What If Light Was Really Slow? Again.
Вставка
- Опубліковано 14 гру 2023
- I decided to rediscover the world in which light is slow. The implications are much more interesting than I thought.
The relativistic raytracing uses Sebastian Lague's raytracing project as the base. Check it out here: • Coding Adventure: Ray ...
If you are feeling generous:
/ worldsinmotion
- - - - - - - - - -
🖥️ Links to my gear: (I get a commission from Amazon if used)
CPU: amzn.to/43hTIkF
GPU: amzn.to/3PnZMlS
RAM: amzn.to/3TB07Ea
Mic: amzn.to/3PnW5ww
Audio Interface: amzn.to/3x6yQk7
Headphones: amzn.to/4cgPTjM
- - - - - - - - - -
📃Sources & Further Reading:
Sources:
github.com/SebLague/Ray-Tracing
Yurtsever, U., & Wilkinson, S. (2015). Limits and Signatures of Relativistic Spaceflight. ArXiv. /abs/1503.05845 - arxiv.org/abs/1503.05845
file.scirp.org/Html/10-750328...
github.com/MITGameLab/OpenRel...
en.wikipedia.org/wiki/Electro...
- - - - - - - - - -
🎵 Music used:
Virtual Roaming Charges - half.cool
Rinse Repeat - DivKid
Icelandic Arpeggios - DivKid
Intellect - Yung Logos
Final Girl - Jeremy Blake
- - - - - - - - - -
For business inquiries: lermand77@gmail.com
The way you so nonchalantly say "So I programmed this super-difficult photon path visualizer with a dynamic POV." I'm impressed.
or just modified an existing one.
it's not at all difficult, and it's very poorly implemented compared to what many write as teenagers
It really hurts to see such high-quality content get so few views
well clearly it didnt have enough time to grow yet
Yeah, instead of using boring buildings and teapots, he should've used 3d animations of some fat african-american girls doing some twerking or something like that.
be glad you found him early and actually get to communicate with him via the comments.
So few views? 45k is a pretty big number I’d be overwhelmed if that many people watched a video i made (not just a Mario kart clip)
You can share instead of complaining
I never thought that "light shockwaves" were a thing. Mainly because you need a _really_ special approach (and idea) to even got to the point where they are non-instant.
I love these videos ♥
Actually there is real-world video of light bouncing off the PET bottle in the same way. Of course femtosecond lasers were involved.
In the Discworld novels, light travels very slowly, thus sunrises are describe as golden light moving across a landscape like honey. I wonder if you could visualize that with a basic 3D landscape and a sun slowly moving across the sky while also sending out sub-sonic photons.
I actually didn't know that about discworld. I'm reading Guards! Guards! right now and that's a cool fun fact
@@serbianspaceforce6873 You're in for a treat, the Night Watch novels on the discworld are absolutely stellar.
But then again, all of Pratchett's work is fun to read.
The early Rincewind novels have a very different feel compared to the later ones, you can really watch Discworld grow and mature as a setting.
The books featuring the witches are also a blast.
In short, you're in for a treat with every book.
@@h.a.9880Prachett definitely has a very unique writing style and tempo, it's great fun to read.
@@h.a.9880 I've gone through 20+ discworld books, as much as I love the seriies, i'm disappointed in the tonal shift from the first few books to the rest of the seriies, I wish we would havegot 20 rincewind focused books, ala Conan by R.E Howard.
Moving Pictures is probably my favorite discworld book.
@@ai_serf As much as I can understand that feeling, I am very glad that the setting branched out and also added some parodies of a city watch, witches, the post and banking systems... it's a lot of variety.
Personally, I would have prefered a more focused middle-ages fantasy setting rather than tha (at times) almost victorian setting here and there, but it's still very enjoyable to me.
Controlling light could easily be the most broken power
This is fascinating. Those low light room simulations remind me of walking about the house in the dark / at night - my eyes only catching a dull grainy, gray image.
And only out of the edges of your vision since that's where the most light-sensitive(but monochrome) receptors are
@@EliSmithyes, indeed.
I believe that is an artifact of the first moments after the light gets on, eventually the whole room will flood with fotons and you will see the same picture as normally, until spmething moves.
