maybe I need to learn Godot now!
Thank you so much! Your videos were actually the first time I saw the effect and I knew I wanted to try and make it myself! I'm glad you liked it I didn't expect to see you here
@@brantagames This video just popped up in my feed, I guess UA-cam figured out the connection. I'm subscribed, I'll look forward to seeing what else you come up with 🙂
This reminds me of a mini game / tech demo years ago called _Lost in the Static_. It was basically impossible to screenshot because any screenshot was black-and-white pixels, and I expect video compression would have struggled greatly too.
Yeah, I looked through the comments to see if the video author was aware of it, and they mentioned liking it a lot on a previous video. It's one of those games that really sticks with you just because of how novel it is.
He doesn't say that the text is invisible unless you have sunglasses. He just says that it pops out more, which it does kind of pop out when I tested this.
This would be epic for a horror game like the concept is when you were a kid and would get mesmerized by the old TV static so much you would almost catch shapes and figures in the static. except this time there WOULD be ACTUAL shapes and figures in the static! Would you survive?!?
It's not about the text not being visible without doing those things, it's about the text seemingly popping out of the screen (like a 3d movie with 3d glasses) when you do.
It's called Pulrich effect, you can kinda feel depth/3D if you use sunglass to your right eye. If you make a small hole with your fingers and put it on your right eye to make your view darker, you will feel the depth
If you pause the video, sometimess you will see the donut while paused due to the slight blurring of the Pixels in the area where the donut was. But then that's because of the fps.
It's probably because video compression uses similarities between nearby frames to reduce how much data is needed. Things that don't change for multiple frames in a row end up keeping more details, and changing parts end up losing details. If a chunk of pixels is the same for multiple frames in a row, the video compressor doesn't need to duplicate that information. It can basically encode, "this block of 32×32 pixels in this frame are the same as last frame," using a few bytes instead of the 3096 bytes it takes to exactly describe it the first time in full, 1024 pixels × 3 colors/pixel × 1 byte / color. If you limit how much data the video compressor is allowed to use to describe each frame, duplicated blocks have their full quality retained with a few bytes, but changed blocks are only able to be described approximately. The compression algorithm has different techniques for describing a block in less data, and it tries to pick the ones that give the closest approximation balanced with how much size that approximation takes for a given block. The smaller the data budget, the more noticeable the sacrifices in quality are. If you only give it a couple words to describe each block, some of them end up as "mostly red-ish green with the top half being darker than the bottom half", or if even shorter, just "red-ish green." As a result, changing parts of the video get blurrier. When you pause, then, you can tell which parts of the image have changed recently because the video compression has smeared those sections. (Sorry if I'm not explaining it well.)
Oh my I'm dumb, I paused and the circle went away and resumed then it came back and I was gonna comment "it doesn't work" but it was just because it's only supposed to disappear while paused >_
If you dont have sunglasses for the ending part, use one hand in front of your eye and move it up and down quickly. Look through the gaps between your fingers and it should darken what you see on one eye. Works surprisingly well.
I can only tell where it is because of compression artifacts. The video codec is very sharp on the non-changing pixels and little softer on the ones that change.
I'll become your 29th subscriber, because in reality you deserve 29 000 000! This is fantastic!!! I can't even process how you managed to do it! Hopefully some day I will though, as I want to learn computer programming to it's full power :)
3:23 - I don't have sun classes, but I get the same effect by forming a pinhole with my fingers (with two thumbs and index fingers) and looking through the hole with my right eye. Through the hole my right eye sees only small part of the picture (2 to 4 or 5 letters at once), and somehow it excites the 3D effect. If I make the hole larger i.e. see larger area at once, at some point the image flattens again.
The sunglasses thing doesn't work at all, but this does! You've taught me one of the only things I've seen that makes 3D effects work for someone with esotropia like me.
I bet you could use this in a VR game to make something only visible to the player when they move their head, but invisible when they don't. it'll also make people super sick but that's their problem.
The donut is using a trick that has to do with motion not the donut itself, what you are observing is the donuts motion not a donut so when you pause it the pixels become static with it and blend in.
