AMD FirePro R5000 AMD FirePro FirePro R-series Servers February 2013 Tidak berlaku 768 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 102.40 102.4 GB/s 150 150W — — — 1x6-Pin
PCIe® 3.0 x16 2x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 768
— — — — — 28nm
— — — — No
— — — — — — — — — 11.0" (279 mm)
Full Height
— Passive
— — — — — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ R5000
Family: FirePro
Series: FirePro R-series
Form Factor: Servers
Launch Date: February 2013
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 768
External Power Connectors: 1x6-Pin
Total Board Power (TBP): 150W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 102.4 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: 2x DisplayPort 1.2
Other Display Outputs: Ethernet RJ45
Board Height: Full Height
Board Length: 11.0" (279 mm)
Board Width: Single Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S4000X AMD FirePro FirePro S-series Servers August 2014 Tidak berlaku 640 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 72 72 GB/s 45 45W — — — — PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 640
— — — — — 28nm
— — — — No
— — — — — — — — — — — — Passive
— — — — — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S4000X
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: August 2014
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 640
Total Board Power (TBP): 45W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 72 GB/s
Memory ECC Support: No
GPU Form Factor: MXM (Type A)
Bus Type: PCIe® 3.0 x16
Cooling: Passive
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S7000 AMD FirePro FirePro S-series Servers August 2012 Tidak berlaku 1,280 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 4 GB 4 GDDR5 — — 154 154 GB/s 150 150W — — — 1x6-Pin
PCIe® 3.0 x16 1x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1280
— — — — — 28nm
— — — — No
— — — — — — — — — 9.5" (241 mm)
Full Height
— Passive
— — — — — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S7000
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: August 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 1280
External Power Connectors: 1x6-Pin
Total Board Power (TBP): 150W
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 154 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: 1x DisplayPort 1.2
Board Height: Full Height
Board Length: 9.5" (241 mm)
Board Width: Single Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S7100X AMD FirePro FirePro S-series Servers May 2016 Tidak berlaku 1,792 — Tidak berlaku — — — — — Graphics Core Next (GCN) 3rd Gen — — — 8 GB 8 GDDR5 — — 160 160 GB/s 100 100W — — — — PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1792
— — — — — 28nm
— — — — No
— — — — — — — — — — — — Passive
— — — — — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— AMD MxGPU Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S7100X
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: May 2016
GPU Architecture: Graphics Core Next (GCN) 3rd Gen
Lithography: 28nm
Stream Processors: 1792
Total Board Power (TBP): 100W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 160 GB/s
Memory ECC Support: No
GPU Form Factor: MXM (Type B)
Bus Type: PCIe® 3.0 x16
Cooling: Passive
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: AMD MxGPU Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S7150 (Active Cooling) AMD FirePro FirePro S-series Servers January 2016 Tidak berlaku 1,792 — Tidak berlaku — — — — — Graphics Core Next (GCN) 3rd Gen — — — 8 GB 8 GDDR5 — — 160 160 GB/s 150 150W — — — 1x6-Pin
PCIe® 3.0 x16 None — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1792
— — — — — 28nm
— — — — No
— — — — — — — — — 9.5" (241 mm)
Full Height
— Passive
— — — — — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— AMD MxGPU Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S7150 (Active Cooling)
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: January 2016
GPU Architecture: Graphics Core Next (GCN) 3rd Gen
Lithography: 28nm
Stream Processors: 1792
External Power Connectors: 1x6-Pin
Total Board Power (TBP): 150W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 160 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: None
Board Height: Full Height
Board Length: 9.5" (241 mm)
Board Width: Single Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: AMD MxGPU Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S7150 (Passive Cooling) AMD FirePro FirePro S-series Servers January 2016 Tidak berlaku 1,792 — Tidak berlaku — — — — — Graphics Core Next (GCN) 3rd Gen — — — 8 GB 8 GDDR5 — — 160 160 GB/s 150 150W — — — 1x6-Pin
PCIe® 3.0 x16 None — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1792
— — — — — 28nm
— — — — No
— — — — — — — — — 9.5" (241 mm)
Full Height
— Passive
— — — — — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— AMD MxGPU Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S7150 (Passive Cooling)
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: January 2016
GPU Architecture: Graphics Core Next (GCN) 3rd Gen
Lithography: 28nm
Stream Processors: 1792
External Power Connectors: 1x6-Pin
Total Board Power (TBP): 150W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 160 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: None
Board Height: Full Height
Board Length: 9.5" (241 mm)
Board Width: Single Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: AMD MxGPU Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S7150 X2 AMD FirePro FirePro S-series Servers January 2016 Tidak berlaku — — Tidak berlaku — — — — — Graphics Core Next (GCN) 3rd Gen — — — 16 GB 16 GDDR5 — — 320 320 GB/s 265 265W — — — 1x6-Pin + 1x8-Pin
PCIe® 3.0 x16 None — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 3584 (2x1792)
— — — — — 28nm
— — — — No
— — — — — — — — — 10.5" (267 mm)
Full Height
— Passive
— — — — — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— AMD MxGPU Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S7150 X2
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: January 2016
GPU Architecture: Graphics Core Next (GCN) 3rd Gen
Lithography: 28nm
Stream Processors: 3584 (2x1792)
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 265W
Dedicated Memory Size: 16 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 320 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: None
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: AMD MxGPU Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S9000 AMD FirePro FirePro S-series Servers August 2012 Tidak berlaku 1,792 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 6 GB 6 GDDR5 — — 264 264 GB/s 225 225W — — — 1x8-Pin
PCIe® 3.0 x16 1x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1792
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Passive
3.23 TFLOPs
— 806 GFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S9000
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: August 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 1792
Peak Single Precision (FP32 Vector) Performance: 3.23 TFLOPs
Peak Double Precision (FP64) Performance: 806 GFLOPs
External Power Connectors: 1x8-Pin
Total Board Power (TBP): 225W
Dedicated Memory Size: 6 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 264 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: 1x DisplayPort 1.2
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S9050 AMD FirePro FirePro S-series Servers August 2014 Tidak berlaku 1,792 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 12 GB 12 GDDR5 — — 264 264 GB/s 225 225W — — — 1x8-Pin
PCIe® 3.0 x16 1x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1792
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Passive
3.23 TFLOPs
— 806 GFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S9050
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: August 2014
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 1792
Peak Single Precision (FP32 Vector) Performance: 3.23 TFLOPs
Peak Double Precision (FP64) Performance: 806 GFLOPs
External Power Connectors: 1x8-Pin
Total Board Power (TBP): 225W
Dedicated Memory Size: 12 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 264 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: 1x DisplayPort 1.2
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S9100 AMD FirePro FirePro S-series Servers October 2014 Tidak berlaku 2,560 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 12 GB 12 GDDR5 — — 320 320 GB/s 225 225W — — — 1x8-Pin
PCIe® 3.0 x16 None — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2560
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Passive
4.22 TFLOPs
— 2.11 TFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S9100
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: October 2014
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 2560
Peak Single Precision (FP32 Vector) Performance: 4.22 TFLOPs
Peak Double Precision (FP64) Performance: 2.11 TFLOPs
External Power Connectors: 1x8-Pin
Total Board Power (TBP): 225W
Dedicated Memory Size: 12 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 320 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: None
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S9150 AMD FirePro FirePro S-series Servers August 2014 Tidak berlaku 2,816 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 16 GB 16 GDDR5 — — 320 320 GB/s 235 235W — — — 1x6-Pin + 1x8-Pin
PCIe® 3.0 x16 None — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2816
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Passive
5.07 TFLOPs
— 2.53 TFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S9150
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: August 2014
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 2816
Peak Single Precision (FP32 Vector) Performance: 5.07 TFLOPs
Peak Double Precision (FP64) Performance: 2.53 TFLOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 235W
Dedicated Memory Size: 16 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 320 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: None
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S9170 AMD FirePro FirePro S-series Servers July 2015 Tidak berlaku 2,816 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 32 GB 32 GDDR5 — — 320 320 GB/s 275 275W — — — 1x6-Pin + 1x8-Pin
PCIe® 3.0 x16 None — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2816
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Passive
5.24 TFLOPs
— 2.62 TFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S9170
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: July 2015
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 2816
Peak Single Precision (FP32 Vector) Performance: 5.24 TFLOPs
Peak Double Precision (FP64) Performance: 2.62 TFLOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 275W
Dedicated Memory Size: 32 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 320 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: None
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S9300 X2 AMD FirePro FirePro S-series Servers March 2016 Tidak berlaku — — Tidak berlaku — — — — — Graphics Core Next (GCN) 3rd Gen — — — 8 GB 8 High Bandwidth Memory (HBM) — — 1,000 1 TB/s 300 300W — — — 2x8-Pin
PCIe® 3.0 x16 None — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 8192 (2x4096)
— — — — — 28nm
— — — — No
— — — — — — — — — 10.5" (267 mm)
Full Height
— Passive
13.93 TFLOPs
13.93 TFLOPs
870 GFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S9300 X2
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: March 2016
GPU Architecture: Graphics Core Next (GCN) 3rd Gen
Lithography: 28nm
Stream Processors: 8192 (2x4096)
Peak Half Precision (FP16 Vector) Performance: 13.93 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 13.93 TFLOPs
Peak Double Precision (FP64) Performance: 870 GFLOPs
External Power Connectors: 2x8-Pin
Total Board Power (TBP): 300W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: High Bandwidth Memory (HBM)
Peak Memory Bandwidth: 1 TB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: None
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S10000 AMD FirePro FirePro S-series Servers November 2012 Tidak berlaku 3,584 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 6 GB 6 GDDR5 — — 480 480 GB/s 375 375W — — — 2x8-Pin
PCIe® 3.0 x16 1x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 3584
— — — — — 28nm
— — — — Yes
— — — — — — — — — 12.0" (305 mm)
Full Height
— Passive
5.91 TFLOPs
— 1.48 TFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S10000
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: November 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 3584
Peak Single Precision (FP32 Vector) Performance: 5.91 TFLOPs
Peak Double Precision (FP64) Performance: 1.48 TFLOPs
External Power Connectors: 2x8-Pin
Total Board Power (TBP): 375W
Dedicated Memory Size: 6 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 480 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: 1x Mini-DisplayPort 1.2
DVI: 1x DVI-I (Dual-link)
Board Height: Full Height
Board Length: 12.0" (305 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S10000 (Active Cooling) AMD FirePro FirePro S-series Servers November 2012 Tidak berlaku 3,584 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 6 GB 6 GDDR5 — — 480 480 GB/s 375 375W — — — 2x8-Pin
PCIe® 3.0 x16 1x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 3584
— — — — — 28nm
— — — — Yes
— — — — — — — — — 12.0" (305 mm)
Full Height
— Passive
5.91 TFLOPs
— 1.48 TFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S10000 (Active Cooling)
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: November 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 3584
Peak Single Precision (FP32 Vector) Performance: 5.91 TFLOPs
Peak Double Precision (FP64) Performance: 1.48 TFLOPs
External Power Connectors: 2x8-Pin
Total Board Power (TBP): 375W
Dedicated Memory Size: 6 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 480 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: 1x Mini-DisplayPort 1.2
DVI: 1x DVI-I (Dual-link)
Board Height: Full Height
Board Length: 12.0" (305 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro S10000 12GB Edition AMD FirePro FirePro S-series Servers November 2013 Tidak berlaku 3,584 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 12 GB 12 GDDR5 — — 480 480 GB/s 375 375W — — — 2x8-Pin
PCIe® 3.0 x16 1x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 3584
