AMD Instinct MI100 vs AMD Radeon PRO W6800 Full Specs

Summary

7,680 Shaders 1.5GHz	3,840 Shaders 2.32GHz
32GB HBM21.23TB/s	32GB GDDR6512GB/s
Peak·FP32·FP16 23.07 TFLOPS	Peak·FP32·FP16 17.82 TFLOPS
Form Factor PCIe Card	Form Factor PCIe Card
TDP 300W	TDP 250W
Power Connectors 2x 8-Pin -	Power Connectors 1x 8-Pin 1x 6-Pin

Highlights

Peak OPS

Peak OPS FP32 FP16 Tensor FP16-16 Tensor FP16-32 Tensor FP32 FP64 Tensor BF16 Tensor INT8 Tensor INT4 Memory Bandwidth TDP

Instinct MI100184.6 TFLOPSFP16-16 Tensor

x5.18

Radeon PRO W680035.64 TFLOPSFP16

x1

Show Top Score

Theoretical Performance

Clock Speed1502MHz UC·Base·Boost·OC	Clock Speed2320MHz UC·Base·Boost·OC
Peak OPS 184.6 TFLOPSFP16-16 Tensor	Peak OPS 35.64 TFLOPSFP16
Tensor FP16-16 184.6 TFLOPS Tensor FP16-32 184.6 TFLOPS	-
Tensor BF16 92.28 TFLOPS	-
FP32 23.07 TFLOPS Tensor FP32 46.14 TFLOPS	FP32 17.82 TFLOPS Tensor FP32 -
FP64 11.54 TFLOPS	FP64 1.11 TFLOPS
Tensor INT4 92.28 TOPS	Tensor INT4 -
Tensor INT8 92.28 TOPS	-
Pixel Rate -	Pixel Rate 222.7 GPixel/s
Texture Rate 721 GTexel/s	Texture Rate 556.8 GTexel/s

Core Config

Shaders 7,680 Shaders	Shaders 3,840 Shaders
TMUs 480 TMUs	TMUs 240 TMUs
ROPs -	ROPs 96 ROPs
Tensor Cores 480 T-Cores	Tensor Cores -
RT Cores -	RT Cores 60 RT-Cores
CUs 120 CUs	CUs 60 CUs

Clocks

Base Clock 1GHz	Base Clock 2.08GHz
Boost Clock 1.5GHz	Boost Clock 2.32GHz

Cache

L0 Cache -	L0 Cache 32KB/WGP
L1 16KB/CU	L1 128KB/Array
L2 Cache 8MB shared	L2 Cache 4MB shared
L3 Cache -	L3 Cache 128MB shared
L3 Bandwidth -	L3 Bandwidth 1.66TB/s

Memory

32GB HBM2	32GB GDDR6
Memory Bus 4096-bit	Memory Bus 256-bit
Memory Speed 2.4GT/s	Memory Speed 16GT/s
Memory Bandwidth 1.23TB/s	Memory Bandwidth 512GB/s
ECC Yes	ECC No

Power & Thermals

TDP 300W

TDP 250W

Video Output

Max Resolution -	Max Resolution 7680x4320
Refresh Rate Calculator -	Refresh Rate Calculator 7680x4320 Resolution8K UHD Refresh Rate 120Hz
Multi-Monitor -	Multi-Monitor 4
Adaptive Sync -	Adaptive Sync FreeSync
DSC Not Supported	DSC Supported
HDCP -	HDCP HDCP 2.3

Ports

-	6x Mini DisplayPort 1.4

API Support

Shader Model -	Shader Model 6.5
-	DirectX DirectX 12 Direct3D 12_2
OpenGL - OpenCL 2.1 Vulkan -	OpenGL 4.6 OpenCL 2.1 Vulkan 1.2
GFX 9.4	GFX 10.3

Encoder

Encoder VCN 2.5	Encoder VCN 3.0
Codec AVC (H.264) HEVC (H.265)	Codec AVC (H.264) HEVC (H.265)

Decoder

Decoder VCN 2.5	Decoder VCN 3.0
Codec MPEG-1 MPEG-2 MPEG-4 JPEG VC-1 VP9 AVC (H.264) HEVC (H.265) -	Codec MPEG-1 MPEG-2 MPEG-4 JPEG VC-1 VP9 AVC (H.264) HEVC (H.265) AV1

Card

Form Factor PCIe Card	Form Factor PCIe Card
PCIe PCIe 4.0 x16 2-Slots	PCIe PCIe 4.0 x16 2-Slots
Height 111mm (4.37") Width 267mm (10.51") Depth 37mm (1.46")	Height 111mm (4.37") Width 267mm (10.51") Depth 40mm (1.57")
Cooling Passive -	Cooling Blower 1x Fan
Power Connectors 2x 8-Pin	Power Connectors 1x 8-Pin, 1x 6-Pin
Multi-GPU Multi-GPU Support Supported Multi-GPU Type Infinity Fabric	Multi-GPU Multi-GPU Support Supported Multi-GPU Type Bridgeless

Chip

Manufacturer AMD	Manufacturer AMD
Chip Designer AMD	Chip Designer AMD
Architecture CDNA 1	Architecture RDNA 2
Family Instinct	Family Radeon PRO W
Branding	Branding
Codename Arcturus	Codename Sienna Cichlid
Chip Variant Arcturus XL	Chip Variant Navi 21 XL
Market Segment Server	Market Segment Workstation
Release Date Nov 16, 2020	Release Date Jun 8, 2021

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node N7	Fabrication Node N7
Die Size 750mm²	Die Size 520mm²
Transistor Count 25.6B	Transistor Count 26.8B
Transistor Density 34.13 MTr/mm²	Transistor Density 51.56 MTr/mm²

Images