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 :)

What is 6K?

Thanavorakit K. said:
Would an anamorphic lens cover 6k?
Click to expand...

Certainly depends on the lens. 6K is an amount of pixels, remember, and says nothing about the physical size of each photoreceptor that makes up the sensor. In the case of Dragon versus MX, Dragon is a little larger than the MX sensor, but has smaller pixels; 5.5K is about the same size, physically, as 5K was on MX. If a lens has a large enough image circle, then it will cover all the pixels. Basically, if it covers Full Frame with no vignetting, it should cover Dragon.
 
  • Thread starter
  • Moderator
  • #62
Thanavorakit K. said:
Would an anamorphic lens cover 6k?
Click to expand...

Nick Pasquariello said:
Certainly depends on the lens. 6K is an amount of pixels, remember, and says nothing about the physical size of each photoreceptor that makes up the sensor. In the case of Dragon versus MX, Dragon is a little larger than the MX sensor, but has smaller pixels; 5.5K is about the same size, physically, as 5K was on MX. If a lens has a large enough image circle, then it will cover all the pixels. Basically, if it covers Full Frame with no vignetting, it should cover Dragon.
Click to expand...

If we're talking about how anamorphic lenses are designed to be used there is a 6K ANA format that is designed to work with 2X anamorphic lenses. If you're talking about full sensor coverage, which would lead to a very, very wide aspect ratio of 3.89:1 with a 2X anamorphic, that will vary on a lens by lens basis.

There aren't many anamorphic lenses designed for larger than S35 formats.
 
ericyoung said:
This statement is incorrect if it meant to say that a Bayer sensor with x physical pixels captures only 1/3 of the data of a 3-chip camera utilising 3 sensors of x physical pixels each. Without more information about all the hardware, software and signal processing for a particular camera, it is impossible to say one is "better" than the other.
Click to expand...

yeah, the statement "a Bayer sensor captures only 1/3 of the data PER PIXEL that a 3 chip does" was rather in regards to a comparison in COLOR ACCURACY @ capture, as that's the main thing why IMO one would be interested in a 3 chip camera.

(and then you'd have to deal with the huge amount of data)

CMOS captures 1 primary PER PIXEL the rest is made up - 3 chip captures all 3 primaries PER PIXEL.
 
Thanks for this great comparison chart, Phil...! What such a progress from the SD times.
 
It was sad to find out that Aaton had terrible problems with fab production of CCD sensors. Prototype camera footage is very nice. Would've been great to have another digital option made by legendary camera company, even if just as rental item. In the meantime Red's 6K CMOS chip looks better and better.
 
Nick Pasquariello said:
the debayering process results in 80% of the original recorded resolution
Click to expand...

I'm pretty sure the 80% is just for luminance, not chrominance. Color resolution is still less than 80% after debayering, and if we go with Nyquist, then there's aliasing at anything above half the resolution being physically capture in each channel anyway—which we try and reduce as much as possible with the optical anti-aliasing filters over the sensor, but it's still there (just reduced). Nyquist is why CDs are at 44.1KHz, even though tween girls can only hear up to about 20KHz. Doubling the intended capture resolution is the gold standard, and we won't reach that until there's 4x the bayer photosites in both horizontal and vertical of the output format.

But none of this matters. The only thing that matters is: does the image look good?

So far, I'm liking the look of Dragon-captured images, not to mention the dynamic range. Skin tones look improved too, and the new color science coming in October can only make things better.
 
  • Thread starter
  • Moderator
  • #69
Late night thoughts.

It's been an interesting few days. Between having that production ready Dragon camera on Friday, doing the lens coverage tests, and seeing material from Peter, Mark, Evin, John, Tom, and Gunleik I'm thinking more and more about what can be done with the 6K resolution and sensor size. Also, I've been speaking to filmmakers and studios since Dragon was announced some time ago, which has also created a lot of interesting conversations.

Moving forward there's clearly two varieties of 4K that are being established, and both have been here for a while.


DCP, REDRAY, and cinema delivery. That fits in a container of 4096x2160 with an aspect ratio of 1.9:1.

Broadcast, home displays, etc... We're looking at the UHD 4K standard. Which is 3840x2160 with an aspect ratio of 16x9.


This places the 6K 6144x3160 Epic Dragon sensor in an interesting place and you can sort of "choose" how to capture your formats within it's 1.94:1 aspect ratio.

On productions I've worked on and some that we've seen behind the scenes on it's been interesting to see exactly "what part" of the whole MX frame people have chosen to use. Some for aesthetic reasons, some for more technical ones.

I can see many ways to tackle and use this extra resolution. Framing, cropping, re-sizing, etc... The good news is for those capturing in 6K HD, pretty much any lens that's covering 5K Mysterium-X today will be just fine. Also it's looking good for 6K WS as well on the coverage front, with a few wide angle lenses really being the only concern. An interesting bit regarding S35 lenses is using them within the format they are designed and optimized for. Luckily, and this is a huge smart move by Red here, 5K pretty much lands exactly in the width of the S35 frame.