— — — — — 28nm
— — — — Yes
— — — — — — — — — 12.0" (305 mm)
Full Height
— Passive
5.91 TFLOPs
— 1.48 TFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ S10000 12GB Edition
Family: FirePro
Series: FirePro S-series
Form Factor: Servers
Launch Date: November 2013
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 3584
Peak Single Precision (FP32 Vector) Performance: 5.91 TFLOPs
Peak Double Precision (FP64) Performance: 1.48 TFLOPs
External Power Connectors: 2x8-Pin
Total Board Power (TBP): 375W
Dedicated Memory Size: 12 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 480 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
DisplayPort: 1x Mini-DisplayPort 1.2
DVI: 1x DVI-I (Dual-link)
Board Height: Full Height
Board Length: 12.0" (305 mm)
Board Width: Double Slot
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro W600 AMD FirePro FirePro Wx000 Series Desktop June 2012 Tidak berlaku 512 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 19.20 19.2 GB/s 75 75W — — — None
PCIe® 3.0 x16 6x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 512
— — — — — 28nm
— — — — No
— — — — — — — — — 6.6" (168 mm)
Full Height
— Active
770 GFLOPs
— 48 GFLOPs
— — — — — — — 12.0 (feature level 11_1)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ W600
Family: FirePro
Series: FirePro Wx000 Series
Form Factor: Desktops
Launch Date: June 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 512
Peak Single Precision (FP32 Vector) Performance: 770 GFLOPs
Peak Double Precision (FP64) Performance: 48 GFLOPs
External Power Connectors: None
Total Board Power (TBP): 75W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 19.2 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 6x Mini-DisplayPort 1.2
Board Height: Full Height
Board Length: 6.6" (168 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: No
DirectX: 12.0 (feature level 11_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro W2100 AMD FirePro FirePro Wx100 Series Desktop August 2014 Tidak berlaku 320 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 DDR3 — — 28.80 28.8 GB/s 26 26W — — — None
PCIe® 3.0 x16 2x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 320
— — — — — 28nm
— — — — No
— — — — — — — — — 6.6" (168 mm)
Half Height
— Active
400 GFLOPs
— 25 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD)
Name: FirePro™ W2100
Family: FirePro
Series: FirePro Wx100 Series
Form Factor: Desktops
Launch Date: August 2014
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 320
Peak Single Precision (FP32 Vector) Performance: 400 GFLOPs
Peak Double Precision (FP64) Performance: 25 GFLOPs
External Power Connectors: None
Total Board Power (TBP): 26W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: DDR3
Peak Memory Bandwidth: 28.8 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 2x DisplayPort 1.2
Board Height: Half Height
Board Length: 6.6" (168 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: No
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD)
— 공식 사양 수치 확인
— — — — — AMD FirePro W4100 AMD FirePro FirePro Wx100 Series Desktop August 2014 Tidak berlaku 512 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 72 72 GB/s 50 50W — — — None
PCIe® 3.0 x16 4x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 512
— — — — — 28nm
— — — — No
— — — — — — — — — 6.6" (168 mm)
Half Height
— Active
650 GFLOPs
— 41 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
Name: FirePro™ W4100
Family: FirePro
Series: FirePro Wx100 Series
Form Factor: Desktops
Launch Date: August 2014
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 512
Peak Single Precision (FP32 Vector) Performance: 650 GFLOPs
Peak Double Precision (FP64) Performance: 41 GFLOPs
External Power Connectors: None
Total Board Power (TBP): 50W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 72 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 4x Mini-DisplayPort 1.2
Board Height: Half Height
Board Length: 6.6" (168 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: No
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
— 공식 사양 수치 확인
— — — — — AMD FirePro W4170M AMD FirePro FirePro Mobility Series Mobile — Tidak berlaku 384 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 64 64 GB/s 25 25W — — — — PCIe® 3.0 x8 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 384
— — — — — 28nm
— — — — No
— — — — — — — — — — — — — 630 GFLOPs
— 40 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD)
Name: FirePro™ W4170M
Family: FirePro
Series: FirePro Mobility Series
Form Factor: Laptops
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 384
Peak Single Precision (FP32 Vector) Performance: 630 GFLOPs
Peak Double Precision (FP64) Performance: 40 GFLOPs
Total Board Power (TBP): 25W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 64 GB/s
Memory ECC Support: No
Bus Type: PCIe® 3.0 x8
10-bit Display Color Output: Yes
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD)
— 공식 사양 수치 확인
— — — — — AMD FirePro W4190M AMD FirePro FirePro Mobility Series Mobile — Tidak berlaku 384 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 64 64 GB/s 25 25W — — — — PCIe® 3.0 x8 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 384
— — — — — 28nm
— — — — No
— — — — — — — — — — — — — 630 GFLOPs
— 40 GFLOPs
— — — — — — — — — — — — OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD)
Name: FirePro™ W4190M
Family: FirePro
Series: FirePro Mobility Series
Form Factor: Laptops
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 384
Peak Single Precision (FP32 Vector) Performance: 630 GFLOPs
Peak Double Precision (FP64) Performance: 40 GFLOPs
Total Board Power (TBP): 25W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 64 GB/s
Memory ECC Support: No
Bus Type: PCIe® 3.0 x8
10-bit Display Color Output: Yes
Supported Technologies: OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD)
— 공식 사양 수치 확인
— — — — — AMD FirePro W4300 AMD FirePro FirePro Wx100 Series Desktop December 2015 Tidak berlaku 768 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 4 GB 4 GDDR5 — — 96 96 GB/s 48 48W — — — None
PCIe® 3.0 x16 4x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 768
— — — — — 28nm
— — — — No
— — — — — — — — — 6.6" (168 mm)
Half Height
— Active
1.43 TFLOPs
— 89 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
Name: FirePro™ W4300
Family: FirePro
Series: FirePro Wx100 Series
Form Factor: Desktops
Launch Date: December 2015
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 768
Peak Single Precision (FP32 Vector) Performance: 1.43 TFLOPs
Peak Double Precision (FP64) Performance: 89 GFLOPs
External Power Connectors: None
Total Board Power (TBP): 48W
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 96 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 4x Mini-DisplayPort 1.2
Board Height: Half Height
Board Length: 6.6" (168 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: No
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
— 공식 사양 수치 확인
— — — — — AMD FirePro W5000 AMD FirePro FirePro Wx000 Series Desktop August 2012 Tidak berlaku 768 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 102 102 GB/s 71 71W — — — None
PCIe® 3.0 x16 2x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 768
— — — — — 28nm
— — — — No
— — — — — — — — — 6.8" (173 mm)
Full Height
— Active
1.27 TFLOPs
— 79 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology
Name: FirePro™ W5000
Family: FirePro
Series: FirePro Wx000 Series
Form Factor: Desktops
Launch Date: August 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 768
Peak Single Precision (FP32 Vector) Performance: 1.27 TFLOPs
Peak Double Precision (FP64) Performance: 79 GFLOPs
External Power Connectors: None
Total Board Power (TBP): 71W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 102 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 2x DisplayPort 1.2
DVI: 1x DVI-I (Dual-link)
Board Height: Full Height
Board Length: 6.8" (173 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: No
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology
— 공식 사양 수치 확인
— — — — — AMD FirePro W5000 DVI AMD FirePro FirePro Wx000 Series Desktop February 2013 Tidak berlaku 768 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 102 102 GB/s 71 71W — — — None
PCIe® 3.0 x16 None — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 768
— — — — — 28nm
— — — — No
— — — — — — — — — 6.8" (173 mm)
Full Height
— Active
1.27 TFLOPs
— 79 GFLOPs
— — — — — — — 12.0 (feature level 11_1)
4.6
2.0
1.0
— AMD DirectGMA Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ W5000 DVI
Family: FirePro
Series: FirePro Wx000 Series
Form Factor: Desktops
Launch Date: February 2013
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 768
Peak Single Precision (FP32 Vector) Performance: 1.27 TFLOPs
Peak Double Precision (FP64) Performance: 79 GFLOPs
External Power Connectors: None
Total Board Power (TBP): 71W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 102 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: None
DVI: 2x DVI-I (Dual-link)
Board Height: Full Height
Board Length: 6.8" (173 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: No
DirectX: 12.0 (feature level 11_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: AMD DirectGMA Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro W5100 AMD FirePro FirePro Wx100 Series Desktop August 2014 Tidak berlaku 768 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 4 GB 4 GDDR5 — — 96 96 GB/s 75 75W — — — None
PCIe® 3.0 x16 4x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 768
— — — — — 28nm
— — — — No
— — — — — — — — — 6.8" (173 mm)
Full Height
— Active
1.43 TFLOPs
— 89 GFLOPs
— — — — — — — 12.0 (feature level 11_1)
4.6
2.0
1.0
— Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ W5100
Family: FirePro
Series: FirePro Wx100 Series
Form Factor: Desktops
Launch Date: August 2014
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 768
Peak Single Precision (FP32 Vector) Performance: 1.43 TFLOPs
Peak Double Precision (FP64) Performance: 89 GFLOPs
External Power Connectors: None
Total Board Power (TBP): 75W
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 96 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 4x DisplayPort 1.2
Board Height: Full Height
Board Length: 6.8" (173 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes (Stereo Connector sold separately)
DirectX: 12.0 (feature level 11_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro W5130M AMD FirePro FirePro Mobility Series Mobile — Tidak berlaku 512 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 64 64 GB/s 35 35W — — — — PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 512
— — — — — 28nm
— — — — No
— — — — — — — — — — — — — 950 GFLOPs
— 59 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
Name: FirePro™ W5130M
Family: FirePro
Series: FirePro Mobility Series
Form Factor: Laptops
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 512
Peak Single Precision (FP32 Vector) Performance: 950 GFLOPs
Peak Double Precision (FP64) Performance: 59 GFLOPs
Total Board Power (TBP): 35W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 64 GB/s
Memory ECC Support: No
Bus Type: PCIe® 3.0 x16
10-bit Display Color Output: Yes
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
— 공식 사양 수치 확인
— — — — — AMD FirePro W5170M AMD FirePro FirePro Mobility Series Mobile — Tidak berlaku 640 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 2 GB 2 GDDR5 — — 72 72 GB/s 45 45W — — — — PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 640
— — — — — 28nm
— — — — No
— — — — — — — — — — — — — 1.18 TFLOPs
— 74 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
Name: FirePro™ W5170M
Family: FirePro
Series: FirePro Mobility Series
Form Factor: Laptops
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 640
Peak Single Precision (FP32 Vector) Performance: 1.18 TFLOPs
Peak Double Precision (FP64) Performance: 74 GFLOPs
Total Board Power (TBP): 45W
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 72 GB/s
Memory ECC Support: No
Bus Type: PCIe® 3.0 x16
10-bit Display Color Output: Yes
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
— 공식 사양 수치 확인
— — — — — AMD FirePro W6150M AMD FirePro FirePro Mobility Series Mobile — Tidak berlaku 768 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 4 GB 4 GDDR5 — — 88 88 GB/s 100 100W — — — — PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 768
— — — — — 28nm
— — — — No
— — — — — — — — — — — — — 1.97 TFLOPs
— 123 GFLOPs
— — — — — — — — — — — — OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
Name: FirePro™ W6150M
Family: FirePro
Series: FirePro Mobility Series
Form Factor: Laptops
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 768
Peak Single Precision (FP32 Vector) Performance: 1.97 TFLOPs
Peak Double Precision (FP64) Performance: 123 GFLOPs
Total Board Power (TBP): 100W
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 88 GB/s
Memory ECC Support: No
Bus Type: PCIe® 3.0 x16
10-bit Display Color Output: Yes
Supported Technologies: OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
— 공식 사양 수치 확인
— — — — — AMD FirePro W7000 AMD FirePro FirePro Wx000 Series Desktop June 2012 Tidak berlaku 1,280 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 4 GB 4 GDDR5 — — 154 154 GB/s 150 150W — — — 1x6-Pin
PCIe® 3.0 x16 4x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1280
— — — — — 28nm
— — — — No
— — — — — — — — — 9.5" (241 mm)
Full Height
— Active
2.43 TFLOPs
— 152 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology , S400 Synchronization Module Support
Name: FirePro™ W7000
Family: FirePro
Series: FirePro Wx000 Series
Form Factor: Desktops
Launch Date: June 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 1280
Peak Single Precision (FP32 Vector) Performance: 2.43 TFLOPs
Peak Double Precision (FP64) Performance: 152 GFLOPs
External Power Connectors: 1x6-Pin
Total Board Power (TBP): 150W
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 154 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 4x DisplayPort 1.2
Board Height: Full Height
Board Length: 9.5" (241 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology , S400 Synchronization Module Support
— 공식 사양 수치 확인
— — — — — AMD FirePro W7100 AMD FirePro FirePro Wx100 Series Desktop August 2014 Tidak berlaku 1,792 — Tidak berlaku — — — — — Graphics Core Next (GCN) 3rd Gen — — — 8 GB 8 GDDR5 — — 160 160 GB/s 150 150W — — — 1x6-Pin
PCIe® 3.0 x16 4x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1792