Seeing the light bouncing making shockwaves really changed how I understand Diffuse Lighting, wow, thank you! So cool
Really cool! I implemented a similar ray tracer for my graphics class, and now I don't have to make a video about it
One thing you might try to increase the resolution is to render all of the frames in one big batch, instead of rendering over and over again (which I assume you are doing?)
Vram becomes a problem, Unity kept wanting to crash so I had to stack some shots :)
@@worldsinmotionWhat GPUs do you have? You could buy four V100 SXM2s, an AOM-SXMV, and all the necessary connectors (PCIe ribbons, 2x RSC-G-6) for another 64GB of VRAM and extra computing power. If you have US$1050 and an extra 1000W to spare that is.
@@worldsinmotion Please share how much VRAM you had so we know what to demand for :)
I would really like to see this, but also with photons behaving as if they had newtonian gravity.
Interestingly, light rays are bent by Earth's gravity, but not quite like other objetcs. Photons's trajectories, because they travel at c, are bent twice as much as massive objects, they effectively feel 2g of force near the surface instead of one
There isn't really such a thing that makes sense because light's defining feature is that it travels at the same speed for all observers. Doing this creates something fundamentally different to a photon that follows different laws of physics and completely changes the properties of light itself. Also, @HunsterMonter is correct. To a very good approximation in general relativity, light bends by roughly twice the angle as something obeying Newtonian gravity and approaching at the speed of light.
Those clips at the end really show how wave-like light appears at these speeds.
This is actually fascinating. Thank you for putting the time into creating this video.
That channel is CRIMINALLY and ILLEGALLY underrated.
Love that video. Take care bro
Thank you for making these. This video and your last one on the topic stand unmatched quality and explanation. Your previous one sits at the top of my all time favorite physics videos.
FINALLY I feel like I understand the whole space/time relativity thing. This is a very good way of illustrating it.
The simulated living room environment is similar to my own. I have a flatscreen on top of a credenza with large stereo speakers on the sides of the TV, integrated such that they always provide the sound. No yucky TV speaker noises. My favorite part was at the end when you were showing the light source move around, pretty neat.
Dude amazing video! Really good explainations
This video was an amazing thank you. I now kinda wana simulate and play around with this stuff myself
Incredible work. Thank you.
I am so happy right now that i found your channel :)
Thankful that people like you exist
love love love this kind of content !!!!
happy new year :)) pls keep making good content
Did you account for the travel time from the last bounce to the camera? Objects closer to the camera should appear faster
interesting.
this concept would make a great cave exploration game
Great content. 1000 times better then TV docus nowadays... 😁
I've had this exact thought before and wondered what it looked like and now finally have an answer
Correction on the space vs time diagram example:
The thing that cannot go below zero is the proper time, which can be defined as dT^2=dt^2-dx^2/c^2.
This is the time the moving object experience when moving in your frame of reference a dx distance over a period of time t. The negative sign means the speed of light is when the proper time interval is the lowest it can go without becoming a complex number, also equivalent to light not experiencing time(even though its time component in our coordinate system is non zero).
The example has the right intuition but the math is a bit misleading
The simulations are really nice though. I think there is an old youtube video where they film a pulse of light passing through a water filled cola bottle irl and it looks just like the example with the teapot.
Thanks for bringing back that cool song at the intro ❤❤
In the next video, he will start talking years after the video started. Because his sound has to reach to the microphone and turn into electrical waves. Then since light speed is slow too, electricity is way slower too. So it will take a lot of time to reach us. Then of course it will be turned into sound again through our speakers. Get ready to watch the video muted a long time before his voice reaches you XD
Beautiful video!
Looking forward to the sound video. I've been watching a lot of videos about sound diffraction and I'd love if you took a deep dive.
Video of the year!
I would REALLY love to see a similar simulation with gravity-affected light
This is way more complicated than it first sounds, unfortunately. It would involve solving the Einstein field equations of general relativity and that's a *very* difficult computational task for almost any physical system except certain special cases. You can't just get away with assuming a Newtonian model of gravity; it would violate light's fundamental property that it travels at the same speed for all observers and it predicts measurably incorrect trajectories of photons in the real world.
Commenting for the algorithm: another great video, keep up the good work :)
1:17 finally someone explained it in a way I _completely_ get it. Why did no-one mention this before?
