Be aware that renderings in the Studio Environment or via the Raytracing option in the View tab are both CPU operations and not related to the graphics card. Raytracing operations, in general, are multi-threaded operations, so the more cores you can throw at them, the better.
Inventor, as a design package, requires a balance of clock speed vs cores for optimum performance. Most modeling tasks are coded for single core computing, though certain operations (FEA, drawing views, renderings, etc.) can capitalize on many cores.
Your system topology really depends on what you're looking to achieve. For my day-to-day modeling, I use an I7 8700k (6c/12t) but for my FEA and rendering, I have a dual Xeon 2696v4 (44c/88t). The Xeon workstation looks like a beast from the outside, but when I'm doing single-threaded operations (things like Dynamic Simulation especially), it is crazy slow. The flip side is that with the Xeon machine I can render your same scene to the same 270,000 iterations in under two hours.
K. Cornett
Generative Design Consultant / Trainer