— — — — — 28nm
— — — — No
— — — — — — — — — 9.5" (241 mm)
Full Height
— Active
3.3 TFLOPs
— 206 GFLOPs
— — — — — — — 12.0 (feature level 11_1)
4.6
2.0
1.0
— S400 Synchronization Module Support , AMD DirectGMA Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ W7100
Family: FirePro
Series: FirePro Wx100 Series
Form Factor: Desktops
Launch Date: August 2014
GPU Architecture: Graphics Core Next (GCN) 3rd Gen
Lithography: 28nm
Stream Processors: 1792
Peak Single Precision (FP32 Vector) Performance: 3.3 TFLOPs
Peak Double Precision (FP64) Performance: 206 GFLOPs
External Power Connectors: 1x6-Pin
Total Board Power (TBP): 150W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 160 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 4x DisplayPort 1.2
Board Height: Full Height
Board Length: 9.5" (241 mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
DirectX: 12.0 (feature level 11_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: S400 Synchronization Module Support , AMD DirectGMA Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro W7170M AMD FirePro FirePro Mobility Series Mobile — Tidak berlaku 2,048 — Tidak berlaku — — — — — Graphics Core Next (GCN) 3rd Gen — — — 4 GB 4 GDDR5 — — 160 160 GB/s 100 100W — — — — PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2048
— — — — — 28nm
— — — — No
— — — — — — — — — — — — — 2.96 TFLOPs
— 185 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
Name: FirePro™ W7170M
Family: FirePro
Series: FirePro Mobility Series
Form Factor: Laptops
GPU Architecture: Graphics Core Next (GCN) 3rd Gen
Lithography: 28nm
Stream Processors: 2048
Peak Single Precision (FP32 Vector) Performance: 2.96 TFLOPs
Peak Double Precision (FP64) Performance: 185 GFLOPs
Total Board Power (TBP): 100W
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 160 GB/s
Memory ECC Support: No
Bus Type: PCIe® 3.0 x16
10-bit Display Color Output: Yes
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE)
— 공식 사양 수치 확인
— — — — — AMD FirePro W8000 AMD FirePro FirePro Wx000 Series Desktop June 2012 Tidak berlaku 1,792 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 4 GB 4 GDDR5 — — 176 176 GB/s 189 189W — — — 2x6-Pin
PCIe® 3.0 x16 4x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1792
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Active
3.2 TFLOPs
— 800 GFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology , S400 Synchronization Module Support
Name: FirePro™ W8000
Family: FirePro
Series: FirePro Wx000 Series
Form Factor: Desktops
Launch Date: June 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 1792
Peak Single Precision (FP32 Vector) Performance: 3.2 TFLOPs
Peak Double Precision (FP64) Performance: 800 GFLOPs
External Power Connectors: 2x6-Pin
Total Board Power (TBP): 189W
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 176 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 4x DisplayPort 1.2
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology , S400 Synchronization Module Support
— 공식 사양 수치 확인
— — — — — AMD FirePro W8100 AMD FirePro FirePro Wx100 Series Desktop June 2014 Tidak berlaku 2,560 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 8 GB 8 GDDR5 — — 320 320 GB/s 225 225W — — — 2x6-Pin
PCIe® 3.0 x16 4x DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2560
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Active
4.22 TFLOPs
— 2.11 TFLOPs
— — — — — — — 12.0 (feature level 11_1)
4.6
2.0
1.0
— S400 Synchronization Module Support , AMD DirectGMA Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ W8100
Family: FirePro
Series: FirePro Wx100 Series
Form Factor: Desktops
Launch Date: June 2014
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 2560
Peak Single Precision (FP32 Vector) Performance: 4.22 TFLOPs
Peak Double Precision (FP64) Performance: 2.11 TFLOPs
External Power Connectors: 2x6-Pin
Total Board Power (TBP): 225W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 320 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 4x DisplayPort 1.2
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
DirectX: 12.0 (feature level 11_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: S400 Synchronization Module Support , AMD DirectGMA Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro W9000 AMD FirePro FirePro Wx000 Series Desktop June 2012 Tidak berlaku 2,048 — Tidak berlaku — — — — — Graphics Core Next (GCN) — — — 6 GB 6 GDDR5 — — 264 264 GB/s 274 274W — — — 1x6-Pin + 1x8-Pin
PCIe® 3.0 x16 6x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2048
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Active
4 TFLOPs
— 1 TFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology , S400 Synchronization Module Support
Name: FirePro™ W9000
Family: FirePro
Series: FirePro Wx000 Series
Form Factor: Desktops
Launch Date: June 2012
GPU Architecture: Graphics Core Next (GCN)
Lithography: 28nm
Stream Processors: 2048
Peak Single Precision (FP32 Vector) Performance: 4 TFLOPs
Peak Double Precision (FP64) Performance: 1 TFLOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 274W
Dedicated Memory Size: 6 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 264 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 6x Mini-DisplayPort 1.2
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology , S400 Synchronization Module Support
— 공식 사양 수치 확인
— — — — — AMD FirePro W9100 AMD FirePro FirePro Wx100 Series Desktop March 2014 Tidak berlaku 2,816 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 16 GB 16 GDDR5 — — 320 320 GB/s 275 275W — — — 1x6-Pin + 1x8-Pin
PCIe® 3.0 x16 6x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2816
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Active
5.24 TFLOPs
— 2.62 TFLOPs
— — — — — — — 12.0 (feature level 11_1)
4.6
2.0
1.0
— S400 Synchronization Module Support , AMD DirectGMA Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
Name: FirePro™ W9100
Family: FirePro
Series: FirePro Wx100 Series
Form Factor: Desktops
Launch Date: March 2014
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 2816
Peak Single Precision (FP32 Vector) Performance: 5.24 TFLOPs
Peak Double Precision (FP64) Performance: 2.62 TFLOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 275W
Dedicated Memory Size: 16 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 320 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 6x Mini-DisplayPort 1.2
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
DirectX: 12.0 (feature level 11_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.0
Supported Technologies: S400 Synchronization Module Support , AMD DirectGMA Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , DirectX® 12 Technology , OpenGL 4.6 , OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD FirePro W9100 32GB AMD FirePro FirePro Wx100 Series Desktop April 2016 Tidak berlaku 2,816 — Tidak berlaku — — — — — Graphics Core Next (GCN) 2nd Gen — — — 32 GB 32 GDDR5 — — 320 320 GB/s 275 275W — — — 1x6-Pin + 1x8-Pin
PCIe® 3.0 x16 6x Mini-DisplayPort 1.2 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2816
— — — — — 28nm
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Active
5.24 TFLOPs
— 2.62 TFLOPs
— — — — — — — — — — — — DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology , S400 Synchronization Module Support
Name: FirePro™ W9100 32GB
Family: FirePro
Series: FirePro Wx100 Series
Form Factor: Desktops
Launch Date: April 2016
GPU Architecture: Graphics Core Next (GCN) 2nd Gen
Lithography: 28nm
Stream Processors: 2816
Peak Single Precision (FP32 Vector) Performance: 5.24 TFLOPs
Peak Double Precision (FP64) Performance: 2.62 TFLOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 275W
Dedicated Memory Size: 32 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 320 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 6x Mini-DisplayPort 1.2
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
Supported Technologies: DirectX® 12 Technology , OpenCL® 2.0 , OpenGL 4.6 , Unified Video Decoder (UVD) , Video Code Engine (VCE) , AMD DirectGMA Technology , S400 Synchronization Module Support
— 공식 사양 수치 확인
— — — — — AMD Instinct MI6 AMD Instinct Instinct MI Series Servers June 2017 Tidak berlaku 2,304 — Tidak berlaku — — — — — Polaris — — — 16 GB 16 GDDR5 — — 224 224 GB/s — — — — — 1x6-Pin
PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2,304
— — — — — 14nm FinFET
— — — — No
— — — 150W Peak
— — — — — 9.5" (241 mm)
Full Height
— Passive
5.73 TFLOPs
5.73 TFLOPs
358 GFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— — Name: Instinct™ MI6
Family: Instinct
Series: Instinct MI Series
Form Factor: Servers
Launch Date: June 2017
GPU Architecture: Polaris
Lithography: 14nm FinFET
Stream Processors: 2,304
Peak Half Precision (FP16) Performance: 5.73 TFLOPs
Peak Single Precision (FP32) Performance: 5.73 TFLOPs
Peak Double Precision (FP64) Performance: 358 GFLOPs
External Power Connectors: 1x6-Pin
Thermal Design Power (TDP): 150W Peak
Dedicated Memory Size: 16 GB
Dedicated Memory Type: GDDR5
Peak Memory Bandwidth: 224 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
Board Height: Full Height
Board Length: 9.5" (241 mm)
Board Width: Single Slot
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
ROCm Open Ecosystem: <p><a href="https://rocm.github.io/index.html" target="_blank">ROCm Open eCosystem</a></p>
Vulkan: 1.0
— 공식 사양 수치 확인
— — — — — AMD Instinct MI8 AMD Instinct Instinct MI Series Servers 06/2017 Tidak berlaku 4,096 — Tidak berlaku — — — — — Graphics Core Next (GCN) 3rd Gen — — — 4 GB 4 High Bandwidth Memory (HBM) — — 512 512 GB/s — — — — — 1x8-Pin
PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 4,096
— — — — — 28nm
— — — — No
— — — 175W Peak
— — — — — 6.0" (152 mm)
Full Height
— Passive
8.9 TFLOPs
8.9 TFLOPs
512 GFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.6
2.0
1.0
— OpenCL® 2.0
Name: Instinct™ MI8
Family: Instinct
Series: Instinct MI Series
Form Factor: Servers
Launch Date: 06/2017
GPU Architecture: Graphics Core Next (GCN) 3rd Gen
Lithography: 28nm
Stream Processors: 4,096
Peak Half Precision (FP16) Performance: 8.9 TFLOPs
Peak Single Precision (FP32) Performance: 8.9 TFLOPs
Peak Double Precision (FP64) Performance: 512 GFLOPs
External Power Connectors: 1x8-Pin
Thermal Design Power (TDP): 175W Peak
Dedicated Memory Size: 4 GB
Dedicated Memory Type: High Bandwidth Memory (HBM)
Peak Memory Bandwidth: 512 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
Board Height: Full Height
Board Length: 6.0" (152 mm)
Board Width: Double Slot
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.6
OpenCL: 2.0
ROCm Open Ecosystem: <p><a href="https://rocm.github.io/index.html" target="_blank">ROCm Open eCosystem</a></p>
Vulkan: 1.0
Supported Technologies: OpenCL® 2.0
— 공식 사양 수치 확인
— — — — — AMD Instinct MI25 AMD Instinct Instinct MI Series Servers June 2017 Tidak berlaku 4,096 64 Tidak berlaku — — — — — Vega — — — 16 GB 16 HBM2 — — 484 484 GB/s — — — — — 2x8-Pin
PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 4,096
64
— — — — 14nm FinFET
— — — — Yes
— — — 300W Peak
— — — — — 10.5" (267mm)
Full Height
— Passive
12.29 TFLOPs
24.6 TFLOPs
768 GFLOPs
— — — — — — — 12.0 (feature level 12_1)
4.6
2.0
1.0
— OpenCL® 2.0 , High Bandwidth Cache (HBC) Controller
Name: Instinct™ MI25
Family: Instinct
Series: Instinct MI Series
Form Factor: Servers
Launch Date: June 2017
GPU Architecture: Vega
Lithography: 14nm FinFET
Stream Processors: 4,096
Compute Units: 64
Peak Half Precision (FP16) Performance: 24.6 TFLOPs
Peak Single Precision (FP32) Performance: 12.29 TFLOPs
Peak Double Precision (FP64) Performance: 768 GFLOPs
External Power Connectors: 2x8-Pin
Thermal Design Power (TDP): 300W Peak
Dedicated Memory Size: 16 GB
Dedicated Memory Type: HBM2
Peak Memory Bandwidth: 484 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267mm)
Board Width: Double Slot
DirectX: 12.0 (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.0
ROCm Open Ecosystem: <p><a href="https://rocm.github.io/index.html" target="_blank">yes</a></p>
Vulkan: 1.0
Supported Technologies: OpenCL® 2.0 , High Bandwidth Cache (HBC) Controller
— 공식 사양 수치 확인
— — — — — AMD Instinct MI50 (16GB) AMD Instinct Instinct MI Series Servers 11/18/2018 Tidak berlaku 3,840 60 Tidak berlaku — — 1,725 — — Vega20 — — 1725 MHz 16 GB 16 HBM2 4,096 4096-bit 1,000 1 TB/s — — — — — 2x8-Pin
PCIe® 4.0 x16 , PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 3,840
60
— — — — TSMC 7nm FinFET
— — — 1 GHz
Yes (Full-Chip)
— — — 300W Peak
— — — — — 10.5" (267mm)
Full Height
— Passive
13.3 TFLOPs
26.5 TFLOPs
6.6 TFLOPs
— — — — — — — — 4.6
2.0
1.0
Linux x86 64-Bit
— Name: Instinct™ MI50 (16GB)
Family: Instinct
Series: Instinct MI Series
Form Factor: Servers
Launch Date: 11/18/2018
GPU Architecture: Vega20
Lithography: TSMC 7nm FinFET
Stream Processors: 3,840
Compute Units: 60
Peak Engine Clock: 1725 MHz
Peak Half Precision (FP16) Performance: 26.5 TFLOPs
Peak Single Precision (FP32) Performance: 13.3 TFLOPs
Peak Double Precision (FP64) Performance: 6.6 TFLOPs
Peak INT8 Performance: 53 TOPs
External Power Connectors: 2x8-Pin
Thermal Design Power (TDP): 300W Peak
Dedicated Memory Size: 16 GB
Dedicated Memory Type: HBM2
Memory Interface: 4096-bit
Memory Clock: 1 GHz
Peak Memory Bandwidth: 1 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16 , PCIe® 3.0 x16
Infinity Fabric Links: 2
Peak Infinity Fabric Link Bandwidth: 92 GB/s
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267mm)
Board Width: Double Slot
RAS Support: Yes
OpenGL: 4.6
OpenCL: 2.0
ROCm Open Ecosystem: <p><a target="_blank">ROCm Open Ecosystem</a></p>
Vulkan: 1.0
OS Support: Linux x86 64-Bit
— 공식 사양 수치 확인
— — — — — AMD Instinct MI50 (32GB) AMD Instinct Instinct MI Series Servers 11/18/2018 Tidak berlaku 3,840 60 Tidak berlaku — — 1,725 — — Vega20 — — 1725 MHz 32 GB 32 HBM2 4,096 4096-bit 1,000 1 TB/s — — — — — 2x8-Pin
PCIe® 4.0 x16 , PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 3,840
60
— — — — TSMC 7nm FinFET
— — — 1 GHz
Yes (Full-Chip)
— — — 300W Peak
— — — — — 10.5" (267mm)
Full Height
— Passive
13.3 TFLOPs
26.5 TFLOPs
6.6 TFLOPs
— — — — — — — — 4.6
2.0
1.0
Linux x86 64-Bit
— Name: Instinct™ MI50 (32GB)
Family: Instinct
Series: Instinct MI Series
Form Factor: Servers
Launch Date: 11/18/2018
GPU Architecture: Vega20
Lithography: TSMC 7nm FinFET
Stream Processors: 3,840
Compute Units: 60
Peak Engine Clock: 1725 MHz
Peak Half Precision (FP16) Performance: 26.5 TFLOPs
Peak Single Precision (FP32) Performance: 13.3 TFLOPs
