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 :)

Enough "K" already!

More K?

I'll tell you why?

Because I want it! The hairs on the back of my neck stand up just thinking about what I could achieve with the RED 28K 617 camera.

Though it does need decent DR to make it really work... hint, hint!!!

JohnF
 
Jim should either found a new company for faster CPUs or stop increasing the Ks :-D Vanawesome, I can hear ya bro!
 
More K... Gimme, gimme, gimme!

I want 5K+ so I can get pristine, oversampled 4K images.
 
I'm good for 5k, even possibly 6k, but let's stop there and work on DR and FPS. I'm all about that 617. It makes me moist... on the lips... from the drooling, you sick bastards... but 617 would be just dandy at 6k, but 3x the well size. Though, I suppose for still photography, that 28k is all kinds of cool for large prints.
 
I think one thing people are overlooking when looking at "the K" is the sensor size and how the photosite sizes compare. Mysterium Monstro is supposed to have a 6 micron photosite size. That is larger than Mysterium and mysterium X. The FF35 Monstro sensor will only have a ~4.5K imaging area when shooting with S35mm glass, so no more K than the current R1. In fact, it will be less per mm of sensor area, but should provide for capturing the full S35mm frame.

On 645 and 617 size sensors, pushing them back down into lower K numbers does not make a whole lot of sense. They are physically very large sensors and for a digital medium to be comparable to film of that size, it needs every bit of this resolution. For the 617, the 28K size makes absolute perfect sense for panoramic and large-format photo and print work. The possibility of shooting up to 25fps on that format is just mind-blowingly cool.
 
I will take more of both, thank you very much, at least untill 8k effective resolution and 16 stops DR.

From there I will take improved signal chain.
 
To open up a can of worms potentially; I've heard on here before (forgive me for forgetting on what thread) that having more resolution does also provide more dynamic range because when you down sample (say 5k to 2k finish), you lose noise and can get away with more shadow detail?
 
Even with the "More than 4K" cams coming out, most of us will still be shooting at 4K anyways...windowed, scaled, whatever...Currently the K's aren't the issue, it's the other "stuff"
like: DR, FPS and Configuration....

...oh yeah, did mention workflow? or "deliverables"?
 
Imagine all the pixels... (yeah, yeah photosites)

Imagine all the pixels... (yeah, yeah photosites)

Increased photosite count is fantastic. And as mentioned above, the 28K can really aid in motion and still background plates for FX/Post work.

Now I present the issues that need to be solved.

1) Downsampling. Kilgroe mentioned downsampled 5/6k for silky smooth 4k images. This is good, but on a 4k projector I'd put money down that inter-cut downsampled and pure 4k would have noticeable image quality differences. Some attention should be paid to developing a downsampling algorithm that will match edge sharpness, gradient quality, etc. to a native 4k image. While the downsampled might be just a tad 'cleaner' in the noise department, I don't think it will be noticeable to most like the line characteristics.

2) Computing technology. RED has a great sensor program, I have full faith in that. It seems that while Canon, Nikon, Sony may have a whole lot of people working on their sensors, the RED team has made a lot of progress with a leaner development group (Although I am not privy to precise headcounts at any of the companies).
Computers, dual quad-cores (regardless of Mac/PC status) and 1-2Gb workstation cards, aren't happy when dealing with full-rez 4k. They can do it, but it reminds me of where we were with HD 5-10 years ago. The newest generation is catching up, and now with the release of 5-28k motion imaging, the computing power isn't there to make these practical for a wider audience.
I'm looking at Intel's processor development and we'll get to the point where this data will flow well. Also, the shift in processing from scalar to parallel will greatly help 2k+ imaging as well as the spread of 64-bit architectures. I'm yammering now, and seeing s I've lost half the readers will shut up now.
 
Graeme Nattress said:
Sure, the optical filtering does lower the measured rez a little, but when you watch RED 4K as 4K, on a nice screen, as I did today, it looks like 4k. It looks superb. I am so impressed with the images you all are making. There's a beautiful smoothness and seemlessness to the images that is just stunning. I can't wait to see what the higher resolutions bring to the image. I can only think of positive things that more resolution will produce.

Graeme
Click to expand...

and when you watch red and HVX 200* footage side by side on a barco projector, as i did today, you wonder why the DR of a 4k camera is comparable to a HD camera. I doubt more "K" will never bring more DR to an image, as many "professionals" claim.



(with m2 adapter and nikon lens)



Justin Kirchhoff said:
They did The Dark Knight film out to IMAX at 4k.
Click to expand...


Real 4k footage. Not pseudo 4k footage,FYI
 
Brian Harbauer said:
If they're bringing more DR, why not more k? Why not both? Why do we have to choose one over the other if both are possible?

I hope there are some projectors planned that will be ably to playback all this k? :love:
Click to expand...

Because you really can't have both.... More dynamic range calls for larger photosites... larger photosites = less photosites = less resolution. A 5.4 micron photosite captures less light (and thus less dynamic range) then a 6 micron photosite.

Which you clearly see stated in the new cameras brochure.

http://makethemoviehappen.com/red.bmp
 
Say you have an image of a particular resolution and dynamic range. You can always trade some resolution for more dynamic range. You can't trade the dynamic range for more resolution...

Graeme
 
Graeme Nattress said:
You can't trade the dynamic range for more resolution..
Click to expand...

Sometimes adding noise to an image makes it appear sharper - i guess the noise gets mistaken for textural detail by our in-brain processor.

This doesn't add true resolving power, but hey, sometimes "it looks nice" is enough ;-)
 
Eki....
you by any chance the author of those nifty little lightwave plugins?
 
You must log in or register to reply here.
Back
Top