Those RT options will be scalable though i.e. low to medium to high to ultra so if RT effects do get dialled up even further where only 40xx/rdna 3 is capable of high/ultra, ampere will be capable of medium where as RDNA 2 won't even be capable of low, we have already seen RDNA 2 having to sacrifice the RT settings even when upscaling is used i.e. same way any potential problematic vram intensive games can be solved by knocking down the texture setting by one notch.Im not as convinced of this. Ray tracing in its current state is being run and in a lot of cases only feasible with upscaling. No problem with that as far as I'm concerned. But ampere at high resolutions with full Ray tracing is not foolproof, and with the rate of expected improvements in Ray tracing coming up, I don't think ampere will age that well.
Point taken with RDNA 2, but if both are struggling to run RT in a few years, the vram may be of more benefit for longer. It'll depend how the long promised direct storage and upscaling tech comes through.
What DF etc. have found is when RT is more complex or/and more effects are used, RDNA 2 completely buckles where as ampere performs even better:
![www.pcgameshardware.de](https://www.pcgameshardware.de/screenshots/original/2021/04/ART-Mark_Raytracing-Reflections-pcgh1.jpg)
Raytracing-Showcase getestet: Ampere deklassiert Turing und RDNA 2 - ein Blick in die Zukunft?
Der Raytracing-Benchmark ART-Mark streckt selbst eine Geforce RTX 3090 und Radeon RX 6900 XT nieder - Zeit für einen genauen Blick und Benchmarks.
The more complex the ray tracing in space, the better Ampere performs.
Alex from DF did a very good video showing how ampere and rnda 2 performs when you add more RT effects into the mix too.