Panasonic going towards 8K to the masses...

[Hrvoje Simic]

A close up of eyes in controlled situation isn't really saying much of the overall capabilities of a sensor or camera. I can fool people with my iPhone camera using my studio flashes, it's a different thing when shooting in uncontrolled situations with practicals and mixed color temps.

I have recently shown in a nearby thread what it possible to achieve with your camera when a shooter knows his stuff and a proper transformation is applied.

Light pollution messes up colour performance of any camera.

 

[Christoffer Glans]

I have recently shown in a nearby thread what it possible to achieve with your camera when a shooter knows his stuff and a proper transformation is applied.

Light pollution messes up colour performance of any camera.

I'm quite seasoned with my camera and I've developed my own process with it that I think look great. I'm not saying that what we have is shite, I'm talking about the progression and improvements of the cameras. All of this is about how the tech evolves and Red's position in this development.

What I meant by light pollution is that, at least in my own opinion, a cameras quality should be measured by how it handles in the worst case scenarios, not the ideal ones. All cameras on the market handles very well under controlled situations, just look at Steve Yedlin's test and you see what I mean. However, the actual quality of a camera should be done using the worst scenarios, because those worst scenarios might happen on many upcoming projects. It's the same idea behind how to write compelling characters in screenplays, their quality is shown by how they handle all the shit that they go through. The worst break down while the best conquer.

Light pollution, practical lights in bad looking locations, low light, rolloff into shadows and highlights behavior, color handling, color look, noise structure and so on, all show what the camera is really capable of. The MX sensor was pure shit in mixed color temp situations, the Sony A7s had great low light capabilities but use it hand held and the rolling shutter rendered the camera useless, the Dragon STH OLPF made strange optical sharpness problems (I demonstrated this a while back) while having a really high noise floor even at ISO800 (with complaints from a lot of VFX people)... and so on. The Alexa sensor however demonstrated (and still does), much better color handling of mixed lighting and color temps, while looking much better in tungsten than Dragon.

This is why I'm generally tired of comparison videos between cameras since they most of the time shoot something in-doors in controlled ideal situations with charts and things you will never see in any real projects. The end result is that you really don't know how it actually looks or behaves in stressful problematic conditions. Say a test is done under daylight lights, most cameras works best at K5000, most sensors are rated there. So what happens when shooting in a low light, moody location using light bulbs that are dimmed down to half? Shoot with the Alexa and the Dragon in such conditions and it's a night and day difference between them.

 

[Zeb B]

A close up of eyes in controlled situation isn't really saying much of the overall capabilities of a sensor or camera. I can fool people with my iPhone camera using my studio flashes, it's a different thing when shooting in uncontrolled situations with practicals and mixed color temps.

How do you know how this, or any shot, was lit? Just curious

 

[Christoffer Glans]

How do you know how this, or any shot, was lit? Just curious

Soft light source reflection in her eyes with dark surroundings, obviously a window or similar. Plus, a close up like that always looks better than say a wide shot putting strain on the resolution of a sensor. It's also one light source, meaning no contaminated color temp differences. It's not a difficult image to pull of nicely with any camera so I don't feel it's proving anything really. A competent photographer with lesser gear can make any images look good, but what this is about is about the highest quality acquisition technology possible, not what's "good enough". 20 stops of 8K and low noise is a huge baseline for acquisition, meaning if it's better than the Monstro but will be available for a fraction of Monstros price, that's a big deal. Yes, it might be available 2020 and yes, Red might have a new sensor in the pipeline until then, but if the price tags are going in the direction they're going for Red, they may very well end up higher than the Monstro sensor investment cost. So if Panasonic release this sensor and it goes into the GH60, that is a big big deal for the future arena of filmmaking technology. It might just be the return of the DSLR era and for any company that isn't Arri, this is gonna change up the way primarily independent filmmakers goes about. Just look at Steven Soderberghs latest film.

 

[David J. Buchanan]

Here is an exampel of such project we did, where things for sure could have been done better.

Blinking red light in a red environment with mirrors. The colorist did a great job to rescue some of the shots but for sure, balancing the light in camera would have brought out tons more quality of the alexa sensor in this case.

So to me the broken skin tones and what not is not a red vs arri thing, it´s about how you light and how you filter as both have their limitations and they need to be watched and cared for, or shit will simply fall of the plate. But to me its quite easy to see that monstro has a way larger plate than what Arri are currently strutting.



https://vimeo.com/231855072

Is this supposed to be a diss on Arri? Because this looks incredibly lol. More like "Shut up, and take my money!"

