On the photo above, using the Seafire's Dark Slate Grey as the control colour which is about 113% of the colour saturation of the real paint's published colourspace values - i.e. it is too strong/intense an olive, it should be a little greyer than that photograph portrays it. If the Dark Slate Grey is oversaturated, then it's reasonable to assume the MS3 or G20 (given the dates of these photos are shortly after the new paints were promulgated and may or may not have needed repainting by this time).
There is also this photo which is slightly different colour balance than the above one.
Attachment:
Indomitable3.jpg
FWIW I think we're looking at B6 there on the island. However, others disagree and believe it's B5. The real issue with B6 is that we have, to my knowledge, 4 surviving references to tell us what B6 looked like. All more or less agree on the chroma or light reflectance value or the order of 25-30%, so lighter than MS3 / G20 but darker than 507C / G45. The trouble is that a surviving paper camouflage design sheet portraying B6 (which may or may not *actually* be B6) agrees rather well with file AD.29 - i.e. PRS' x,y,z colour values for what they received from the existing paints to study for incorporation into the B & G series. They both show a light blue with moderate colour saturation resulting in a colour any sane person would casually describe as "light blue". The weakness of those are that we only assume they are B6 - but the design sheet might not actually be B6 and the colour we think is B6 in AD.29 is referred to as B.30 in that work which predates the official promulgation of the B&G series by half a year. So, both are good clues, but neither are proof in themselves.
Then we have the other two references - a pair of surviving real paint samples. One is held in file ADM 212/124 at Kew, and the other is held by the Admiralty Library at Portsmouth. The providence of those chips is certain - the way they are stamped or glued onto thin leaf paper in a 1942 vintage document pretty much assures that they are what they say they are. The trouble is that whilst they closely resemble each other, they look nothing like any of these photos above nor the first pair of references I described. Both Kew and Portsmouth's samples are so unsaturated that any casual observer would describe them as "grey".
Does that mean I believe the samples more than the first pair? No, it doesn't, as it happens. Whilst the formulations of B5, B6 etc are unknown, the favoured pigments used by the RN are known, and from the mid 1930s onwards we only see Ultramarine Blue used for blue colouring. Ultramarine is either natural lapis lazuli mineral ground down, or it's made synthetically. Acid in particular is well known to desaturate ultramarine leaving it a grey. Some manufacturing processes for paper leave them acidic, and both of those samples described above are painted onto card, not metal.
B6 has caused me more hand wringing than the rest combined, frankly, but I personally believe AD.29's "B30" recorded by PRS colourometer work was indeed a sample of B6. The belief that B6 was a light blue reconciles more conflicts with rational argument than does the belief that the 76 year old samples in Kew and Portsmouth are representative does. Degredation of ultramarine dominated paint on card stock can be explained with understood observational science. The existence of another enigmatic light blue of the same tone appearing in multiple forms in multiple places yet not being mentioned in writing in any of the surviving documents requires a lot more assumptions to be made than the facts support, in my opinion.