He doesn't say that the text is invisible unless you have sunglasses. He just says that it pops out more, which it does kind of pop out when I tested this.
Cool to see GODOT be used like this! I my self experimented with creating a "fractal flight-sim" after watching a random YT recommendation which was about the cool feedback loop effect you get when you point a camera at a screen which displays what the camera sees! I started it with a Control scene, that is set to scale with the viewport. Then I added 3 siblings: a Camera 3D, a WorldEnvironment and a TextureRect. I set the sky on the WorldEnviroment as Canvas, and created a script for the TextureRect so that it copies the image from the root viewport, and additional controls for rotating, scaling and moving the TextureRect (note: To fly, the rect should be slightly bigger than the viewport/game window. Though making it smaller gives other cool effects!). Then I added two, black and white, plane meshes so that they are at the bottom and to of the visible area of the camera. By moving the rect away from the white mesh (or in the POV of a plane, turning towards it), you can start the feetback loop from there and hopefully see some fractals form, that you can "fly" into! Oh and if you need to display anything in top of all this, just parent the whole scene on a SubViewportContainer's SubViewport, and parent whatever you want to display as a sibling of the SubViewportContainer! Otherwise any labels and such would be caught in the feedback loop too!
@@brantagames That is what I'm wondering too! It is also the reason why I did this in 3D, as I was hoping that by randomly putting objects for one frame onto the world, it would allow me to draw trees and such! Though it has its own problems, such as that: As the camera never rotates, any 3D object in the world wouldn't match where the "ground" or "sky" is in the feedback loop. It would probably be a lot easier to program a shader that can detect where the point between "ground" and "sky" is to draw them, which means you could do all this in 2D! Another problem are the filtering options for textures. I used the Nearest option before, but it creates these "chunks" when you don't move, which cause the most center one to not update in anyway (best seen in person!). For the other options, it creates a blurry image that takes over, so you no longer have pure black and white as sky and ground. But by giving the Camera3D a CameraAttributesPractical, and setting the Auto Exposure on, fixes this! (Also would probably allow for some other cool effects! Just gotta fiddle around with it to see what)
He doesn't say that the text is invisible unless you have sunglasses. He just says that it pops out more, which it does kind of pop out when I tested this.
My low IQ brain just assumed that the animation would be gone from the video even after I’ve replayed it. It turns out I misunderstood the instruction completely.
If you want to return to the original background static, I recommend having a single-pixel (or larger, depending on your step size), wide stroke around each object. You'll have to remember the original pixel values, but this way you can make the randomization happen in the stroke area, and use those values for the 'rotation'. Stepping out of the stroke area, you return back to your normal value. Stepping into the stroke area from inside the object, you randomize as well, btw. Has to go both ways. This can be useful when you want to, say, show an 'invisible' creature moving through an area. Set your randomization to have some severe constraints on the input values, vary about those points, and it'll look like something isn't there when it's not moving. :-D
Optical illusion perspective When the donut (or othe shape of pixels) move u are differentiate between the ones that's movie and the ones that aren't or moving in other direction, but when u pause everything blends in, But in HQ and if u pause in yhe right time u would still see a lil Boarder between the shape of the donut and the other pixels
Only 20 seconds in but have to comment this is the first one of these I’ve seen where I actually can’t see the shape when I pause, other ones I’ve always still been able to make out the outlines and find the edges, this time I couldn’t and it’s really cool to experience the real deal
never really understood when people would say a creature's "eyesight is based on movement" but tbh this helped me understand perfectly how a creature could see like this
The spinning donut does NOT disappear, it just becomes hard to see. It is composed of edge effects in the static, and it is hard to see a slight shift in a chaotic pattern.
Part of the fun of this is that when paused the shapes don't disappear instantly, they fade out. This shows gives me a clue how much of the illusion is in the eye and the brain.
