Resizable BAR on AMD and Nvidia = smoke and mirrors?
the Resizable BAR makes the news by giving the impression that a miracle update would deliver “free” performance to almost all users.
What could be better than a marathon test to find out. Our colleagues from Guru3D used an AMD X570 + Ryzen 9 5950X platform and all the graphics cards that the market could offer them:
- GeForce RTX 3060
- GeForce RTX 3060 Ti
- GeForce RTX 3070
- GeForce RTX 3080
- GeForce RTX 3090
- Radeon RX 6700 XT
- Radeon RX 6800
- Radeon RX 6800 XT
- Radeon RX 6900 XT
Does the resizable BAR really improve performance?
With almost 500 individual tests, the results on the graphs are indisputable: yes, enabling this technology will allow you to improve performance a bit. But in the real world, only the test graphics will show you this gain because visually you will never notice it in your games.
Assassins Creed: Valhalla the only game where the payoff is significant
So the conclusion is a bit of a paradox: rBAR is a feature that has been overlooked for years and is suddenly back in fashion again, probably thanks to the magic wand of marketing.
If your setup is fresh, with the latest technology, there is no reason not to enable rBAR, which will get better over time.
The facts are boring but the numbers can hardly be criticized. The vast majority of games will show virtually no difference with exceptions here and there. Obviously, even though the small improvement is only noticeable in benchmarks, we can’t spit out extra free performance here and there. But the magic potion is unfortunately a myth …