GPU Comparison

A

AMD

RADEON

AMD Radeon Instinct MI250

CORE STATE Aldebaran

VRAM 128 GB

CLOCK SPEED 1700 MHz

TDP 500 W

BUS WIDTH 8192 bit

ARCHITECTURE CDNA 2.0

nm

PROCESS 6 nm

LAUNCH DATE 2021

VS

NVIDIA

GEFORCE

GeForce RTX 3080 Ti

CORE STATE GA102

VRAM 12 GB

CLOCK SPEED 1665 MHz

TDP 350 W

BUS WIDTH 384 bit

ARCHITECTURE Ampere

nm

PROCESS 8 nm

LAUNCH DATE 2021

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12

N/A

5,077

geekbench_opencl

N/A

191,309

geekbench_vulkan

N/A

171,798

passmark_directx_10

N/A

184

passmark_directx_11

N/A

223

passmark_directx_12

N/A

110

passmark_directx_9

N/A

274

passmark_g2d

N/A

1,091

passmark_g3d

N/A

26,896

passmark_gpu_compute

N/A

15,282

DETAILED SPECIFICATIONS

SPECIFICATION

Instinct MI250

RTX 3080 Ti

Core Specs

Shading Units

13,312

10,240 -23.1%

Shaders

13,312

10,240 -23.1%

TMUs

832

320 -61.5%

ROPs

0

112 +∞%

Compute Units

208

—

SM Count

—

80

Clocks

Base Clock

1000 MHz

1365 MHz

Boost Clock

1700 MHz

1665 MHz

Memory Clock

1600 MHz 3.2 Gbps effective

1188 MHz 19 Gbps effective

Memory

Memory Size

128 GB

12 GB

VRAM (MB)

131,072

12,288 -90.6%

Memory Type

HBM2e

GDDR6X

Memory Bus

8192 bit

384 bit

Bandwidth

3.28 TB/s

912.4 GB/s

Cache

L1 Cache

16 KB (per CU)

128 KB (per SM)

L2 Cache

16 MB

6 MB

Performance

Pixel Rate

0 MPixel/s

186.5 GPixel/s

Texture Rate

1,414.4 GTexel/s

532.8 GTexel/s

FP32 (TFLOPS)

45.26 TFLOPS

34.10 TFLOPS

FP64 (TFLOPS)

45.26 TFLOPS (1:1)

532.8 GFLOPS (1:64)

FP16 (TFLOPS)

362.1 TFLOPS (8:1)

34.10 TFLOPS (1:1)

AI/RT

RT Cores

—

80

Tensor Cores

—

320

Power

TDP

500 W

350 W

TDP (W)

500

350 -30.0%

Suggested PSU

900 W

750 W

Power Connectors

2x 8-pin

1x 12-pin

Architecture

Architecture

CDNA 2.0

Ampere

GPU Name

Aldebaran

GA102

Generation

Radeon Instinct (MIx)

GeForce 30

Process Size

6 nm

8 nm

Transistors

58,200 million

28,300 million

Die Size

724 mm²

628 mm²

Foundry

TSMC

Samsung

Density

80.4M / mm²

45.1M / mm²

AMD MCM

MCM

2

—

API Support

DirectX

—

12 Ultimate (12_2)

OpenGL

—

4.6

Vulkan

—

1.4

OpenCL

3.0

3.0

CUDA

—

8.6

Shader Model

—

6.8

Physical

Slot Width

Dual-slot

Dual-slot

Length

267 mm 10.5 inches

285 mm 11.2 inches

Height

111 mm 4.4 inches

112 mm 4.4 inches

Outputs

No outputs

1x HDMI 2.13x DisplayPort 1.4a

Bus Interface

PCIe 4.0 x16

PCIe 4.0 x16

Other

Launch Price

—

1,199 USD

Production

End-of-life

End-of-life

Predecessor

FirePro Data Center

GeForce 20

Successor

—

GeForce 40

View Radeon Instinct MI250 Details View GeForce RTX 3080 Ti Details