Welcome to our community

Be a part of something great, join today!

  • 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 - examples of natural, "filmic" color?

I think that depends on your perspectives and prejudices. I've seen enough epic vs alexa tests (they just won't stop with the tests) where epic has been as good or better than alexa.

There one particular test where people insisted that they didn't need to test epic because alexa had better skintones. They tested and epic won the skintone test. This was for Victoria's Secret, Every skin tone and shade you could imagine on real life models in real shooting environments under different lighting, daylight, strobe, tungsten and HMI.

If for Victoria's secret skintones are better on the Epic why keep going with this thread: it's a clear winner!
 
Deanan- It'd probably be good for you guys to publish some of those tests. Worth a shot and might save some time.

Almost certainly not Red's tests to publish.

I've noticed a certain reluctance amongst our crowd to share the detailed results of tests.
 
If for Victoria's secret skintones are better on the Epic why keep going with this thread: it's a clear winner!

Because this isn't an Epic vs Alexa thread.

This is an "examples of natural filmic color" thread... and I have so far seen only one actual example of a guy making RED footage in the style that I like so far - Gunleik!

And it's not some wacky style that I want... just the skin tones of awesome movies :)

Gotta find some more examples of RED-shot stuff before shutting up :P

Also, why should RED be happy with only beating Alexa? The goal is to beat film in every way - and that includes capturing the color data needed to make any filmic look that we want from the raw file IMHO!

Bruce Allen
www.boacinema.com
 
Because this isn't an Epic vs Alexa thread.

This is an "examples of natural filmic color" thread... and I have so far seen only one actual example of a guy making RED footage in the style that I like so far - Gunleik!

And it's not some wacky style that I want... just the skin tones of awesome movies :)

Gotta find some more examples of RED-shot stuff before shutting up :P

Also, why should RED be happy with only beating Alexa? The goal is to beat film in every way - and that includes capturing the color data needed to make any filmic look that we want from the raw file IMHO!

Bruce Allen
www.boacinema.com


It's not a single look taht we're after but the flexibility of creating the look desired.
By starting with high resolution , you can dumb it down as needed.
By including a removable olpf we can provide deep red cutoffs on the ir/olpf
or very early red cutoff or somewhere inbetween.
 
What I can understand from this thread is that nobody has problems with the way RED footage looks. Being RAW it means that it can look like almost anything.

The 'problem' is that we are missing an optional "look" or colorscience in RC-X or in-camera than will give to the footage that 'magic' film tonality that is so immediately achievable with the Alexa. I am 100% certain that something like this is possible.

It seems that the majority of the writers in this thread think that all RED owners or RED users are master graders. This is far from reality and RED has to understand that many shooters need that extra help with their footage.
 
What I can understand from this thread is that nobody has problems with the way RED footage looks. Being RAW it means that it can look like almost anything.

The 'problem' is that we are missing an optional "look" or colorscience in RC-X or in-camera than will give to the footage that 'magic' film tonality that is so immediately achievable with the Alexa. I am 100% certain that something like this is possible.

It seems that the majority of the writers in this thread think that all RED owners or RED users are master graders. This is far from reality and RED has to understand that many shooters need that extra help with their footage.

I have to agree on this statement. I tried very hard to explain RAW to production people myself, minority gets it, after many many arguments, but most dont. They are still stuck with the norm of digital works (WYSIWYG), they also have very very little understanding of what color space, gamma space and LUT is all about. They are the people who is behind the camera and doesnt spend much time in post. Alexa hype boost because it seems pleasant to them out of the box and not many control comparing to R3D. it is simple for them, prores, grade and done. I have also know some of the post here absolutely have zero idea about R3D, they are still stuck with build21 software !! and they come to ask why RED footage is extremely yellowish !

Rivai
 
one small problem I'd like to point out about where this thread has veered off to: everyone is talking about the epic and its skin tones vs. Alexa right now. But the thread started off with the original poster asking for examples of films (feature films etc) that have good color, and since Epic is just being released and to my knowledge no feature films are even out yet that have been shot entirely on Epic, this means by its very definition the OP is asking about Red MX vs. Alexa not Red Epic.
So the question stands, how does Red MX compare to Alexa color-wise?
 
Wrong, not red vs alexa. Examples of natural looking skin tones shot on Red.
That's it. You got an example, show it to us.
 
I repeat

I repeat

Are we trying to reproduce what the eye sees or the way film looks? Two different targets.

Bruce has an issue with subtle variations in skin tones getting lost on R1/Epic footage, particularly at the upper and lower reaches of the exposure curve.

It has even been suggested that pre-DI films from the 80's are the premier examples of great skin tones. It's an interesting point about how independent trajectories in the evolution of motion imaging technology led to a "golden age" of skin tone reproduction. By the 80's the film stocks had improved substantially from the 60's/70's in terms of color accuracy, grain structure, etc and the lens designers were figuring out how to use advanced coatings to get more micro contrast, less flare and overall improved IQ. Until the 90's DI for features was still a rarity, so precisely managed photochemical finishing biased toward making bankable stars look their best was SOP.

It has also been suggested that film, and perhaps Alexa, have a better ability to maintain color accuracy in hues that are not a part of any skin tone management done in grading. It's an interesting supposition that by intentionally modulating Bayer pattern decoding algorithms to get desired skin tones you run the risk of compromising overall color accuracy. If true, wouldn't that impact the Alexa in a similar way?

