We can't have the cake and eat it too, even though we've made the F-14 more performant in many cases (e.g. put 20 F-18s on the Ramp and compare the FPS versus if they were F-14s).
Rendering triangles is extremely cheap. Nothing about bolts on the front canopy frame is relevant.
Performance is mostly impacted by drawcalls and lighting passes in DCS. That is why (looking out over terrain) your FPS drops dramatically.
Comparing the F-14 to the F-18 in a 1:1 comparison is rather unrealistic unfortunately, especially in VR. Even if we created it to the same standard as the F-18 - you'd still take a performance hit with the F-14.
It has animated crew, a 2-person cockpit (that in general is far more geometrically complex in reality as well, just compare the back walls in each cockpit!), is much bigger, etc.
The changing settings not impacting your frames is not that odd- but it does mean you're bottlenecked somewhere.
I understand that there are people with issues. Our tester team included everything from a GTX760 (!) to a 2080 Ti, and I think we've been very prudent about measuring performance.
The F-14 performs worse than the F-18 1:1- yes- but unplayable stutters have not been encountered on any widespread basis yet. I know that it sounds dismissive,
but when we have thousands of customers on similar hardware reporting no issues; we have to evaluate accordingly.
Also, I'd reiterate deleting the fxo and metashaders folders on every graphics change. We're not sure why the precompiled shaders are behaving so erratically in DCS.
As dismissive as this post sounds, I emphasize with your issues. We're simply in the position of trying to raise fidelity - which is (as always) impossible without reducing performance.