Anywho, not as impressive as the Pana news but Sony is cooking up something as well.

http://www.canonrumors.com/industry...nverter-that-enables-global-shutter-function/

 

[Björn Benckert]

Is this supposed to be a diss on Arri? Because this looks incredibly lol. More like "Shut up, and take my money!"

Anywho, not as impressive as the Pana news but Sony is cooking up something as well.

http://www.canonrumors.com/industry...nverter-that-enables-global-shutter-function/

Yes its a dis on Arri preforming under heavy tinted light, just as when shoting red you can not dial it back and as its pretty much only one color channel present so resolution will be cut in a third. Image will look low detailed and noise will surface.

Not talking about the beginning and end where the light is more in ballance bit the red mid part. But sure as long as there is girls dancing everything looks smashing.

 

[Hrvoje Simic]

All of this is about how the tech evolves and Red's position in this development.

You didn't see a bunch of colour modes and gammas in Red raw settings ?
You missed five sensors in the last 12 years, each bringing some improvement ?

So what happens when shooting in a low light, moody location using light bulbs that are dimmed down to half?

If it's a poor quality light image typically turns to crap with every camera, colorist gets the thick end of the stick and some tend to have unrealistic expectations and blame the camera.

Light pollution messes up every camera performance and Alexa also goes bonkers with some spectral frequencies.
Camera stress test is based on signal extremes, not on messed up scene, because there are too many variables. Practicals on a serious shoot have controlled and tested quality bulbs inside, not the cheap bulbs from the nearest store.

Also, in matters of colour, with typical tests you are never testing a camera alone, but the transformation to viewing standard as well.

 

[Zeb B]

Soft light source reflection in her eyes with dark surroundings, obviously a window or similar. Plus, a close up like that always looks better than say a wide shot putting strain on the resolution of a sensor. It's also one light source, meaning no contaminated color temp differences. It's not a difficult image to pull of nicely with any camera so I don't feel it's proving anything really. A competent photographer with lesser gear can make any images look good, but what this is about is about the highest quality acquisition technology possible, not what's "good enough". 20 stops of 8K and low noise is a huge baseline for acquisition, meaning if it's better than the Monstro but will be available for a fraction of Monstros price, that's a big deal. Yes, it might be available 2020 and yes, Red might have a new sensor in the pipeline until then, but if the price tags are going in the direction they're going for Red, they may very well end up higher than the Monstro sensor investment cost. So if Panasonic release this sensor and it goes into the GH60, that is a big big deal for the future arena of filmmaking technology. It might just be the return of the DSLR era and for any company that isn't Arri, this is gonna change up the way primarily independent filmmakers goes about. Just look at Steven Soderberghs latest film.

You sound upset bro. Like maybe not related to whatever RED is doing or not doing. I wouldn't put too much investment or faith in any corporation or government. Be who you are with little regard to external events like what RED is doing with future product. Heck, I'll bet if you pulled back the curtain at RED you would find a dude reacting day to day without a bunch of precision . . . not a magic wizard who has it all figured out ;-)

 

[Christoffer Glans]

You didn't see a bunch of colour modes and gammas in Red raw settings ?
You missed five sensors in the last 12 years, each bringing some improvement ?

What? I'm talking about this development, the Alexa LF and the Panasonic sensor this thread is about and their development and how Red counters it. Five sensors in 12 years while Arri completely dominated the narrative film industry is saying the opposite of good. And increasing the price more and more each cycle moves further away from the initial philosophy Red had about democratizing filmmakers. I didn't miss a thing, I see things a little bit too clearly.

If it's a poor quality light image typically turns to crap with every camera, colorist gets the thick end of the stick and some tend to have unrealistic expectations and blame the camera.

Light pollution messes up every camera performance and Alexa also goes bonkers with some spectral frequencies.
Camera stress test is based on signal extremes, not on messed up scene, because there are too many variables. Practicals on a serious shoot have controlled and tested quality bulbs inside, not the cheap bulbs from the nearest store.

Also, in matters of colour, with typical tests you are never testing a camera alone, but the transformation to viewing standard as well.

