Red One 2.8K and Epic 3.5K Maximun Output Resolution?

[Christian Munoz D]

Does anyone knows where come from all this numbers?

Executive Producer of Telenova Productions, one of the founders of the Jackson Hole Wildlife Film Festival & and unofficial called guru of new technology, Barry Clark said:

"The CMOS chips in the Red One and Red Epic are rated at 4K and
5K respectively, but the Bayer-pattern data from these chips must be converted it to RGB colorspace
for post-processing. As a result, the output of these cameras has only about 70% of the resolution of
the chips—amounting to an estimated output resolution of 2.8K in the case of the Red One and 3.5K
in the case of the Red Epic. Regardless of this, the Red One is commonly referred to as a “4K”
camera and the Red Epic is referred to as a “5K” camera. By contrast, the Sony F65 has an 8K CMOS
chip, but outputs a 4K signal. In spite of this, Sony refers to the F65 as a 4K camera, basing its
advertised resolution on its output rather than the pixel-count of its chip." .

More Here: http://jhwildlifefilm.wordpress.com/2011/05/05/barry-clarks-manifesto/

No source and no explanation whatsoever about the origin of his statements. And not only on Red cameras... Arri, Sony, Canon, etc. too.

Can someone explain how these “gurus of the new technology” gets their conclusions?

 

[David Rasberry]

The guy is making assumptions based on generalizations about debayer algorithms. But MTF and resoluton charts posted by Graeme show otherwise. Doesn't change the fact that Reds are still currently the highest effective resolution cameras on the market right now.

 

[David Mullen ASC]

First of all, I think Red has tested and found better than 70% resolution retained after conversion from RAW to RGB -- I seem to recall a measurable resolution of 3.4K from a 4K RAW recording? I don't know, but it was more like 80% to 85%, not 70%.

Second, the Red One can record 4.5K RAW, which would probably come close to 3.7K in measurable resolution. It follows that the 5.1K RAW image from the Epic would resolve about 4K.

Third, all the other digital cameras on the market have similar limitations between the number of pixels on the sensor versus the measurable resolution of the recorded image, partly due to Bayer filter limitations, but also due to limitations caused by being limited to 1080P recording, as with cameras like the Genesis, F35, or Alexa when recorded to 1080P -- in a sense, you want the measurable resolution to be below the recorded resolution otherwise you'd run into aliasing. So until the Sony F65 comes out, there isn't really a camera with more resolution than the Epic. And a 5.1K Epic recording probably resolves 4K if Red is getting something like 80% of the original RAW pixel resolution after debayering.

So the notion that the Sony F65 will be the only "true" 4K camera is a bit bogus, though I'm sure it will create a great 4K image too when it comes out.

Fourth, it's always amazing how cameras that aren't on the market yet are better on paper than cameras that are currently on the market. Say it aint so!

It's amusing that as soon as a manufacturer makes a camera with more resolution than their competition, suddenly resolution "matters", and when they lagged behind the competition in terms of resolution, resolution didn't matter. I guess one should always talk up their strengths and Sony clearly feels that resolution is going to be the main selling point for the F65.

But the fact that a Bayer CFA sensor has less measurable resolution once converted from RAW to RGB is nothing new. Dalsa went through the same controversy before Red when they came out with the 4K Origin camera, the "it's not really 4K" argument. Well, one can argue that 35mm negative isn't really 4K either, so I'm not sure why being exactly 4K is some sort of magic number to hit, but measuring out to 3.5K or something isn't acceptable. Few people mention that a 1920 x 1080 camera is probably going to measure out at 1.5K or so. So I don't get what the big deal is, there is no conspiracy going on. I'm glad that Sony finally realizes the benefits of oversampling for beyond 2K resolution after the Dalsa Origin, Red One, Epic, and Phantom 65 already pointed the way!

 

[Christian Munoz D]

David R. and David M. I completely agree with your argument, but my question still without an answer.

We all in this forum know, or at least the majority, what is the measurable resolution of our Red and Epic.

My question is how this people make this kind of assumptions. I am not worry if my camera is 3. something and not an exact 4K.

