RED ONE footage

[Thomas Worth]

I don't see why you are worrying about the Nyquist theorem at all. Nyquist-Shannon sampling theorm is just that a sampling theorem. You are trying to relate the DISCRETE signal from the sensor to the DISCRETE signal of the display without any re-sampling.

Thanks for the helpful explanation, Tripurari. I wish I knew more so I could keep the fire going, but unfortunately the discussion is starting to exceed my knowledge of the subject. :huh:

[Ralph Oshiro]

NO, NO, NO?WAIT . . . I know what I'd like to see RED shoot . . . the next VICTORIA'S SECRET commercial! Duh, I was just on the set of a Victoria's Secret commercial! And yes, Adriana Lima is THAT hot. Wish I could clear the BTS footage I shot to show you guys. She smiled at my lens like all day. Adriana Lima . . . pure SEX.

Edited September 12, 2007 by Ralph Oshiro

[Max Jacoby]

Very interesting -- So big anamorphic zooms today are still soft enough to mask the 2:1 ratio between vertical and horizontal resolution at the film plane.

Ananmorphic zooms are just as big as spherical, in fact they are spherical zooms with an anamorphot added at the back. In their case the anamorphot has to be much smaller than if it were at the front or the middle of the lene. Hence the quality is not as good as with primes and they lose one stop compared to the spherical version.

[Graeme Nattress]

Thomas, listen to Tripurari as he's trying to educate you on your mis-understanding of sampling theory. I've tried to post two explanations here, but my crappy internet connection in the hotel screwed me up both times.

 

Graeme

[Michel Hafner]

Thanks for the helpful explanation, Tripurari. I wish I knew more so I could keep the fire going, but unfortunately the discussion is starting to exceed my knowledge of the subject. :huh:

Try to see some 4k projection of Red footage and you will see that it's not 1K or 2K. :)

[Thomas Worth]

Thomas, listen to Tripurari as he's trying to educate you on your mis-understanding of sampling theory. I've tried to post two explanations here, but my crappy internet connection in the hotel screwed me up both times.

Yes, I'm glad he contributed. I'm going to read up some more on the subject. And thanks for trying to post -- sorry you got lost in cyberspace!

[Graeme Nattress]

Putting it simply, 4096 samples is enough to sample and obey Nyquist 4096 / 2 = 2048 cycles. 2048 cycles are made up of 2048 line pairs (need a pair of lines to represent a cycle) and hence 2048 line pairs are made up of 4096 lines.

 

So, you need 4096 samples to get 4096 lines and you're still obeying sample theory.

 

However, there is no such thing as a brick wall optical low pass filter. So we use a practical filter that does low pass filtering. This stops aliasing, but will limit your measured resolution. We can measure around just over 3000 lines, with minimum of nasty aliasing artifacts. That's luma. Chroma you get anywhere from 2000 to 3000 depending on image content.

 

Now, with a 3 chip system, you still need that low pass filter or you'll get some nasty aliasing. You also get prism artifacts too. You'll get a bit higher luma resolution, perhaps, because you don't need to demosaic, you may get slightly higher chroma resolution, but that might be affected by the prism and chip alignment.

 

Pixels and sensors and lenses and prisms don't tell the whole story. What counts is the image. We've stated the size of our chip, it's bayer pattern, the number of photosites, the size of the photosites and the bit depth of the a-to-d. They're hard specs you can unambiguously count. Measured resolution is a soft spec as you've to measure it - and each way of measuring will give different answers. It depends on a vast number of variables.

[Graeme Nattress]

So, why do we not use a 3 chip and prism system? Cost, and we'd use the ability to use standard cinema lenses, because we've chosen to use a large sensor, so we get high resolution and resonable sized photosites. Why bayer pattern? Well, it's very well understood. Why do we use a well specified optical low pass filter? We don't like nasty aliasing.

 

Graeme

[Phil Rhodes]

That's all absolutely fine. To be fair you end up throwing a lot more away in the LPF on a Bayer chip than you do on a monochrome one because you have such big gaps between each channel's pixels, so the situation is considerably less good.

 

If you've measured 3000 lines, why not publish the test shots? It's a conventional enough thing to do. I really do stand ready to be silenced on this!