Bottom line is still that Alexa is favored by the best cinematographers and directors in the narrative film industry, meaning the people who's "eyes" may be considered the most competent in deciding image quality are the ones not choosing Red most of the time. You are looking at this way too technical, it's not about the technology it's about how the image looks in different conditions. The Alexa looks way better in more conditions than Red, the tech talk doesn't change the fact most experienced and seasoned professionals in the creative field think like that. The only way to respond for Red is to look at why these people like the Alexa more than Red. It's not how many stops or how "accurate" the colors are, it's how the colors look and how the stops are handled. I'm still waiting for a Red sensor that tries to focus on feeling like film. Film has over hundred years of development time behind it, the look has been experimented into absurdity, but Red wants to be rebels and not go for replicating that look, like Alexa does? This is one of the main reasons most cinematographers choose Alexa over Red, there are too many cases of this just to dismiss it as nonsense, even I who own a Weapon 6K like the Alexa colors better. I just needed the resolution, if Arri had a 6K camera I would have gone for that in a second.

You sound upset bro. Like maybe not related to whatever RED is doing or not doing. I wouldn't put too much investment or faith in any corporation or government. Be who you are with little regard to external events like what RED is doing with future product. Heck, I'll bet if you pulled back the curtain at RED you would find a dude reacting day to day without a bunch of precision . . . not a magic wizard who has it all figured out ;-)

Upset? No, you can't read emotion into a text like that, I'm to the point and straight forward with what I think, that's it, no bullshit, straight talk
My interest in Red's development is that of caring for Red keeping up with Arri, since I've been on board with Red since 2007, I've invested a lot of time in influencing people in the industry here to look at Red as a serious option (and far too many lack knowledge of Red as a system, dismissing it because of it). But I'm not interested in being the most popular member of Reduser, I'm probably more vocal than most of the Red veterans, it's just in my nature to adress what I think are issues instead of praising everything that's presented. The day I become an optimistic extrovert fanboy is the day I kill myself :tongue_smilie:

 

[Zeb B]

Not sure that being a dedicated Brand Advocate is worth the effort anymore. In 2018, why not just fill a market need what ever that might be? RED is a privately held, not for profit company that is run by a couple guys who are going to do what they think is fun and cool while having a excellent life adventure. Perhaps it used to be that engaging folks like yourself personally on RED User was the fun and interesting thing to do for them. Maybe they moved on. Maybe there are different mountains to climb. Maybe the phone project is the new conquest since RED pretty much did all they set out to do in the Movie Industry. That has been explicitly stated here by JJ

Everything changes

 

[Wayne Morellini]

Sorry, not for profit. When did that happen?

 

[Karim D. Ghantous]

Not sure that being a dedicated Brand Advocate is worth the effort anymore.

IMHO, it is. Because brands have meanings. Leica is the number one brand for me, because the products are second to none in terms of desirability, and often superior in pure functional terms. Leica also has a meaning, a vibe, that goes beyond the mere hardware.

A brand is not a label that you put on a product. It isn't even a promise. A brand is a story, and you cannot fake it, rush it or buy it.

This does not mean that I approve of everything Leica does. In fact I don't know why they put video in the M 240. But they got smart and left it out of the M10. What other company would be so swashbuckling, even though the decision was obviously the right one, and advocated by many photographers?

 

[Hrvoje Simic]

What? I'm talking about this development, the Alexa LF and the Panasonic sensor this thread is about and their development and how Red counters it.

What about them ?

Alexa LF is the same tech packed in a larger format.
Organic CMOS tech has only recently been discovered and breakthroughs like these don't grow on trees and appear every year.

What do you know about Red developments over the years, the challenges they had to meet to bring things to reality or what is next to follow ? A stop increase in DR is "marginal" ? You design sensors and electronic processing or run the production so can objectively valorize ?

Five sensors in 12 years while Arri completely dominated the narrative film industry is saying the opposite of good.

Fixations tend to distort the perception of reality.

In reality people have an 16 bit raw ultra high definition high DR cinema quality imaging device. And tend to get ungrateful for what they have wanting more for less.
In reality developments are welcome and their frequency varies depending on company and its philosophy.

And in a wider context, a 12 y.o. project started by a small group of camera enthusiasts who have never built a camera before, reaching a high end domain, ending up on the same set as Arri, being adopted in Hollywood with tech implemented in Panavision is nothing short of outstanding achievement.

And with accurate Logic if Johnny does something best and Marc does something good and more frequently moves a step further does not mean that Marc sucks. It means he has a different approach. The approach which in this case allowed you and many others to have something you could otherwise only dream about.

And increasing the price more and more each cycle moves further away from the initial philosophy Red had about democratizing filmmakers. I didn't miss a thing, I see things a little bit too clearly.

