Welcome to our community

Be a part of something great, join today!

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 :)

Calibration software and tools - which ones

M Most said:
The X'Y'Z TIFF files that constitute the DCDM are assumed to have been created from material graded at 2.6 gamma using the P3 color space.
Click to expand...

Steve Shaw said:
As a result the only projectors that can really do XYZ are high-end projectors.
Click to expand...

I must be misinterpreting what both of you have written because there are no projectors that can display XYZ. They accept XYZ input but they do a color space conversion internally to a projector native, P3, or other spaces.
 
The problem, as you say, is that P3 can display colours that Rec709 can't, and the conversion from P3 back to rec709 can cause these out of gamut colours to display incorrectly, if the conversion is not handles well.

A good conversion LUT should actually not require any 'trim' pass - that is a sign that the conversion LUT is not good.

It's making a good LUT that can be less than straight forward...
 
Steve Shaw said:
A good conversion LUT should actually not require any 'trim' pass - that is a sign that the conversion LUT is not good.
Click to expand...

Horses for courses - but I 100% disagree. Even if you map it to an extremely close color - if your "perfect LUT" - it's not the SAME color and sometimes it will just "feel different".

Sure, 90% of folks won't even see a difference - I like to try to make the 10% happy. And we can always get there in a fast trim pass.
 
The problem is that the out of gamut colours will never be able to look 'the same' - obviously - those colours just can't be displayed in the smaller colour gamut of Rec709..

So, yep, a creative alteration may be needed - this is not a trim pass as you will probably have to use secondary controls - back to being a pain again...
 
Steve Shaw said:
So, yep, a creative alteration may be needed - this is not a trim pass as you will probably have to use secondary controls - back to being a pain again...
Click to expand...

We use secondary controls on our 709 trim passes. I was not aware that folks don't call a trim pass a trim pass when you use secondaries.
 
Deanan said:
I must be misinterpreting what both of you have written because there are no projectors that can display XYZ. They accept XYZ input but they do a color space conversion internally to a projector native, P3, or other spaces.
Click to expand...

It is, of course, true that projectors must of necessity display RGB. However, if the proper matrix math is used to go from the color graded material, presumed to be graded while monitoring at 2.6 gamma in P3 color space, to XYZ, the translation to XYZ and thus the projector's translation back to RGB should be relatively transparent - and on every DI I've been involved with, that has been the case.
 
Mark L. Pederson said:
Horses for courses - but I 100% disagree. Even if you map it to an extremely close color - if your "perfect LUT" - it's not the SAME color and sometimes it will just "feel different".

Sure, 90% of folks won't even see a difference ...
Click to expand...

True. Color space conversions are all about perceptual differences, since almost all color spaces will contain some colors that are out of gamut for other color spaces. The art - and I do mean art - of proper gamut mapping is to minimize the perceptual differences in "real" images, not to try to achieve exact matches on chip charts. All deliverables should "look" the same, not necessarily be the same.
 
Yep, Deanan is correct. Bad use of descriptive terms on my part!

However, it is not necessary to use P3 and a gamma of 2.6 for the DI - it is easy to conver to the original to a DCI deliverable XYZ format from any smaller colour gamut and different gamma. It just means the final available colours will not be as wide ranging as they coud be, but they will still be 'right' compared to the original grading intent.

What P3 gives is a large colour pallett, but as said this can cause problems when making a deliverable in a smaller colour gamut. It is a help for a project that has film (or DCI) as the major deliverable for obvious reasons.

Don't you just love this!!!
 
Steve Shaw said:
It just means the final available colours will not be as wide ranging as they coud be, but they will still be 'right' compared to the original grading intent.
Click to expand...

But you did limit the "grading intent" (to rec709) in the first place.

I don't know about most folks, but when I see a movie in the cinema ... I want the biggest screen ... the HIGHEST resolution ... and the widest color gamut.
 
well you can have all 3 of those and still have a crap movie :-)
Story is king and isn't a grade there to add to the story and mood.

So what is the norm then.. grade in P3 if you're going film out then do a separate graded pass for rec709 for any tapes to eventually go on DVD or BR?

And if you're not going film out then just stick to rec709?
 