 

Of course film aliases too, just to random blobs rather than various vectors of a grid, and while film aliasing seems to be OK, for some reason video aliasing is not...

 

Phil

[Graeme Nattress]

You do need a strong OLPF on a bayer sensor so as to avoid nasty chroma fringing artifacts. There are some bayer cameras around that posted test images that were absolutely riddled with such nasties, so it's vitally important to get that right. The OLPF, I think it set to the pixel pitch of the sensor, and would be the same that we'd use if it were a monochrome sensor.

 

I didn't perform the analysis on the 4k zone plate that I shot, but the 2k patterns are pretty darn near perfect for luma and green, with the slightest hint of aliasing on the red and blue. The 4k patterns show resolution out to just over 75% of the way, or just over 3000 lines for the luma, with slight aliasing beyond that showing that the lens we used, I think a Cooke 65mm if I remember rightly had enough resolution for this chart.

 

After using the zone plate chart, I'd love to point some other cameras at it and see how they perform.

 

I think that the alises you get on film are not so much of a problem because of the way that the the sampling pattern is changing every frame, and is essentially random. On any system of regular samples, it's going to be easier to spot the aliases. I wonder if grain also has a masking effect?

 

Graeme

[Phil Rhodes]

> After using the zone plate chart, I'd love to point some other cameras at it and see how they perform.

 

Me too.

 

Why not publish the shots?

 

I'd particularly like to see a zone plate after it's been through your wavelet codec, at least one channel of it.

 

Phil

[Tripurari Singh]

That's all absolutely fine. To be fair you end up throwing a lot more away in the LPF on a Bayer chip than you do on a monochrome one because you have such big gaps between each channel's pixels, so the situation is considerably less good.

 

[snip]

 

Phil

 

You don't need to account for the gaps between R, B pixels when designing the OLPF. See

 

E. Dubois, “Frequency-domain methods for demosaicking of Bayer-sampled color images,” IEEE Signal Processing Letters, vol. 12, pp. 847-850, Dec. 2005. doi: 10.1109/LSP.2005.859503

 

for the spectrum of a Bayer image. Your OLPF just needs to keep the luma and chroma separate.

 

Tripurari

Edited September 12, 2007 by Tripurari Singh

[Graeme Nattress]

The chart was specially shot for an engineer to perform some analysis for me. I'd prefer that it we are to publish the results that it's performed again, under more controlled conditions, and on production cameras rather than the hand-built prototypes with the older sensor boards. I'd also like to shoot other cameras alongside to add perspective to the proceedings.

 

Graeme

[Phil Rhodes]

Mr. Nattress:

 

> I'd also like to shoot other cameras alongside to add perspective to the proceedings.

 

Marvellous, let's do it.

 

Dr. Singh:

 

> You don't need to account for the gaps between R, B pixels when designing the OLPF.

> E. Dubois

 

Okay, but doesn't that (again) rely on assumptions regarding the limited chromacity of typical scenes, and the overlap in filter passbands? I appreciate you're compromising the desire to extract information from the HF components of the signal against the (frankly much greater) likelihood that the highest frequencies present are actually aliasing, and perhaps I'm just being cautious on the basis that this is intrinsically unfixable, and to some extent a matter of judgement.

 

Phil

[Graeme Nattress]

When I get the chance, I will. BTW, "Graeme" please - this "Mr. Nattress stuff" reminds me of when I taught math :-)

 

Graeme

[Stephen Williams]

The 4k patterns show resolution out to just over 75% of the way, or just over 3000 lines for the luma, with slight aliasing beyond that showing that the lens we used, I think a Cooke 65mm if I remember rightly had enough resolution for this chart.

 

Graeme

 

Hi Graeme,

 

From what I saw on the screen at IBC I have no reason to doubt those numbers.

 

Stephen

[Tripurari Singh]

Mr. Nattress:

 

> I'd also like to shoot other cameras alongside to add perspective to the proceedings.

 

Marvellous, let's do it.

 

Dr. Singh:

 

> You don't need to account for the gaps between R, B pixels when designing the OLPF.

