What did I do Exactly?

To be absolutely transparent about what you’re gonna get if you wish to purchase the Portra 400 profile here is what I did:

• I shot a ton of color charts (color checker classic and color checker sg) under various lighting conditions side by side on Portra 400 and Sony FX3 (Sgamut3/Slog3)

• I scanned the negatives with a Reflecta rps 10m film scanner using Vuescan to get linear raw scans of the negative

• The Raw scans are then inverted using the Cineon Math using the film base of the negative obtaining effectively a Cineon Scan of the negative

• The Cineon Scans have been then balanced on the middle grey patch using offset as it’s done in the classic photochemical process

• Once the scans were balanced, the color patches values have been fed to the neural network to match the Sony FX3 to Portra 400

• A film print is then applied to prepare the scans and the digital images (once matched to the negative scans) for the display and output rec709 gamma 2.4

What you are going to get:

With the download you’re gonna get 3 LUT’s and 2 PowerGrades, a 4K grain plate for video and a 4K grain plate for stills (one frame).

LUT’s:

• NEGATIVE + PRINT LUT

• NEGATIVE + PRINT LUT + Extra Saturation

• NEGATIVE alone (without the print component so that you can apply any print LUT of choice of even grade it manually as if you were grading a Cineon Film Scan)

Powergrades: the 2 PowerGrade’s are identical with the only difference that one uses a matrix to fit the still photos workflow (more on that below)

The NEGATIVE+PRINT LUT expects SGamut3/Slog3 not SGamut3.cine as input and it outputs Rec709 gamma 2.4

The NEGATIVE only LUT expects SGamut3/Slog3 not SGamut3.cine as input and it outputs Cineon Film Log

The workflow for photos is explained below.

The 4K Grain Plate has been generated by utilizing statistical models to reproduce the appearance, distribution and randomness of film grain. I wrote a python script to generate Boolean distribution noise based on this paper here: https://www.ipol.im/pub/art/2017/192/article_lr.pdf

The grain behaves as real grain would. This type of approach is similar if not the same as the one Steve Yedlin uses, as explained by his colorist Walter Volpatto.

Does it work with ACES, Davici WideGamut color management?

No, these LUT’s have not been built with these color management systems in mind. Honestly I’m not a big fan of ACES and Davinci Wide Gamut. Does this mean that you can’t use color management? No, you are still using color management, only a manual one instead of ACES or DWG. If you know what the input is supposed to be what the output is, you are color managing correctly

Normal LUT and even more ACES and DWG have big problems with out of gamut color that my LUTs don’t suffer from. Why? Before starting the matching process I undo the colorimetric matrix that the camera manufacturer applies post debayer. This matrix causes highly saturated colors to “explode” becoming even more saturated and causing out of gamut issues, the shouldn’t be there in the fist place.

Does it work with any camera?

Yes, the LUT’s work with any camera, but have to use them correctly. Here’s how:

The LUT’s expect SGamut3 (not SGamut3.cine) as color space and Slog3 as gamma. As long as you shoot log with your camera and you’re camera is supported by your software of choice with by a Color Space Transform, you will be able to use the LUT’s accurately. Digital cameras are very similar to one another even if manufactures would like us to think otherwise. A color space transform is a mathematically accurate way to transform between color spaces and gammas and make sure you can use LUT’s that were created using different sensors. It’s just math and it works!!!!

N.B: If you’re a Sony shooter the LUT’s will work natively with your camera, but keep in mind that the color space to be used is not SGamut3.Cine but Sgamut3. SGamut3 as opposed to SGamut3.cine is closer to the native space of the camera ad what the LUT’s have been built upon. If you don’t shoot directly with Sgamut3 you ca use a color space transform.

Does it work with still cameras?

YES, AND YOU’RE GONNA LOVE IT, but a couple of things probably need to change in regards as to how you edit your photos., workflow wise. (trust me it’s totally worth it)

The premise: Lightroom is a black box that doesn’t allow to fully take advantage of nuanced color science. It’s honestly mind blowing how it came to be such a popular tool despite the fact that you can’t really decide what happens to your raw images.

Hopefully this will change today!

Lightroom uses camera profiles to debayer the raw image into prophoto color space linear gamma. This is lightroom working space and this is where you’re doing all your color editing in lightroom.

The problem with Lightroom is that it does some manipulations under the hood that can’t be undone which compromises the creation of a reliable color pipeline. In order to escape from this adobe imposition we need to use a software that allows to do exactly what we need and want to our images.

The workflow: RAW image debayering in RAWTherapee (free) -16bits linear tiff image export- Import image in Davinci Resolve-(custom, reliable and nuanced color science)-EXPORT FINAL IMAGE

With the download there is a video explanation with the exact process to follow.

Here a quick overview of the process (which is incredibly simple). You just need to batch export all your images with some specific settings from RawTherapee, open them in resolve, apply the PowerGrade (photo version) and you’re good to go!

In the gallery below, you can see some images exported from Lightroom in adobe standard, along side the same image exported from RAW therapee and with the Portra PowerGrade applied in Resolve.. The images are straight from Lightroom and and straight Resolve with the Portra Profile applied (No extra color grading has been done)