ReLive Performance

Next, we ran some tests to check on its performance impact.


First, we tested what effect different recording and encoding settings in the game panel have. As you can see, the performance hit is somewhere in the 3.5-4 % region, which is quite good. Also, it's surprising that the different options have so little effect on gaming performance.


Next, we tested what changes when different output resolutions for the recorded video are picked. The game was always running at 1920x1080 for these tests. With Radeon ReLive, 4K recording is only supported at up to a recording rate of 30 FPS. If you pick 60 FPS, the game will switch to 1920x1080 60 FPS, which might not always be what you want. Also, when you select "In-Game" for Recording Resolution and play a 4K game, it will switch to 1080p recording. The fix is to select "2160" in that dropdown menu, and a 30 FPS rate, which will net you a proper 4K 30 FPS video.


We then moved on to see what performance impact recording has when the game is running at different resolutions. Using the "High" encoding settings every time, ReLive was set to always capture the native game resolution.


We also ran through a bunch of games to see the performance impact. All testing was performed at 1080p and with maximum game settings, like in our regular reviews.

I find by how much the performance hit varies interesting. With the exception of Arkham Knight, the performance penalty for recording is negligible and pretty much made up for by the driver performance improvements we'll cover on a later page.



Here's a recording we made using a single Radeon R9 290, rendering "Just Cause 3" at 1440p (while recording at 1080p@60Hz).