報告 #1211 跑分
已認證，裝置已鎖定 有效
- GPU
- Adreno (TM) 725
- 驅動程式
- turnip_faux123_v1.2.0 Mesa 26.3.0-devel (git-dd45fd0baa)
- 裝置
- Samsung Galaxy Z Fold4
- 晶片
- SM8475
- CPU
- 4 x 2.02 GHz + 3 x 2.75 GHz + 1 x 3.19 GHz, arm64-v8a
- 記憶體
- 11.0 GB
- 回報為
- samsung / SM-F936B
- 收到時間
- Android
- 16
- 函式庫檔案 SHA-256
- 47761a30ffdb063938c354d83246534c7f905807e0116a0402b5d88e73d5ad89
測試條件
- seconds_per_scene
- 6
- warmup_frames
- 30
- warmup_max_ms
- 2000
- warmup_min_frames
- 5
- render_width
- 1920
- render_height
- 1080
- heavy_width
- 3840
- heavy_height
- 2160
- heavy_below_ms
- 8
- heavy_particles
- 2
- cool_rule
- status_below_2_and_full_gpu_limit
- cooling_pauses
- 0
- cooling_s
- 0
- cooling_capped
- 1
- test_budget_s
- 180
- test_s
- 355
- system_log_unreadable
- False
- display_width
- 1920
- display_height
- 1080
- display_refresh_hz
- 60
- rounds
- 3
- scene_set
- 6
- hdr_format
- R16G16B16A16_SFLOAT
- 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. hdr runs at the larger picture only. Compare drivers on one device; across devices, compare only rows that kept the GPU busy.
測試期間繪圖晶片頻率低於峰值的 90%（GPU 頻率）：compute_nbody, pipeline_barriers。可能是裝置壓住了繪圖晶片（發熱或省電模式），也可能是場景太輕，用不著更高頻率。這些資料只適合和頻率相近的測試比較。
場景
| 場景 | 中位數 ms | 平均 ms | p95 ms | 輪次 | 開測時已發熱 | GPU 頻率 |
|---|---|---|---|---|---|---|
| oit_linked_lists | 21.88 | 22.05 | 23.42 | 3 | 0 | 900 / 900 MHz |
| subpasses | 4.16 | 4.18 | 4.67 | 3 | 0 | 900 / 900 MHz |
| compute_nbody | 9.92 | 10.38 | 16.47 | 3 | 0 | 591 / 900 MHz |
| pipeline_barriers | 2.95 | 3.76 | 8.15 | 3 | 0 | 716 / 900 MHz |
| subpasses_2160p | 15.96 | 15.97 | 16.78 | 3 | 0 | 900 / 900 MHz |
| pipeline_barriers_2160p | 8.75 | 8.84 | 9.52 | 3 | 0 | 900 / 900 MHz |
| hdr_2160p | 28.79 | 28.87 | 29.85 | 3 | 0 | 900 / 900 MHz |
單次執行（21）
| 場景 | 輪 | 狀態 | 中位數 ms | 幀數 |
|---|---|---|---|---|
| oit_linked_lists | 0 | ok | 21.94 | 236 |
| subpasses | 0 | ok | 4.16 | 1381 |
| compute_nbody | 0 | ok | 9.92 | 553 |
| pipeline_barriers | 0 | ok | 4.26 | 1030 |
| oit_linked_lists | 1 | ok | 21.88 | 237 |
| subpasses | 1 | ok | 8.23 | 684 |
| compute_nbody | 1 | ok | 10.67 | 529 |
| pipeline_barriers | 1 | ok | 2.95 | 1524 |
| oit_linked_lists | 2 | ok | 21.95 | 232 |
| subpasses | 2 | ok | 5.23 | 927 |
| compute_nbody | 2 | ok | 12.16 | 415 |
| pipeline_barriers | 2 | ok | 3.83 | 1291 |
| subpasses_2160p | 0 | ok | 15.96 | 341 |
| pipeline_barriers_2160p | 0 | ok | 9.16 | 598 |
| hdr_2160p | 0 | ok | 28.81 | 160 |
| subpasses_2160p | 1 | ok | 16.23 | 334 |
| pipeline_barriers_2160p | 1 | ok | 9.19 | 591 |
| hdr_2160p | 1 | ok | 42.15 | 116 |
| subpasses_2160p | 2 | ok | 23.23 | 247 |
| pipeline_barriers_2160p | 2 | ok | 8.75 | 630 |
| hdr_2160p | 2 | ok | 28.79 | 173 |