Benchmark #1827

Download
Nine Sols: BPFLAND-PREV-NEXT vs BPFLAND-NEXT vs BPFLAND-NEXT-NO-STICKY

GPU Limited static scene + No Background Load | 30 second frame recording with 0ms interval | Performance Governor/EPP | Proton GE 10.17 Wayland Native | SCX Prev Next: 7dbacfe Next: b368aa1 (390228e) | kernel-cachyos-lto-6.17.1-cachyos1.lto.fc44.x86_64 with Staging SCX Patches.

Submitted 2 weeks ago by summedkibbles36

Specifications
Label OS GPU CPU RAM OS specific
Next 1 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
Next 2 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
Next 3 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
Next --no-sticky-tasks 1 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
Next --no-sticky-tasks 2 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
Next --no-sticky-tasks 3 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
Prev Next 1 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
Prev Next 2 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
Prev Next 3 Steam Runtime 3 (sniper) AMD Radeon RX 7900 GRE (RADV NAVI31) AMD Ryzen 7 7800X3D 8-Core Processor 33 GB 6.17.1-cachyos1.lto.fc44.x86_64 performance
# Top runs: * **Highest FPS**: `Next --no-sticky-tasks 3` and this run achieved the highest average FPS, with over a 0.1% increase compared to the majority of other runs. * **Smoothest FPS**: `Next 2` and this run demonstrated the lowest frame time variance and standard deviation, making it the smoothest run. * **Best overall**: `Next 2` with a balance of high average FPS and smooth frame time, it provides optimal consistency and performance. # Issues: * The run labeled as `Prev Next 1` showed a performance decrease in terms of FPS, with about 0.1% lower than most of the higher-performing runs and higher FPS variability compared to others. * Compared to the top-performing runs, the run labeled as `Next --no-sticky-tasks 2` displayed a slight reduction in frame rate stability along with a marginal drop in average FPS. # Summary This benchmark evaluates the performance differences in a GPU limited static scene using variations of the `BPFLAND` scheduler configurations. The configurations included both transitioning and non-sticky task handling. The run using `BPFLAND-NEXT` with no sticky tasks showcased the highest FPS performance. Meanwhile, the same `BPFLAND-NEXT` configuration demonstrated optimal frame time smoothness, highlighting its advantageous stability. Overall, while all configurations run with identical hardware and minimal environmental changes besides scheduler configurations, marginal variations in FPS and consistency were observed favoring the `BPFLAND-NEXT` with default and no-sticky-task setups.