Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

5K 3000FPS 128K ISO. Whats Next?

T

Tommi Hares

Well-known member
Joined
Sep 19, 2009
Messages
56
Reaction score
0
Points
6
So what happens next?

It's just a matter of time (5 to 10 years?) when we have a camera (all cameras) with basicly unlimited FPS, unlimited ISO and plenty of resolution. What does one sell then? Glass? I think so! Your thoughts?
 
Stories and the way they are told ;)

But on gear, I suppose you are right in that optics, and support gear, will probably continue to evolve for a very long time. I want to see holographic cameras.
 
There is never an ending point in the market. The market will just continue to grow and evolve in different ways.

Resolution, FPS, ect. will of course continue to improve over the years. The next thing to improve will be availability/price (which RED is starting to be a leader in). With these two factors, I think the market will change drastically, in ways we can't yet see.

20 years ago people weren't looking for the phone that had the best camera or biggest screen. I think something similar will happen with the film market.
 
The End Game is volume capture.

Once you get all of the geometry, lights and shaders into a 3D scene and then render it your optics, sensors and such are all irrelevant.
 
Shaders to render them, but what do you read it with?

Or are we talking animation + motion-capture instead of some sort of futuristic camera without a conventional sensor?
 
Gavin Greenwalt said:
The End Game is volume capture.

Once you get all of the geometry, lights and shaders into a 3D scene and then render it your optics, sensors and such are all irrelevant.
Click to expand...

I concur. I've blathered on about it quite a bit on these boards but I really do see volume capture as where we're headed.

It doesn't mean that films will be made in sterile studios necessarily. I imagine rocking up to a location and going through hair and makeup/ set dressing as normal.

the camera department deploys a number of LIDAR/Kinect like sensor/camera arrays on every corner of the set and you're good to go.

Has anyone here seen the "technology behind the performance" featureete for L.A Noir?
 
If you add nano technology to cameras line of development it is possible that we could have nano cameras (and recording system) behind our own ayes shooting in the same, or greater, resolution all the time while we are not sleaping. But, this is not so near future. In near future we should expect some sort of nano optical sensors that would dramatically enhance its sensitivity and resolution with dynamic range similar to human ayes.
 
Tommi Hares said:
So what happens next?

It's just a matter of time (5 to 10 years?) when we have a camera (all cameras) with basicly unlimited FPS, unlimited ISO and plenty of resolution. What does one sell then? Glass? I think so! Your thoughts?
Click to expand...

I think the ultimate technological development in cameras is a device that can capture and imprison the human soul.

R.
 
Imagine a sensor where every pixel would have a variable ND filter, thus each pixel would expose at the sensor's sweet point, and the exposure value would become meta data.

Or imagine a sensor that could record more than the 3 primary colors.

Or imagine a sensor that would also record the physical distance of each pixel.
 
Eren Ozkural said:
I concur. I've blathered on about it quite a bit on these boards but I really do see volume capture as where we're headed.

It doesn't mean that films will be made in sterile studios necessarily. I imagine rocking up to a location and going through hair and makeup/ set dressing as normal.

the camera department deploys a number of LIDAR/Kinect like sensor/camera arrays on every corner of the set and you're good to go.

Has anyone here seen the "technology behind the performance" featureete for L.A Noir?
Click to expand...

Kick-ass Eren. Thanks for recommending that.

I think you just nailed it. You give the actors the motivation from being in real locations, you remove all crew from the room.

You have two or three teams leap-frogging the setups, and you move with the actors from space to space. The your DP, after consulting on set, gets to do anything he could imagine, with light and camera movement. And the model of the locations can be tweaked as much as needed. Avatar meets Benjamin Button meets Polar Express, on location.
 
Roberto Lequeux said:
Kick-ass Eren. Thanks for recommending that.

I think you just nailed it. You give the actors the motivation from being in real locations, you remove all crew from the room.

You have two or three teams leap-frogging the setups, and you move with the actors from space to space. The your DP, after consulting on set, gets to do anything he could imagine, with light and camera movement. And the model of the locations can be tweaked as much as needed. Avatar meets Benjamin Button meets Polar Express, on location.
Click to expand...

Exactly.

Have you guys seen what the modding community has been doing with the kinect? The results are quite iffy right now but the data generated via the ir sensor/stereo cameras available in the kinect have shown the first steps of what I imagine are the future.

E.g.: CLICK HERE I mean, keep in mind that this is one of the early hacks from a low priced consumer video game peripheral...

I'm not sure why no ones tried to connect 2+ Kinects and programmed an application that combines the data yet.

Imagine a kinect with 2 Monstro sensors, a more powerful IR sensor...and combine 8 of them on any given set to create your volume.

Add to this a massive library of pre scanned objects, a library of shaders, object/mesh/armature detection and wrap it all up in a drag and drop interface... it's a wonderful future ahead of us :)

I love what's being achieved with plenoptic lenses but I calculated that you'd need to use a sensor with the resolution roughly matching the RED 617 to end up with ~4K resolution RGB image.

EDIT: Actually the guy in the video I posted above has modded 2 kinects together. Once again, the results are really rough but they're promising. The 2 kinect video can be found HERE and the guy's official page is HERE. Apparently, his program can work with as many kinects as the PC's USB bus can handle.

If I had a lot of money I'd buy this guy off right now and set him to work on concentrating on motion picture applications of this technology.
 
Eren Ozkural said:
I love what's being achieved with plenoptic lenses but I calculated that you'd need to use a sensor with the resolution roughly matching the RED 617 to end up with ~4K resolution RGB image.
Click to expand...

Why do you want to add plenoptic sensors into the mix? Are you trying to suggest that we could also use plenoptic lenses on these cameras to make it easier to gather enough information in tight spaces?
 
Roberto Lequeux said:
Why do you want to add plenoptic sensors into the mix? Are you trying to suggest that we could also use plenoptic lenses on these cameras to make it easier to gather enough information in tight spaces?
Click to expand...

Not really, I was just trying to illustrate the fact that you need a lot of equipment and computing horsepower to achieve a rather limited stereo image.

Although now that you mention it, if adding plenoptic cameras to these future Volume Capture devices could yield better results then I'd be all for it. As it stands I think (with my VERY limited knowledge of image sciences, math etc.) going the kinect route over plenoptics would yield a more substantial leap in freedom and image quality.
 
You must log in or register to reply here.
Back
Top