Havent watched the whole vid yet but in the beginning, the donut's pixels are blacker in appearance so you can definitely spot them with some difficulty.
if all you were interested in was the objecy disappearing when paused, you could try flipping the brightness/hues/ and saturations randomly of pixels on the silhouette of the object. So long as the action performed is a symmetry over the entropy of your noise, the still images themselves will appear completely random. I really want this to be something that you could use in a horror game but it heavily utilises the fact that the background has very high entropy, so designing an environment seems rather difficult. I suppose you can have the background noise be dependent on the environment, but still have high entropy, i.e the average brightness/hue/saturation of the noise over the silhouette of an object matches the colour of that object, but is still inherently noisy, and when you have a monster over that object, they randomlu invert the hue etc of some pixels about the averages of their respective distribution(so that the entropies are equal). However, this effect is heavily dependent on the fact that the background noise is consistent over time, expect for the monster/object you wish to disappear when paused. This seems impossible with a moving camera. I suppose you could have the noise be part of the texture, but every time you take a step towards a coherent environment, the effect diminishes. The only way I could see this working in a coherent environment is if you had texture as a function, whereby a local symmetry in entropy is a global symmetry in entropy. This is the case for independently distributed pixels. Well actually so long as they are independently distributed then it could work. Maybe it could work, im not much of a programmer, so i wouldnt be able to do something like this, but i'd love to see someone give it a shot.
This is reminding me of another video that talk about pixels, where how simple static on and off of pixels can create an "illusion" of moving object in our eyes.
This is literally what the objects in my vision look like whenever I get a migraine. Whatever I'm looking at gets distorted with this kind of repeating moving pattern, like when you look at the sun for too long. It looks like how pins and needles feel.
for that last bit you can just half close one eye and it works the same!
Follow up: what if I’m cross eyed?
i can do it without doing anything
@@H_fromDiscord_real same
@@nyphakosi also wait what hi nyphakosi!!
@@Ciytheprotogen I'm cross eyed and I saw it without doing anything
edit: wait is it like that 3D shit everyone talks about?
This would work really well for a horror game as it would preserve the creatures unknowable nature, can't pause to study the enemy
Alr bet hopping on my PC rn and trying (probably gonna fail because I'm stupid) to make a horror game with that
@@v0idbyt3 Tell us if you succeed.
@@hanneskarlbom6644 I'm being honest idk how to implement it, like should the whole screen be random noise and have that effect? That's the only good idea I have rn
@@v0idbyt3wait 1 frickin second you were on my video and i just hearted it and now you’re here too
@@v0idbyt3 You can try something like rain or fog? Motion blur might work too.
There are also those blind games where you use a scanner/LIDAR or sound to see, it might give some inspiration.
The effect is basically that we see a contrasting movement, but when nothing is moving there's no contrast to be seen.
Something like Super Hot(The game), time/the environmental effects only move when you don't or do move or look around allowing you to see it, you could also add some extra flavour of being able to move slowly to sneak but the environment/time won't move so you won't be sure where it is.
You can also make it so the ¨monster¨ can still move when invisible by having the effect only apply to its texture like a chameleon.
You could put in the effort to make it FP or just make something like Pacman.
Hope this helps.
0:06 When you pause sound also disappears too!
that's crazy!! i guess this effect also applies to sound or something!1
If you pause, time stops
Einstein would have been so proud
SOUND IS STOP
😱
any second now someone's going to recreate it in source engine
"guys how do I fix this?" **sends a screenshot**
screenshots dont work on it.
@ATypicalLand327
Whoooooooosh
@goodfunnisbadart oh...
Bash it with a hammer :D
Me who sent a video: pathetic
Good use of "baba is you" music. Works very well for tech demos.
PUSH IS WIN
BITRATE ON STATIC IS DEFEAT
flag is defeat
Skull Is Push
Push Is Pull
This is what is needed for the next level of Captcha, if you can program it. It would be difficult, even for AI, to see what is being shown.
This is a really good example of text that computers aren’t able to analyze as easily as current computer vision technologies prosess video one frame at a time.
A computer could be able to tell which pixels changed.