The Red camera makes a beatuful images and is affordable. If Arri, Sony, GE, Westinghouse, Toyota or any company makes a camera with similar characteristics or better and is affordable I will buy it.
My whole point is that professionals writing blogs or "White Papers" like this, about a product, need to support with solid sources what they said and at least do a little research before spread wrong information to the masses.

 

[Deanan]

Much of the info in the pdf doesn't look right... from f65 to alexa to dalsa origin to red one.
Not sure I could agree with most of what he's written.

 

[Jannard]

The RED ONE 4K capture measures conservatively 3.2K... same as slow speed film from a 4K scan. 4.5K measures conservatively 3.6K. 5K EPIC measures 4K. Think 80% and you are safe. Old rules of thumb don't apply anymore. Early algorithms weren't written by Graeme.

BTW... fast film 4K scan = 2.8K
F35 = 1.7K (generous)
Alexa = 1.6K (generous)

1.6K measured is 1/4 of 3.2K measured.

Our graphs (proof) have been posted more than one time.

Jim

 

[Jeff Kilgroe]

My question is how this people make this kind of assumptions. I am not worry if my camera is 3. something and not an exact 4K.

Same old story... They often buy into the FUD and manufacturer hype and regurgitate it as if it were fact. Many people don't take the time to understand a lot of the underlying technology that affects similar devices from multiple manufacturers, ie; Bayer Color Filter Array Sensors. People don't take the time to do their own testing or even to see proper demonstrations by which they can make an informed decision. RED is fighting an up-hill battle, always has been since the beginning. They're the new kid on the block who stirred the pot with disruptive technology. Not only that, but they're selling it at a price point, which is much lower than what a lot of the old industry guard charges for lower specs. There were some issues in the beginning or reports of issues... RED One was released as an unfinished product, in "beta" status. All of those early reported issues that kept people away from RED have been addressed and fixed, yet there is FUD surrounding those issues that persists.

 

[David Mullen ASC]

I think to some degree, salesmen assume they are talking to idiots, or at least, the clueless, newbies, etc. -- it's 2011 and they are still trying to talk about how the measured resolution after RAW to RGB conversion is lower than the original pixel dimensions??? That's sooo 2004...

 

[Christian Munoz D]

Much of the info in the pdf doesn't look right... from f65 to alexa to dalsa origin to red one.
Not sure I could agree with most of what he's written.

Absolutelly. That is why I am trying to understand on what grownds these people write blogs and articles like this.

 

[Jannard]

A lack of testing knowledge? A lack of testing? Here-say? Old rule of thumb? Half-rez debayer? Who knows?

David Mullen just shot EPIC. His opinion counts. A ton of others are shooting EPIC features because "it compares with Vista Vision".

Jim

Absolutelly. That is why I am trying to understand on what grownds these people write blogs and articles like this.

 

[Christian Munoz D]

Same old story... They often buy into the FUD and manufacturer hype and regurgitate it as if it were fact. Many people don't take the time to understand a lot of the underlying technology that affects similar devices from multiple manufacturers, ie; Bayer Color Filter Array Sensors. People don't take the time to do their own testing or even to see proper demonstrations by which they can make an informed decision.

That makes sense...bad sense! Is unfortunate that there are many of this kind of professionals on the industry with access to tools like the Internet to spread wrong or outdated information. And worst, there are other professionals that believe what they are saying.

 

[Jannard]

Fortunately there are plenty that know for themselves...

Jim

That makes sense...bad sense! Is unfortunate that there are many of this kind of professionals on the industry with access to tools like the Internet to spread wrong or outdated information. And worst, there are other professionals that believe what they are saying.

 

[Christian Munoz D]

A lack of testing knowledge? A lack of testing? Here-say? Old rule of thumb? Half-rez debayer? Who knows?

David Mullen just shot EPIC. His opinion counts. A ton of others are shooting EPIC features because "it compares with Vista Vision".

Jim

I think to some degree, salesmen assume they are talking to idiots, or at least, the clueless, newbies, etc. -- it's 2011 and they are still trying to talk about how the measured resolution after RAW to RGB conversion is lower than the original pixel dimensions??? That's sooo 2004...

I have the impression that have lots to do with personal ego. The democratization of internet has developed a silent battle for who is the most knowledgeable and popular on this and other areas. Without care on what “weapon” they are using or the harm that it may generate.