NVIDIA GB200 vs NVIDIA GeForce RTX 4090 D Full Specs - HMC

37,888 Shaders 2.11GHz	14,592 Shaders 2.52GHz
384GB HBM3e16TB/s	24GB GDDR6X1.01TB/s
·· 160 TFLOPS	·· 73.54 TFLOPS
Form Factor Superchip	Form Factor PCIe Card
TDP 2700W	TDP 450W
Power Connectors -	Power Connectors 1x 16-Pin 12VHPWR

GB200

GB20040 PFLOPSFP4 Tensor Sparse (2x 20 PFLOPS)

x17.00

GeForce RTX 4090 D

GeForce RTX 4090 D2.35 POPSINT4 Tensor Sparse

x1

Clock Speed ··	Clock Speed ···
Peak OPS 40 PFLOPSFP4 Tensor Sparse	Peak OPS 2.35 POPSINT4 Tensor Sparse
Tensor FP4 20 PFLOPS FP4 Tensor Sparse 40 PFLOPS	-
Tensor FP8-16 10 PFLOPS FP8-16 Tensor Sparse 20 PFLOPS Tensor FP8-32 10 PFLOPS FP8-32 Tensor Sparse 20 PFLOPS	Tensor FP8-16 588.3 TFLOPS FP8-16 Tensor Sparse 1.18 PFLOPS Tensor FP8-32 294.2 TFLOPS FP8-32 Tensor Sparse 588.3 TFLOPS
Tensor FP16-16 5 PFLOPS FP16-16 Tensor Sparse 10 PFLOPS Tensor FP16-32 5 PFLOPS FP16-32 Tensor Sparse 10 PFLOPS	Tensor FP16-16 294.2 TFLOPS FP16-16 Tensor Sparse 588.3 TFLOPS Tensor FP16-32 147.1 TFLOPS FP16-32 Tensor Sparse 294.2 TFLOPS
BF16 320.1 TFLOPS Tensor BF16 5 PFLOPS BF16 Tensor Sparse 10 PFLOPS	BF16 73.54 TFLOPS Tensor BF16 147.1 TFLOPS BF16 Tensor Sparse 294.2 TFLOPS
Tensor TF32 2.5 PFLOPS	Tensor TF32 73.54 TFLOPS
FP32 160 TFLOPS	FP32 73.54 TFLOPS
FP64 80.02 TFLOPS Tensor FP64 78.13 TFLOPS	FP64 1.15 TFLOPS Tensor FP64 -
Tensor INT4 -	Tensor INT4 1.18 POPS
Tensor INT8 10 POPS	Tensor INT8 588.3 TOPS
Ray -	Ray 170 TOPS
Pixel Rate 67.6 GPixel/s	Pixel Rate 443.5 GPixel/s
Texture Rate 1.25 TTexel/s	Texture Rate 1.15 TTexel/s

Shaders 37,888 Shaders	Shaders 14,592 Shaders
TMUs 1,184 TMUs	TMUs 456 TMUs
ROPs 64 ROPs	ROPs 176 ROPs
Tensor Cores 1,184 T-Cores	Tensor Cores 456 T-Cores
RT Cores -	RT Cores 114 RT-Cores
SMs 296 SMs	SMs 114 SMs

Base Clock -	Base Clock 2.23GHz
Boost Clock 2.11GHz	Boost Clock 2.52GHz
Tensor Clock 2.06GHz	Tensor Clock -

L1 64KB/SM Tex	L1 64KB/SM Tex
L2 Cache 64MB shared	L2 Cache 72MB shared

384GB HBM3e	24GB GDDR6X
Memory Bus 8192-bit	Memory Bus 384-bit
Memory Speed 7.8GT/s	Memory Speed 21GT/s
Memory Bandwidth 16TB/s	Memory Bandwidth 1.01TB/s
ECC No	ECC No

TDP 2700W	TDP 450W
Max Temp -	Max Temp 90°C Max

Max Resolution -	Max Resolution 7680x4320
Refresh Rate Calculator -	Refresh Rate Calculator 7680x4320 Resolution8K UHD Refresh Rate 60Hz
Multi-Monitor -	Multi-Monitor 4
Adaptive Sync G-Sync FreeSync	Adaptive Sync G-Sync FreeSync
DSC Not Supported	DSC Supported
HDCP -	HDCP HDCP 2.3

-	3x DisplayPort 1.4 1x HDMI 2.1

Shader Model -	Shader Model 6.6
-	DirectX DirectX 12 Direct3D 12_3
OpenGL - OpenCL 3 Vulkan -	OpenGL 4.6 OpenCL 3 Vulkan 1.3
CUDA 10 PureVideo HD VP13 VDPAU Feature Set M	CUDA 8.9 PureVideo HD VP12 VDPAU Feature Set L

-	Encoder 2x NVENC 8
Codec -	Codec AVC (H.264) HEVC (H.265) AV1

Decoder 7x NVDEC 6	Decoder NVDEC 5
Codec MPEG-1 MPEG-2 MPEG-4 VC-1 VP8 VP9 AVC (H.264) HEVC (H.265) AV1	Codec MPEG-1 MPEG-2 MPEG-4 VC-1 VP8 VP9 AVC (H.264) HEVC (H.265) AV1

Form Factor Superchip	Form Factor PCIe Card
PCIe PCIe 6.0 x16 -	PCIe PCIe 4.0 x16 3-Slots
-	Height 137mm (5.39") Width 304mm (11.97") Depth 61mm (2.4")
Cooling Passive -	Cooling Open-Air 2x Fans
Power Connectors -	Power Connectors 1x 16-Pin 12VHPWR
Multi-GPU Multi-GPU Support Supported Multi-GPU Type NVLink	Multi-GPU -

Manufacturer NVIDIA	Manufacturer NVIDIA
Chip Designer NVIDIA	Chip Designer NVIDIA
Architecture Blackwell	Architecture Ada Lovelace
Family Server	Family GeForce 40
Branding	Branding
Codename NV190	Codename NV182
Chip Variant Umbriel	Chip Variant AD102-200-A1
Market Segment Server	Market Segment Desktop
Release Date Mar 18, 2024	Release Date Dec 28, 2023

Foundry TSMC	Foundry TSMC
Fabrication Node 4NP	Fabrication Node 4N
Die Size 1620mm²	Die Size 608mm²
Transistor Count 208B	Transistor Count 76.3B
Transistor Density 128.4 MTr/mm²	Transistor Density 125.4 MTr/mm²