Peak Double Precision (FP64) Performance: 6.6 TFLOPs
Peak INT8 Performance: 53 TOPs
External Power Connectors: 2x8-Pin
Thermal Design Power (TDP): 300W Peak
Dedicated Memory Size: 32 GB
Dedicated Memory Type: HBM2
Memory Interface: 4096-bit
Memory Clock: 1 GHz
Peak Memory Bandwidth: 1 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16 , PCIe® 3.0 x16
Infinity Fabric Links: 2
Peak Infinity Fabric Link Bandwidth: 92 GB/s
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267mm)
Board Width: Double Slot
RAS Support: Yes
OpenGL: 4.6
OpenCL: 2.0
ROCm Open Ecosystem: <p><a href="https://www.amd.com/en/graphics/servers-solutions-rocm-hpc">ROCm Open Ecosystem</a></p>
Vulkan: 1.0
OS Support: Linux x86 64-Bit
— 공식 사양 수치 확인
— — — — — AMD Instinct MI60 AMD Instinct Instinct MI Series Servers 11/18/2018 Tidak berlaku 4,096 64 Tidak berlaku — — 1,800 — — Vega20 — — 1800 MHz 32 GB 32 HBM2 4,096 4096-bit 1,000 1 TB/s — — — — — 1x6-Pin + 1x8-Pin
PCIe® 4.0 x16 , PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 4,096
64
— — — — TSMC 7nm FinFET
— — — 1 GHz
Yes (Full-Chip)
— — — 300W Peak
— — — — — 10.5" (267mm)
Full Height
— Passive
14.7 TFLOPs
29.5 TFLOPs
7.4 TFLOPs
— — — — — — — — 4.6
2.0
1.0
Linux x86 64-Bit
— Name: Instinct™ MI60
Family: Instinct
Series: Instinct MI Series
Form Factor: Servers
Launch Date: 11/18/2018
GPU Architecture: Vega20
Lithography: TSMC 7nm FinFET
Stream Processors: 4,096
Compute Units: 64
Peak Engine Clock: 1800 MHz
Peak Half Precision (FP16) Performance: 29.5 TFLOPs
Peak Single Precision (FP32) Performance: 14.7 TFLOPs
Peak Double Precision (FP64) Performance: 7.4 TFLOPs
Peak INT8 Performance: 59 TOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Thermal Design Power (TDP): 300W Peak
Dedicated Memory Size: 32 GB
Dedicated Memory Type: HBM2
Memory Interface: 4096-bit
Memory Clock: 1 GHz
Peak Memory Bandwidth: 1 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16 , PCIe® 3.0 x16
Infinity Fabric Links: 2
Peak Infinity Fabric Link Bandwidth: 92 GB/s
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267mm)
Board Width: Double Slot
RAS Support: Yes
OpenGL: 4.6
OpenCL: 2.0
ROCm Open Ecosystem: <p><a href="https://rocm.github.io/" target="_blank">yes</a></p>
Vulkan: 1.0
OS Support: Linux x86 64-Bit
— 공식 사양 수치 확인
— — — — — AMD Instinct MI100 AMD Instinct Instinct MI100 Series Servers 11/16/2020 Tidak berlaku 7,680 120 Tidak berlaku — — 1,502 — — CDNA — — 1502 MHz 32 GB 32 HBM2 4,096 4096-bit 1,200 1.2 TB/s — — — — — PCIe Powered
PCIe® 4.0 x16 , PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 7,680
120
— — — — TSMC 7nm FinFET
— — — 1.2 GHz
Yes (Full-Chip)
— — — 300W Peak
— — — — — 10.5" (267mm)
Full Height
— Passive
23.1 TFLOPs
184.6 TFLOPs
11.5 TFLOPs
— — — — — — — — — — — — AMD Infinity Architecture , AMD CDNA™ Architecture , AMD ROCm™ - Ecosystem without Borders
Name: AMD Instinct™ MI100
Family: Instinct
Series: Instinct MI100 Series
Form Factor: Servers
Launch Date: 11/16/2020
GPU Architecture: CDNA
Lithography: TSMC 7nm FinFET
Stream Processors: 7,680
Compute Units: 120
Peak Engine Clock: 1502 MHz
Peak Half Precision (FP16) Performance: 184.6 TFLOPs
Peak Single Precision Matrix (FP32) Performance: 23.1 TFLOPs
Peak Single Precision (FP32) Performance: 23.1 TFLOPs
Peak Double Precision (FP64) Performance: 11.5 TFLOPs
Peak INT4 Performance: 92.3 TOPs
Peak INT8 Performance: 92.3 TOPs
Peak bfloat16: 92.3 TFLOPs
External Power Connectors: PCIe Powered
Thermal Design Power (TDP): 300W Peak
Dedicated Memory Size: 32 GB
Dedicated Memory Type: HBM2
Memory Interface: 4096-bit
Memory Clock: 1.2 GHz
Peak Memory Bandwidth: 1.2 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16 , PCIe® 3.0 x16
Infinity Fabric Links: 3
Peak Infinity Fabric Link Bandwidth: 92 GB/s
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267mm)
Board Width: Double Slot
RAS Support: Yes
Page Retirement: Yes
Supported Technologies: AMD Infinity Architecture , AMD CDNA™ Architecture , AMD ROCm™ - Ecosystem without Borders
— 공식 사양 수치 확인
— — — — — AMD Instinct MI210 AMD Instinct Instinct MI200 Series Servers 3/22/22 Tidak berlaku 6,656 104 Tidak berlaku — — 1,700 — — CDNA2 — — 1700 MHz 64 GB 64 HBM2e 4,096 4096-bit 1,600 1.6 TB/s — — — — — — PCIe® 4.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 6,656
104
— — — — — — — — 1.6 GHz
Yes (Full-Chip)
— — — — — — — — — 10.5" (267mm)
Full Height
— Passive
22.6 TFLOPs
181 TFLOPs
22.6 TFLOPs
— — — — — — — — — — — — AMD CDNA™ 2 Architecture , AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders
Name: AMD Instinct™ MI210
Family: Instinct
Series: Instinct MI200 Series
Form Factor: Servers
Launch Date: 3/22/22
GPU Architecture: CDNA2
Stream Processors: 6,656
Compute Units: 104
Peak Engine Clock: 1700 MHz
Peak Half Precision (FP16) Performance: 181 TFLOPs
Peak Single Precision Matrix (FP32) Performance: 45.3 TFLOPs
Peak Single Precision (FP32) Performance: 22.6 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 45.3 TFLOPs
Peak Double Precision (FP64) Performance: 22.6 TFLOPs
Peak INT4 Performance: 181 TOPs
Peak INT8 Performance: 181 TOPs
Peak bfloat16: 181 TFLOPs
Dedicated Memory Size: 64 GB
Dedicated Memory Type: HBM2e
Memory Interface: 4096-bit
Memory Clock: 1.6 GHz
Peak Memory Bandwidth: 1.6 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16
Infinity Fabric Links: 3
Peak Infinity Fabric Link Bandwidth: 100 GB/s
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267mm)
Board Width: Double Slot
RAS Support: Yes
Page Retirement: Yes
Supported Technologies: AMD CDNA™ 2 Architecture , AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders
— 공식 사양 수치 확인
— — — — — AMD Instinct MI250 AMD Instinct Instinct MI200 Series Servers 11/08/2021 Tidak berlaku 13,312 208 Tidak berlaku — — 1,700 — — CDNA2 — — 1700 MHz 128 GB 128 HBM2e 8,192 8192-bit 3,200 3.2 TB/s — — — — — — PCIe® 4.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 13,312
208
— — — — TSMC 6nm FinFET
— — — 1.6 GHz
Yes (Full-Chip)
— — — 500W | 560W Peak
— — — — — — — — Passive OAM
45.3 TFLOPs
362.1 TFLOPs
45.3 TFLOPs
— — — — — — — — — — — — AMD CDNA™ 2 Architecture , AMD ROCm™ - Ecosystem without Borders , AMD Infinity Architecture
Name: AMD Instinct™ MI250
Family: Instinct
Series: Instinct MI200 Series
Form Factor: Servers
Launch Date: 11/08/2021
GPU Architecture: CDNA2
Lithography: TSMC 6nm FinFET
Stream Processors: 13,312
Compute Units: 208
Peak Engine Clock: 1700 MHz
Peak Half Precision (FP16) Performance: 362.1 TFLOPs
Peak Single Precision Matrix (FP32) Performance: 90.5 TFLOPs
Peak Single Precision (FP32) Performance: 45.3 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 90.5 TFLOPs
Peak Double Precision (FP64) Performance: 45.3 TFLOPs
Peak INT4 Performance: 362.1 TOPs
Peak INT8 Performance: 362.1 TOPs
Peak bfloat16: 362.1 TFLOPs
Thermal Design Power (TDP): 500W | 560W Peak
Dedicated Memory Size: 128 GB
Dedicated Memory Type: HBM2e
Memory Interface: 8192-bit
Memory Clock: 1.6 GHz
Peak Memory Bandwidth: 3.2 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: OAM Module
Bus Type: PCIe® 4.0 x16
Infinity Fabric Links: 8
Peak Infinity Fabric Link Bandwidth: 100 GB/s
Cooling: Passive OAM
RAS Support: Yes
Page Retirement: Yes
Supported Technologies: AMD CDNA™ 2 Architecture , AMD ROCm™ - Ecosystem without Borders , AMD Infinity Architecture
— 공식 사양 수치 확인
— — — — — AMD Instinct MI250X AMD Instinct Instinct MI200 Series Servers 11/08/21 Tidak berlaku 14,080 220 Tidak berlaku — — 1,700 — — CDNA2 — — 1700 MHz 128 GB 128 HBM2e 8,192 8192-bit 3,200 3.2 TB/s — — — — — — PCIe® 4.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 14080
220
— — — — TSMC 6nm FinFET
— — — 1.6 GHz
Yes (Full-Chip)
— — — 500W | 560W Peak
— — — — — — — — Passive OAM
47.9 TFLOPs
383 TFLOPs
47.9 TFLOPs
— — — — — — — — — — — — AMD CDNA™ 2 Architecture , AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders
Name: AMD Instinct™ MI250X
Family: Instinct
Series: Instinct MI200 Series
Form Factor: Servers
Launch Date: 11/08/21
GPU Architecture: CDNA2
Lithography: TSMC 6nm FinFET
Stream Processors: 14080
Compute Units: 220
Peak Engine Clock: 1700 MHz
Peak Half Precision (FP16) Performance: 383 TFLOPs
Peak Single Precision Matrix (FP32) Performance: 95.7 TFLOPs
Peak Single Precision (FP32) Performance: 47.9 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 95.7 TFLOPs
Peak Double Precision (FP64) Performance: 47.9 TFLOPs
Peak INT4 Performance: 383 TOPs
Peak INT8 Performance: 383 TOPs
Peak bfloat16: 383 TFLOPs
Thermal Design Power (TDP): 500W | 560W Peak
Dedicated Memory Size: 128 GB
Dedicated Memory Type: HBM2e
Memory Interface: 8192-bit
Memory Clock: 1.6 GHz
Peak Memory Bandwidth: 3.2 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: OAM Module
Bus Type: PCIe® 4.0 x16
Infinity Fabric Links: 8
Peak Infinity Fabric Link Bandwidth: 100 GB/s
Cooling: Passive OAM
RAS Support: Yes
Page Retirement: Yes
Supported Technologies: AMD CDNA™ 2 Architecture , AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders
— 공식 사양 수치 확인
— — — — — AMD Instinct MI300A AMD Instinct Instinct MI300 Series Servers 12/06/2023 Tidak berlaku 14,592 228 Tidak berlaku — — 2,100 — — CDNA3 — — 2100 MHz 128 GB 128 HBM3 8,192 8192-bit 5,300 5.3 TB/s — — — — — — PCIe® 5.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 14,592
228
— — — — TSMC 5nm | 6nm FinFET
146 Billion
— — 5.2 GHz
Yes (Full-Chip)
Yes
— — 550W | 760W Peak
— — — — — — — — Passive & Active
122.6 TFLOPs
980.6 TFLOPs
61.3 TFLOPs
— — — — — — — — — — — — AMD CDNA™ 3 Architecture , AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders
Name: AMD Instinct™ MI300A
Family: Instinct
Series: Instinct MI300 Series
Form Factor: Servers
Launch Date: 12/06/2023
GPU Architecture: CDNA3
Lithography: TSMC 5nm | 6nm FinFET
Stream Processors: 14,592
Matrix Cores: 912
Compute Units: 228
Peak Engine Clock: 2100 MHz
Peak Eight-bit Precision (FP8) Performance (E5M2, E4M3): 1.96 PFLOPs
Peak Eight-bit Precision (FP8) Performance with Structured Sparsity (E5M2, E4M3): 3.92 PFLOPs
Peak Half Precision (FP16) Performance: 980.6 TFLOPs
Peak Half Precision (FP16) Performance with Structured Sparsity: 1.96 PFLOPs
Peak Single Precision (TF32 Matrix) Performance: 490.3 TFLOPs
Peak Single Precision (TF32) Performance with Structured Sparsity: 980.6 TFLOPs
Peak Single Precision Matrix (FP32) Performance: 122.6 TFLOPs
Peak Single Precision (FP32) Performance: 122.6 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 122.6 TFLOPs
Peak Double Precision (FP64) Performance: 61.3 TFLOPs
Peak INT8 Performance: 1.96 POPs
Peak INT8 Performance with Structured Sparsity: 3.92 POPs
Peak bfloat16: 980.6 TFLOPs
Peak bfloat16 with Strutured Sparsity: 1.96 PFLOPs
Transistor Count: 146 Billion
AMD EPYC CPU Architecture: Zen 4
CPU Cores: 24
CPU Peak Engine Clock: 3700 MHz
Thermal Design Power (TDP): 550W | 760W Peak
Last Level Cache (LLC): 256 MB
Dedicated Memory Size: 128 GB
Dedicated Memory Type: HBM3
Infinity Cache: Yes
Memory Interface: 8192-bit
Memory Clock: 5.2 GHz
Peak Memory Bandwidth: 5.3 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: APU SH5 Socket
Bus Type: PCIe® 5.0 x16
Infinity Fabric Links: 8
Peak Infinity Fabric Link Bandwidth: 128 GB/s
Cooling: Passive & Active
RAS Support: Yes
Page Retirement: Yes
Page Avoidance: Yes
SR-IOV: Yes
Supported Technologies: AMD CDNA™ 3 Architecture , AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders
— 공식 사양 수치 확인
— — — — — AMD Instinct MI300X AMD Instinct Instinct MI300 Series Servers 12/06/2023 Tidak berlaku 19,456 304 Tidak berlaku — — 2,100 — — CDNA3 — — 2100 MHz 192 GB 192 HBM3 8,192 8192-bit 5,300 5.3 TB/s — 750W Peak — — — 54V UBB
PCIe® 5.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 19,456
304
— — — — TSMC 5nm | 6nm FinFET
153 Billion
— — 5.2 GHz
Yes (Full-Chip)
Yes
— — — — — — — — — — — Passive OAM
163.4 TFLOPs
1.3 PFLOPs
81.7 TFLOPs
— — — — — — — — — — — — AMD CDNA™ 3 Architecture , AMD ROCm™ - Ecosystem without Borders , AMD Infinity Architecture
Name: AMD Instinct™ MI300X
Family: Instinct
Series: Instinct MI300 Series
Form Factor: Servers
Launch Date: 12/06/2023
GPU Architecture: CDNA3
Lithography: TSMC 5nm | 6nm FinFET
Stream Processors: 19,456
Matrix Cores: 1216
Compute Units: 304
Peak Engine Clock: 2100 MHz
Peak Eight-bit Precision (FP8) Performance (E5M2, E4M3): 2.61 PFLOPs
Peak Eight-bit Precision (FP8) Performance with Structured Sparsity (E5M2, E4M3): 5.22 PFLOPs
Peak Half Precision (FP16) Performance: 1.3 PFLOPs
Peak Half Precision (FP16) Performance with Structured Sparsity: 2.61 PFLOPs
Peak Single Precision (TF32 Matrix) Performance: 653.7 TFLOPs
Peak Single Precision (TF32) Performance with Structured Sparsity: 1.3 PFLOPs
Peak Single Precision Matrix (FP32) Performance: 163.4 TFLOPs
Peak Single Precision (FP32) Performance: 163.4 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 163.4 TFLOPs
Peak Double Precision (FP64) Performance: 81.7 TFLOPs
Peak INT8 Performance: 2.6 POPs
Peak INT8 Performance with Structured Sparsity: 5.22 POPs
Peak bfloat16: 1.3 PFLOPs
Peak bfloat16 with Strutured Sparsity: 2.61 PFLOPs
Transistor Count: 153 Billion
External Power Connectors: 54V UBB
Typical Board Power (TBP): 750W Peak
Last Level Cache (LLC): 256 MB
Dedicated Memory Size: 192 GB
Dedicated Memory Type: HBM3
Infinity Cache: Yes
Memory Interface: 8192-bit
Memory Clock: 5.2 GHz
Peak Memory Bandwidth: 5.3 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: OAM Module
Bus Type: PCIe® 5.0 x16
Infinity Fabric Links: 8
Peak Infinity Fabric Link Bandwidth: 128 GB/s
Cooling: Passive OAM
RAS Support: Yes
Page Retirement: Yes
Page Avoidance: Yes
SR-IOV: Yes
Supported Technologies: AMD CDNA™ 3 Architecture , AMD ROCm™ - Ecosystem without Borders , AMD Infinity Architecture
— 공식 사양 수치 확인
— — — — — AMD Instinct MI325X AMD Instinct Instinct MI300 Series Servers 10/10/2024 Tidak berlaku 19,456 304 Tidak berlaku — — 2,100 — — CDNA3 — — 2100 MHz 256 GB 256 HBM3E 8,192 8192-bit 6,000 6 TB/s — 1000W Peak — — — 54V UBB
PCIe® 5.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 19,456
304
— — — — TSMC 5nm | 6nm FinFET
153 Billion
— — 6 GHz
Yes (Full-Chip)
Yes
— — — — — — — — — — — Passive OAM
163.4 TFLOPs
1.3 PFLOPs
81.7 TFLOPs
— — — — — — — — — — — — AMD CDNA™ 3 Architecture , AMD ROCm™ - Ecosystem without Borders , AMD Infinity Architecture
Name: AMD Instinct™ MI325X
Family: Instinct
Series: Instinct MI300 Series
Form Factor: Servers
Launch Date: 10/10/2024
GPU Architecture: CDNA3
Lithography: TSMC 5nm | 6nm FinFET
