报告 #44 跑分
参考设备 有效 

至尊支持者
- GPU
- Mali-G57 MC4
- 驱动
- panvk_faux123_rg556_v0.12.0 Mesa 26.3.0-devel (git-7886eada08)
- 设备
- Anbernic RG556
- 芯片
- T820
- CPU
- 4 x 2.18 GHz + 3 x 2.30 GHz + 1 x 2.70 GHz, arm64-v8a
- 内存
- 7.7 GB
- 上报为
- Anbernic / RG556
- 收到时间
- Android
- 13
- 库文件 SHA-256
- cba2560ea6e4eb2e0e160f675eadce523a2e02c9d33f82de5f70d7416de98ef3
测试条件
- seconds_per_scene
- 6
- warmup_frames
- 30
- render_width
- 1920
- render_height
- 1080
- heavy_width
- 3840
- heavy_height
- 2160
- heavy_below_ms
- 8
- heavy_particles
- 2
- cool_rule
- status_below_2
- cool_start_ref_c
- None
- cooling_pauses
- 0
- cooling_s
- 0
- cooling_capped
- 0
- part_redone
- False
- cooling_gave_up
- False
- display_width
- 1080
- display_height
- 1920
- display_refresh_hz
- 60
- rounds
- 3
- scene_set
- 3
- vsync
- OFF
Every scene draws at 1920 by 1080, whatever the screen. A scene under 8.0 ms there is too light for the chip and runs again at a heavier load: at 3840 by 2160, as the row named with _2160p, or, for compute_nbody, with 2 times the particles, as the row named with _2x. Compare drivers on one device; across devices, compare only rows that kept the GPU busy.
场景
| 场景 | 中位数 ms | 平均 ms | p95 ms | 轮次 | 开测时已发热 | GPU 频率 |
|---|---|---|---|---|---|---|
| oit_linked_lists | 143.41 | 145.46 | 152.10 | 3 | 0 | 未记录 |
| subpasses | 51.45 | 54.27 | 60.51 | 3 | 0 | 未记录 |
| compute_nbody | 30.15 | 30.31 | 31.50 | 3 | 0 | 未记录 |
| pipeline_barriers | 34.75 | 33.70 | 40.52 | 3 | 0 | 未记录 |
单次运行(12)
| 场景 | 轮 | 状态 | 中位数 ms | 帧数 |
|---|---|---|---|---|
| oit_linked_lists | 0 | ok | 153.45 | 8 |
| subpasses | 0 | ok | 51.45 | 80 |
| compute_nbody | 0 | ok | 30.20 | 163 |
| pipeline_barriers | 0 | ok | 35.10 | 139 |
| oit_linked_lists | 1 | ok | 145.14 | 9 |
| subpasses | 1 | ok | 52.44 | 80 |
| compute_nbody | 1 | ok | 30.15 | 160 |
| pipeline_barriers | 1 | ok | 35.48 | 141 |
| oit_linked_lists | 2 | ok | 143.41 | 10 |
| subpasses | 2 | ok | 53.35 | 80 |
| compute_nbody | 2 | ok | 30.40 | 162 |
| pipeline_barriers | 2 | ok | 34.75 | 141 |