Tagging is not converting.

The numbers in a video file mean nothing on their own. 0.5 is a code value, and what colour it is depends entirely on which space and which curve you say the file is in. Change the tag and every pixel changes meaning while not one byte of the file moves. A conversion moves the pixels so the appearance survives; a tag moves the meaning and leaves the pixels. Both are legitimate and they are opposites — and doing the wrong one produces a file that is byte-identical to the correct one, which is why every diagnosis that starts by comparing files finds nothing.

The engine is free and needs no account. Write the stages your material passes through and every join between them is checked here. A relabel between two pure power curves is exactly v^(encode/display), so the shift is a number rather than an argument — and there are two of them: the absolute error peaks in the midtones, which is why a chart with a black patch and a white patch passes the test a face fails, while the relative error grows toward black, which is why the colourist quotes a quarter and the client quotes six percent and both are right. Gamuts are triangles, so what survives a delivery is an area: P3 into Rec.709 survives 73.7%, and Rec.709 into Rec.2020 survives 100% while filling 52.9% — two different questions, and quoting one for the other is how a bad delivery gets signed off. Every lane ships a worked example that costs nothing. Writing the report needs credits.