Last time I checked Red has has spread its market in two directions, a more affordable segment and more high end.
And now you can get a 16 bit raw UHD cinema quality Red camera for less then ever before. And superior products then before.

Meanwhile you have a problem with rental. Of cameras designed for multifunctionality and dominantly ownership. And use 60-100K specialized products of a 100 y.o. company with a different philosophy and market approach as a reference for your valorization. So sorry, but the clarity you mention seems questionable.

Bottom line is still that Alexa is favored by the best cinematographers and directors in the narrative film industry, meaning the people who's "eyes" may be considered the most competent in deciding image quality are the ones not choosing Red most of the time.

Yes, and they are choosing Arri because of:

a) the image
b) philosophy
c) functionality
d) standardized accessories
e) brand perception, experience & heritage from 100 years of Cinema

So not just the image. The image can get very close.

You are looking at this way too technical, it's not about the technology

Five factors mentioned above.

This is one of the main reasons most cinematographers choose Alexa over Red, there are too many cases of this just to dismiss it as nonsense, even I who own a Weapon 6K like the Alexa colors better. I just needed the resolution, if Arri had a 6K camera I would have gone for that in a second.

Yes it would be cool to have best color and resolution and DR handling and a 16 bit raw codec in a compact body and a film philosophy and sexyness of a 100 year brand. For half the price. Meanwhile unapproachable to others so it rents more.

 

[Christoffer Glans]

Yes, and they are choosing Arri because of:

a) the image
b) philosophy
c) functionality
d) standardized accessories
e) brand perception, experience & heritage from 100 years of Cinema

So not just the image. The image can get very close.



Five factors mentioned above.

As this thread is about just the sensor, that is what this thread is about, but yes and as I've mentioned in the thread about the Alexa LF, there is a problem with the current philosophy Red has, which is different from what they went into the cinema camera business with. The lineup is confusion for most people, the workflow is confusing, camera modularity is confusing and failures tend to happen more often with cables and side-gear and so on. All these reasons are a problem for Red that they seem to adress less than they focus on bragging about their next technological leap. People also grow tired of this. But if we keep to the sensor, yes, it comes close, but it isn't as good and it doesn't feel and behave like 35mm film, which is what most want. Even with IPP2, that should have opened up the possibility for output LUTs that convert everything into 35mm looking images, Red didn't release such LUTs. And answers by them about this feels more in line of "well you can do your own LUTs"; that's not how you market a camera to cinematographers.

Yes it would be cool to have best color and resolution and DR handling and a 16 bit raw codec in a compact body and a film philosophy and sexyness of a 100 year brand. For half the price. Meanwhile unapproachable to others so it rents more.

It's not unapproachable for Red. They recently showed a limited edition Gemini camera with larger photosites. If they released a 6K Vista Vision with base ISO of 1600 and made that sensor to behave on it's base level, closer to how Kodak film stocks behave with the addition of an output LUT that resembles how 35mm film looks and feels both in DR and colors, then releasing it with the bold name of "Red 35" or "Red Celluloid", for around $40 000 I will guarantee you that most high profile cinematographers would stop and actually look at Red for the first time in a long time.

 

[Milano Questa]

Yes, and they are choosing Arri because of:

a) the image
b) philosophy
c) functionality
d) standardized accessories
e) brand perception, experience & heritage from 100 years of Cinema
f) reliability
g) professionality (no skulls / weird juvenile names)
h) no wavelet codec needed (although Red has added ProRes since a few years, for good reason)

fix'd

 

[Hrvoje Simic]

I get it but please don't add your words under my name.

 

[Björn Benckert]

As this thread is about just the sensor, that is what this thread is about, but yes and as I've mentioned in the thread about the Alexa LF, there is a problem with the current philosophy Red has, which is different from what they went into the cinema camera business with. The lineup is confusion for most people, the workflow is confusing, camera modularity is confusing and failures tend to happen more often with cables and side-gear and so on. All these reasons are a problem for Red that they seem to adress less than they focus on bragging about their next technological leap. People also grow tired of this. But if we keep to the sensor, yes, it comes close, but it isn't as good and it doesn't feel and behave like 35mm film, which is what most want. Even with IPP2, that should have opened up the possibility for output LUTs that convert everything into 35mm looking images, Red didn't release such LUTs. And answers by them about this feels more in line of "well you can do your own LUTs"; that's not how you market a camera to cinematographers.