Also, I'm very fond of shooting 2:1 and I think there's something nice about the 6000x3000 resolution that can be extracted from the sensor. While not exactly the full width, it's certainly a nice number and a nice format.

I'm particularly looking forward to shooting content in some lesser used aspect ratios as well. Likely being derived from about 6000 pixels wide perhaps. In terms of finishing/optimizing the material for theatrical or broadcast release I'll certainly need to do some tests, and much of the reason this is on my mind is I've already been experimenting with exactly this concept during this weekend. However, having all of this additional resolution is going to provide a great deal of flexibility. And this is something I really, really like.

Can't wait to get down and gritty with the camera. The more time I spend around Dragon it seems more and more provocative questions formulate in my mind.
 
Exactly “Also, I'm very fond of shooting 2:1 and I think there's something nice about the 6000x3000 resolution that can be extracted from the sensor. While not exactly the full width, it's certainly a nice number and a nice format.” Phil Holland

The format flits well within the Master Primes at the wider Focal Length, if the 25 mm “ZEISS MASTER PRIME” has an Image Circle of 35 Ø mm, and the Red 6K Dragon Epic has an Image Circle of 34.5 Ø mm! So then it stands to reason that a 6000 X 3000 image at 2:1 has an Image Circle of 33.52 Ø mm. So the 16 mm “ZEISS MASTER PRIME” which has an Image Circle of 34 Ø mm will fit well within the 33.52 Ø mm (image circle) of the 6000 (Horizontal) X 3000 (Vertical) which is also 29.98 (Horizontal) x 14.99 (Vertical).

I know it’s kind of hard to read such a technical explanation (at least for me)! Just saying!

Humberto Rivera
 
I have a question about lens sharpness. Will 6K take us beyond the resolvability for many standard DSLR lenses? In other words, will lenses that look sharp at 5K 1:1 still look sharp at 6K 1:1? I am sure it depends on the lens, so I will get specific. I have come to depend on Nikon pro zooms like the 15-35, 28-70, 80-200 and the Tokina 11-16.

Also, how about the PL mount Optimo zooms?

Thanks,
Phil
 
  • Thread starter
  • Moderator
  • #72
Phil Bates said:
I have a question about lens sharpness. Will 6K take us beyond the resolvability for many standard DSLR lenses? In other words, will lenses that look sharp at 5K 1:1 still look sharp at 6K 1:1? I am sure it depends on the lens, so I will get specific. I have come to depend on Nikon pro zooms like the 15-35, 28-70, 80-200 and the Tokina 11-16.
Click to expand...

There's a couple of ways to look at this. If you shoot at 6K, 5K, or whatever 1:1 pixel sharpness will be identical. Dragon's pixels are smaller and from what I understand the final Dragon LP/AA filter is "thinner". So right away we should see a slight increase in resolving power at a 100% pixel level.

Now that's at the pixel level, what about at the 4K finish level. In an example of a 6K->4K finish (or even a 5K->4K finish) when running a debayer and combined with scaling down having more pixels and "squishing them together" will increase the resolving power. A perfect debayer is typically 80% of captured resolution, Dragon gets you 66.6% at full resolution.

Combining all of this stuff; super sampling, smaller pixels, weaker LP/AA filter = better looking images from all of your lenses.

This is certainly one of those "fine hair" things to test, but soon we'll see jut what all of this means and how it effects our image quality.
 
Thanks Phil for all the great info, much appreciated!
 
  • Thread starter
  • Moderator
  • #75
Late night thought.

It seems odd these days, but for over 5 years I've been getting asked the same question and now I'd say it's increasing in frequency now that everybody is catching on.


"But why 6K? Why 4K? Why UHD?"


However, this evening's answer wasn't the technically driven response or even the psychological one that I seem to have loaded in the chamber.

Today's answer was a bit more simple and really from the core of the drive forward.


Because it brings our dreams closer to reality.
 
Hey Phil,

After debayer, what is the actual measured resolution of the Dragon material captured in 6KFF? Would it equate to approximately 5.2K? If R1 was 4K and after debayer 3.2K measured, I would assume Dragon is 5.2K+ after debayer due to the smaller pixel arrangement, higher pixel count, and improved IR+OLPF. I know it's early but just trying to wrap my head around it and get an approximate number.
 
  • Thread starter
  • Moderator
  • #78
Matt Ryan said:
Hey Phil,

After debayer, what is the actual measured resolution of the Dragon material captured in 6KFF? Would it equate to approximately 5.2K?
Click to expand...

Purely mathematically speaking it comes out to 4915x2528 if we're using typical logic and theory. However, there's a couple other things going on. Like Dragon's pixel size physically being smaller and super sampling down to 4K which will effect resolved detail at 4K. We're pretty much nitpicking there and there's a variety of other things that effect image quality. Glass, IROLPF, LP/AA, etc...


Gunleik Groven said:
Hi ohil.

Really dig what you are doing here.

Hope we get a chance to meet one day!
Click to expand...

Thanks. Me too man. I think we'd get along just fine :)
 
You must log in or register to reply here.
Back
Top