As you say about the video on sound... Have you ever wonder why the sound from a far away lightning stars low in volume and then progressively gets louder instead of starting out really low in the beginning? I always asked myself that but I couldn't get an answer for it. Though I guess it has to do with the bounces in the landscape as the soundwave cross the long distance.
Awesome Work!
Hey! Thanks a bunch! Much appreciated :)
So this is how we would experience light if we were cosmic giants. Love it
Damn this is a cool explanation! This'll get more views in no time 😊
I saw your previous video a while back, and both that one and this one are great stuff! I just wanted to let you know (and I know English isn't your native language and that's totally fine) that the word "causality" is pronounced "koz-ael-it-ee", like "cause-ality". It's not like the word "casual" (kaz-yew-all). "Causality" isn't "casuality". Just some constructive feedback, your videos are super awesome! I was just telling my oldest daughter on our way home from her school yesterday how I'd like to do a proper high-speed motion simulation that shows what light would appear like if we could move really fast, because everything I'd seen so far either just distorted space or warped the colors, but didn't seem to capture all of the actual effects that would take place. I think you beat me to it!
Or in British english: "kor-zal-ee-tee"
Also I would have thought the american would be better spelt as "kah-zal-ee-dee" (non american's actually pronounce t's in the middle of the word)
This (pun intended) illuminated some things about light for me. I especially liked the depiction of the rays bouncing!
Keep it up until your channel blows up like a supernova!
This reminds me of that Slow Mo Guys video "Filming the Speed of Light at 10 Trillion FPS" where they used insanely short pulses of laser light to visualise light passing through a scene.
I'm mulling over the shape of the buildings, and I'm not convinced they would bend away from a moving observer under continuous lighting. Gonna go through a few scenarios
1) The observer is stationery, the world is dark, and the building glows for just a moment. Since the light is slow, the observer would see the building's glow rise start at eye level and then spread up and down, but mostly up. Shape clearly not affected. Just the latency of seeing the top of the building. When it reaches the observer, the light from the base is more recent, where the light from the top is older.
2) Observer is moving towards the building, the world is dark, building emits short glow. The glow would spread similarly, but more quickly, as the later light now has less distance to travel, as the observer has moved closer since the beginning of the pulse. So, the last bit of light would still be from the top of the tower, but at an angle of incident for the observer being closer. In other words, the observer would see the base of the building earlier and at a further lateral distance than the top of the building. In this scenario the shape may seem weird if we think of the shape of the pulse, but every part of the pulse shows the angle of the building based on the observer's location.
Okay so 3) Observer is stationery, the world is lit continuously, but super slow. Since the light is continuous, the observer sees all heights of the building but from different times. So at an initial distance from the building, the observer sees the base reflecting 12:00 sunlight, the second story reflecting 10:00 am sunlight, and the top of the building reflecting sunrise. The light and coloring of the building will look wildly different from our world.
Last scenario 4) Same as 3 but observer is moving towards the building. Much like 2 vs 1, while the observed originating light time shifts during movement, the angle of the light observed is still dependent of the position of the observer. So the observer should see a normally shaped building, but their movement changes which time of day the see each height of the building.
This is pretty cool.
Your computer is on a whole nother level
Dude i thought this video would havw line 3mil views. You've gained a subscriber
This is incredible. Thank you so much... I hope this gets to the whole world!
Real Eyes
Realize
Real Lies
Dope vid!
great video
24 subs? You deserve more
This is awesome 👏
It is a god tier underrated YT
2:00 Star trek and so on are "using" the concept of a warp drive and therefore are not moving through space but moving space itself to achieve faster than light travel
Awesome!!
Yea
Changing the speed of light is literally like trying to modify the code that already works, only to find out that everything broke and now you have millions of errors and a ton of bugs that's hard to catch
killer visuals!
"Casuality"? It's Causality.
Thank you. Pet peeve
Clearly it's not his first language, give him a break
Hey looks like the algorithm has decided this is in fact a good video
Cool got me subed
Really cool. I know that probably took a lot of processing power, but maybe if you could outsource the rendering to a render farm, it’d be really cool to see that simulated with higher samples, and maybe run through a denoise algorithm.
Damn, really cool!
3:10 Light is a wave front, and the observed would determine light direction to be othogonal to the fronts, so this part of argument (as such) does not apply. The wave fronts actually do appear tilted due to the shift in simultaneity: waves further ahead are ahead in time. Interestingly the angle is just in accordance to this photon particle explanation.
what you used to render this, looks AMAZING!!!
