Matching cameras with a color checker

Hi all,
Just checking this is working the way it's supposed to... I'm using a colour checker (Spydercheckr) to match some cameras. I can manually correct to the vectorscope targets using Hue-hue / Hue-sat curves and get a pretty spot-on result, or use the automatic color matching tool in Resolve and get a result that looks ok (perhaps a little more vivid), but is not hitting the vectorscope targets.
Do I assume Resolve is calibrated to the color swatches? Or that the color swatches should be calibrated to be correct in the vectorscope? (vectorscope is at 2x zoom)
Thanks

auto matched color

auto color match.png (145.31 KiB) Viewed 4155 times

manual matched color

manual color match.png (134.61 KiB) Viewed 4155 times

My opinion: the Colorchecker chart is not very accurate, particularly in terms of saturation levels. I think the Colorchecker Video is closer, and the Colorchecker Digital is also good. None are 100% accurate.

I think you can get generally close in terms of hue vectors, but not saturation. The DSC Chroma DuMonde is a lot closer, and they provide you with a user guide telling you exactly where the color patches should be in terms of the vectorscope, and where the grayscale should be in terms of video.

I think if you have 2 (or more) cameras and they shoot the same chart under the same lighting at the same time, and then all the cameras match that chart, they should "in general" match each other. The important question is whether the pictures will look good, and if the lighting is even and there's no issues with color temperature. What specific cameras are you using?

BTW, if you haven't read it already, I highly recommend Tektronix' free guide on using scopes:

"Using Waveform Monitors as Artistic Tools for Color Grading"
https://www.tek.com/document/primer/gui ... resolution

Do I assume Resolve is calibrated to the color swatches? Or that the color swatches should be calibrated to be correct in the vectorscope?

The Spyderchecker doesn't reference 709 primaries. Alongside the CDM it's pretty obvious how much more saturated the red and blue primaries are on the Spyderchecker. Keep in mind the colors are at printed at 50% saturation (when using 2x gain) so they should appear muted. The auto match function presents a more accurate representation, red and blue are more saturated and green is less so. One can bend the Spyderchecker colors to hit 709 primaries but all accuracy is gone at that point.

My opinion: the Colorchecker chart is not very accurate

If the goal is referencing 709 colorimetry then it's not even a little bit accurate.

Good Luck

spydervcdm.jpg

The ColorChecker 24 and SpyderChecker targets will never really hit the vectorscope targets because their swatches weren't meant to. Their primary targets are for Photography. You will get close on the vectorscope with the RGB CMY swatches, but it will never be an exact match, nor should you attempt to manually match them.

The ColorChecker Passport Video and ColorChecker Video on the other hand DO hit the targets because they were meant to.

The Photography oriented charts can be used to for Resolve's Auto Color Match function though, which you will likely get better results with compared to manual methods.

All the charts have defined color swatches (with some margin of error) that Resolve will attempt the color match on.

Thanks all - really helpful,
I'm going to use the Resolve auto-match because it's faster and I now assume it knows better where the colours should end up, given they clearly don't line up with a video vectorscope.
As a little extended test, I did pull a still image from a comparison video (

) and ran the colour match on both in the same frame. So, for anyone reading this for future reference, the way this guy approaches it (manually lining both up with the vector targets) is WRONG - just use the Resolve match function with both. Manually lining up will give you 'better' results with the x-rite video passport, but your saturation still won't be right, so just don't overthink it and use the automatic tools with these prosumer-level cards.
What my test did confirm for me is the Spydercheckr delivers essentially the same results as the video passport when both are automated (the x-rite swatch hues, post-matching, are just better aligned for video scopes).
For my purposes I'm happy with the Datacolor SpyderCheckr I have (based on the bigger, better case and tripod mount on it, but not the dumb product name) - just be aware that you should definately be using the automatic matching for accurate results.

For my purposes I'm happy with the Datacolor SpyderCheckr I have (based on the bigger, better case and tripod mount on it, but not the dumb product name) - just be aware that you should definately be using the automatic matching for accurate results.

Anything Camera Match can do, you can also do yourself. It just takes time and effort. We've done it that way for more than 40 years. There will always be cases where you get material with no charts and still have to make it match. The danger of bad charts is they can be worse than no charts.

My conclusion was the automatic tool doesn't do what it claims on the tin.

viewtopic.php?f=21&t=148895&hilit=+checker

Only option is manual hue-hue, Hue-Sat curves.