Jussi Rovanperä

At 50 the curve will have no effect, at 0-50 the curve will be inverted...

Not in Resolve, because the color space transform can do the transforms I need. I use matrixes in Fusion quite a bit.

6 minutes ago, Amada Daro said:

Thank you Mitch! Anyone knows how to apply a 3x3 image transform matrix or a function by node in Davinci? I have never bothered to apply a matrix only because I have always wanted the 'film' curve (that is not transformed in a matrix) and not primary and white point corrections.

You can write a matrix/function as DCTL and save that in the LUTs folder, and apply to a node. There's an example of a matrix dctl code in the Resolve manual on page 1248.

Also Paul Dore has written a matrix ofx that can be found here

I'd start with finding the right contrast curve for the footage, pastel stuff is usually low contrast with lifted midtones, and that kind of curve will desat the colors to start with.

I think one problem with profiling Portra or Ektar is that those don't come in 400ft rolls...

Steve Yedlin has been doing film profiling that's impressive, I've understood that he's using a remote controlled Arri Skypanel as the color source, shooting without a lens in camera, and reading the developed film with an i1 spectrometer. Also I think one reason why he likes the Alexa is that with .ARI files you can also access the raw data before demosaicing/wb/color transforms etc.

This is in Fusion:

First node LUTcubecreator created a HALD pattern (the color grid)

Second node applies a color matrix (in this case just a unit matrix that does nothing)

Third node analyzes what has changed in the HALD pattern and writes the LUT file.

 A technical LUT that transforms the color space is created by running values thru a matrix, and then writing the results to a LUT.

It's linear algebra, a 3x3 matrix multiplication, that will transform 3 values into another 3 values.

Typically the color matrixes are color space or color gamut transformations, like RGB-to-XYZ or XYZ-to-RGB transformations.

So if you want to transform color from one RGB gamut to another RGB gamut, you would use 2 matrixes: rgb1-to-XYZ and XYZ-to-rgb2.

This is just a super simplified explanation...

Fusion is free, and Fusion Studio is 300 bucks...

Isn't Temp/Tint is just a gain control?

The Arri log-c to 709 lut is designed to make your footage look nice,

the OFX is just a technical transform from one gamma to another. The log-c to 709 OFX does not have highlight rolloff, and will push values over 1 easily, but can also be perfectly reversed with a 709 to log-c OFX...

CIE is the organization that is behind most of the "modern" color science. Rlab is just baselight's slightly modified version of CIE Lab colorspace.

There's a strong contrast curve and the colors are desaturated. Not a film thing and pretty common look.

In Resolve 14 the color space transform ResolveFx has now tone mapping and gamut mapping options. I've been using the transform nodes a lot, for example converting slog2 to log-c in the first node, and then converting log-c to rec709 with the tone mapping in a later node.

The same could be done with using "Resolve YRGB color managed" and setting input to slog2, timeline to log-c, and output to rec709 with tone mapping enabled. 

I was not fond of this look, but it was interesting to read what was the rationale behind it, as if the film was shot in the 60's, left in the rolls, and found 50 years later.

Printer points only work "correctly" on log footage, under a log-video transform.

Usually printer points introduce a tint in the shadows when pushed far, because most log footage is not purely logarithmic in the shadows.

11 hours ago, Nicolas Hanson said:

I will try LAB colorspace as suggested, as it seems like the chroma is affected with both lgg luminance only adustments and y curve. Most times this will not be a problem, but I need total isolation for a particular project I'm working on. 

You can't really isolate Luma from color, those will always affect each other, because the output is still RGB, and it's easy to push values out of gamut with Luma/Chroma corrections. For example you can't have dark saturated yellow or light saturated blue, those are way out of gamut, but the Luma/Chroma tools will happily try to push the color there, and clip the color channels.

Okay, different methods for sure, these's HSL/HSB/HSV/HCL etc... While the hue is changing, Resolve is preserving Y, while Photoshop is preserving Lightness?

You mean these hue controls?

I just use the Resolvefx grain...

LG might be the only manufacturer of big OLED panels... But FSI is picking the best (and expensive) panels, and test each of them for quality and performance. And the rest of the monitor is their own electronics.

9 hours ago, Orash Rahnema said:

Hi Louis!

There's been a little misunderstanding, i wish was as you'r saying

No, unfortunatelly i don't have a film to grade and in the market where i work i don't think i will ever have the chance. 

So i wanted to do it myself shoting motion pic film in a still camera and see if i can replicate the workflow. 

I will buy the vision3 from where @Jussi Rovanperä suggested, probably the hard part would come with the proccessing and scanning. 

I Remembered reading that there is Vision for still cameras, googled it and that was one of the first links I found, so I have not ordered vision from that ebay seller or others, and can't really endorse any seller.

I think you get rolls of Kodak Vision for still cameras, but not officially from Kodak.

https://www.ebay.com/itm/35mm-Film-Kodak-Vision-3-500T-for-your-35mm-still-camera-/152545145856

Another thing is that if levels in clip attributes has been set to "auto", the auto behavior might have changed with some file formats in 14 vs 12.5. Someone had that issue.

But this doesn't really look like levels change.

Yes, you can use BMD cards for full screen playback with Premiere.