Abby Bader

It is a common issue with retrieving digital highlights that it can introduce tints along the line of magenta or cyan. To eliminate the tint you can desaturate the highlight with a gentle lum vs sat curve. 

Jussi, seems like we posted our posts simultaneous 

In the 'color management' settings you can check "use mac display color profile" and see if that solves the issue. Unfortunately, it might not take you all the way as matching the Resolve viewer with a player that doesn't read video attributes on an Imac could be a lost case. Also check your video vs. full range settings.

DCM is a non-destructive way to transform to another color space, meaning you can go back and forth as many times as you want without experiencing any quality loss.

LUTs are normally built with a more pleasing tonal curve than what DCM provides. Designed for a better visual result. 

LUTs are destructive and values that are clipped can't be recovered in nodes after the LUT. That is not the case with DCM transforms. 

2 hours ago, dermot.shane said:

i have accurate filmout luts from many vendors, Technicolor, Deluxe etc, but they are only used in the monitoring chain, and removed from the rendering path.

Are you still delivering for print?

The % is a work space only limitation and can be extended by holding the mouse cursor over and sliding to the side. Then you will be able to retime the speed with no restrictions. 

The use of printer lights in digital processing is not an accurate science. The photomechanical process was affected by so many random parameters that we don't know how that mathematically transfers to digital data. The important thing is that you can make precise corrections in the color channel of choice, both if the correction is altered through a LUT or not. But, it certainly helps if the input data is log.

Darkest parts, middle tones, brightest parts?

You could stack a couple of layers, create a mask around the blemish and ofset the first layer. That's a quick fix that works on some shots inside your color corrector. 

Panasonic chose to unveil its new TVs at a Deluxe facility is that both Deluxe and Technicolor – which between them deliver post production on around 70% of Hollywood’s blockbuster films each year – now officially use 2017 Panasonic EZ1000 OLED TVs as one of their main large-screen mastering monitor solutions. Deluxe alone has more than 60 EZ1000s installed in its mastering suites around the world.

https://www-forbes-com.cdn.ampproject.org/c/s/www.forbes.com/sites/johnarcher/2018/01/06/panasonic-unveils-two-new-hollywood-tuned-oled-tv-ranges/amp/

I guess @Daniele Siragusano and @Andy Minuth could answer that as they are both working for FilmLight. 

Many colorists use the offset controls to balance because it's global controls that doesn't break the relationship between lift, gamma and gain. 

In this video @John Daro talks about FotoKem's ACES workflow on the Extreme Weather project . Worth looking at. 

Welcome to this forum Matt!

The Print Emulation LUTs were originally applied to the end out the chain to see how the grade would look when printed on a specific film stock. This was necessary because the look of the digital film master was radically changed once printed on a stock. E.g a Kodak 2393 LUT were used to look through if the destination stock was Kodak 2393.

Today, we don't need to use the film LUTs this way, simply because the digital master files are not printed on film.

What  you are describing is a modern approach where the same data is used creatively in the color process. If you want that to work properly, it still means that the LUT should be applied at the end because if you do any corrections after the LUT is applied, the calibration will break or the pre-defined and intended look will be changed. That also means that the LUT will be applied after the initial base correction with a log to linear transform, which means that the image isn't logarithmic anymore. That is a problem because the typical so called 'creative' print emulation LUT's have a built in ormalization curve. 

So, as suggested above, you need to use an invert LUT to transform the image back to log state after your initial correction. Or, you can use a LUT ment for linear input data.

There are many inaccurate representations of printer points online, but this resource explains the concept very well. In DaVinci Resolve you can use the offset controls in the primary tab.

I just want to share this with you, colored by Peter Doyle. What an artistry, beautiful work. 

Texture is love!

The different channels got a numerical value that is pre-set in every color corrector. That means if you change one of those numerical values without changing the others you will break the natural relationship between them. The other way around, by grouping them all together you will retain the relationship. That said, it will only be as natural as your initial starting point, that often points to how the camera is set up to reflect the real world. 

Babel!

You could use a splitter combine node and reduce some noise in the different color chanels. Temporal noise reduction will most times give the best results. I would not crush the blacks compltely but spread the information through the low IRE range.

Other takes in low light scenes could be to sharpen the eyes and/or add some small color nuances in the skin to gain a better separation from the background since the luminance levels are so close. 

Great thoughts guys!

Look at the emboss and edge tools in the openFX menu. You can also degrade the image by changing the texture in the grain parameter and also adjusting the values in the grain advanced controls. And you can check out the JPG damage and lens distortion too.

I work in A/B and C/L for the most and use regular curves (not 2D Curves) mainly for sat/luma (and opposite) corrections. Most tasks can be solved in the regular color correctors, but pushing the grid can solve some challenges faster so it can be a great supplement. I use HSPe btw.

Can you elaborate more on the Arri workflow @Tom Tomlinson?