As others have said, it is fairly well known in the Vue community that their render quality presets are pretty far off the mark of what their names imply. So you'll probably have to do some custom settings to get comparable (and fair, to both apps) results.
What I think makes most sense, at least for a start, is to compare based entirely on external assets, so that there is no element of procedural differentiation, etc. So this means using a heightfield that you can load in both apps, and at least one object available in OBJ format (e.g. Xfrog tree). This gives you a better opportunity for direct comparison. Then setup the sun angle the same, and adjust colors and strength and/or exposure to match lighting. Then you'll have a decent starting point for comparison. Now I will say this type of render does not highlight TG2's strengths with procedural displacement, but it would still be an interesting starting comparison.
After that it would be interesting indeed to test procedurals in a more relative way (since absolute direct comparison will not be possible). Create "infinite" procedural terrains in both, with similar levels of detail octaves, and standard (e.g. Perlin) noise functions, and see what results.
Clouds would be another interesting one, but again only comparable in a relative way, and even more variably than with terrain due to volumetric shading differences and different lighting models. This area would probably be most challenging to compare, but I'm quite interested in the results.
Also, I think it's fair to publish your level of experience in each app. It seems relevant to compare the learning curve of each as you go, too.
I'm glad to see someone has undertaken this project and I'm quite curious to see the results. I hope you stick with it long enough to get a couple of different comparable scenes rendered, and to create a good write-up on your experiences.
- Oshyan