Benchmarks found: 101
Varying scene from a clip, with background load composed of Discord Call + 1080p60fps CPU encoded streaming. 40 second frame recording with 0ms interval. Performance Governor/EPP | Performance mode used in all tests besides LAVD --autopilot and P2DQ (not provided).
By summedkibbles36CPU Limited scene from the first stage with a stable number of enemies, no background load. 30 second frame recording with 0ms interval. All schedulers were running in performance mode except LAVD --autopilot. Performance Governor/EPP
By summedkibbles36CPU Limited scene from the first stage with a stable number of enemies and the boss, no background load. 30 second frame recording with 0ms interval. Performance Governor/EPP | BORE default settings | LAVD --performance | BPFLAND -m performance | FLASH | P2DQ | RUSTY
By summedkibbles36CPU Limited static scene from the second mission, with background load composed of Discord Call + 1080p60fps CPU encoded streaming. Remastered Graphics are used to not hit the engine frame cap of ~1000fps. 30 second frame recording with 0ms interval. Performance Governor/EPP | BORE default settings | BPFLAND -m performance | LAVD --autopilot | LAVD --performance
By summedkibbles36CPU Limited static scene from the first mission, no significant background load (Discord). 30 second frame recording with 0ms interval. Performance Governor/EPP | BORE default settings | BPFLAND -m performance | LAVD --autopilot | LAVD --performance
By summedkibbles36Close to GPU limited scene (not fully saturated) in Gatefront Ruins (AI heavy) with background load composed of Discord Call + 1080p60fps CPU encoded streaming and 720p30fps CPU decoded streaming. 30 second frame recording with 0ms interval. Performance Governor/EPP | BORE default settings | scx_bpfland --performance --slice-us 3000 | sudo scx_lavd --slice-max-us 1000
By summedkibbles36Synchronization/CPU limited scene from a somewhat realistic static scene; There were no active AI interactions since entities can kill eachother, causing significant increases in performance; Ideally to test we'd create an artificial scene where entities can interact but not kill eachoth. Tested with background load composed of a Discord Call + 1080p60fps CPU encoded streaming. 30s at 0ms. BORE | BPFLAND -m performance | BPFLAND -m performance --local-pcpu | LAVD --autopilot | LAVD --performance
By summedkibbles36Synchronization limited scene from one of the heaviest areas of the game. Tested with background load composed of a Discord Call + 1080p60fps CPU encoded streaming. 30s at 0ms. Data from this shouldn't be particular representative of regular gameplay since there are a lot more entities than the ones the engine can comfortably handle, still it's a worst case scenario. BORE | BPFLAND -m performance | BPFLAND -m performance --local-pcpu | LAVD --autopilot | LAVD --performance
By summedkibbles36Synchronization limited scene from one of the heaviest areas of the game. Tested with background load composed of a Discord Call + 1080p60fps CPU encoded streaming. Data from this shouldn't be particular representative of regular gameplay since there are a lot more entities than the ones the engine can comfortably handle, still it's a worst case scenario. BORE | BPFLAND -m performance | BPFLAND -m performance --local-pcpu | LAVD --autopilot | LAVD --performance
By summedkibbles36Static mostly GPU limited scene (not fully saturated) with no background load. 30 second frame recording with 0ms interval to get the most accurate results; single run only. Performance Governor/EPP | BORE default settings | BPFLAND --performance | FLASH | LAVD --performance | P2DQ | P2DQ --autoslice | various max timeslices for BPFLAND and LAVD. BPFLAND 3000 v2 is a re-run at the end to account for variations in the environment since it's impossible to control them.
By summedkibbles36