Is it a G a better lens? or simply it's optimized for long distance work?
It's designed for a specific purpose outwith the optimal range of regular enlarging lenses. That's all there is to it. There are some who claim that the performance spec of the 105 G is (in its range) at least as good as the 105 Apo EL Nikkor in its own optimal performance range - but I'd suggest that's an apples/ oranges comparison as they're designed to do quite different things.
We have several tricks to make a regular lens behave like a G for murals:
> using a longer focal and increasing projection distance, this is named poorman's G,
> hacking the optimal magnification by shimming the front or rear cell.
You'd be better off getting an optical engineer to see what tweaking an (Apo) Sironar N for flatness of field (at expense of coverage) and optimal performance at a wide aperture (rather than f22) would get relative to a Rodagon-G. The G series has more in common with taking lenses than enlarging lenses in terms of optimisation, design, glass etc. These alterations are not a million miles off how certain classic LF lens designs were tweaked for use with MF digital sensors - coverage traded for wide open performance and flatness of field. The costs of going so are however going to be vastly greater than a WA-Rodagon 120mm, especially if you aren't going above 15x - and you'd better have some exceptional work that needs to go to 1.2x1.5m...
> buying-selling several samples of the No-letter versuion until you have the contrary of a dog.
> overpowering enlarger with LEDs to be able to expose at optimal aperture.
The reason the Rodagon-G is optimised for a wider aperture is pretty simple: if you want to maintain a theoretical minimum of 5 cyc/mm at 40x, you need to use f8. At 20x, you'll potentially be able to hold 10 cyc/mm (theoretically) whereas with a lens outwith its correction range and stopped to f11, you probably won't even be getting close to the theoretical 7 cyc/mm because various aberrations will be getting in the way destroying the ability to hold good contrast in fine detail areas. And yes, paper can hold some resolution out to 10 cyc/mm or beyond, but the necessary resolution for a truly sharp image has been generally regarded as 7 or 8 cyc/mm for a very long time, but it is a good idea to be able to outresolve your material in order to be able to absorb the inevitable losses to various factors. For that matter, most regular f5.6 enlarging lenses from the 70's onwards peak barely a stop down (within optimisation range of course). So as we can see, it's not just pure resolution, it's about the contrast transfer, distortion etc - they all matter and interconnect. Reductive reliance on high contrast resolution alone will get nowhere fast - all that gives is a threshold marker and no indications of overall performance en route to that point.
I have a 1.12µm DIY optical bench, around 400lp/mm: This is an old Sony Z2 smartphone with the lens removed (6.17mm/5248pix), so projecting directly on sensor. I see that camera in a big monitor for total convenience.
Not up to the task I don't think.
Unless you have the wherewithal and time to do it properly, don't bother trying to do it half-assed - real world imaging will tell a more realistic story much faster about actual optical performance.