Steve Shaw said:
However, it is not necessary to use P3 and a gamma of 2.6 for the DI - it is easy to conver to the original to a DCI deliverable XYZ format from any smaller colour gamut and different gamma. It just means the final available colours will not be as wide ranging as they coud be, but they will still be 'right' compared to the original grading intent.
Click to expand...

They won't if the projector is set up for P3, with gamma and white point based on that spec - which it almost always is in a DCI compliant theater. If the projector is using a DCI derived setup, Rec709 material will exhibit crushed blacks and be a bit too blue. Conversely, if P3 graded material is projected on a projector that is set up for Rec709, it will be too light and too pink. To imply that these differences are inconsequential is to lead people astray.
 
roryhinds said:
So what is the norm then.. grade in P3 if you're going film out then do a separate graded pass for rec709 for any tapes to eventually go on DVD or BR?

And if you're not going film out then just stick to rec709?
Click to expand...

Unfortunately, there is no "norm." Many different facilities do DI work in many different ways, and a lot of the methods are based on the most common source materials they're working with. If the source material is primarily Red, grading in P3 and sourcing everything from that via a combination of conversion transforms and trim passes is not uncommon, and is a good approach, but it depends on a number of factors, including the way the DI theater is set up, the projector being used, the lab being used for either recording or printing or both, the primary target deliverable (has been film for years, that is not always the case today), and what systems are being used and how to set them up most efficiently.
 
roryhinds said:
And if you're not going film out then just stick to rec709?
Click to expand...

Hey Rory -

You mean digital cinema exhibition or film out.

More cinemas are going digital every day. Those projectors can, and do, display the wider gamut of P3.

Yes - story is king and you can make a finish and deliver a great movie in rec709.
 
okay so you would grade to P3 for DCI and if it wan't going theatrical then rec709 would be best suited.
If you're going film out then P3 is still the better option unless you have a projector that can reach XYZ?
 
roryhinds said:
okay so you would grade to P3 for DCI and if it wan't going theatrical then rec709 would be best suited.
If you're going film out then P3 is still the better option unless you have a project that can reach XYZ?
Click to expand...

Unless it is a television episodic - I would grade in P3. Period.

You don't know where you project is going to end up - and the DCI spec DCDM deliverable is now requested even on low, low budget feature films as a delivery requirement.

Yes, you can make a DCDM from a 709 file - but - the whole point of "archive" is to preserve your project is the highest quality possible. I would include gamut/color space in that.

In the future - all our displays will show more colors that rec709.

If you don't have a projector that can be set up for P3 - then you need to grade in 709 - as you don't really have a choice.
 
roryhinds said:
okay so you would grade to P3 for DCI and if it wan't going theatrical then rec709 would be best suited.
If you're going film out then P3 is still the better option unless you have a project that can reach XYZ?
Click to expand...

You seem to be looking for rules, and you're also assuming these things are the filmmaker's choice. That is rarely the case. Deliverables are determined and specified by the distributor, whomever that might be. When a project is sold and thus becomes "real", multiple deliverables are always required. If that is anticipated, your post path should take it into account right from the start, hence Mark's comment about using a format with the widest possible color gamut, which these days is probably best represented by P3. Everything else can be and almost always is derived from that without a whole lot of trouble (despite Steve's "pain in the neck" comment to the contrary..). The primary exception to this would be a project that is a television series, which will likely live its entire life on an HDTV format.

OOPS - just noticed that Mark said exactly the same thing while I was typing this. Great minds think alike....
 
I have heard that the new Panasonic Plasma's get very close to P3 and that the JVC HD projector can do P3
 
M Most said:
They won't if the projector is set up for P3, with gamma and white point based on that spec - which it almost always is in a DCI compliant theater. If the projector is using a DCI derived setup, Rec709 material will exhibit crushed blacks and be a bit too blue. Conversely, if P3 graded material is projected on a projector that is set up for Rec709, it will be too light and too pink. To imply that these differences are inconsequential is to lead people astray.
Click to expand...

Nope - that is so missing the point of colourspace conversion it's no longer funny... I refer you to my earlier posts. Conversion from a narrow gamut to a wider one is easy - the other way round is a problem. Gamma is easy...

If you don't convert the material the failure of the results are of your own doing...

Got to smile :)
 
You must log in or register to reply here.
Back
Top