On a specific note, if each electronic camera system has to pick a strategy in terms of IR and near IR reds for both filtering and decoding matrices then as image makers we need to understand how to dance that particular step most artfully. It has been suggested that on R1/Epic adding some blue when shooting tungsten (1/4 CTB for example) will improve skin tones. On the original Mysterium sensor with the first two generations of color science we used to call it "feeding the blue beast", but it was more of a noise mitigation strategy than a quest for better skin tones. My 1/4 CTB filter has been gathering dust since MX and RC2 came along but perhaps I'll break it out next time I shoot people under tungsten and see if I can "fool" the algorithm :-0

I wonder if 16bit Epic material will give us more of the subtle shades Bruce seeks? In theory, 16bits should also carry more chroma information at the exposure extremes. In any case, I believe the Epic creates a terrific data set and with proper utilization of that image information it should be possible to reproduce the complex nuances of human skin without dinging other colors in the scene. I'll check with John Galt and see if that's doable ;-) hehe

Cheers - #19
 
Here is an example of some 80's inspired saturated skin of various shades. This is not so much a comparison and an Alexa/Epic thing, this was shot as an A/B cam setup on a shoot I did recently. It was graded on set by my DIT Stephen Latty on a Resolve to MATCH. Can you guess which is which? Why? What are the visual cues?

Alexa was shot 60fps Prores422 and Epic 120fps 4KHD.

Download them and watch them in a proper viewer (like Photoshop) as they appear a bit desaturated in Firefox and Safari...

askin.jpg

bskin.jpg
 
Remember, in art not necessarily what is most "accurate" is most beautiful. Case in point, I was reading an explanation of bokeh in lenses and how scientifically speaking the most scientifically/mathematically "perfect" construction of a lens would actually result in what's considered very ugly rigid/sharp bokeh. So engineers have to go against the principles of mathematical 'perfection' to actually design lenses in such a way that is most artistically pleasing to our eyes.
So with that in mind, it's not necessarily what is most accurate to real life, or closest mathematically/scientifically to what our eyes see, it's a more abstract question of purity and beauty and the Alexa seems to create images/skin tones that are simply considered by a majority of people to be more pleasing in their color rendition. I have no clue if the Red is actually more true to life, or etc, but this is the point I myself am making.
 
Florian, i would say Top : Alexa, Bottom : Epic

There is some obvious difference on the pillar at the back between each other, highlight characteristic is different. And the one on top has more red on overall skintone and more subtle red on shades of the skin.

Cross finger !

Rivai
 
Further to Florian's example, here are some examples of Kodak Vision 3 250D and Red MX.

7_filmconvert.jpg

7_film.jpg


If you think you can tell which one is digital, I have created a page where you can take a test and see if you are right.

There are 7 more examples here, and it will tell you the percentage you got right at the end of the test:

http://www.filmconvert.com/film_vs_digital_test.asp
 
I hope the top one is EPIC as the bottom shot has green skin tones on my monitor
 
Nigel...interesting test. The 3rd one fool me completely, i nailed the rest. The one that u show in the forum has some subtle differences in the close up of the mouth

Result :
filmdigtest.jpg
 
I must get my eye's checked (or my monitor re-calibrated) in Florian's example I see no hue, and if anything it shows just how well they both can look/reproduce skin tones (both look nearly identical in the example to me). And Nigel's example here almost looks like the same shot, now I'm just viewing these on Firefox(not the best) and on my LED monitor (for viewing not grading), and I'm by no means a professional colorist... so take my comments with a grain of salt.

thanks for the examples!
 
Here is an example of some 80's inspired saturated skin of various shades. This is not so much a comparison and an Alexa/Epic thing, this was shot as an A/B cam setup on a shoot I did recently. It was graded on set by my DIT Stephen Latty on a Resolve to MATCH. Can you guess which is which? Why? What are the visual cues?

Alexa was shot 60fps Prores422 and Epic 120fps 4KHD.

Download them and watch them in a proper viewer (like Photoshop) as they appear a bit desaturated in Firefox and Safari...

I haven't downloaded them but from what I can see in the browser I would guess that the top is ALEXA and the bottom is EPIC.
 
Hi Nigel
On all the digital shots the blacks look a little flat, so that is a bit of a giveaway.
On the swimming pool shots the Alexa is the bottom picture. A good example of how in my opinion how the skins tone look more filmic. With access to raw camera files there shouldn't be a specific look associated with the red camera it should be a choice of looks, pick which one gets you closest to the look your after. Hate to keep repeating it but a red look that mimics the tonality of film that can be set in camera would really change how the camera is perceived on set and in post.
 
Nigel...interesting test. The 3rd one fool me completely, i nailed the rest. The one that u show in the forum has some subtle differences in the close up of the mouth

Result :
filmdigtest.jpg

Yeah those examples are not the best, as the batch they were shot on was not the same as the one that created the LUTs, and there are slight differences between batches. But I put it up anyway as I personally think that even with differences, it's pretty hard to say one looks better than the other. I'm working on some better examples for the test which give a bit more variety visually and with different stocks.
 
Back
Top