samsung SM-S938N vs samsung SM-S938N
| samsung SM-S938N | samsung SM-S938N | Difference | |
|---|---|---|---|
| Single Precision | 557 | 557 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Half Precision | 27064 | 27064 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Quantized | 60522 | 60522 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Geekbench AI 1.2.0 | Geekbench AI 1.2.0 | ||
System Information
| samsung SM-S938N | samsung SM-S938N | |
|---|---|---|
| AI Framework | TensorFlow Lite | TensorFlow Lite |
| AI Backend | QNN | QNN |
| AI Device | Qualcomm ARMv8 | Qualcomm ARMv8 |
| Operating System | Android 15 | Android 15 |
| Model | samsung SM-S938N | samsung SM-S938N |
| Processor |
Qualcomm ARMv8 @ 3.53 GHz
1 Processor, 8 Cores |
Qualcomm ARMv8 @ 3.53 GHz
1 Processor, 8 Cores |
| Processor ID | ARM implementer 81 architecture 8 variant 3 part 1 revision 4 | ARM implementer 81 architecture 8 variant 3 part 1 revision 4 |
| Motherboard | sun | sun |
| Memory | 10.86 GB | 10.86 GB |
AI Performance
| samsung SM-S938N | samsung SM-S938N | Difference | |
|---|---|---|---|
| Image Classification (SP) | 450 | 450 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Image Classification (HP) | 21470 | 21470 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Image Classification (Q) | 49077 | 49077 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Image Segmentation (SP) | 563 | 563 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Image Segmentation (HP) | 13519 | 13519 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Image Segmentation (Q) | 42154 | 42154 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Pose Estimation (SP) | 637 | 637 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Pose Estimation (HP) | 123426 | 123426 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Pose Estimation (Q) | 477345 | 477345 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Object Detection (SP) | 312 | 312 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Object Detection (HP) | 18912 | 18912 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Object Detection (Q) | 21864 | 21864 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Face Detection (SP) | 761 | 761 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Face Detection (HP) | 41461 | 41461 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Face Detection (Q) | 140803 | 140803 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Depth Estimation (SP) | 668 | 668 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Depth Estimation (HP) | 83771 | 83771 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Depth Estimation (Q) | 144072 | 144072 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Style Transfer (SP) | 1235 | 1235 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Style Transfer (HP) | 98760 | 98760 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Style Transfer (Q) | 490533 | 490533 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Image Super-Resolution (SP) | 340 | 340 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Image Super-Resolution (HP) | 52851 | 52851 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Image Super-Resolution (Q) | 132083 | 132083 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Text Classification (SP) | 395 | 395 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Text Classification (HP) | 3872 | 3872 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Text Classification (Q) | 7850 | 7850 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Machine Translation (SP) | 683 | 683 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Machine Translation (HP) | 4433 | 4433 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
| Machine Translation (Q) | 2960 | 2960 | 100.0% |
| samsung SM-S938N | |||
| samsung SM-S938N | |||