@@tecnicrazeComputers can calculate value deltas, would not be that effective. They analyze things frame by frame, but can keep the context of previous frames in memory for comparison
ai can literally draw a shape basing on changing pixels and then analyse it
take 2 frames, compare the difference, boom instant picture of the thing to food to an algorithm. only useful if its obscure so people dont bother writing somenhing for it
the donut no desappears if the quality is 360p or 240p or 144p
SODA NE!! DONUTO NO FUKKATSU!!
yeah I noticed it too at 480p
It does
I think that may be due to compression artifacts that are spread out by a set number of pixels, therefore becoming less prevalent at higher resolutions. That means it might not be there if you played the project instead of watching the video.
@@UA-camWatcherIg on mobile I still see it thanks to compression artifacts
Now we need someone to animate bad apple using this
someone already did, its like bad apple but its in noise or something
its called bad apple but its magic eye
here's the earliest I can find watch?v=WnecZgwhHVg
ua-cam.com/video/Myeatl2Q3wM/v-deo.htmlsi=AeFFxKlaVSX5Lz8h close enough
I think I've already seen three Bad Apple implementations of this 😂
Not me pausing 300 times trying to find a single mistake!
I can juuuust barely make out some of the distortion of the donut when paused.
It's probably really good for UA-cam algorithm, lots of engagement
Code doesn't make mistakes.
take a screenshot on every pauses
@@extrapathosThen you don’t know code that well.
As a child I discovered the Pulfrich Effect by accident. As my parents were driving on long trips, I had a pair of sunglasses. Looking out the passenger seat window at the passing scenery, if I covered one eye with one of the lenses, I saw what I could only explain as covering my right* eye everything looked smaller, yet closer, while covering my left eye everything looked giant and further away -- that is except for cars which were driving alongside us.
* I may have the eyes flipped as to which caused things to look near or far, as this was something I discovered several decades ago.
Years later, some TV shows used this effect to create 3D. They had 3-D glasses you could get from 7-11 before the show which were effectively the same thing, tinting only one eye. The show was staged so that foreground movement moved left to right while background movement moved right to left. Static elements of the set or actors were centered in the depth field. This gave the sense that there was actual depth in a scene by making parts of the stage appear closer to the camera or further away.
It's a really neat effect and it's great that you've used it. If you really want to explore it, you can have the static moving faster or slower depending on the depth in the scene, and have the static for objects which extend outward going the opposite direction. To really take advantage of this, you can apply that shift on things like the spinning donut, so that the velocity of the static changes at different depth of the surfaces and control the overall depth by changing it as the object moves closer and further in the field.
Thank you for the comment! I'll have to try doing that next time I'm on a long drive. I'll play around with the shader and see if I can add more depth to the effect. I feel like a side scrolling game would be really good with this. The player could be on a moving train, the background could move in one direction, and a train in the foreground could move in the other
There's a Doctor Who / Eastenders crossover charity special from 1993 that made use of the effect using judicious camera movement and VFX animation. It's up on UA-cam - "Doctor Who: Dimensions in Time". That's how I learned of the effect several years ago... very entertaining!
The effect being taken advantage of here is called persistence of vision. Your eyes aren't like your phone camera, taking individual frames. Instead, your eyes are more like a radio sending a continuous analog signal to your brain. But there's billions of radios, one for each of your rods and cones. This allows for a certain degree of "smearing" in our vision which allows us to see patterns over time that a frame by frame image won't capture
Also takes advantage of the human brain's amazing ability to detect subtle changes very effectively, makes the thing pop out when it's not paused.
I believe some animals actually see only the change (or see it much better). It would make sense that some animals stand still not to be detected. Movement in general is easy to see
This is the basis behind animation too. In this case, we can differ the part that moves as the foreground, and the static part as the background (or vice versa). When paused, everything is static so there's nothing to differentiate.
If you look deep into the B&W object, that'll become the background.
i believe it is actually debated whether our perception is "continuous" or if it happens in "discrete frames". it's been a while since i read any research on that though, so i don't remember the arguments for it other than the "wagon wheel effect" (the whole "a wheel in motion appearing to slow down or move backwards). it's not a great argument though as i believe it is mostly explained by lighting conditions
I would love to see the "edges" of a room also have this effect applied to them. I think it would really help to retain a sense of depth.