Stream Processors: 19,456
Matrix Cores: 1216
Compute Units: 304
Peak Engine Clock: 2100 MHz
Peak Eight-bit Precision (FP8) Performance (E5M2, E4M3): 2.61 PFLOPs
Peak Eight-bit Precision (FP8) Performance with Structured Sparsity (E5M2, E4M3): 5.22 PFLOPs
Peak Half Precision (FP16) Performance: 1.3 PFLOPs
Peak Half Precision (FP16) Performance with Structured Sparsity: 2.61 PFLOPs
Peak Single Precision (TF32 Matrix) Performance: 653.7 TFLOPs
Peak Single Precision (TF32) Performance with Structured Sparsity: 1.3 PFLOPs
Peak Single Precision Matrix (FP32) Performance: 163.4 TFLOPs
Peak Single Precision (FP32) Performance: 163.4 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 163.4 TFLOPs
Peak Double Precision (FP64) Performance: 81.7 TFLOPs
Peak INT8 Performance: 2.6 POPs
Peak INT8 Performance with Structured Sparsity: 5.22 POPs
Peak bfloat16: 1.3 PFLOPs
Peak bfloat16 with Strutured Sparsity: 2.61 PFLOPs
Transistor Count: 153 Billion
External Power Connectors: 54V UBB
Typical Board Power (TBP): 1000W Peak
Last Level Cache (LLC): 256 MB
Dedicated Memory Size: 256 GB
Dedicated Memory Type: HBM3E
Infinity Cache: Yes
Memory Interface: 8192-bit
Memory Clock: 6 GHz
Peak Memory Bandwidth: 6 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: OAM Module
Bus Type: PCIe® 5.0 x16
Infinity Fabric Links: 8
Peak Infinity Fabric Link Bandwidth: 128 GB/s
Cooling: Passive OAM
RAS Support: Yes
Page Retirement: Yes
Page Avoidance: Yes
SR-IOV: Yes
Supported Technologies: AMD CDNA™ 3 Architecture , AMD ROCm™ - Ecosystem without Borders , AMD Infinity Architecture
— 공식 사양 수치 확인
— — — — — AMD Instinct MI350P AMD Instinct Instinct MI350 Series Servers — Tidak berlaku 8,192 128 Tidak berlaku — — 2,200 — — CDNA4 — — 2200 MHz 144 GB 144 HBM3E 4,096 4096-bit 4,000 4 TB/s — 600W TBP (Max); 450W TBP configurable — — — 12V-2x6
PCIe® 5.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 8,192
128
— — — — TSMC 3nm | 6nm FinFET
73 Billion
— — — Yes (Full-Chip)
Yes
— — — — — — — — 10.5" (267 mm)
Full Height
— Passive
72 TFLOPs
72 TFLOPs
36 TFLOPs
— — — — — — — — — Yes
— Linux x86 64-Bit
AMD CDNA™ 4 Architecture , 4th Gen AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders
Name: AMD Instinct™ MI350P
Family: Instinct
Series: Instinct MI350 Series
Form Factor: Servers
GPU Architecture: CDNA4
Lithography: TSMC 3nm | 6nm FinFET
Stream Processors: 8,192
Matrix Cores: 512
Compute Units: 128
Peak Engine Clock: 2200 MHz
Peak Microscaling Four-bit Precision Matrix (MXFP4) Performance: 4.6 PFLOPs
Peak Microscaling Six-bit Precision Matrix (MXFP6) Performance: 4.6 PFLOPs
Peak Microscaling Eight-bit Precision Matrix (MXFP8) Performance: 2.3 PFLOPs
Peak Open Compute Project Eight-bit Precision Matrix (OCP-FP8) Performance (E5M2, E4M3): 2.3 PFLOPs
Peak Open Compute Project Eight-bit Precision Matrix (OCP-FP8) Performance with Structured Sparsity (E5M2, E4M3): 4.6 PFLOPs
Peak Half Precision Matrix (FP16) Performance: 1.15 PFLOPs
Peak Half Precision Matrix (FP16) Performance with Structured Sparsity: 2.3 PFLOPs
Peak Half Precision (FP16) Performance: 72 TFLOPs
Peak Single Precision Matrix (FP32) Performance: 72 TFLOPs
Peak Single Precision (FP32) Performance: 72 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 36 TFLOPs
Peak Double Precision (FP64) Performance: 36 TFLOPs
Peak INT8 Matrix Performance: 2.3 POPs
Peak INT8 Matrix Performance with Structured Sparsity: 4.6 POPs
Peak bfloat16 (BF16) Matrix Performance: 1.15 PFLOPs
Peak bfloat16 (BF16) Matrix Performance with Structured Sparsity: 2.3 PFLOPs
Transistor Count: 73 Billion
External Power Connectors: 12V-2x6
Typical Board Power (TBP): 600W TBP (Max); 450W TBP configurable
Last Level Cache (LLC): 128 MB
Dedicated Memory Size: 144 GB
Dedicated Memory Type: HBM3E
Infinity Cache: Yes
Memory Interface: 4096-bit
Peak Memory Bandwidth: 4 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 5.0 x16
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
RAS Support: Yes
Page Retirement: Yes
Page Avoidance: Yes
SR-IOV: Yes
OpenMP: Yes
OpenCL: Yes
HIP: Yes
ROCm Open Ecosystem: <p>Yes</p>
TensorFlow: Yes
PyTorch: Yes
ONYX-RT: Yes
SGLang: Yes
JAX: Yes
Triton: Yes
Kokkos: Yes
RAJA: Yes
OS Support: Linux x86 64-Bit
Supported Technologies: AMD CDNA™ 4 Architecture , 4th Gen AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders
— 공식 사양 수치 확인
— — — — — AMD Instinct MI350X AMD Instinct Instinct MI350 Series Servers 6/12/2025 Tidak berlaku 16,384 256 Tidak berlaku — — 2,200 — — CDNA4 — — 2200 MHz 288 GB 288 HBM3E 8,192 8192-bit 8,000 8 TB/s — 1000W (TBP) — — — 54V UBB
PCIe® 5.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 16,384
256
— — — — TSMC 3nm | 6nm FinFET
185 Billion
— — 8 GHz
Yes (Full-Chip)
Yes
— — — — — — — — — — — Passive OAM
144.2 TFLOPs
144.2 TFLOPs
72.1 TFLOPs
— — — — — — — — — — — — 4th Gen AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders , AMD CDNA™ 4 Architecture
Name: AMD Instinct™ MI350X
Family: Instinct
Series: Instinct MI350 Series
Form Factor: Servers
Launch Date: 6/12/2025
GPU Architecture: CDNA4
Lithography: TSMC 3nm | 6nm FinFET
Stream Processors: 16,384
Matrix Cores: 1024
Compute Units: 256
Peak Engine Clock: 2200 MHz
Peak Microscaling Four-bit Precision Matrix (MXFP4) Performance: 9.2 PFLOPs
Peak Microscaling Six-bit Precision Matrix (MXFP6) Performance: 9.2 PFLOPs
Peak Microscaling Eight-bit Precision Matrix (MXFP8) Performance: 4.6 PFLOPs
Peak Open Compute Project Eight-bit Precision Matrix (OCP-FP8) Performance (E5M2, E4M3): 4.6 PFLOPs
Peak Open Compute Project Eight-bit Precision Matrix (OCP-FP8) Performance with Structured Sparsity (E5M2, E4M3): 9.2 PFLOPs
Peak Half Precision Matrix (FP16) Performance: 2.3 PFLOPs
Peak Half Precision Matrix (FP16) Performance with Structured Sparsity: 4.6 PFLOPs
Peak Half Precision (FP16) Performance: 144.2 TFLOPs
Peak Single Precision Matrix (FP32) Performance: 144.2 TFLOPs
Peak Single Precision (FP32) Performance: 144.2 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 72.1 TFLOPs
Peak Double Precision (FP64) Performance: 72.1 TFLOPs
Peak INT8 Matrix Performance: 4.6 POPs
Peak INT8 Matrix Performance with Structured Sparsity: 9.2 POPs
Peak bfloat16 (BF16) Matrix Performance: 2.3 PFLOPs
Peak bfloat16 (BF16) Matrix Performance with Structured Sparsity: 4.6 PFLOPs
Transistor Count: 185 Billion
External Power Connectors: 54V UBB
Typical Board Power (TBP): 1000W (TBP)
Last Level Cache (LLC): 256 MB
Dedicated Memory Size: 288 GB
Dedicated Memory Type: HBM3E
Infinity Cache: Yes
Memory Interface: 8192-bit
Memory Clock: 8 GHz
Peak Memory Bandwidth: 8 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: OAM Module
Bus Type: PCIe® 5.0 x16
Infinity Fabric Links: 7
Peak Infinity Fabric Link Bandwidth: 153 GB/s
Scale-out AMD Infinity Fabric Links (Per GPU): 1
Scale-up (Peak) AMD Infinity Fabric Link Bandwidth: 153 GB/s
Scale-out (Peak) AMD Infinity Fabric Link Bandwidth: 128 GB/s
Cooling: Passive OAM
RAS Support: Yes
Page Retirement: Yes
Page Avoidance: Yes
SR-IOV: Yes
Supported Technologies: 4th Gen AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders , AMD CDNA™ 4 Architecture
— 공식 사양 수치 확인
— — — — — AMD Instinct MI355X AMD Instinct Instinct MI350 Series Servers 6/12/2025 Tidak berlaku 16,384 256 Tidak berlaku — — 2,400 — — CDNA4 — — 2400 MHz 288 GB 288 HBM3E — — 8,000 8 TB/s — 1400W (TBP) — — — — PCIe® 5.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 16,384
256
— — — — TSMC 3nm | 6nm FinFET
185 Billion
— — — Yes (Full-Chip)
Yes
— — — — — — — — — — — Passive & Active
157.3 TFLOPs
157.3 TFLOPs
78.6 TFLOPs
— — — — — — — — — — — — 4th Gen AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders , AMD CDNA™ 4 Architecture
Name: AMD Instinct™ MI355X
Family: Instinct
Series: Instinct MI350 Series
Form Factor: Servers
Launch Date: 6/12/2025
GPU Architecture: CDNA4
Lithography: TSMC 3nm | 6nm FinFET
Stream Processors: 16,384
Matrix Cores: 1024
Compute Units: 256
Peak Engine Clock: 2400 MHz
Peak Microscaling Four-bit Precision Matrix (MXFP4) Performance: 10.1 PFLOPs
Peak Microscaling Six-bit Precision Matrix (MXFP6) Performance: 10.1 PFLOPs
Peak Microscaling Eight-bit Precision Matrix (MXFP8) Performance: 5 PFLOPs
Peak Open Compute Project Eight-bit Precision Matrix (OCP-FP8) Performance (E5M2, E4M3): 5 PFLOPs
Peak Open Compute Project Eight-bit Precision Matrix (OCP-FP8) Performance with Structured Sparsity (E5M2, E4M3): 10.1 PFLOPs
Peak Half Precision Matrix (FP16) Performance: 2.5 PFLOPs
Peak Half Precision Matrix (FP16) Performance with Structured Sparsity: 5 PFLOPs
Peak Half Precision (FP16) Performance: 157.3 TFLOPs
Peak Single Precision Matrix (FP32) Performance: 157.3 TFLOPs
Peak Single Precision (FP32) Performance: 157.3 TFLOPs
Peak Double Precision Matrix (FP64) Performance: 78.6 TFLOPs
Peak Double Precision (FP64) Performance: 78.6 TFLOPs
Peak INT8 Matrix Performance: 5 POPs
Peak INT8 Matrix Performance with Structured Sparsity: 10.1 POPs
Peak bfloat16 (BF16) Matrix Performance: 2.5 PFLOPs
Peak bfloat16 (BF16) Matrix Performance with Structured Sparsity: 5 PFLOPs
Transistor Count: 185 Billion
Typical Board Power (TBP): 1400W (TBP)
Last Level Cache (LLC): 256 MB
Dedicated Memory Size: 288 GB
Dedicated Memory Type: HBM3E
Infinity Cache: Yes
Peak Memory Bandwidth: 8 TB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: OAM Module
Bus Type: PCIe® 5.0 x16
Infinity Fabric Links: 7
Peak Infinity Fabric Link Bandwidth: 153 GB/s
Scale-out AMD Infinity Fabric Links (Per GPU): 1
Scale-up (Peak) AMD Infinity Fabric Link Bandwidth: 153 GB/s
Scale-out (Peak) AMD Infinity Fabric Link Bandwidth: 128 GB/s
Cooling: Passive & Active
RAS Support: Yes
Page Retirement: Yes
Page Avoidance: Yes
SR-IOV: Yes
Supported Technologies: 4th Gen AMD Infinity Architecture , AMD ROCm™ - Ecosystem without Borders , AMD CDNA™ 4 Architecture
— 공식 사양 수치 확인
— — — — — AMD Instinct MI455X AMD Instinct Instinct MI400 Series — 7/23/2026 Tidak berlaku — — Tidak berlaku — — 2,400 — — CDNA5 — — 2400 MHz 432 GB 432 HBM4 — — 23,300 23.3 TB/s — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — TSMC 2nm | 3nm FinFET
320 Billion
— — — Yes
— 192 MB
— — — — — — — — — — Direct Liquid Cooling (DLC)
315 TFLOPs
315 TFLOPs
5 TFLOPs
— — — — — — — — — Yes
No
Linux x86 64-Bit
AMD CDNA™ 5 Architecture , AMD ROCm™ - Ecosystem without Borders , UALink™, UALoE
Name: AMD Instinct™ MI455X
Family: Instinct
Series: Instinct MI400 Series
Launch Date: 7/23/2026
GPU Architecture: CDNA5
Lithography: TSMC 2nm | 3nm FinFET
Work Group Processors: 256
Peak Engine Clock: 2400 MHz
Peak OCP MXFP4 Performance: 40.3 PFLOPs
Peak OCP MXFP6 Performance: 20.1 PFLOPs
Peak OCP MXFP8 Performance: 20.1 PFLOPs
Peak OCP FP8 Performance: 20.1 PFLOPs
Peak Vector FP16 Performance: 315 TFLOPs
Peak Matrix FP16 Performance: 5 PFLOPs
Peak Matrix FP16 Performance with Structured Sparsity: 10.1 PFLOPs
Peak Matrix FP32 Performance: 315 TFLOPs
Peak Vector FP32 Performance: 315 TFLOPs
Peak Matrix FP64 Performance: 5 TFLOPs
Peak Vector FP64 Performance: 5 TFLOPs
Peak INT8 Matrix Performance: 5 POPs
Peak INT8 Matrix Performance with Structured Sparsity: 10.1 POPs
Peak bfloat16 (BF16) Matrix Performance: 5 PFLOPs
Peak bfloat16 (BF16) Matrix Performance with Structured Sparsity: 10.1 PFLOPs
Transistor Count: 320 Billion
Dedicated Memory Size: 432 GB
Dedicated Memory Type: HBM4
L2 Cache: 192 MB
Stacks of Memory: 12
Peak Memory Bandwidth: 23.3 TB/s
Memory ECC Support: Yes
GPU Form Factor: Enhanced Accelerator Module (EAM)
Scale-up (Peak) UALoE Bi-directional Bandwidth: 3.6 TB/s
Scale-out (Peak) UALink Bi-directional Bandwidth: 600 GB/s
Scale-out Bus Type: UALink™
Cooling: Direct Liquid Cooling (DLC)
RAS Support: Yes
SR-IOV: Yes
OpenMP: Yes
OpenCL: Yes
HIP: Yes
ROCm Open Ecosystem: <p>Yes</p>
Vulkan: No
TensorFlow: Yes
PyTorch: Yes
ONYX-RT: Yes
SGLang: Yes
JAX: Yes
Triton: Yes
Kokkos: Yes
RAJA: Yes
OS Support: Linux x86 64-Bit
Supported Technologies: AMD CDNA™ 5 Architecture , AMD ROCm™ - Ecosystem without Borders , UALink™, UALoE
— 공식 사양 수치 확인
— — — — — AMD Radeon AI PRO R9600 AMD Radeon AI PRO Radeon AI PRO R9000 Series Desktop — Tidak berlaku 3,072 48 Tidak berlaku 48 — — — — AMD RDNA™ 4 — — — 32 GB 32 GDDR6 256 256-bit 640 640 GB/s 150 150W — 450 450W 12V-2x6
PCIe® 5.0 x16 Configuration varies by partner design. GPU supports up to 4x display out. — — — Not listed — — 단일 모델·변형
— — 96 Tidak berlaku 3072
48
— — — — — 53.9 Billion
— — — Yes (Linux Only)
64 MB
— — — — — — — Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC, 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed, 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC, 1x 12288 x 6912 (12K) @ 120Hz with DSC
— — — Active
24.8 TFLOPs
24.8 TFLOPs
— — — — — — — — — — — — Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit
AMD Software Adrenalin Edition , AMD Fluid Motion Frames , AMD Radeon™ Super Resolution , AMD Smart Access Memory , AMD Smart Access Video , AMD Radeon™ Boost , AMD Radeon™ Anti-Lag , AMD Radeon™ Image Sharpening 2 , AMD Enhanced Sync Technology , AMD FreeSync™ Technology , AMD Radeon™ Chill , AMD Virtual Super Resolution , AMD FSR™
Name: AMD Radeon™ AI PRO R9600
Family: Radeon AI PRO
Series: Radeon AI PRO R9000 Series
Form Factor: Desktops
GPU Architecture: AMD RDNA™ 4
Ray Accelerators: 48
AI Accelerators: 96
Stream Processors: 3072
Compute Units: 48
Peak Half Precision (FP16 Vector) Performance: 24.8 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 24.8 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance: 397 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance: 199 TOPs
Peak Half Precision (FP16 Matrix) Performance with Structured Sparsity: 199 TFLOPs
Peak Half Precision (FP16 Matrix) Performance: 99 TFLOPs
Peak 8-bit Precision (FP8 Matrix) Performance with Structured Sparsity (E5M2, E4M3): 397 TFLOPs
Peak 8-bit Precision (FP8 Matrix) Performance (E5M2, E4M3): 199 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance with Structured Sparsity: 794 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance with Structured Sparsity: 397 TOPs
Transistor Count: 53.9 Billion
External Power Connectors: 12V-2x6
Total Board Power (TBP): 150W
Minimum PSU Recommendation: 450W
Dedicated Memory Size: 32 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 64 MB
Memory Interface: 256-bit
Peak Memory Bandwidth: 640 GB/s
Memory ECC Support: Yes (Linux Only)
Bus Type: PCIe® 5.0 x16
Cooling: Active
DisplayPort: Configuration varies by partner design. GPU supports up to 4x display out.
