Trillion-Frame-Per-Second Video

Incredible. I am amazed at the speed at which new inventions in imaging is developing. This is a little too futuristic for me, and I freaking love it.

Did he use the word "consumer" ?!!! Time to start saving :D
 
I saw alot of nothing in that to be honest..... am I alone in that?
 
Llian said:
I saw alot of nothing in that to be honest..... am I alone in that?
Click to expand...

No. Most of the time I see one of these things and say "that's cool, you could use it for this and that and that" or at least afte rthey say what you can use it for go "ah, I can see how that works." This one, I'm like how the heck do you get form point a to point b.

matchstick said:
So where are the photos taken by this camera? all I have seen are CGI approximations.
Click to expand...

I don't believe they are cgi approximations. I believe they are using CGI to turn the single line of blinky bits into something you cna visual parse to try and make it useful. I cna sort of see how you can create ever more accurate physics based light models for CGI with it. You analyze your line of blinky bits and build lighting algorithms until they produce a very close match to your line of blinky bits when simulating th real world conditions that generated your blinky bits. How you get to optical "sonigrams" I'm not seeing so much. I guess if you know how your light source is producing photons, you cna measure deflection and changes in deflection mean changes in optical characteristics of a substance. However, I think there's a lot more variability in your optical qualities from tissue type to tissue type. Density wise, they are mostly water, and I see how that might make sonigrams easier to do calculations on. To sound we are more homogeneous.
 
This has got to be horseshit. A trillion is a far larger number than most people are capable of comprehending. For instance 1 billion seconds equals 31.7 years but a trillion seconds is 31,688 years.
 
I guess it's time for a new video card so I can get that trillion fps playback. It's just not smooth at anything less that 900 billion fps and I need the overhead for post processing effects... IMHO
 
snarky comment 1: "Well, I don't see the point, since the human eye can't perceive anything over 30fps anyway.. sheesh"

snarky comment 2: "Almost as fast as my gaming rig."
 
matchstick said:
So where are the photos taken by this camera? all I have seen are CGI approximations.
Click to expand...

video, not photos @36seconds and 1:45second during the video, did you blink?

PeaKr said:
This has got to be horseshit. A trillion is a far larger number than most people are capable of comprehending. For instance 1 billion seconds equals 31.7 years but a trillion seconds is 31,688 years.
Click to expand...

in the 1930's they could capture 1000 frames a second moving CELULOID film thru a camera, welcome to the 21st century and computers, BIG not = impossible
 
holy fuk, that was a real video of a point of light traveling through the bottle? (and the apple scene)
 
PeaKr said:
This has got to be horseshit. A trillion is a far larger number than most people are capable of comprehending. For instance 1 billion seconds equals 31.7 years but a trillion seconds is 31,688 years.
Click to expand...

It's horseshit because a trillion is a big number? I guess since the national deficit is 15 trillion dollars, it must not exist because that's "too big" a number, right? :rolleyes:
 
It has also been proven If you have an indian accent or look indian that will make you sound smarter when talking about protons
 
To clarify, they are using a "pulsing" technique to generate video of an EQUIVALENT of 1 trillion frames per second. They are not actually capturing 1 trillion frames of a single one second event.

(By the way, playing back one millisecond of 1T FPS video, playing back at 60 fps, would take over 30 years.)

They repeat the same event thousands of times, with the camera shuttering a trillionth of a second later each repeat. In the video, they show a "light wave" going across an apple. This was actual footage, but it was not footage of a single light wave crossing an apple. It was many repeated 'pulses' of light, each pulse captured slightly later each time. It's more of a strobe effect where the strobe is sufficient to show the equivalent of 1T FPS.

(Sort of like filming a prop plane at normal frame rate, only with a strobe light. You watch the prop turning slowly, but you're not watching a single spin, you're watching many many spins, each spin frozen in time slightly later in the spin.)
 
Other than having a crystal-clear clarity of a slow motion video, what else does it do?
 
CharonPDX said:
(By the way, playing back one millisecond of 1T FPS video, playing back at 60 fps, would take over 30 years.)
Click to expand...

but how long would it take you to realize the video'd subject was moving. They have selectively edited out frames to give the light wave the appearance of motion. Stop motion photography.

