NVIDIA GeForce RTX 3070 Ti vs NVIDIA GeForce GTX 970

VS

NVIDIA GeForce RTX 3070 Ti NVIDIA GeForce GTX 970

We compared two Desktop platform GPUs: 8GB VRAM GeForce RTX 3070 Ti and 4GB VRAM GeForce GTX 970 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

NVIDIA GeForce RTX 3070 Ti 's Advantages

Released 6 years and 8 months late

Boost Clock has increased by 50% (1770MHz vs 1178MHz)

More VRAM (8GB vs 4GB)

Larger VRAM bandwidth (608.3GB/s vs 224.4GB/s)

4480 additional rendering cores

NVIDIA GeForce GTX 970 's Advantages

Lower TDP (148W vs 290W)

Score

Benchmark

FP32 (float)

GeForce RTX 3070 Ti +454%

21.75 TFLOPS

GeForce GTX 970

3.92 TFLOPS

3DMark Time Spy

GeForce RTX 3070 Ti +312%

14971

GeForce GTX 970

3625

Shadow of the Tomb Raider 2160p

GeForce RTX 3070 Ti +366%

70

GeForce GTX 970

15

Shadow of the Tomb Raider 1440p

GeForce RTX 3070 Ti +316%

125

GeForce GTX 970

30

Shadow of the Tomb Raider 1080p

GeForce RTX 3070 Ti +345%

178

GeForce GTX 970

40

GTA 5 2160p

GeForce RTX 3070 Ti +83%

77

GeForce GTX 970

42

GTA 5 1440p

GeForce RTX 3070 Ti +159%

114

GeForce GTX 970

44

GTA 5 1080p

GeForce RTX 3070 Ti +67%

164

GeForce GTX 970

98

GeForce RTX 3070 Ti

VS

GeForce GTX 970

Graphics Card

May 2021

Release Date

Sep 2014

GeForce 30

Generation

GeForce 900

Desktop

Type

Desktop

PCIe 4.0 x16

Bus Interface

PCIe 3.0 x16

Clock Speeds

1575 MHz

Base Clock

1050 MHz

1770 MHz

Boost Clock

1178 MHz

1188 MHz

Memory Clock

1753 MHz

Memory

8GB

Memory Size

4GB

GDDR6X

Memory Type

GDDR5

256bit

Memory Bus

256bit

608.3GB/s

Bandwidth

224.4GB/s

Render Config

48

SM Count

-

Compute Units

-

6144

Shading Units

1664

192

TMUs

104

96

ROPs

56

192

Tensor Cores

-

48

RT Cores

-

128 KB (per SM)

L1 Cache

48 KB (per SMM)

4 MB

L2 Cache

2 MB

Theoretical Performance

169.9 GPixel/s

Pixel Rate

65.97 GPixel/s

339.8 GTexel/s

Texture Rate

122.5 GTexel/s

21.75 TFLOPS

FP16 (half)

-

21.75 TFLOPS

FP32 (float)

3.920 TFLOPS

339.8 GFLOPS

FP64 (double)

122.5 GFLOPS

Graphics Processor

GA104

GPU Name

GM204

GA104-400-A1

GPU Variant

GM204-200-A1

Ampere

Architecture

Maxwell 2.0

Samsung

Foundry

TSMC

8 nm

Process Size

28 nm

17.4 billion

Transistors

5.2 billion

392 mm²

Die Size

398 mm²

Board Design

290W

TDP

148W

600 W

Suggested PSU

300 W

1x HDMI 2.1 3x DisplayPort 1.4a

Outputs

1x DVI 1x HDMI 2.0 3x DisplayPort 1.4a

1x 12-pin

Power Connectors

2x 6-pin

Graphics Features

12 Ultimate (12_2)

DirectX

12 (12_1)

4.6

OpenGL

4.6

3.0

OpenCL

3.0

1.3

Vulkan

1.3

8.6

CUDA

5.2

6.6

Shader Model

6.4

Related GPU Comparisons

1

NVIDIA GeForce RTX 3070 Ti vs NVIDIA GeForce RTX 4060

2

NVIDIA GeForce RTX 3070 Ti vs NVIDIA GeForce RTX 4060 Ti 8 GB

3

NVIDIA GeForce RTX 4050 vs NVIDIA GeForce RTX 3070 Ti

4

NVIDIA GeForce RTX 4070 SUPER vs NVIDIA GeForce RTX 3070 Ti

5

NVIDIA GeForce RTX 4060 Ti vs NVIDIA GeForce RTX 3070 Ti

6

NVIDIA GeForce RTX 4070 Ti vs NVIDIA GeForce RTX 3070 Ti

7

NVIDIA GeForce RTX 3070 Ti vs Intel Arc Pro A50

8

NVIDIA GeForce RTX 3070 Ti vs ATI FirePro V3750

9

NVIDIA GeForce RTX 3070 Ti vs AMD FirePro W9100

10

NVIDIA GeForce RTX 3070 Ti vs NVIDIA GeForce 8400 GS PCI

Related News

1

NVIDIA Grace CPU Benchmark Results Released, Stunning Performance

2

Minisforum Launches UM890 Pro Mini PC with R9 8945HS and Dual USB4 Ports

3

Does the Apple A18 Neural Engine Boast Superior Performance? Computational Power May Surpass M4 Chip

4

AMD's Commitment to Pioneering Each Generation with Cutting-Edge Process Technology and Continuous Innovation in Design, Packaging, and Assembly

5

NVIDIA to Ramp Up Orders and Add New Suppliers to Meet Explosive Demand for Blackwell Products

6

NVIDIA Informs Brands of Reduced Supply for RTX 40 Series GPUs

7

Only 3 Weeks Left! AMD Zen5 Launching Soon

8

Ilya Departs OpenAI to Launch New Company Aiming at Superintelligence!

9

Comprehensive Overview of AMD Ryzen 9000 Zen5 Desktop CPUs

10

Intel May Increase Base Frequencies for Arrow Lake-S / H / HX, High-End Chips to Reach 3GHz or Higher