Home
GPUs
GeForce RTX 3070

GPU • NVIDIA • AMPERE

GeForce RTX 3070

Q: When was the GeForce RTX 3070 released?

The GeForce RTX 3070 was officially released in October 29, 2020. It was launched as a graphics solution by NVIDIA.

Q: What is the 3DMark Time Spy score of GeForce RTX 3070?

The GeForce RTX 3070 scores 13,644 on 3DMark Time Spy, indicating its solid capabilities in modern gaming scenarios. This benchmark is the industry standard for measuring DirectX 12 gaming performance.

Q: How much VRAM does the GeForce RTX 3070 have?

The GeForce RTX 3070 is equipped with 8 GB of VRAM, which is currently the sweet spot for smooth 1440p gaming and modern texture quality.

Q: What is the TGP (power consumption) of GeForce RTX 3070?

The GeForce RTX 3070 has a Total Graphics Power (TGP) of 220 W. This value dictates your power supply (PSU) requirements, so ensure your system can handle this draw effectively.

Q: What is the base clock speed of the GeForce RTX 3070?

The GeForce RTX 3070 operates at a base clock speed of 1500 MHz. This is the minimum clock frequency the graphics card runs at under standard workloads before boosting.

Q: What is the boost clock speed of the GeForce RTX 3070?

The GeForce RTX 3070 features a boost clock speed of 1725 MHz. This clock speed represents the maximum frequency the graphics card can dynamically reach during intense gaming or rendering workloads.

Q: What architecture is the GeForce RTX 3070 based on?

The GeForce RTX 3070 is built on the Ampere architecture. This foundational design determines the GPU's power efficiency and support for modern features like hardware Ray Tracing and AI upscaling (DLSS/FSR).

Q: How does the GeForce RTX 3070 rank in gaming performance?

With a performance score of 70%, the GeForce RTX 3070 is ranked as a high-performance graphics card in our extensive GPU database.

Q: What bus interface does the GeForce RTX 3070 use?

The GeForce RTX 3070 uses the PCIe 4.0 x16 interface, ensuring high-speed data transfer and minimal bottlenecking between the graphics card and the motherboard.

Q: How many shading units does the GeForce RTX 3070 have?

The GeForce RTX 3070 features 5,888 shading units (cores), which are directly responsible for rendering complex visual effects and lighting in modern games.

GA104 Released: October 29, 2020

Compare

GeForce RTX 3070

GA104 Released: October 29, 2020

Memory

GB GDDR6

Boost Clock

1725

MHz

TGP

220

Technical Highlights

Architecture : Ampere

Generation : GeForce 30

Shading Units : 5888

RT Cores : 46

About GeForce RTX 3070

GeForce RTX 3070 is a discrete graphics card from NVIDIA released on October 29, 2020 based on the Ampere architecture with 8 GB GDDR6 memory on a PCIe 4.0 x16 bus interface. It runs at 1500 MHz base clock, reaches up to 1725 MHz boost clock, and has a 220 W power rating. It supports DirectX 12 Ultimate (12_2), Vulkan 1.3, and OpenGL 4.6.

Last updated: August 07, 2026

Jump to

Overall Scores General Info Graphics Processor Raw Performance Memory Board Design APIs Benchmarks Performance Charts FAQ

Overall Score Of GeForce RTX 3070

This summarizes how GeForce RTX 3070 scores across key areas like gaming, rendering, AI, and Energy Efficiency.

Overall

Gaming Performance

AI & Machine Learning

Hardware Architecture

Workstation

Energy Efficiency

General Info of GeForce RTX 3070

Vendor

NVIDIA

Release Date

October 29, 2020

Foundry

Samsung

Architecture

Ampere

Process Size

Good

8 nm

50% of the current best Process Size score

Transistors

17 billion

Launch Price

499

Graphics Processor of GeForce RTX 3070

Base Clock

1500 MHz

56% of the current best Base Clock score

Boost Clock

1725 MHz

55% of the current best Boost Clock score

Shading Units

5888

27% of the current best Shading Units score

Compute Units

27% of the current best Compute Units score

Ray Tracing Cores (RTCs)

