Welcome to our community

Be a part of something great, join today!

Register Log in
  • 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 :)

LIGHT IRON COLOUR MATRIX / IPP2

James Miller 72

James Miller 72

Well-known member
Joined
Jul 29, 2014
Messages
105
Reaction score
2
Points
18
Location
West Sussex, UK
I've been messing around with a whack together of the Matrix of Light Iron Colour. It's a bit ad hoc with conversion from DC1 RG2 in the mix. Separated from any gamma curves just the colour.

If you want to try them load them as IPP2 luts in camera or in RCX. Download here https://we.tl/53GrS4ODca

I've been using as in 2k ProRes Baked-in.

Rec.709
Low Contrast
Very Soft

F003_C097_0304WZ_001.jpg
 
Peter Lyons Collister said:
James,

How are you backwards engineering this? Do you have some r3ds from DXL2? Just curious.
Click to expand...

I used the .r3ds that they released for DC1/RG2 then cross converted, removed gamma curves and adjusted for RWG LOG3G10. But undertone is Iron Colour from what Panavision supply for post. DC1/RG2 even when in Legacy mode and just monitoring, doesn't sit right, well at least with Helium. So thats when I wanted to just use matrix and make something IPP2 based.

I do wish that RED would allow an option of a LUT to be end of chain and not IPP2 colour constrained but keeping curves and roll.
 
James Miller 72 said:
I used the .r3ds that they released for DC1/RG2 then cross converted, removed gamma curves and adjusted for RWG LOG3G10. But undertone is Iron Colour from what Panavision supply for post. DC1/RG2 even when in Legacy mode and just monitoring, doesn't sit right, well at least with Helium. So thats when I wanted to just use matrix and make something IPP2 based.

I do wish that RED would allow an option of a LUT to be end of chain and not IPP2 colour constrained but keeping curves and roll.
Click to expand...

There's not much stopping you from adding a LUT to the end of the chain, just not in RCX or in camera. There's advantages to where IPP2 Creative Cubes are implemented now in the grading chain for sure.

Also, there's new LI color for DXL2 fyi. This is mostly relevant for Dragon DXLs.
 
Phil Holland said:
There's not much stopping you from adding a LUT to the end of the chain, just not in RCX or in camera. There's advantages to where IPP2 Creative Cubes are implemented now in the grading chain for sure.

Also, there's new LI color for DXL2 fyi. This is mostly relevant for Dragon DXLs.
Click to expand...

Yes of course, easy to do in Legacy but you can't bake the LUT in that way. And getting to nitty gritty banding issues using ProRes below 4444 have problems when not baking in IPP2 or a creative lut in the pipeline. Totally see why its done like that to maximise different delivery colour gamut/gamma options but I still miss not having an option for the end of chain without secondary gamut adjustment. I use IPP2 with creative Luts all the time, its easier in the long run and gets picked up in RCX and somehow red player does a nice just of keeping the creative lut for preview without loading, just from meta.

I transformed the LI matrix from DC1 RG2 Dragon over to RWG LOG3G10 just really keeping original influences in colour mapping from the LI lut.
I didn't have the second generation LI2 but that was talked about only to fall within IPP2 etc. I was just moving rec to rec and all things being equal (not that they are) mapping from a balanced image to another balance image will have same relative shift of colours.

Over the past few years I have profiled many cameras so I used my dragon > helium data for initial primary transform.

All just for fun though :)


James
 
Love the effort of this LUT! But playing around with it I have seen serious issues of banding and noise. It is not useable in a grading scenario- I guess as a viewing LUT on set it could be fine. Would be great for someone with the know how to reverse engineer a solid LUT that is similar to the Light Iron color for DXL2 but that doesn't fall apart like this.
 
Eric Haase said:
Love the effort of this LUT! But playing around with it I have seen serious issues of banding and noise. It is not useable in a grading scenario- I guess as a viewing LUT on set it could be fine. Would be great for someone with the know how to reverse engineer a solid LUT that is similar to the Light Iron color for DXL2 but that doesn't fall apart like this.
Click to expand...

Where and how are you applying this? Trilinear has been the culprit for me in the past. This looks great on 8k footage when pixel peeping in both RCX and Davinci for me. Wonder what's causing that.
 
While not knocking James effort or generosity it should be noted that a modified LUT will give you only a facsimile of the DXL look. It might be close but the biggest issue is that Panavision replaced the colour filter Array in front of the chip with what they believe is a superior one. So DXL sees colour information differently. Unsure how big a difference it makes but enough for Panavision to believe their more expensive CF Array is worth it.
 
James,

I've been testing these LUTs a lot, and I wanted to say they're awesome! I'll try to show some samples soon. In the mean time, thanks!!

-Eric
 
Thanks James! I messed with these, and the colors really did look nice. I had some pretty bad banding in low contrast areas though. I'm not sure what would cause that, but it definitely doesn't appear until the LUT is added. I can send you a still privately if this is something you want to try to work out. Your efforts are appreciated!
 
It works with IPP2 (e.g. colour space = RWG and gamma = Log3G10), but I think James was trying to separate the colour properties from the gamma... As in, this will get your colours similar to LIColour, and then you can dial in the contrast to taste off of Log3G10 or experiment with the other RG1-4.
 
You must log in or register to reply here.
Back
Top