(I hear the next Wallace and Grommet they ride a beam of light to Pluto! :D )
 
The shutter can takes a picture in a trillionth of a second, it’s not a trillion fps. The processing of the picture(each frame) actually ends up taking a long time. I think the MIT guys call it "the slowest, fastest camera". The way they got the ultimate series of pictures was by redoing the experiment over and over and over again.

They had to redo the experiment a crazy number of times in order to create the complete video. None the less this is amazing in of itself because just the timing of redoing the experiment and matching it with each frame, with a shutter speed as high as it is, is pretty unbelievably crazy. :eek:
 
MavericK96 said:
It's horseshit because a trillion is a big number? I guess since the national deficit is 15 trillion dollars, it must not exist because that's "too big" a number, right? :rolleyes:
Click to expand...

Wait until he finds out how many bits are in a 1TB HD. Heck, the number of atoms that compose earth. Now that's some imaginary horseshit right there. :eek:
 
okashira said:
holy fuk, that was a real video of a point of light traveling through the bottle? (and the apple scene)
Click to expand...

There's no such thing as "a point of light traveling."

They did an awful job explaining their invention in the video, but it seems like they've got something that can record a pulse of light in slow motion. Their claim that "we can image a photon" is retarded, you can't image a photon unless it hits your detector. It's like trying to detect a sound wave with sonar, the whole point of light waves is they travel through each other. So you cannot image a photon that is traveling somewhere else.
 
I aint too smart and such, really, not a physicist, but the example of trying to detect a sound wave with sonar was good...at first, maybe, but light does interfere with itself...like the classic double slit experiments, so maybe light could be used to detect light, no? I ceretainly am not arguing...more like wildly speculating...would someone who reads books and stuff tell me if this would also violate the Shrodigers cat thing....I am not sure this applies to photons...but how can you measure something that small without altering it...like if you could bounce a photon off another photon...would there be some imparted energy that would change both the wavelength and the trajectory or the original photon? My head hurts. Please help.
 
Wshmaster0 said:
I used to move at a trillion m/s, then I took a titanium sapphire laser to the knee.
Click to expand...

I see what you did thar :p

*leaves thread to go play Skyrim*

On topic - it's pretty interesting what they've done, but they are being a tad sensationalist about it. As someone said, they can't actually record a moving photon, so they're being kind of misleading in that sense. However, what they WERE able to do is capture the immediate refraction effects and qualities of the light as it came into contact with and passed through the bottle. Still pretty neat.

Reminds me of the story from a year or so ago where scientists reportedly found a way to effectively halt photons completely. What does a photon that's been stopped actually look like? I don't imagine we would be able to see halted photons with our eyes due to the fact that, well duh, the photons themselves wouldn't be hitting our senses. Such strange mysteries plague this damned universe!
 
matchstick said:
So where are the photos taken by this camera? all I have seen are CGI approximations.
Click to expand...
The bottle one wasn't CG.

Of course they can do a trillion, they're only recording one pixel in height! For anyone who does high-speed shooting, you know that the faster you shoot, the more light you need and the lower the resolution gets. They simply record one line of pixels each shot, then another, then another, etc. and combine all the videos into one complete video. It's not practical at all, but it is fast enough to basically slow down light.

Personally, I shoot at 240 fps then slow it down with software so I can still manage near-HD images at high speeds without a camera costing hundreds of thousands of dollars. Enjoy!

http://www.youtube.com/watch?v=ZJ7SKVaziEA
http://www.youtube.com/watch?v=CxgLIRO4cGk
 
schoenda said:
I aint too smart and such, really, not a physicist, but the example of trying to detect a sound wave with sonar was good...at first, maybe, but light does interfere with itself...like the classic double slit experiments, so maybe light could be used to detect light, no? I ceretainly am not arguing...more like wildly speculating...would someone who reads books and stuff tell me if this would also violate the Shrodigers cat thing....I am not sure this applies to photons...but how can you measure something that small without altering it...like if you could bounce a photon off another photon...would there be some imparted energy that would change both the wavelength and the trajectory or the original photon? My head hurts. Please help.
Click to expand...