27% of the current best Ray Tracing Cores (RTCs) score

Tensor Cores

184

27% of the current best Tensor Cores score

Textured Mapping Units (TMUs)

184

27% of the current best Textured Mapping Units (TMUs) score

Render Output Units (ROPs)

38% of the current best Render Output Units (ROPs) score

Die Size

392 mm²

Density

43.37 MTr/mm²

L1 Cache

128 KB (per SM)

50% of the current best L1 Cache score

L2 Cache

4 MB

4% of the current best L2 Cache score

Raw Performance of GeForce RTX 3070

Pixel Fill Rate

165.6 GPixel/s

35% of the current best Pixel Fill Rate score

Texture Fill Rate

317.4 GTexel/s

19% of the current best Texture Fill Rate score

FP32 Performance

20.31 TFLOPs

19% of the current best FP32 Performance score

Memory of GeForce RTX 3070

Memory Size

Good

8 GB

25% of the current best Memory Size score

Memory Type

Good

GDDR6

70% of the current best Memory Type score

Bus Width

256-bit

Memory Clock

Good

1750 MHz

67% of the current best Memory Clock score

Effective Speed

14000 Mbps

47% of the current best Effective Speed score

Bandwidth

448 GB/s

25% of the current best Bandwidth score

Board Design of GeForce RTX 3070

Bus Interface

Excellent

PCIe 4.0 x16

80% of the current best Bus Interface score

PCIe Version

Excellent

4.0

80% of the current best PCIe Version score

PCIe Lanes

TGP

220 W

Length

242 mm

Width

40 mm

Height

112 mm

APIs of GeForce RTX 3070

DirectX

Top tier

12 Ultimate (12_2)

100% of the current best DirectX score

OpenGL

Top tier

4.6

100% of the current best OpenGL score

OpenCL

Top tier

3.0

100% of the current best OpenCL score

Vulkan

Top tier

1.3

91% of the current best Vulkan score

HDMI

Top tier

2.1

100% of the current best HDMI score

DisplayPort

Excellent

1.4a

80% of the current best DisplayPort score

CUDA

8.6

Ray Tracing

DLSS

DLSS 2

Benchmarks of GeForce RTX 3070

3D Mark

A standardized suite that approximates real-world gaming workloads to compare graphics performance.

Steel Nomad Lite:

14152

27% of max score 27% of best score for Steel Nomad Lite

Time Spy:

13644 (37th)

29% of max score 29% of best score for Time Spy

Solar Bay:

64452

28% of max score 28% of best score for Solar Bay

Port Royal:

8312

22% of max score 22% of best score for Port Royal

Fire Strike:

34056

39% of max score 39% of best score for Fire Strike

Wild Life Extreme:

28488

26% of max score 26% of best score for Wild Life Extreme

Night Raid:

122159

59% of max score 59% of best score for Night Raid

GeekBench 6 OpenCL

GPU compute benchmark focused on workloads like image processing and parallel compute tasks.

Compute Score:

126195 (39th)

30% of max score 30% of best score for Compute Score

Background Blur:

161.6 img/sec

38% of max score 38% of best score for Background Blur

Face Detection:

117.2 img/sec

39% of max score 39% of best score for Face Detection

Horizon Detection:

5.89 Gpixels/sec

27% of max score 27% of best score for Horizon Detection

Edge Detection:

8.05 Gpixels/sec

24% of max score 24% of best score for Edge Detection

Gaussian Blur:

6.65 Gpixels/sec

19% of max score 19% of best score for Gaussian Blur

Feature Matching:

1.25 Gpixels/sec

49% of max score 49% of best score for Feature Matching

Stereo Matching:

550.6 Gpixels/sec

19% of max score 19% of best score for Stereo Matching

Particle Physics:

17613.1 FPS

30% of max score 30% of best score for Particle Physics

Passmark Graphics

A multi-test benchmark that evaluates raster, memory bandwidth, and compute performance.