It's not unapproachable for Red. They recently showed a limited edition Gemini camera with larger photosites. If they released a 6K Vista Vision with base ISO of 1600 and made that sensor to behave on it's base level, closer to how Kodak film stocks behave with the addition of an output LUT that resembles how 35mm film looks and feels both in DR and colors, then releasing it with the bold name of "Red 35" or "Red Celluloid", for around $40 000 I will guarantee you that most high profile cinematographers would stop and actually look at Red for the first time in a long time.

Im not sure what film look is as it´s very subjective. Most have probably never really looked at a film negative... it is ofcourse negative in colors and even if you flip them the look is further away from a normal looking image than IPP2 is even without a lut. Then you need to differentiate between the look of a negative and a positive as they look very different, then you got different stocks...

So you got the negative, the developing process then you pick your print stock and strike a positive, then a ton of creative choices have been made. And if you go film to digital then even more so, a film scanner adds a lot, the colourist and the whole process. A raw negative is pretty far from looking appealing.

But I guess you are referring to the dynamic rounding that happens in the highlights of film. So far no Cmos sensor has a feature like dynamic rounding, they all catch light linear to clip. So to mimic such look all that can be done is not to over expose and make sure the highlights are put in a spot below clip where color accuracy is still good and then use digital gain to lift the highlights to white. Then you can choose to round them off and make them white and have good accuracy all the way to the result picture white point.


The bigger DR you have the more stops you can use for rounding. Arri has had a lot of highlight coverage, i.e they used a lot of digital gain in their color science and had good amount of DR to still hold up in the darks. But helium and especially monstro got greater DR so if used right you can have the same nice fallow to clip as arri and have more coverage in the low end. The rest is just creative grading.

 

[Eric Lange]

Just a minor side note: "Celluloid" ?

Celluloid, chemically is cellulose-nitrate... "gun cotton"/ Parkesine.

Explosive and dangerous. Amateur use stopped around the 1930's. Kodak continued its use until about 1951 (for professional/cinematic use).

Meanwhile Cellulose acetate film was used (cellulose tri-acetate) after that for professional use, but was inherently unstable but no where near as explosive as cellulose nitrate/"Celluloid". Hence the name "Safety film".

All of these have been replaced by a Polyester base or substrate that is far more stable.

I find the term "Celluloid" just slightly creepy somehow... (Probably just me). "RED Celluloid" Maybe just too many L's ... CelllyoullololololoiiiiYYd.

Steve Yedlin's article as featured on the "other thread" was interesting as he introduced/mentioned the concept of stochastic and probabilistic methods that could be dialed in / spatial frequencies and particle distribution for "Micro-contrast" for film grain effects. In other words to take certain aspects of film to a more fundamental set of variables. In other words to "abstract" what we like about certain properties of film and then take it somewhere else (in a new direction) if we want to.


_________________________________________________________________________________________________________________________________________________________________________________________________

https://en.wikipedia.org/wiki/Celluloid

https://en.wikipedia.org/wiki/Cellulose_acetate_film

https://en.wikipedia.org/wiki/Film_stock

 

[Hrvoje Simic]

It's not unapproachable for Red. They recently showed a limited edition Gemini camera with larger photosites. If they released a 6K Vista Vision with base ISO of 1600 and made that sensor to behave on it's base level, closer to how Kodak film stocks behave with the addition of an output LUT that resembles how 35mm film looks and feels both in DR and colors, then releasing it with the bold name of "Red 35" or "Red Celluloid", for around $40 000 I will guarantee you that most high profile cinematographers would stop and actually look at Red for the first time in a long time.

We don't know the specs of Gemini's sensor. If it is more sensitive it doesn't necessarily mean it has all other image aspects superior. Though some properties could be assumed.

I agree on the notion of larger pixel pitch. I'd also go with 4K S35 /6K Vista Vision. Makes sense in the context of a bunch of factors. I have zero worries about sensitivity or colour. One more stop in highlights and that's it. 4K S35 16 bit lossless compressed 2.5:1 imagery from Monstro tech with a larger pixel pitch seems just about right. Now that Red won the resolution race it would be cool to offer other options for those who don't need the extra acuity.

 

[Blair S. Paulsen]

I contend that we are rapidly approaching the practical limits of motion capture using CMOS Bayer pattern sensors. I expect the next truly significant breakthrough to be sensors the can read both photon levels and wavelength at every photo site.

Cheers - #19