0:06 I paused it and still see it because of compression artifacts
@@Monkeymario. same
The idea is it's not visible if you screenshot the game itself, not a recording of the game.
omg delete this video it's been a lie!
Woah neat, when I turn the resolution down it also makes it subtly visible when paused
Yep
Awesome work, Jack, and thanks for linking to my videos in your description. What you've done here is way more sophisticated than my simple XOR-based code and gives a really interesting effect. Looks like maybe I need to learn Godot now!
Thank you so much! Your videos were actually the first time I saw the effect and I knew I wanted to try and make it myself! I'm glad you liked it I didn't expect to see you here
@@brantagames This video just popped up in my feed, I guess UA-cam figured out the connection. I'm subscribed, I'll look forward to seeing what else you come up with 🙂
1:24 paused here and it didnt disappear. sad.
yes it does?
It only works with higher resolution youtibe crunches moving part making moving area slightly gray
I don’t see it
Yeah good catch. (The answer has nothing to do with UA-cam artifacts).
Guys I think he was talking about the red square…
3:02 holy album art generator
Secret documents... unless you make a .gif
Specifically a .gif with more than 1 distinct images.
I'd suggest to consistently use hue, saturation and value as terms. If you use "color" for "hue" it will quickly become confusing.
Yes you're absolutely right about that. I'll try my best to be more concise in the future!
*precise, though hue is shorter than color.
@@pcenero What do you mean by "hue is shorter than color"?
@@ooqui hue has 3 letters and color has 5 lol
@@thepastarat I don't understand why you're mentioning this. What does the length of the words have to do with the comment?
This reminds me of a mini game / tech demo years ago called _Lost in the Static_. It was basically impossible to screenshot because any screenshot was black-and-white pixels, and I expect video compression would have struggled greatly too.
That's what I thought this video was going to be about. _Lost in the Static_ was a trip, but so short
Yeah, I looked through the comments to see if the video author was aware of it, and they mentioned liking it a lot on a previous video. It's one of those games that really sticks with you just because of how novel it is.
Very cool! the stylistic implications of this are very exciting.
3:46 i mean i can see it normaly without any sunglasses
Same here
Same- is there a subset of people who see otherwise?
With the sunglasses, the text appears raised slightly from the background, giving the image a 3d appearance.
@@NG-VQ37VHR intresting
He doesn't say that the text is invisible unless you have sunglasses. He just says that it pops out more, which it does kind of pop out when I tested this.
1:04 MY EYES!!!!
It's most likely your brain, since the total brightness is lower, for you brain tho it makes a lot more contraste and information to process.
Really cool stuff, my fellow Godot user! Keep doing what you love, and post when you feel like it. Let the world see this wonderful art of yours.
Will do!
Doesn't work, paused at 0:24 and could still see the cubes
Hi, this is because the cubes are not part of the effect. Hope this helps.
Hi, this is because the cubes are not part of the effect. Hope this helps.
Hi, this is because the cubes are not part of the effect. Hope this helps.
Hi, this is because the cubes are not part of the effect. Hope this helps.
Hi, this because the cubes are not part of the effect. Hope this helps.
0:11 "And if you stay till the very end, I'll show you a really cool 3D pop-out effect"
Me: **Skips to the very end**
This would be epic for a horror game like the concept is when you were a kid and would get mesmerized by the old TV static so much you would almost catch shapes and figures in the static. except this time there WOULD be ACTUAL shapes and figures in the static! Would you survive?!?
3:42 the text is still visible even without closing an eye, using sunglasses, or doing weird handtricks others have pointed out. I wonder why this is?
It's not about the text not being visible without doing those things, it's about the text seemingly popping out of the screen (like a 3d movie with 3d glasses) when you do.
It's called Pulrich effect, you can kinda feel depth/3D if you use sunglass to your right eye. If you make a small hole with your fingers and put it on your right eye to make your view darker, you will feel the depth
If you pause the video, sometimess you will see the donut while paused due to the slight blurring of the Pixels in the area where the donut was. But then that's because of the fps.
It's probably because video compression uses similarities between nearby frames to reduce how much data is needed. Things that don't change for multiple frames in a row end up keeping more details, and changing parts end up losing details.