Display Configurations: Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC, 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed, 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC, 1x 12288 x 6912 (12K) @ 120Hz with DSC
HDR Support: Yes
8K Support: Yes
10K Support: Yes
12K Support: Yes
Board Width: Single Slot
OS Support: Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit
Supported Video Formats: H264 Encode and Decode , H265∕HEVC Encode and Decode , AV1 Encode & Decode
Supported Technologies: AMD Software Adrenalin Edition , AMD Fluid Motion Frames , AMD Radeon™ Super Resolution , AMD Smart Access Memory , AMD Smart Access Video , AMD Radeon™ Boost , AMD Radeon™ Anti-Lag , AMD Radeon™ Image Sharpening 2 , AMD Enhanced Sync Technology , AMD FreeSync™ Technology , AMD Radeon™ Chill , AMD Virtual Super Resolution , AMD FSR™
TPP: 1589
TPP Density: 4.62
— 공식 사양 수치 확인
— 48
— 96
— AMD Radeon AI PRO R9600D AMD Radeon AI PRO Radeon AI PRO R9000 Series Desktop — Tidak berlaku 3,072 48 Tidak berlaku 48 — — — — AMD RDNA™ 4 — — — 32 GB 32 GDDR6 256 256-bit 640 640 GB/s 150 150W — 450 450W 12V-2x6
PCIe® 5.0 x16 Configuration varies by partner design. GPU supports up to 4x display out. — — — Not listed — — 단일 모델·변형
— — 96 Tidak berlaku 3072
48
— — — — — 53.9 Billion
— — — Yes (Linux Only)
64 MB
— — — — — — — Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC, 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed, 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC, 1x 12288 x 6912 (12K) @ 120Hz with DSC
— — — Passive
24.8 TFLOPs
24.8 TFLOPs
— — — — — — — — — — — — Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit
AMD Software Adrenalin Edition , AMD Fluid Motion Frames , AMD Radeon™ Super Resolution , AMD Smart Access Memory , AMD Smart Access Video , AMD Radeon™ Boost , AMD Radeon™ Anti-Lag , AMD Radeon™ Image Sharpening 2 , AMD Enhanced Sync Technology , AMD FreeSync™ Technology , AMD Radeon™ Chill , AMD Virtual Super Resolution , AMD FSR™
Name: AMD Radeon™ AI PRO R9600D
Family: Radeon AI PRO
Series: Radeon AI PRO R9000 Series
Form Factor: Desktops
GPU Architecture: AMD RDNA™ 4
Ray Accelerators: 48
AI Accelerators: 96
Stream Processors: 3072
Compute Units: 48
Peak Half Precision (FP16 Vector) Performance: 24.8 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 24.8 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance: 397 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance: 199 TOPs
Peak Half Precision (FP16 Matrix) Performance with Structured Sparsity: 199 TFLOPs
Peak Half Precision (FP16 Matrix) Performance: 99 TFLOPs
Peak 8-bit Precision (FP8 Matrix) Performance with Structured Sparsity (E5M2, E4M3): 397 TFLOPs
Peak 8-bit Precision (FP8 Matrix) Performance (E5M2, E4M3): 199 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance with Structured Sparsity: 794 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance with Structured Sparsity: 397 TOPs
Transistor Count: 53.9 Billion
External Power Connectors: 12V-2x6
Total Board Power (TBP): 150W
Minimum PSU Recommendation: 450W
Dedicated Memory Size: 32 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 64 MB
Memory Interface: 256-bit
Peak Memory Bandwidth: 640 GB/s
Memory ECC Support: Yes (Linux Only)
Bus Type: PCIe® 5.0 x16
Cooling: Passive
DisplayPort: Configuration varies by partner design. GPU supports up to 4x display out.
Display Configurations: Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC, 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed, 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC, 1x 12288 x 6912 (12K) @ 120Hz with DSC
HDR Support: Yes
8K Support: Yes
10K Support: Yes
12K Support: Yes
Board Width: Single Slot
OS Support: Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit
Supported Video Formats: H264 Encode and Decode , H265∕HEVC Encode and Decode , AV1 Encode & Decode
Supported Technologies: AMD Software Adrenalin Edition , AMD Fluid Motion Frames , AMD Radeon™ Super Resolution , AMD Smart Access Memory , AMD Smart Access Video , AMD Radeon™ Boost , AMD Radeon™ Anti-Lag , AMD Radeon™ Image Sharpening 2 , AMD Enhanced Sync Technology , AMD FreeSync™ Technology , AMD Radeon™ Chill , AMD Virtual Super Resolution , AMD FSR™
TPP: 1589
TPP Density: 4.62
— 공식 사양 수치 확인
— 48
— 96
— AMD Radeon AI PRO R9700 AMD Radeon AI PRO Radeon AI PRO R9000 Series Desktop — Tidak berlaku 4,096 64 Tidak berlaku 64 — — — — AMD RDNA™ 4 — — — 32 GB 32 GDDR6 256 256-bit 640 640 GB/s 300 300W — 750 750W 12V-2x6
PCIe® 5.0 x16 Configuration varies by partner design. GPU supports up to 4x display out. — — — Not listed — — 단일 모델·변형
— — 128 Tidak berlaku 4096
64
— — — — — 53.9 Billion
— — — Yes (Linux Only)
64 MB
— — — — — — — Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC, 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed, 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC, 1x 12288 x 6912 (12K) @ 120Hz with DSC
— — — Active
47.8 TFLOPs
47.8 TFLOPs
— — — — — — — — — — — — Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit
AMD Software Adrenalin Edition , AMD FSR™ "Redstone" , AMD Fluid Motion Frames , AMD Radeon™ Super Resolution , AMD Smart Access Memory , AMD Smart Access Video , AMD Radeon™ Boost , AMD Radeon™ Anti-Lag , AMD Radeon™ Image Sharpening 2 , AMD Enhanced Sync Technology , AMD FreeSync™ Technology , AMD Radeon™ Chill , AMD Virtual Super Resolution
Name: AMD Radeon™ AI PRO R9700
Family: Radeon AI PRO
Series: Radeon AI PRO R9000 Series
Form Factor: Desktops
GPU Architecture: AMD RDNA™ 4
Ray Accelerators: 64
AI Accelerators: 128
Stream Processors: 4096
Compute Units: 64
Peak Half Precision (FP16 Vector) Performance: 47.8 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 47.8 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance: 766 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance: 383 TOPs
Peak Half Precision (FP16 Matrix) Performance with Structured Sparsity: 383 TFLOPs
Peak Half Precision (FP16 Matrix) Performance: 191 TFLOPs
Peak 8-bit Precision (FP8 Matrix) Performance with Structured Sparsity (E5M2, E4M3): 766 TFLOPs
Peak 8-bit Precision (FP8 Matrix) Performance (E5M2, E4M3): 383 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance with Structured Sparsity: 1531 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance with Structured Sparsity: 766 TOPs
Transistor Count: 53.9 Billion
External Power Connectors: 12V-2x6
Total Board Power (TBP): 300W
Minimum PSU Recommendation: 750W
Dedicated Memory Size: 32 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 64 MB
Memory Interface: 256-bit
Peak Memory Bandwidth: 640 GB/s
Memory ECC Support: Yes (Linux Only)
Bus Type: PCIe® 5.0 x16
Cooling: Active
DisplayPort: Configuration varies by partner design. GPU supports up to 4x display out.
Display Configurations: Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC, 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed, 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC, 1x 12288 x 6912 (12K) @ 120Hz with DSC
HDR Support: Yes
8K Support: Yes
10K Support: Yes
12K Support: Yes
Board Width: Double Slot
OS Support: Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit
Supported Video Formats: H264 Encode and Decode , H265∕HEVC Encode and Decode , AV1 Encode & Decode
Supported Technologies: AMD Software Adrenalin Edition , AMD FSR™ "Redstone" , AMD Fluid Motion Frames , AMD Radeon™ Super Resolution , AMD Smart Access Memory , AMD Smart Access Video , AMD Radeon™ Boost , AMD Radeon™ Anti-Lag , AMD Radeon™ Image Sharpening 2 , AMD Enhanced Sync Technology , AMD FreeSync™ Technology , AMD Radeon™ Chill , AMD Virtual Super Resolution
TPP: 3062
TPP Density: 8.91
— 공식 사양 수치 확인
— 64
— 128
— AMD Radeon AI PRO R9700S AMD Radeon AI PRO Radeon AI PRO R9000 Series Desktop — Tidak berlaku 4,096 64 Tidak berlaku 64 — — — — AMD RDNA™ 4 — — — 32 GB 32 GDDR6 256 256-bit 640 640 GB/s 300 300W — 750 750W 12V-2x6
PCIe® 5.0 x16 Configuration varies by partner design. GPU supports up to 4x display out. — — — Not listed — — 단일 모델·변형
— — 128 Tidak berlaku 4096
64
— — — — — 53.9 Billion
— — — Yes (Linux Only)
64 MB
— — — — — — — Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC, 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed, 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC, 1x 12288 x 6912 (12K) @ 120Hz with DSC
— — — Passive
47.8 TFLOPs
47.8 TFLOPs
— — — — — — — — — — — — Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit
AMD Software Adrenalin Edition , AMD Fluid Motion Frames , AMD Radeon™ Super Resolution , AMD Smart Access Memory , AMD Smart Access Video , AMD Radeon™ Boost , AMD Radeon™ Anti-Lag , AMD Radeon™ Image Sharpening 2 , AMD Enhanced Sync Technology , AMD FreeSync™ Technology , AMD Radeon™ Chill , AMD Virtual Super Resolution , AMD FSR™
Name: AMD Radeon™ AI PRO R9700S
Family: Radeon AI PRO
Series: Radeon AI PRO R9000 Series
Form Factor: Desktops
GPU Architecture: AMD RDNA™ 4
Ray Accelerators: 64
AI Accelerators: 128
Stream Processors: 4096
Compute Units: 64
Peak Half Precision (FP16 Vector) Performance: 47.8 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 47.8 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance: 766 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance: 383 TOPs
Peak Half Precision (FP16 Matrix) Performance with Structured Sparsity: 383 TFLOPs
Peak Half Precision (FP16 Matrix) Performance: 191 TFLOPs
Peak 8-bit Precision (FP8 Matrix) Performance with Structured Sparsity (E5M2, E4M3): 766 TFLOPs
Peak 8-bit Precision (FP8 Matrix) Performance (E5M2, E4M3): 383 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance with Structured Sparsity: 1531 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance with Structured Sparsity: 766 TOPs
Transistor Count: 53.9 Billion
External Power Connectors: 12V-2x6
Total Board Power (TBP): 300W
Minimum PSU Recommendation: 750W
Dedicated Memory Size: 32 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 64 MB
Memory Interface: 256-bit
Peak Memory Bandwidth: 640 GB/s
Memory ECC Support: Yes (Linux Only)
Bus Type: PCIe® 5.0 x16
Cooling: Passive
DisplayPort: Configuration varies by partner design. GPU supports up to 4x display out.
Display Configurations: Up to: 4x 4096 x 2160 (4K DCI) @ 120Hz with DSC, 2x 6144 x 3456 (6K) 12-bit HDR @ 60Hz Uncompressed, 2x 7680 x 4320 (8K) 12-bit HDR @ 60Hz with DSC, 1x 12288 x 6912 (12K) @ 120Hz with DSC
HDR Support: Yes
8K Support: Yes
10K Support: Yes
12K Support: Yes
Board Width: Double Slot
OS Support: Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit
Supported Video Formats: H264 Encode and Decode , H265∕HEVC Encode and Decode , AV1 Encode & Decode
Supported Technologies: AMD Software Adrenalin Edition , AMD Fluid Motion Frames , AMD Radeon™ Super Resolution , AMD Smart Access Memory , AMD Smart Access Video , AMD Radeon™ Boost , AMD Radeon™ Anti-Lag , AMD Radeon™ Image Sharpening 2 , AMD Enhanced Sync Technology , AMD FreeSync™ Technology , AMD Radeon™ Chill , AMD Virtual Super Resolution , AMD FSR™
TPP: 3062
TPP Density: 8.91
— 공식 사양 수치 확인
— 64
— 128
— AMD Radeon HD 6450 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku 160 — Tidak berlaku — — 750 — — — — — Up to 750 MHz 1 GB 1 — — — — — 30 30 W — — — — — Yes Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 160
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 6450
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Boost Frequency: Up to 750 MHz
Stream Processors: 160
Typical Board Power (Desktop): 30 W
Max Memory Size: 1 GB
DisplayPort: Yes
Dual Link DVI: Yes
HDMI: Yes
VGA: Yes
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 6570 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku 480 — Tidak berlaku — — 650 — — — — — Up to 650 MHz 1 GB 1 GDDR5 — — 29 Up to 29 GB/s 40 40 W — — — — — Yes Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 480
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , Unified Video Decoder (UVD) , DirectX® 11
Name: AMD Radeon™ HD 6570
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Boost Frequency: Up to 650 MHz
Stream Processors: 480
Typical Board Power (Desktop): 40 W
Max Memory Size: 1 GB
Memory Type: GDDR5
Memory Bandwidth: Up to 29 GB/s
DisplayPort: Yes
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , Unified Video Decoder (UVD) , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 6670 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku — — Tidak berlaku — — 800 — — — — — Up to 800 MHz 1 GB 1 — — — 64 Up to 64 GB/s — — — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 6670
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Boost Frequency: Up to 800 MHz
Max Memory Size: 1 GB
Memory Bandwidth: Up to 64 GB/s
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 6750 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku — — Tidak berlaku — — 900 — — — — — Up to 900 MHz 1 GB 1 — — — — — — — — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 6750
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Boost Frequency: Up to 900 MHz
Max Memory Size: 1 GB
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 6770 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku — — Tidak berlaku — — 900 — — — — — Up to 900 MHz 1 GB 1 — — — — — — — — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 6770
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Boost Frequency: Up to 900 MHz
Max Memory Size: 1 GB
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 6850 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku — — Tidak berlaku — — — — — — — — — 1 GB 1 — — — — — — — — — — — — No — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 6850
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Max Memory Size: 1 GB
DisplayPort: No
Dual Link DVI: No
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 6870 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku — — Tidak berlaku — — 900 — — — — — Up to 900 MHz 1 GB 1 — — — — — — — — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 6870
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Boost Frequency: Up to 900 MHz
Max Memory Size: 1 GB
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 6950 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku — — Tidak berlaku — — 800 — — — — — Up to 800 MHz 2 GB 2 — — — — — 500 500 W — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 6950
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Boost Frequency: Up to 800 MHz
Typical Board Power (Desktop): 500 W
Max Memory Size: 2 GB
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 6970 AMD Radeon HD Radeon HD 6000 Series Component — Tidak berlaku — — Tidak berlaku — — 880 — — — — — Up to 880 MHz 2 GB 2 — — — — — 550 550 W — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 6970
Family: Radeon HD
Series: Radeon HD 6000 Series
Board Type: Component
Boost Frequency: Up to 880 MHz
Typical Board Power (Desktop): 550 W
Max Memory Size: 2 GB
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 7730 AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software
Name: AMD Radeon™ HD 7730
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
DisplayPort: No
Dual Link DVI: No
VGA: No
Supported Technologies: Catalyst Software
— 공식 모델 목록 확인 · 상세 사양 미확인
— — — — — AMD Radeon HD 7750 AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku 512 — Tidak berlaku — — 800 — — — — — Up to 800 MHz 1 GB 1 — — — 72 Up to 72 GB/s 75 75 W — — — — — Yes Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 512
