3090 Reg x8 - AfterB +200MHz +750MHz mem 50%v+ CRASH Stress Generic
Geekbench AI Score
17628
Single Precision Score
22950
Half Precision Score
12074
Quantized Score
Geekbench AI 1.2.0 for Windows x86 (64-bit)
Result Information
| User | krofa |
| Upload Date | January 28 2025 05:25 PM |
| Views | 13 |
| Notes | 250+ MHz for g-clocks didn't work for even 100% v+. 10500 is the mem clock of Ti |
System Information
| AI Information | |
|---|---|
| Framework | ONNX |
| Backend | DirectML |
| Device | NVIDIA GeForce RTX 3090 |
| System Information | |
|---|---|
| Operating System | Microsoft Windows 10 Pro (64-bit) |
| Model | Generic |
| Motherboard | ASRock X79 Professional |
| Power Plan | Balanced |
| CPU Information | |
|---|---|
| Name | Intel Xeon E5-1680 v2 |
| Topology | 1 Processor, 8 Cores, 16 Threads |
| Identifier | GenuineIntel Family 6 Model 62 Stepping 4 |
| Base Frequency | 4.00 GHz |
| Cluster 1 | 8 Cores |
| Memory Information | |
|---|---|
| Size | 32.00 GB |
AI Performance
| Workload | Accuracy | Score | |
|---|---|---|---|
|
Image Classification (SP)
|
100% |
6639
1.23 KIPS |
|
|
Image Classification (HP)
|
100% |
8632
1.61 KIPS |
|
|
Image Classification (Q)
|
100% |
5046
938.3 IPS |
|
|
Image Segmentation (SP)
|
100% |
7320
118.7 IPS |
|
|
Image Segmentation (HP)
|
100% |
8148
132.1 IPS |
|
|
Image Segmentation (Q)
|
98% |
6212
101.0 IPS |
|
|
Pose Estimation (SP)
|
100% |
167065
194.9 IPS |
|
|
Pose Estimation (HP)
|
100% |
280431
327.2 IPS |
|
|
Pose Estimation (Q)
|
96% |
126695
148.4 IPS |
|
|
Object Detection (SP)
|
100% |
8707
690.6 IPS |
|
|
Object Detection (HP)
|
100% |
10719
850.2 IPS |
|
|
Object Detection (Q)
|
85% |
6477
521.0 IPS |
|
|
Face Detection (SP)
|
100% |
18570
220.7 IPS |
|
|
Face Detection (HP)
|
100% |
21504
255.5 IPS |
|
|
Face Detection (Q)
|
97% |
14490
172.8 IPS |
|
|
Depth Estimation (SP)
|
100% |
55349
426.4 IPS |
|
|
Depth Estimation (HP)
|
99% |
84483
652.7 IPS |
|
|
Depth Estimation (Q)
|
77% |
34692
277.6 IPS |
|
|
Style Transfer (SP)
|
100% |
259442
333.5 IPS |
|
|
Style Transfer (HP)
|
100% |
488311
627.7 IPS |
|
|
Style Transfer (Q)
|
98% |
189497
244.4 IPS |
|
|
Image Super-Resolution (SP)
|
100% |
33443
1.23 KIPS |
|
|
Image Super-Resolution (HP)
|
100% |
41712
1.54 KIPS |
|
|
Image Super-Resolution (Q)
|
99% |
19786
732.6 IPS |
|
|
Text Classification (SP)
|
100% |
2365
3.16 KIPS |
|
|
Text Classification (HP)
|
100% |
2706
3.61 KIPS |
|
|
Text Classification (Q)
|
97% |
1289
1.73 KIPS |
|
|
Machine Translation (SP)
|
100% |
1944
33.5 IPS |
|
|
Machine Translation (HP)
|
96% |
1915
33.1 IPS |
|
|
Machine Translation (Q)
|
70% |
1054
19.8 IPS |