If a chunk of pixels is the same for multiple frames in a row, the video compressor doesn't need to duplicate that information. It can basically encode, "this block of 32×32 pixels in this frame are the same as last frame," using a few bytes instead of the 3096 bytes it takes to exactly describe it the first time in full, 1024 pixels × 3 colors/pixel × 1 byte / color.
If you limit how much data the video compressor is allowed to use to describe each frame, duplicated blocks have their full quality retained with a few bytes, but changed blocks are only able to be described approximately.
The compression algorithm has different techniques for describing a block in less data, and it tries to pick the ones that give the closest approximation balanced with how much size that approximation takes for a given block. The smaller the data budget, the more noticeable the sacrifices in quality are. If you only give it a couple words to describe each block, some of them end up as "mostly red-ish green with the top half being darker than the bottom half", or if even shorter, just "red-ish green."
As a result, changing parts of the video get blurrier. When you pause, then, you can tell which parts of the image have changed recently because the video compression has smeared those sections. (Sorry if I'm not explaining it well.)
yooo baba is you music
why do I see you everywhere!?
@ magnets
@@aadenboy ikr
ngl I only watched until 0:35, my eyes started hurting like hell
Yeah I stopped there too, but mine was from motion sickness. Ugh.
Weak
@@kitrod GET OUT-
How are you in every vid im watching lmao do we have the same algorithms
Oh my I'm dumb, I paused and the circle went away and resumed then it came back and I was gonna comment "it doesn't work" but it was just because it's only supposed to disappear while paused >_
Instructions unclear. There is a hole in my sunglasses and my eye
If you dont have sunglasses for the ending part, use one hand in front of your eye and move it up and down quickly. Look through the gaps between your fingers and it should darken what you see on one eye. Works surprisingly well.
He's sort of right about the doughnut, but I still see it when it's paused
I can only tell where it is because of compression artifacts. The video codec is very sharp on the non-changing pixels and little softer on the ones that change.
I'll become your 29th subscriber, because in reality you deserve 29 000 000! This is fantastic!!! I can't even process how you managed to do it! Hopefully some day I will though, as I want to learn computer programming to it's full power :)
I was NOT prepared for that 3D thing. It looks like it's about 1cm in depth. Pretty cool!
3:23 - I don't have sun classes, but I get the same effect by forming a pinhole with my fingers (with two thumbs and index fingers) and looking through the hole with my right eye. Through the hole my right eye sees only small part of the picture (2 to 4 or 5 letters at once), and somehow it excites the 3D effect. If I make the hole larger i.e. see larger area at once, at some point the image flattens again.
Woah, you’re right, it is bizarre
Hah! Love this
The sunglasses thing doesn't work at all, but this does! You've taught me one of the only things I've seen that makes 3D effects work for someone with esotropia like me.
I don't have sun classes either, but I did the same thing and it kinda works
@@dado__ Thanks. Nice to hear.
Looks like an 90's game! Ah those good days...
I love the Baba is you ost
Yoo I just saw your post on reddit about this yesterday! Did not expect it to show up in my youtube recommendations!
I thought that this was a channel with 100k + subscribers, keep up the great work!
Aaah. Fellow baba is you music enjoyer. You have my respect.
BABA IS YOU MUSIC 🔥🔥🔥
That really cool, 3D pop out effect also works if you look through your fingers!
If I pause at 3:02 I see it, smh my head
Shaking my head my head
I bet you could use this in a VR game to make something only visible to the player when they move their head, but invisible when they don't. it'll also make people super sick but that's their problem.
Paused at 0:46 and could see it, what a scam (jk this is cool)
Lol
Idiot he was showing it going wrong
Usually cause internet is bad.
gotta appreciate this video not getting destroyed by bitrate, thats astounding.
The epic-ness grows with each new video…
This is actually such an awesome aesthetic, it’s like futuristic retro
the bitrate explodes my pc ☠☠💀💀
The donut is using a trick that has to do with motion not the donut itself, what you are observing is the donuts motion not a donut so when you pause it the pixels become static with it and blend in.
