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 :)

Red not true 4k

Status
Not open for further replies.
Ummm,

Bit like asking a veggie which sausages to buy.

Mike C
 
Daniel S Williams said:
Excellent article cutting through the marketing of pixels to reveal the truth. John Galt is an expert.

here's the link
http://magazine.http://www.teenink....le/the-truth-about-2k-4k-the-future-of-pixels
Click to expand...

John Galt is a smart guy, but he is by far one of the most biased sources for camera information I can think of. I certainly wouldn't take my "gospel" of "What is 4K?" from him when he's talking about competitive products.

Lucas
-----
ASSIMILATE, inc.
LA, CA, USA
 
Thanks Luki,

Just what I was thinking...
 
Still, it's not a true 4k camera.


its like saying the hvx200 is a true 1080p camera.

Pixels are tricolored data sets.
Galt is not entirely biased in this article. He provides technical evidence of his claims. He's more knowledgeable than half the people on this board.
 
So on one hand, you have a Genesis that measures < 33% of it's horizontal pixel count and a RED that measures > 78% of it's horizontal pixel count. Both use colour filter arrays.

Graeme
 
The camera has a 4K sensor.

It records a 4K RAW file.

That is converted to 4K RGB.

But in practical measurement of line resolution after debayering, it's around 3K.

So saying that "RED is not 4K" is so vague as to be useless because one hasn't defined what they mean by "4K". RED is a 4K RAW camera, it can create 4K RGB files, so I think as long as you don't drop the word "RAW" next to "4K" you are accurate to describe the camera that way.

But the other problem I have with this fixation on the term "4K" -- both for those that promote the RED and those that criticize it -- is that what exactly are you comparing the RED to? What is our frame of reference for "4K"?

You can't say that it is a 4K RGB scan of a 35mm film negative because that might not measure out to having 4K resolution either. You generally want to scan film at a higher resolution than what it records, hence why 35mm film is scanned at 4K, even 6K, though it may not produce more than 3K at times in terms of measurable resolution.

So what other frame of reference is there? Is there a 3-sensor camera out there, 4K per sensor?
 
I had a producer say to me the other day, "Why is the genesis so expensive to rent, i mean, you got the Red cam with 4 times the resolution"

It's bloody crazy. There is no relationship between photosites and pixels anymore. It's all marketing. And it's tiring.
 
I define 4k as in PIXELS, tricolored data sets. There is no true 4k camera. Not even IMAX is full 4k. It's MTF is 0% at 4k.
 
Daniel S Williams said:
Still, it's not a true 4k camera.
Click to expand...

I'm going to go out on a limb here...

< heavily opinionated mode />

Who gives a crap? Let John argue his point. Let Graeme and Deanan and Rob refute it. At the end of the day, they all know more than me. And past a certain point, some people will believe John. Some people will believe RED.

At the end of the day, I ask these questions:

1) Does the camera work for you?
2) Does it look good?

Everything else, past a certain extent is superfluous. Thanks, next subject please.

< /heavily opinionated mode >

Lucas
------
VP of Resolution and Free Opinions
ASSIMILATE, inc.
LA, CA, USA
 
"Everything else, past a certain extent is superfluous."

Not true, especially if you're supervising post production on a RED production. You need to understand the camera back to front. You need to know the output, the relationship between photosite and pixel
 
All that counts is measured resolution. And all that detracts is aliasing. Point a RED at a zone plate and you'll see around 3.2k resolution and minimal aliasing. Point a Genesis at a zone plate (and as far as I know, nobody has every published such a test) and you must measure < 1920 lines and who knows about the aliasing?

There has not been any relationship between photosites and resolution of the final image in such a long time... What counts is what you can measure. There are so many different and valid ways of sensing an image that all you can do to compare them is use good test charts and measurements of the end result image. At RED we've said, since day one, exactly how many photosites the sensor has, and that it uses a Bayer colour filter array. We also employ a good optical low pass filter and again, right from day one said you'd get >75% of the horizontal photosite count as measured resolution. You can probably dig up the old DVXuser post from me stating that. We've been more than brutally honest about how all this works, and what the advantages and disadvantages of our approach is.

Graeme
 
Daniel, if you could point out a camera that's recorded image measures it's rated resolution without significant aliasing, we'd all be interested - but no camera does that. That's just basic sampling theory at work.

Graeme
 
To paraphrase James Carville, "It's the picture, stupid." Cinematographers took one look at the images from the Red camera and promptly bought it; that's why over 5000 entered the field in such a short time.

Is 1080 larger than 720? Of course it is. Not trying to resurrect a tired old argument, but why do so many regard the pictures from VariCam as superior to F900? Because all that really matters is the way the images look.

Why does the Nikon D3 at 12.1 Megapixels slightly out score the 21 Megapixel Canon 1Ds MkIII on the DxO Mark?

You can't see the genius in the Mona Lisa by counting the brush strokes.

Frankly, this is just another inflammatory titled post from trollville and trusting the analysis of a very less than impartial huckster is problematic.

Good shooting and best regards,

Leo
 
Daniel S Williams said:
Anyways, thanks all, I'll stop posting as everyone gets so defensive.
Click to expand...

Give me a break. Someone like Graeme takes the time to give a reasoned argument to dispute your initial statement, and all of a sudden you're taking off because "everyone gets so defensive."

We're not defensive - we've just all been through these arguments before and the path you're going down serves no real purpose other than to inflame.

Lucas
 
Daniel S Williams said:
Still, it's not a true 4k camera.
Click to expand...

Yes, that is correct that it doesn't deliver a nominal 4K image within it's 4K of pixels. Then again, it will be a true measured 4K (or very close) once RED enables the 4.5K shooting mode. Show me one digital system that actually delivers a measured resolution equal to pixel count. Seriously.

Here's another news flash for ya... I think most people here already know this.

its like saying the hvx200 is a true 1080p camera.
Click to expand...

No, that's something quite different.

Pixels are tricolored data sets.
Galt is not entirely biased in this article. He provides technical evidence of his claims. He's more knowledgeable than half the people on this board.
Click to expand...

Where he becomes biased is the claims he makes in regards to the products he endorses. He doesn't paint an equally detailed picture on the flip side of the canvas here.

Anyway, the article has been discussed to death, this is one dead horse that has been beaten to a bloody pulp.
 
Status
Not open for further replies.
Back
Top