Nice video
commenting for the algorithm, great video!
The rainbow rings are awesome. Sci-fi should get on that
I'm gonna call that trade off of spatial speed and temporal speed the luminal compass. Magnitude doesn't change, but we can change direction (kinda).
Nice video 👍
So amazing
Imagine how cool that would be implemented in Cycles render engine in Blender and combine it with Open Image Denoise to get rid of the noise. Those are free and open source btw, so that's not impossible at all.
Brillant!
There's a fundamental mistake in your video. It's high-quality & thought provoking.
One small nitpick - near the beginning, you said “casuality” (like a casual outfit) instead of “causality” (like cause & effect). I’ve run into that issue before too
How does this affect transparent material? Is the index of refraction more extreme? Do total internal reflections happen at a different angle? And would the sun be able to sustain fusion with the lesser energy resulting from the lower speed of light?
youtube compression was not kind (this was a reference to how in one scene it went 'youtube compression be kind']
Interesting recommendation by the algorithm
since you said in the teapot part that light can only bounce so many times as it loses energy with each bounce, would that inturn infer as it loses energy it will change wavelength and inturn change color
yo, incredible.
You would not, in fact, get a rainbow effect from traveling through the universe close to the speed of light. Most natural light sources, including the CMB, give off a thermal spectrum, not a single frequency, so you'd see "white" light at different color temperatures.
How about showing the photon trajectories from less distance or bounces to more? I think it would show beautiful and smooth shapes
A follow up yay 🎉
Fantastic
My question is, how can light bounce or reflect? Why is it not absorbed by all materials thus absorbed or canceled out?
best vid today
god have mercy on those compressions 🙏🙏
it really looks like scanner in the end or some echolocation visualization.
just got recommended this, good shit mate. saw a bunch of doomposters in the comments so i felt it prudent to let you know that im in fact randomly stumbling in
sick!!!!
Do you think you could make like a really small game out of this?
Even if it's really basic?
Something is up with the algorithm.. your video should get millions of views more! Makes me sad. But just having seen this video, my mood is still considerably better than it was 15 minutes ago :)
Cheer mate and have a great new year!
Single-source sound wave in your video? can I think of it that way?
Ill do my part and comment for the algorithmz
>can't move faster than light
But what if **clear throat** we played a video showing a photon at 2x speed
You’d just be doubling the speed of light then
I think you're actually getting at something similar to the "what if you point a laser at the moon and wiggle it back and forth really fast?" question
i remember seeing the “speed of light” slow video and wooooah…
What if the light source moved faster than the reduced speed of light? Would the result potentially be similar to what happens with sound in a sonic boom?
what program are you using for these renders?
This made me wonder if time can be explained as a result of moving through space slower than the speed of light.
Potentially, but you have to accept that "time" in relativity is really just a mathematical coordinate. It is not a "real" quantity and neither is it unique. You can use any set of coordinates you like in relativity, including ones that have no analogue of time as we normally know it, and all the physics is equally as valid.
Proper time is the only time that is really concrete and that passes the same for all observers.
Breaking causality is, like, kinda my bag, baby 😎💯
I'm really curious about how you actually implemented this. Could you please go through all of that in a more detailed video or at least share some of your code? I assume you're sending out some number of rays from the light source, then more rays at every spot they hit that isn't the empty void/background or the camera, and then keep doing it until some of them eventually hit the camera or you reach the max bounces and have them all deliberately target the camera?
As opposed to going from the camera into the scene until you hit the light source. Since the normal camera to light approach assumes an instant speed of light while light to camera would allow you to have a slower speed of light, but then you've got remarkably lengthy and detailed animations/scenes given how insanely inefficient light to camera should be.
So what were the any optimizations you figured out that meant you could do light to camera more efficiently/faster than what it should be? I suppose you could always just be like "now go hit the camera" after every single bounce, but you'd still have to send insane amounts of rays from the light source and bounces that wouldn't be directed at the camera and might not even end up in a part of the scene that's visible to the camera. Did you somehow cull them? Maybe if they ended up somewhere that wasn't visible by the camera they'd stopped existing, or did you just accept the inefficiency and throw as much computation/time at the inefficient approach as necessary.