3:34 if you also see it without sunglasses, reply (so i know if im weird, or hes lying)
i saw it too lol
Yeah i don't think you really need sunglasses
Too :/
same
He doesn't say that the text is invisible unless you have sunglasses. He just says that it pops out more, which it does kind of pop out when I tested this.
This must be how a trex feels when you stop moving. Cool demo, very simple rules for the effect.
this is just amazing, great work
I paused the video and I still see the donut.
Cool to see GODOT be used like this! I my self experimented with creating a "fractal flight-sim" after watching a random YT recommendation which was about the cool feedback loop effect you get when you point a camera at a screen which displays what the camera sees!
I started it with a Control scene, that is set to scale with the viewport. Then I added 3 siblings: a Camera 3D, a WorldEnvironment and a TextureRect.
I set the sky on the WorldEnviroment as Canvas, and created a script for the TextureRect so that it copies the image from the root viewport, and additional controls for rotating, scaling and moving the TextureRect (note: To fly, the rect should be slightly bigger than the viewport/game window. Though making it smaller gives other cool effects!).
Then I added two, black and white, plane meshes so that they are at the bottom and to of the visible area of the camera.
By moving the rect away from the white mesh (or in the POV of a plane, turning towards it), you can start the feetback loop from there and hopefully see some fractals form, that you can "fly" into!
Oh and if you need to display anything in top of all this, just parent the whole scene on a SubViewportContainer's SubViewport, and parent whatever you want to display as a sibling of the SubViewportContainer!
Otherwise any labels and such would be caught in the feedback loop too!
That's really cool! I've never thought about using viewports in a recursive way like that! I wonder what kind of gameplay you could get with that
@@brantagames That is what I'm wondering too! It is also the reason why I did this in 3D, as I was hoping that by randomly putting objects for one frame onto the world, it would allow me to draw trees and such!
Though it has its own problems, such as that: As the camera never rotates, any 3D object in the world wouldn't match where the "ground" or "sky" is in the feedback loop.
It would probably be a lot easier to program a shader that can detect where the point between "ground" and "sky" is to draw them, which means you could do all this in 2D!
Another problem are the filtering options for textures. I used the Nearest option before, but it creates these "chunks" when you don't move, which cause the most center one to not update in anyway (best seen in person!).
For the other options, it creates a blurry image that takes over, so you no longer have pure black and white as sky and ground. But by giving the Camera3D a CameraAttributesPractical, and setting the Auto Exposure on, fixes this! (Also would probably allow for some other cool effects! Just gotta fiddle around with it to see what)
This hurts my brain a lot and I love it
i see the donut when i paused the video 😅😅😅
Bro this is insane, I find it fascinating that no matter how hard you try you can't recover the image while paused but it slowly fades out
3:41 I could already read it so I’m confused
He doesn't say that the text is invisible unless you have sunglasses. He just says that it pops out more, which it does kind of pop out when I tested this.
Very good choice of music haha, love Baba!
3:45 I Didn’t Need Sunglasses To See It, It Says
Like
Comment
Subscribe
The 3d popout effect was lost to you then. I used sunglasses and it popped out at me. You should try it next time before posting a comment like this.
@@traugdoragreed
This is really cool! Thanks for the video :D
My low IQ brain just assumed that the animation would be gone from the video even after I’ve replayed it. It turns out I misunderstood the instruction completely.
One of my cats really enjoyed the spinning ring, she was pawing it as it moved around.
The real magic, when you think about it, isn’t in the software. It’s in the human mind.
If you want to return to the original background static, I recommend having a single-pixel (or larger, depending on your step size), wide stroke around each object. You'll have to remember the original pixel values, but this way you can make the randomization happen in the stroke area, and use those values for the 'rotation'. Stepping out of the stroke area, you return back to your normal value. Stepping into the stroke area from inside the object, you randomize as well, btw. Has to go both ways.
This can be useful when you want to, say, show an 'invisible' creature moving through an area. Set your randomization to have some severe constraints on the input values, vary about those points, and it'll look like something isn't there when it's not moving. :-D
An MRI scan would look terrifying done with this.
