My only beef with dE is that it only tells you the magnitude of the error and not the direction of the error(s).

Say you have two points (to keep it simple: 10% and 20% on your greyscale.)

Say both have a dE of 2.5, but the 10% point is +Blue while the 20% point is +Red ... now you have an effective relative dE of 5.0 ...

Now imagine the chaos that could happen when you rely on multiple dE calculations around the gamut to figure out if you have your CMS setup "right."

Say you have two points (to keep it simple: 10% and 20% on your greyscale.)

Say both have a dE of 2.5, but the 10% point is +Blue while the 20% point is +Red ... now you have an effective relative dE of 5.0 ...

Now imagine the chaos that could happen when you rely on multiple dE calculations around the gamut to figure out if you have your CMS setup "right."