> E. Dubois

 

Okay, but doesn't that (again) rely on assumptions regarding the limited chromacity of typical scenes, and the overlap in filter passbands? I appreciate you're compromising the desire to extract information from the HF components of the signal against the (frankly much greater) likelihood that the highest frequencies present are actually aliasing, and perhaps I'm just being cautious on the basis that this is intrinsically unfixable, and to some extent a matter of judgement.

 

Phil

 

Ha! You have quickly come to the crux of the matter. A system that relies only on filtration and modulation/demodulation will achieve 3K resolution only if high frequencies of color difference signals (G-R, G-B) decay rapidly. Such a system will exhibit cross-color and cross-luminance artifacts on saturated images rather than aliasing.

 

These can be avoided using an adaptive system. If you read that paper carefully, you will find that color difference signals either:

 

1. have multiple copies and only parts of 1 copy is subjected to crosstalk depending on local features (ie is there a horizontal or vertical edge in the neighborhood) or

 

2. are centered about (pi, pi). This center is diagonally across from the luminance center (0,0). The "distance" between the two centers is sqrt(2) units which is adequate separation for a non-astigmatic image.

 

2 is not a problem, and 1 can be tackled (with varying degrees of success) by adaptive systems.

 

The bottom line is that the Bayer pattern is a devilishly clever system that is still not fully understood. I doubt if the inventor himself appreciated the cleverness of his invention. I have yet to see an upper bound on the resolution that can be extracted from it, and as demosaicing algorithms improve more REAL picture information is being extracted.

 

Tripurari

PS: Thanks for your wonderful question

[Adam Thompson]

Hey I'm with you on the 16mm angle. And I think the red is a hec of a toll as well, just not right for current feature projects I'm attached to right now.

 

If you have anything in the works, let me know! Have Aaton, with primes, will travel! :P

 

 

 

...........

Just to say a bit about myself, I am a film major at college, focussed on directing, but I have a long number of years in the media using everythign from svhs, betacam sp, miniDV, and digi-betacam to 35mm for commercials, and broadcast programming. Most recently I've sort of fallen in love with 16mm, while I've been at college, and have an eclair npr, which I would part with if there were a comparable video camera which could produce the raw, evocative images 16mm can create (no way I could afford a red unfortunately...).

 

Mr. Jannard. Congratulations on your vision and faith. I am sure, you and your team can say

 

"...I took the one less traveled by..." borrowed from Robert Frost.

 

Krystian.

[John Sprung]

.... sampling a scene at the exact same rate as the output does not observe the Nyquist-Shannon theorem.

It might or might not -- depending on whether the optical filtration ahead of the sensor is done correctly.

 

 

 

-- J.S.

[John Holland]

This so boring technical poop , dont care if its S8 to 65mm including Red and the rest i just expect to put a camera on sticks, shoulder ,crane whatever film or Video HD just would like good pictures which is up to me as long as i light it correctly and get exposure right . thats whats all about as far as i am concerned . John .

[Phil Rhodes]

You don't have to read the thread :)

 

Nobody complains about discussions of the photochemistry of silver salts. This stuff matters.

 

Phil

[John Holland]

True Phil but hope you understand my drift ? John.

[Chad Stockfleth]

John, I feel you. That's what it's all about for me too.

 

However, this has really turned into a rather fair and courteous discussion. Cheers all around!

[Evan Owen]

John, I feel you. That's what it's all about for me too.

 

However, this has really turned into a rather fair and courteous discussion. Cheers all around!

 

Wow, yeah. Actually, that's quite surprising to me.

 

Here's to even more constructive dialog. B)

[Matthew W. Phillips]

You don't have to read the thread :)

 

Nobody complains about discussions of the photochemistry of silver salts. This stuff matters.

 

Phil

 

To be quite honest, Phil, you're the one who is off-topic. The original purpose of this thread was that Manny put up a screen grab and asked what people thought of it. All of your ridiculous technical rambling ever since has been nothing more than an academic hi-jack of the thread. Not saying you don't have valid points, but ultimately your bantering is bordering on annoying. I agree with John that it's all about the look, not the technical specs. Can specs make you a better Cinematographer than an ASC member?