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 7750
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
Boost Frequency: Up to 800 MHz
Stream Processors: 512
Typical Board Power (Desktop): 75 W
Max Memory Size: 1 GB
Memory Bandwidth: Up to 72 GB/s
DisplayPort: Yes
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 7770 GHz Edition AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku 640 — Tidak berlaku — — 1,000 — — — — — Up to 1000 MHz 1 GB 1 — — — 72 Up to 72 GB/s 125 125 W — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 640
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 7770 GHz Edition
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
Boost Frequency: Up to 1000 MHz
Stream Processors: 640
Typical Board Power (Desktop): 125 W
Max Memory Size: 1 GB
Memory Bandwidth: Up to 72 GB/s
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 7790 AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software
Name: AMD Radeon™ HD 7790
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
DisplayPort: No
Dual Link DVI: No
VGA: No
Supported Technologies: Catalyst Software
— 공식 모델 목록 확인 · 상세 사양 미확인
— — — — — AMD Radeon HD 7850 AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku 1,200 — Tidak berlaku — — 1,000 — — — — — Up to 1000 MHz 2 GB 2 — — — 154 Up to 154 GB/s 225 225 W — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1200
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 7850
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
Boost Frequency: Up to 1000 MHz
Stream Processors: 1200
Typical Board Power (Desktop): 225 W
Max Memory Size: 2 GB
Memory Bandwidth: Up to 154 GB/s
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 7870 GHz Edition AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku 1,024 — Tidak berlaku — — 1,000 — — — — — Up to 1000 MHz 2 GB 2 — — — 154 Up to 154 GB/s 225 225 W — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1024
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 7870 GHz Edition
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
Boost Frequency: Up to 1000 MHz
Stream Processors: 1024
Typical Board Power (Desktop): 225 W
Max Memory Size: 2 GB
Memory Bandwidth: Up to 154 GB/s
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 7950 AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku 1,792 28 Tidak berlaku — — 1,250 — — — — — Up to 1250 MHz 3 GB 3 GDDR5 — — 240 Up to 240 GB/s 300 300 W — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1792
28
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — AMD App Acceleration , Catalyst Software , AMD PowerTune Technology , AMD ZeroCore Power Technology , DirectX® 11
Name: AMD Radeon™ HD 7950
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
Compute Units: 28
Boost Frequency: Up to 1250 MHz
Stream Processors: 1792
Typical Board Power (Desktop): 300 W
Max Memory Size: 3 GB
Memory Type: GDDR5
Memory Bandwidth: Up to 240 GB/s
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: AMD App Acceleration , Catalyst Software , AMD PowerTune Technology , AMD ZeroCore Power Technology , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 7970 AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku 2,048 — Tidak berlaku — — 925 — — — — — Up to 925 MHz 3 GB 3 — — — 264 Up to 264 GB/s 300 300 W — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2048
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 7970
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
Boost Frequency: Up to 925 MHz
Stream Processors: 2048
Typical Board Power (Desktop): 300 W
Max Memory Size: 3 GB
Memory Bandwidth: Up to 264 GB/s
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 7970 GHz Edition AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku 2,048 — Tidak berlaku — — 1,050 — — — — — Up to 1050 MHz 3 GB 3 — — — 288 Up to 288 GB/s 250 250 W — — — — — No Yes — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2048
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11
Name: AMD Radeon™ HD 7970 GHz Edition
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
Boost Frequency: Up to 1050 MHz
Stream Processors: 2048
Typical Board Power (Desktop): 250 W
Max Memory Size: 3 GB
Memory Bandwidth: Up to 288 GB/s
DisplayPort: No
Dual Link DVI: Yes
HDMI: Yes
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 7990 AMD Radeon HD Radeon HD 7000 Series Component — Tidak berlaku — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , DirectX® 11.2
Name: AMD Radeon™ HD 7990
Family: Radeon HD
Series: Radeon HD 7000 Series
Board Type: Component
DisplayPort: No
Dual Link DVI: No
VGA: No
Supported Technologies: Catalyst Software , DirectX® 11.2
— 공식 모델 목록 확인 · 상세 사양 미확인
— — — — — AMD Radeon HD 8570M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 5 Tidak berlaku — — 650 — — — — — Up to 650 MHz 2 GB 2 GDDR5 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 5
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8570M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 5
Boost Frequency: Up to 650 MHz
Max Memory Size: 2 GB
Memory Type: GDDR5
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8590M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 5 Tidak berlaku — — 750 — — — — — Up to 750 MHz 2 GB 2 GDDR5 64 64-bit 36 Up to 36 GB/s — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 5
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software
Name: AMD Radeon™ HD 8590M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 5
Boost Frequency: Up to 750 MHz
Max Memory Size: 2 GB
Memory Type: GDDR5
Memory Interface: 64-bit
Memory Bandwidth: Up to 36 GB/s
Supported Technologies: Catalyst Software
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8670M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 6 Tidak berlaku — — 775 — — — — — Up to 775 MHz 2 GB 2 DDR3 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 6
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8670M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 6
Boost Frequency: Up to 775 MHz
Max Memory Size: 2 GB
Memory Type: DDR3
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8690M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 6 Tidak berlaku — — 775 — — — — — Up to 775 MHz 2 GB 2 GDDR5 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 6
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8690M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 6
Boost Frequency: Up to 775 MHz
Max Memory Size: 2 GB
Memory Type: GDDR5
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8730M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 6 Tidak berlaku — — 650 — — — — — Up to 650 MHz 2 GB 2 DDR3 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 6
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8730M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 6
Boost Frequency: Up to 650 MHz
Max Memory Size: 2 GB
Memory Type: DDR3
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8750M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 6 Tidak berlaku — — 775 — — — — — Up to 775 MHz 2 GB 2 GDDR5 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 6
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8750M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 6
Boost Frequency: Up to 775 MHz
Max Memory Size: 2 GB
Memory Type: GDDR5
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8770M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 6 Tidak berlaku — — 775 — — — — — Up to 775 MHz 2 GB 2 GDDR5 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 6
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8770M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 6
Boost Frequency: Up to 775 MHz
Max Memory Size: 2 GB
Memory Type: GDDR5
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8790M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 6 Tidak berlaku — — 850 — — — — — Up to 850 MHz 2 GB 2 GDDR5 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 6
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8790M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 6
Boost Frequency: Up to 850 MHz
Max Memory Size: 2 GB
Memory Type: GDDR5
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8830M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 10 Tidak berlaku — — 575 — — — — — Up to 575 MHz 2 GB 2 DDR3 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 10
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8830M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 10
Boost Frequency: Up to 575 MHz
Max Memory Size: 2 GB
Memory Type: DDR3
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8850M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 10 Tidak berlaku — — 725 — — — — — Up to 725 MHz 2 GB 2 GDDR5 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 10
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8850M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 10
Boost Frequency: Up to 725 MHz
Max Memory Size: 2 GB
Memory Type: GDDR5
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8870M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 10 Tidak berlaku — — 725 — — — — — Up to 725 MHz 2 GB 2 GDDR5 — — — — — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 10
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , OpenGL 4.4 , DirectX® 11
Name: AMD Radeon™ HD 8870M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 10
Boost Frequency: Up to 725 MHz
Max Memory Size: 2 GB
Memory Type: GDDR5
Supported Technologies: Catalyst Software , OpenGL 4.4 , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon HD 8970M Series GPU AMD Radeon HD Radeon HD 8000M Series Component — Tidak berlaku — 20 Tidak berlaku — — 850 — — — — — Up to 850 MHz 4 GB 4 GDDR5 — — 154 Up to 154 GB/s — — — — — — — — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku — 20
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — Catalyst Software , AMD PowerTune Technology , AMD ZeroCore Power Technology , DirectX® 11
Name: AMD Radeon™ HD 8970M Series GPU
Family: Radeon HD
Series: Radeon HD 8000M Series
Board Type: Component
Compute Units: 20
Boost Frequency: Up to 850 MHz
Max Memory Size: 4 GB
Memory Type: GDDR5
Memory Bandwidth: Up to 154 GB/s
Supported Technologies: Catalyst Software , AMD PowerTune Technology , AMD ZeroCore Power Technology , DirectX® 11
— 공식 사양 수치 확인
— — — — — AMD Radeon Pro Duo AMD Radeon PRO Radeon PRO Series Desktop April 2017 Tidak berlaku — 72 Tidak berlaku — — — — — Polaris — — — 32 GB 32 GDDR5 256 256-bit 448 448 GB/s 250 250W — — — 1x6-Pin + 1x8-Pin
PCIe® 3.0 x16 3x DisplayPort 1.4 1x HDMI 2.0 — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 4608 (2x2304)
72
— — — — 14nm FinFET
— — — — No
— — — — — — — — — 12.0" (305 mm)
Full Height
— Active
11.5 TFLOPs
— 719 GFLOPs
— — — — — — — 12.0 (feature level 12_0)
4.5
2.0
1.0
Windows 11 - 64-Bit Edition
AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , Unified Video Decoder (UVD) , Video Code Engine (VCE) , High Bandwidth Cache (HBC) Controller , AMD DirectGMA Technology , S400 Synchronization Module Support
Name: Radeon™ Pro Duo
Family: Radeon PRO
Series: Radeon PRO Series
Form Factor: Desktops
Launch Date: April 2017
GPU Architecture: Polaris
Lithography: 14nm FinFET
Stream Processors: 4608 (2x2304)
Compute Units: 72
Peak Single Precision (FP32 Vector) Performance: 11.5 TFLOPs
Peak Double Precision (FP64) Performance: 719 GFLOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 250W
Dedicated Memory Size: 32 GB
Dedicated Memory Type: GDDR5
Memory Interface: 256-bit
Peak Memory Bandwidth: 448 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 3x DisplayPort 1.4
HDMI: 1x HDMI 2.0
Board Height: Full Height
Board Length: 12.0" (305 mm)
Board Width: Double Slot
10-bit Display Color Output: Yes
3D Stereo Support: No
DirectX: 12.0 (feature level 12_0)
OpenGL: 4.5
OpenCL: 2.0
Vulkan: 1.0
OS Support: Windows 11 - 64-Bit Edition
Supported Technologies: AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , Unified Video Decoder (UVD) , Video Code Engine (VCE) , High Bandwidth Cache (HBC) Controller , AMD DirectGMA Technology , S400 Synchronization Module Support
— 공식 사양 수치 확인
— — — — — AMD Radeon Pro SSG AMD Radeon PRO Radeon PRO Series Desktop August 2017 Tidak berlaku 4,096 64 Tidak berlaku — — — — — Vega — — — 16 GB 16 HBM2 2,048 2048-bit 484 484 GB/s 260 260W — — — 1x6-Pin + 1x8-Pin
PCIe® 3.0 x16 6x Mini-DisplayPort 1.4 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 4096
64
— — — — 14nm FinFET
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Active
12.3 TFLOPs
— 769 GFLOPs
— — — — — — — 12.0 (feature level 12_1)
4.5
2.0
1.0
Windows 11 - 64-Bit Edition
AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD XConnect™ Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , High Bandwidth Cache (HBC) Controller , AMD DirectGMA Technology , S400 Synchronization Module Support
Name: Radeon™ Pro SSG
Family: Radeon PRO
Series: Radeon PRO Series
Form Factor: Desktops
Launch Date: August 2017
GPU Architecture: Vega
Lithography: 14nm FinFET
Stream Processors: 4096
Compute Units: 64
Peak Single Precision (FP32 Vector) Performance: 12.3 TFLOPs
Peak Double Precision (FP64) Performance: 769 GFLOPs
Solid State Graphics (SSG): Yes (2 TB onboard SSG)
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 260W
Dedicated Memory Size: 16 GB
Dedicated Memory Type: HBM2
Memory Interface: 2048-bit
Peak Memory Bandwidth: 484 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 3.0 x16
Cooling: Active
DisplayPort: 6x Mini-DisplayPort 1.4
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
DirectX: 12.0 (feature level 12_1)
OpenGL: 4.5
OpenCL: 2.0
Vulkan: 1.0
OS Support: Windows 11 - 64-Bit Edition
Supported Technologies: AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD XConnect™ Technology , Unified Video Decoder (UVD) , Video Code Engine (VCE) , High Bandwidth Cache (HBC) Controller , AMD DirectGMA Technology , S400 Synchronization Module Support
— 공식 사양 수치 확인
— — — — — AMD Radeon PRO V520 AMD Radeon PRO Radeon PRO V Series Servers 1/10/2020 Tidak berlaku 2,304 36 Tidak berlaku — — 1,600 — — RDNA — — 1.6 GHz 8 GB 8 HBM2 2,048 2048-bit 512 512 GB/s 225 225W — — — 1x8-Pin
PCIe® 4.0 x16 , PCIe® 3.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2304
36
— — — — TSMC 7nm FinFET
— — — 1 GHz
Yes (Full-Chip)
— — — — — — — — — 10.5" (267 mm)
Full Height
— Passive
7.4 TFLOPs
14.75 TFLOPs
460 GFLOPs
— — — — — — — 12.0 (feature level 12_1)
4.6
2.2
1.1
Windows Server 2022 - 64-Bit Edition , Windows Server 2019 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Ubuntu® 22.04 LTS
AMD RDNA Architecture , AMD Software PRO Edition , Virtualization , AMD High Bandwidth Memory , AMD MxGPU Technology
Name: AMD Radeon™ PRO V520
Family: Radeon PRO
Series: Radeon PRO V Series
Form Factor: Servers
Launch Date: 1/10/2020
GPU Architecture: RDNA
Lithography: TSMC 7nm FinFET
Stream Processors: 2304
Compute Units: 36
Peak Engine Clock: 1.6 GHz
Peak Half Precision (FP16 Vector) Performance: 14.75 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 7.4 TFLOPs
Peak Double Precision (FP64) Performance: 460 GFLOPs
External Power Connectors: 1x8-Pin
Total Board Power (TBP): 225W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: HBM2
Memory Interface: 2048-bit
Memory Clock: 1 GHz
Peak Memory Bandwidth: 512 GB/s
Memory ECC Support: Yes (Full-Chip)
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16 , PCIe® 3.0 x16
