Hello!

i'm using luts+ to make luts for DCP XYZ conversion tests. I made P3 to XYZ and XYZ to P3 and applied them in two nodes in Davinci resolve. I compare the original and after, and i don't have the same images. It seems all is down a bit.

Is someone have an idea?

Thank you

That's exactly s expected.
The colour space white points are different, so to prevent colour channel clipping the conversion has to drop the peak white to compensate.
You can see this looking at the 1D LUT graphs, where the difference in white point is obvious.

Steve

Hi Steven,

thanks for you answer. If i had a DCP made with the p3toxyz lut and i want to make a prores output of it for example, how can i have the original image?

Vincent

If you need to perform a colour space conversion between two colour spaces with different white points, you can't maintain the same peak white level. If you do, you will have clipping of one or two of the colour channels.

Steve

So for p3 to XYZ i can't maintain the same peak white level too?

In other words,

how can i get the original image from the dcp?

Thank you,

Vincent

You would have to use Peak Luma for the return trip.
That will return the peak luma values, but depending on how the P3 to DCI/XYZ was performed you 'may' get colour channel clipping.
That is all dependent on the original conversion process used...

Steve

Ok i will test that.

Thank you!