In some ways, there is a sense in which each of those three, as well as the more modern Marteau, Camoin and Hadar renditions, are 'true' to the Marseille deck.
In some way, it is a little like taking a number of roses from various rose-bushes and asking which is the 'truest' rose. They are all roses - yet the question also makes sense, for we can then really look carefully at each and see how particular growths may be... how shall I put it, deformations from an ideal which cannot manifest. Each will exhibit such.
Similarly, in my view, with Tarot - hence some of my earlier comments on the Ür-Tarot, the arche-typos which finds its ways in various manifestations or 'incarnations' of the Tarot.
The Noblet seems to take the opposite spectrum of the Vieville: the former naming every card, the latter none. The Noblet also has certain characteristics which tends to show the difficulty in some of the art of early wood-carving, with some of the images having rough-cuts or ambiguities which may not have been intended - as an obvious example, it could easily be that the Bateleur's raised hand first had its whole hand and wand, but that the upper portion broke off. The orientation of, for example, XIII (LAMORT - which sounds so much like 'L'Amour' - and thus a good reason, if nothing else, for it to remain un-named) is not consistent with Marseille patterning. Again, it could be that the carver began with an error, but, having worked on it, would either have to start again (on another piece of pear wood, perhaps), or determine that it shouldn't matter.
To my eyes, the Noblet has all the elements of a very fine deck, with, apart from some of these minute details, very careful rendering of what is 'important' - and J-C Flornoy has provided us with a means to hold at least the Major Arcana in our hands (as a limited edition, I suggest to all those interested to obtain a copy!).
The Conver has, for most of us, been the only full replica of an early and full Tarot deck obtainable. In that sense alone, it has attained a reputation and importance. For myself, it remains a deck from which I will continue to use as a means of comparison. For example, the triple nipple on one of the figures tied to the Devil's envil seems not only prominant, but of importance. This is to be balanced with details which seem to also be missing from the deck: does, staying for the same card as an example, the Devil properly need a facial depiction upon its belly as is the case in some other Marseille decks?
Nonetheless, the clarity is superb.
As for the Dodal, I have increasingly also come to value it for its seeming closeness to having incorporated nearly all the elements of central value in the Marseille - including numbering the Hanged Man as IIX; having a clear anvil as the Devil's pedestal, having one of the two depicted on the Sun with eyes closed; leaving XIII untitled; & numerous other details of note.
Which is 'truer'?
Each, as each rose, is a 'true' Marseille (despite not being from that maritime city), and each contains certain elements seemingly 'truer' than others.
One element I personally prefer from the Noblet is the ambiguity of the item on the Bateleur's table being an open book, not a bag; from the Dodal, the Devil card seems so potently clear; and from each of the others - modern or older - so many details!
If I had to design my own deck to reflect what I consider closest to the ideal, I would probably, at this stage, replicate more the Dodal than any other Marseille - yet also incorporate elements not in that deck.