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
DirectX: 12.0 (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.2
Vulkan: 1.1
OS Support: Windows Server 2022 - 64-Bit Edition , Windows Server 2019 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Ubuntu® 22.04 LTS
Supported Technologies: AMD RDNA Architecture , AMD Software PRO Edition , Virtualization , AMD High Bandwidth Memory , AMD MxGPU Technology
— 공식 사양 수치 확인
— — — — — AMD Radeon PRO V620 AMD Radeon PRO Radeon PRO V Series Servers 9/1/2023 Tidak berlaku 4,608 72 Tidak berlaku 72 — 2,200 — — AMD RDNA™ 2 — — 2.2 GHz 32 GB 32 GDDR6 256 256-bit 512 512 GB/s 300 300W — — — 2x8-Pin
PCIe® 4.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 4608
72
— — — — TSMC 7nm FinFET
— — — — Yes
128 MB
— — — — — — — — 10.5" (267 mm)
Full Height
— Passive
20.28 TFLOPs
40.55 TFLOPs
— — — — — — — — 12.0 (feature level 12_1)
4.6
2.2
1.3
Windows Server 2022 - 64-Bit Edition , Windows Server 2019 - 64-Bit Edition , Linux® Red Hat , CentOS , Ubuntu® Linux
AMD RDNA™ 2 Architecture , Virtualization , AMD MxGPU Technology , AMD Software Cloud Edition
Name: AMD Radeon™ PRO V620
Family: Radeon PRO
Series: Radeon PRO V Series
Form Factor: Servers
Launch Date: 9/1/2023
GPU Architecture: AMD RDNA™ 2
Hardware Raytracing: Yes
Lithography: TSMC 7nm FinFET
Ray Accelerators: 72
Stream Processors: 4608
Compute Units: 72
Peak Engine Clock: 2.2 GHz
Peak Half Precision (FP16 Vector) Performance: 40.55 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 20.28 TFLOPs
External Power Connectors: 2x8-Pin
Total Board Power (TBP): 300W
Dedicated Memory Size: 32 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 128 MB
Memory Interface: 256-bit
Peak Memory Bandwidth: 512 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16
Cooling: Passive
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
DirectX: 12.0 (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.2
Vulkan: 1.3
OS Support: Windows Server 2022 - 64-Bit Edition , Windows Server 2019 - 64-Bit Edition , Linux® Red Hat , CentOS , Ubuntu® Linux
Supported Technologies: AMD RDNA™ 2 Architecture , Virtualization , AMD MxGPU Technology , AMD Software Cloud Edition
— 공식 사양 수치 확인
— 72
— — — AMD Radeon PRO V710 AMD Radeon PRO Radeon PRO V Series Servers 10/3/2024 Tidak berlaku 3,456 54 Tidak berlaku 54 — 2,000 — — AMD RDNA™ 3 — — 2 GHz 28 GB 28 GDDR6 224 224-bit 448 448 GB/s 158 158W — — — 1x8-Pin
PCIe® 4.0 x16 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 3456
54
— — — — TSMC 5nm | 6nm FinFET
28.1 Billion
— — — Yes
54 MB
— — — — — — — — 10.5"(267 mm)
Full Height
— Passive
27.65 TFLOPs
27.65 TFLOPs
— — — — — — — — 12.0 (feature level 12_1)
4.6
2.2
1.3
Windows Server 2022 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Linux® Red Hat , CentOS , Ubuntu® Linux
AMD RDNA™ 3 Architecture , Virtualization , AMD MxGPU Technology
Name: AMD Radeon™ PRO V710
Family: Radeon PRO
Series: Radeon PRO V Series
Form Factor: Servers
Launch Date: 10/3/2024
GPU Architecture: AMD RDNA™ 3
Hardware Raytracing: Yes
Lithography: TSMC 5nm | 6nm FinFET
Ray Accelerators: 54
Stream Processors: 3456
Compute Units: 54
Peak Engine Clock: 2 GHz
Peak Half Precision (FP16 Vector) Performance: 27.65 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 27.65 TFLOPs
Peak 4-bit Precision (INT4 Matrix) Performance: 110.59 TOPs
Peak 8-bit Precision (INT8 Matrix) Performance: 55.3 TOPs
Transistor Count: 28.1 Billion
External Power Connectors: 1x8-Pin
Total Board Power (TBP): 158W
Dedicated Memory Size: 28 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 54 MB
Memory Interface: 224-bit
Peak Memory Bandwidth: 448 GB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16
Cooling: Passive
Board Height: Full Height
Board Length: 10.5"(267 mm)
Board Width: Single Slot
DirectX: 12.0 (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.2
Vulkan: 1.3
OS Support: Windows Server 2022 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Linux® Red Hat , CentOS , Ubuntu® Linux
Supported Technologies: AMD RDNA™ 3 Architecture , Virtualization , AMD MxGPU Technology
— 공식 사양 수치 확인
— 54
— — — AMD Radeon Pro VII AMD Radeon PRO Radeon PRO VII Series — 5/13/2020 Tidak berlaku 3,840 60 Tidak berlaku — — — — — Vega20 — — — 16 GB 16 HBM2 4,096 4096-bit 1,000 1 TB/s 250 250W — — — 1x6-Pin + 1x8-Pin
PCIe® 4.0 x16 6x DisplayPort 1.4 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 3840
60
— — — — TSMC 7nm FinFET
— — — — Yes
— — — — — — — — — 10.5" (267 mm)
Full Height
— Active
13.1 TFLOPs
— 6.5 TFLOPs
— — — — — — — 12.0 (feature level 12_1)
4.6
2.0
1.1
Windows 11 - 64-Bit Edition
AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
Name: Radeon™ Pro VII
Family: Radeon PRO
Series: Radeon PRO VII Series
Launch Date: 5/13/2020
GPU Architecture: Vega20
Lithography: TSMC 7nm FinFET
Stream Processors: 3840
Compute Units: 60
Peak Single Precision (FP32 Vector) Performance: 13.1 TFLOPs
Peak Double Precision (FP64) Performance: 6.5 TFLOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 250W
Dedicated Memory Size: 16 GB
Dedicated Memory Type: HBM2
Memory Interface: 4096-bit
Peak Memory Bandwidth: 1 TB/s
Memory ECC Support: Yes
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16
Cooling: Active
DisplayPort: 6x DisplayPort 1.4
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: Yes
DirectX: 12.0 (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.1
OS Support: Windows 11 - 64-Bit Edition
Supported Technologies: AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
— 공식 사양 수치 확인
— — — — — AMD Radeon Pro W5500 AMD Radeon PRO Radeon PRO W5000 Series — — Tidak berlaku 1,408 22 Tidak berlaku — — — — — RDNA — — — 8 GB 8 GDDR6 128 128-bit 224 224 GB/s 125 125W — — — 1x6-Pin
PCIe® 4.0 x16 (x8 electrical) 4x DisplayPort 1.4 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1408
22
— — — — TSMC 7nm FinFET
— — — — No
— — — — — — — — — 9.5" (241 mm)
Full Height
— Active
5.35 TFLOPs
— 330 GFLOPs
— — — — — — — 12.0 (feature level 12_1)
4.6
2.0
1.1
Windows 11 - 64-Bit Edition
AMD RDNA Architecture , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
Name: Radeon™ Pro W5500
Family: Radeon PRO
Series: Radeon PRO W5000 Series
GPU Architecture: RDNA
Lithography: TSMC 7nm FinFET
Stream Processors: 1408
Compute Units: 22
Peak Single Precision (FP32 Vector) Performance: 5.35 TFLOPs
Peak Double Precision (FP64) Performance: 330 GFLOPs
External Power Connectors: 1x6-Pin
Total Board Power (TBP): 125W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: GDDR6
Memory Interface: 128-bit
Peak Memory Bandwidth: 224 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16 (x8 electrical)
Cooling: Active
DisplayPort: 4x DisplayPort 1.4
Board Height: Full Height
Board Length: 9.5" (241 mm)
Board Width: Single Slot
3D Stereo Support: Yes
DirectX: 12.0 (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.1
OS Support: Windows 11 - 64-Bit Edition
Supported Technologies: AMD RDNA Architecture , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
— 공식 사양 수치 확인
— — — — — AMD Radeon PRO W5500M (Mobile) AMD Radeon PRO Radeon PRO W5000 Mobile Series Mobile — Tidak berlaku 1,408 22 Tidak berlaku — — — — — RDNA — — — 4 GB 4 GDDR6 128 128-bit 224 224 GB/s — — 65 - 85W — — — PCIe® 4.0 — — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1408
22
— — — — TSMC 7nm FinFET
— — — — No
— — — — 65 85 — — — — — — — 4.79 TFLOPs
— 299 GFLOPs
— — — — — — — — — — — Windows 11 - 64-Bit Edition
AMD RDNA Architecture , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator
Name: Radeon™ PRO W5500M (Mobile)
Family: Radeon PRO
Series: Radeon PRO W5000 Mobile Series
Form Factor: Laptops
GPU Architecture: RDNA
Lithography: TSMC 7nm FinFET
Stream Processors: 1408
Compute Units: 22
Peak Single Precision (FP32 Vector) Performance: 4.79 TFLOPs
Peak Double Precision (FP64) Performance: 299 GFLOPs
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR6
Memory Interface: 128-bit
Peak Memory Bandwidth: 224 GB/s
Memory ECC Support: No
Bus Type: PCIe® 4.0
TGP: 65 - 85W
10-bit Display Color Output: Yes
OS Support: Windows 11 - 64-Bit Edition
Supported Technologies: AMD RDNA Architecture , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator
— 공식 사양 수치 확인
— — — — — AMD Radeon Pro W5700 AMD Radeon PRO Radeon PRO W5000 Series — 11/19/2019 Tidak berlaku 2,304 36 Tidak berlaku — — — — — RDNA — — — 8 GB 8 GDDR6 256 256-bit 448 448 GB/s 205 205W — — — 1x6-Pin + 1x8-Pin
PCIe® 4.0 x16 5x Mini-DisplayPort 1.4 — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 2304
36
— — — — TSMC 7nm FinFET
— — — — No
— — — — — — — — — 10.5" (267 mm)
Full Height
— Active
8.89 TFLOPs
— 556 GFLOPs
— — — — — — — 12.0 (feature level 12_1)
4.6
2.0
1.1
Windows 11 - 64-Bit Edition
AMD RDNA Architecture , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
Name: Radeon™ Pro W5700
Family: Radeon PRO
Series: Radeon PRO W5000 Series
Launch Date: 11/19/2019
GPU Architecture: RDNA
Lithography: TSMC 7nm FinFET
Stream Processors: 2304
Compute Units: 36
Peak Single Precision (FP32 Vector) Performance: 8.89 TFLOPs
Peak Double Precision (FP64) Performance: 556 GFLOPs
External Power Connectors: 1x6-Pin + 1x8-Pin
Total Board Power (TBP): 205W
Dedicated Memory Size: 8 GB
Dedicated Memory Type: GDDR6
Memory Interface: 256-bit
Peak Memory Bandwidth: 448 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x16
Cooling: Active
DisplayPort: 5x Mini-DisplayPort 1.4
Other Display Outputs: 1x USB-C
Board Height: Full Height
Board Length: 10.5" (267 mm)
Board Width: Double Slot
3D Stereo Support: Yes
DirectX: 12.0 (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.0
Vulkan: 1.1
OS Support: Windows 11 - 64-Bit Edition
Supported Technologies: AMD RDNA Architecture , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
— 공식 사양 수치 확인
— — — — — AMD Radeon PRO W6300M AMD Radeon PRO Radeon PRO W6000 Mobile Series Mobile Workstations 01/19/2022 Tidak berlaku 768 12 Tidak berlaku 12 — — — — AMD RDNA™ 2 — — — 2 GB 2 GDDR6 32 32-bit 64 64 GB/s — — 25W — — — PCIe® 4.0 x4 (3.0 Backward Compatible) Specifications may vary with implementation. GPU Support for up to 5x outputs. — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 768
12
— — — — TSMC 6nm FinFET
— — — — No
8 MB
— — — — — — — — — — — — 3.4 TFLOPs
6.75 TFLOPs
— — — — — — — — 12.0 Ultimate (feature level 12_1)
4.6
2.2
1.2
Windows 11 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Linux x86 64-Bit
AMD FSR™ , AMD Viewport Boost , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
Name: AMD Radeon™ PRO W6300M
Family: Radeon PRO
Series: Radeon PRO W6000 Mobile Series
Form Factor: Mobile Workstations
Launch Date: 01/19/2022
GPU Architecture: AMD RDNA™ 2
Lithography: TSMC 6nm FinFET
Ray Accelerators: 12
Stream Processors: 768
Compute Units: 12
Peak Half Precision (FP16 Vector) Performance: 6.75 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 3.4 TFLOPs
Dedicated Memory Size: 2 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 8 MB
Memory Interface: 32-bit
Peak Memory Bandwidth: 64 GB/s
Memory ECC Support: No
Bus Type: PCIe® 4.0 x4 (3.0 Backward Compatible)
TGP: 25W
DisplayPort: Specifications may vary with implementation. GPU Support for up to 5x outputs.
10-bit Display Color Output: Yes
3D Stereo Support: Yes
DirectX: 12.0 Ultimate (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.2
Vulkan: 1.2
OS Support: Windows 11 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Linux x86 64-Bit
Supported Video Formats: 4K H264 Decode , H265∕HEVC Decode
Supported Technologies: AMD FSR™ , AMD Viewport Boost , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
— 공식 사양 수치 확인
— 12
— — — AMD Radeon PRO W6400 AMD Radeon PRO Radeon PRO W6000 Series Workstations 01/19/2022 Tidak berlaku 768 12 Tidak berlaku 12 — — — — AMD RDNA™ 2 — — — 4 GB 4 GDDR6 64 64-bit 128 128 GB/s — — 50W — — None
PCIe® 4.0 x4 (3.0 Backward Compatible) 2x DisplayPort™ 1.4 with DSC and Audio Support — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 768
12
— — — — TSMC 6nm FinFET
— — — — No
16 MB
— — — — — — — — 6.6" (168mm)
Half Height
— Active
3.5 TFLOPs
7.07 TFLOPs
— — — — — — — — 12.0 Ultimate (feature level 12_1)
4.6
2.2
1.2
Windows 11 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Linux x86 64-Bit
AMD FSR™ , AMD Viewport Boost , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
Name: AMD Radeon™ PRO W6400
Family: Radeon PRO
Series: Radeon PRO W6000 Series
Form Factor: Workstations
Launch Date: 01/19/2022
GPU Architecture: AMD RDNA™ 2
Lithography: TSMC 6nm FinFET
Ray Accelerators: 12
Stream Processors: 768
Compute Units: 12
Peak Half Precision (FP16 Vector) Performance: 7.07 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 3.5 TFLOPs
External Power Connectors: None
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 16 MB
Memory Interface: 64-bit
Peak Memory Bandwidth: 128 GB/s
Memory ECC Support: No
GPU Form Factor: PCIe® Add-in Card
Bus Type: PCIe® 4.0 x4 (3.0 Backward Compatible)
TGP: 50W
Cooling: Active
DisplayPort: 2x DisplayPort™ 1.4 with DSC and Audio Support
Board Height: Half Height
Board Length: 6.6" (168mm)
Board Width: Single Slot
10-bit Display Color Output: Yes
3D Stereo Support: Yes
DirectX: 12.0 Ultimate (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.2
Vulkan: 1.2
OS Support: Windows 11 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Linux x86 64-Bit
Supported Video Formats: 4K H264 Decode , H265∕HEVC Decode
Supported Technologies: AMD FSR™ , AMD Viewport Boost , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
— 공식 사양 수치 확인
— 12
— — — AMD Radeon PRO W6500M AMD Radeon PRO Radeon PRO W6000 Mobile Series Mobile Workstations 01/19/2022 Tidak berlaku 1,024 16 Tidak berlaku 16 — — — — AMD RDNA™ 2 — — — 4 GB 4 GDDR6 64 64-bit 128 128 GB/s — — 35 - 50W — — — PCIe® 4.0 x4 (3.0 Backward Compatible) Specifications may vary with implementation. GPU Support for uo to 5x outputs. — — — Not listed — — 단일 모델·변형
— — — Tidak berlaku 1024
16
— — — — TSMC 6nm FinFET
— — — — No
16 MB
— — — 35 50 — — — — — — — 5.3 TFLOPs
10.6 TFLOPs
— — — — — — — — 12.0 Ultimate (feature level 12_1)
4.6
2.2
1.2
Windows 11 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Linux x86 64-Bit
AMD FSR™ , AMD Viewport Boost , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
Name: AMD Radeon™ PRO W6500M
Family: Radeon PRO
Series: Radeon PRO W6000 Mobile Series
Form Factor: Mobile Workstations
Launch Date: 01/19/2022
GPU Architecture: AMD RDNA™ 2
Lithography: TSMC 6nm FinFET
Ray Accelerators: 16
Stream Processors: 1024
Compute Units: 16
Peak Half Precision (FP16 Vector) Performance: 10.6 TFLOPs
Peak Single Precision (FP32 Vector) Performance: 5.3 TFLOPs
Dedicated Memory Size: 4 GB
Dedicated Memory Type: GDDR6
AMD Infinity Cache Technology: 16 MB
Memory Interface: 64-bit
Peak Memory Bandwidth: 128 GB/s
Memory ECC Support: No
Bus Type: PCIe® 4.0 x4 (3.0 Backward Compatible)
TGP: 35 - 50W
DisplayPort: Specifications may vary with implementation. GPU Support for uo to 5x outputs.
10-bit Display Color Output: Yes
3D Stereo Support: Yes
DirectX: 12.0 Ultimate (feature level 12_1)
OpenGL: 4.6
OpenCL: 2.2
Vulkan: 1.2
OS Support: Windows 11 - 64-Bit Edition , Windows 10 - 64-Bit Edition , Linux x86 64-Bit
Supported Video Formats: 4K H264 Decode , H265∕HEVC Decode
Supported Technologies: AMD FSR™ , AMD Viewport Boost , AMD Remote Workstation , AMD Radeon™ Media Engine , AMD Software PRO Edition , AMD Radeon™ VR Ready Creator , AMD Radeon™ ProRender
— 공식 사양 수치 확인
— 16
— — — 
Indonesia