Just awesome work man! Love it
Optical illusion perspective
When the donut (or othe shape of pixels) move u are differentiate between the ones that's movie and the ones that aren't or moving in other direction, but when u pause everything blends in,
But in HQ and if u pause in yhe right time u would still see a lil Boarder between the shape of the donut and the other pixels
This would be absolutely BONKERs for a puzzle game
Only 20 seconds in but have to comment this is the first one of these I’ve seen where I actually can’t see the shape when I pause, other ones I’ve always still been able to make out the outlines and find the edges, this time I couldn’t and it’s really cool to experience the real deal
"Keep absolutely still. It's vision is based on movement."
These kinds of effects would be great for a music video. The thumbnail would just look like noise.
I had a similar idea for a survival horror game mechanic years ago, cool that you actually implemented this!
The spinning donut will disappear, because it will no longer spin, it will just be a donut
never really understood when people would say a creature's "eyesight is based on movement" but tbh this helped me understand perfectly how a creature could see like this
The spinning donut does NOT disappear, it just becomes hard to see. It is composed of edge effects in the static, and it is hard to see a slight shift in a chaotic pattern.
This is like looking at the bumps on the popcorn ceiling, and once you look away from a certain spot, the shape you saw disappears
Watching this video frame by frame is pretty cool!
Part of the fun of this is that when paused the shapes don't disappear instantly, they fade out. This shows gives me a clue how much of the illusion is in the eye and the brain.
broooo that's so cool. I am a big fan of this idea
I don’t think this video could’ve found me at a better time
I find it massively more comfortable (and easier to see the objects) when the background pixels are moving and the pixels of the objects are not.
I’d love 10 hour loop of these effects. It would make an awesome game for my cat.
now I know how cat feels when they thought they saw something when suddenly somebody stop moving
you should make a very long video, your voice is so relaxing to listen to
Earned a subscribe, this is mind blowing
Havent watched the whole vid yet but in the beginning, the donut's pixels are blacker in appearance so you can definitely spot them with some difficulty.
the showcase part of the video feels like im watching a posy video
if all you were interested in was the objecy disappearing when paused, you could try flipping the brightness/hues/ and saturations randomly of pixels on the silhouette of the object. So long as the action performed is a symmetry over the entropy of your noise, the still images themselves will appear completely random. I really want this to be something that you could use in a horror game but it heavily utilises the fact that the background has very high entropy, so designing an environment seems rather difficult. I suppose you can have the background noise be dependent on the environment, but still have high entropy, i.e the average brightness/hue/saturation of the noise over the silhouette of an object matches the colour of that object, but is still inherently noisy, and when you have a monster over that object, they randomlu invert the hue etc of some pixels about the averages of their respective distribution(so that the entropies are equal). However, this effect is heavily dependent on the fact that the background noise is consistent over time, expect for the monster/object you wish to disappear when paused. This seems impossible with a moving camera. I suppose you could have the noise be part of the texture, but every time you take a step towards a coherent environment, the effect diminishes. The only way I could see this working in a coherent environment is if you had texture as a function, whereby a local symmetry in entropy is a global symmetry in entropy. This is the case for independently distributed pixels. Well actually so long as they are independently distributed then it could work. Maybe it could work, im not much of a programmer, so i wouldnt be able to do something like this, but i'd love to see someone give it a shot.
Thanks you UA-cam video compression, the donut is still kinda there!
it becomes less defined but i can still see the exact area it is.
ngl that red cube reminded me of Persona 5
i love the baba is you music in the background
This is reminding me of another video that talk about pixels, where how simple static on and off of pixels can create an "illusion" of moving object in our eyes.
1:10 my eyes literally hurt when I see that background
BRO, THIS DAMAGED MY EYES. IM NOT JOKING.
DA HELL
Lmao bro this is the most genius implementation for a concept of preventing anyone of using screenshot as an advantage or paused frames.
This is literally what the objects in my vision look like whenever I get a migraine. Whatever I'm looking at gets distorted with this kind of repeating moving pattern, like when you look at the sun for too long. It looks like how pins and needles feel.