When you are speaking macroscopically, you're talking less about EM radiation being photons and more about it being alternative EM fields. Mathematically, these fields obey simple wave equations where the amplitudes of the waves can add constructively and destructively, giving rise to EM interference.

When you're talking about the more precise versions of the double slit experiment where they send particles one at a time, you begin to enter the regime of quantum mechanics. The wave function that represents the photon in quantum mechanics can interfere with itself (just like an electron or a proton can interfere with itself). It is not intuitive, but the math is consistent with the experimental results.

This is going to get technical now.

When you talk about things on the level of quantum electrodynamics (quantum field theory) you'll think of a quantized spin 1 photon field. In the Lagrangian for QED, this field does not interact with itself directly. In the parlance of Feynman diagrams, you never see two photon lines intersect directly. A photon does not interact with another photon to first order in the expansion of the path integral.

BUT, past the first order "tree level" expansion you can have photon-photon scattering through something akin to vacuum polarization. Essentially, this can be seen as two photons creating electron-positron pairs which then annihilate one another into two new photons that do not necessarily have the same trajectories/energies as the initial photons that went into the interaction. Without knowing about the virtual interactions on the interior of these scattering events, it looks like photon-photon scattering from the outside.
 
When I see how much space raw 120fps 1080p footage takes up...they probably aren't storing this on an SSD...
 
krameriffic said:
When you are speaking macroscopically, you're talking less about EM radiation being photons and more about it being alternative EM fields. Mathematically, these fields obey simple wave equations where the amplitudes of the waves can add constructively and destructively, giving rise to EM interference.

When you're talking about the more precise versions of the double slit experiment where they send particles one at a time, you begin to enter the regime of quantum mechanics. The wave function that represents the photon in quantum mechanics can interfere with itself (just like an electron or a proton can interfere with itself). It is not intuitive, but the math is consistent with the experimental results.

This is going to get technical now.

When you talk about things on the level of quantum electrodynamics (quantum field theory) you'll think of a quantized spin 1 photon field. In the Lagrangian for QED, this field does not interact with itself directly. In the parlance of Feynman diagrams, you never see two photon lines intersect directly. A photon does not interact with another photon to first order in the expansion of the path integral.

BUT, past the first order "tree level" expansion you can have photon-photon scattering through something akin to vacuum polarization. Essentially, this can be seen as two photons creating electron-positron pairs which then annihilate one another into two new photons that do not necessarily have the same trajectories/energies as the initial photons that went into the interaction. Without knowing about the virtual interactions on the interior of these scattering events, it looks like photon-photon scattering from the outside.
Click to expand...

Ahh yes..all clear now. (Sticks finger in nose and goes back to Trine 2) Seriously though thanks, I need to dig out some textbooks. I think that it is so easy to be sensationalistic and not quite true to the science when reporting on some of the lesser understood (to laypersons) quantum physics observations...as someone mentioned about the "stopping the photon" experiment etc...I blame the media...they try to "bring science to the masses" but instead just distort and misinform, and by definition their target audience is susceptible to this.
 
I'm trully disapointed in all of you!
No one asks: BUT CAN YOU OCLK IT?
 
Misleading bullshit alert.

When you say trillion frame per second video, you imagine actually being able to record at least one second on continuous footage, and have a trillion frames to it.

This is not capable of that, its just taking pictures very slowly over a long period of time and thus incapable of truly capturing a trillion frames per second as the title implies.
 
Well seems most of you missed the "point" of how interesting this is and instead chose to focus on hot button "numbers".

Being able to see the refraction of the light passing through the bottle was pretty damn cool. I realize its not a "Wow" effect like Mythbusters blowing up a cement truck with 500 pounds of explosives that detonate at 30,000 feet per second and all while being recorded at 1000 fps and that's probably why most of you are unimpressed. But it is quite a feat , another notch MIT gets to add to its belt of accomplishments.
 
You must log in or register to reply here.
Back
Top