G3D Mark Score:

22112 (33rd)

57% of max score 57% of best score for G3D Mark Score

G2D Mark Score:

1005

71% of max score 71% of best score for G2D Mark Score

Good

DirectX 12:

47% of max score 47% of best score for DirectX 12

GPU Compute:

11039 Ops/s

42% of max score 42% of best score for GPU Compute

GFXBench 5

Graphics benchmark using modern GPU APIs and game-like scenes across multiple workloads.

Aztec Ruins High (4K):

170.20 FPS

14% of max score 14% of best score for Aztec Ruins High (4K)

Aztec Ruins Normal:

856.00 FPS

44% of max score 44% of best score for Aztec Ruins Normal

Car Chase:

668.00 FPS

39% of max score 39% of best score for Car Chase

Manhattan 3.1 (1440p):

592.90 FPS

15% of max score 15% of best score for Manhattan 3.1 (1440p)

T-Rex:

3078.40 FPS

46% of max score 46% of best score for T-Rex

Blender

Rendering performance benchmark using Blender Cycles throughput as a practical content-creation metric.

Render Score:

3364 (31st)

22% of max score 22% of best score for Render Score

GeekBench 6 ML

AI benchmark measuring common ML workloads like classification, detection, and super resolution.

ML Single Precision:

19789 (45th)

37% of max score 37% of best score for ML Single Precision

ML Half Precision:

33013

41% of max score 41% of best score for ML Half Precision

ML Quantized:

12443

32% of max score 32% of best score for ML Quantized

Image Classification:

7624

40% of max score 40% of best score for Image Classification

Object Detection:

10304

37% of max score 37% of best score for Object Detection

Image Segmentation:

23420

42% of max score 42% of best score for Image Segmentation

Face Detection:

40535

35% of max score 35% of best score for Face Detection

Image Super Resolution:

17898

32% of max score 32% of best score for Image Super Resolution

Style Transfer:

180089

26% of max score 26% of best score for Style Transfer

Text Classification:

2724

52% of max score 52% of best score for Text Classification

Machine Translation:

4481

52% of max score 52% of best score for Machine Translation

Depth Estimation:

30418

36% of max score 36% of best score for Depth Estimation

Pose Estimation:

83624

24% of max score 24% of best score for Pose Estimation

FurMark

Stress test designed to push sustained load for thermal stability and cooling behavior.

1080p Vulkan:

10183

30% of max score 30% of best score for 1080p Vulkan

1440p Vulkan:

6735

26% of max score 26% of best score for 1440p Vulkan

2160p Vulkan:

3282

23% of max score 23% of best score for 2160p Vulkan

1080p OpenGL:

9063

31% of max score 31% of best score for 1080p OpenGL

1440p OpenGL:

6557

28% of max score 28% of best score for 1440p OpenGL

2160p OpenGL:

3419

22% of max score 22% of best score for 2160p OpenGL

Performance Charts Of GeForce RTX 3070

These charts give a quick view of FPS scaling across resolutions and how this GPU compares against market top benchmark values.

Gaming FPS Analysis

Ultra settings performance of GeForce RTX 3070

1080p Ultra

1440p Ultra

4K Ultra

Performance Benchmarks

Grey is the market top value. Colored bar is GeForce RTX 3070.

Market Top

GeForce RTX 3070

Try to Compare GeForce RTX 3070 With Another GPU

Search for another gpu to compare.

GPU 1

Select GPU 2

More GPUs You Might Like to See

If you are still deciding, these are great options to look at next.

Radeon RX 590 GME

AMD

GeForce RTX 2080 SUPER Mobile

NVIDIA

Radeon RX 5600 XT

AMD

GeForce RTX 3060 Ti

NVIDIA

GeForce RTX 3080

NVIDIA

Radeon RX 6800

AMD

GeForce RTX 3090

NVIDIA

Radeon RX 6800 XT

AMD

Radeon RX 6900 XT

AMD

GeForce RTX 3050 Ti Laptop

NVIDIA

Discover More GPUs

Frequently Asked Questions about GeForce RTX 3070

The GeForce RTX 3070 was officially released in October 29, 2020. It was launched as a graphics solution by NVIDIA.

The GeForce RTX 3070 scores 13,644 on 3DMark Time Spy, indicating its solid capabilities in modern gaming scenarios. This benchmark is the industry standard for measuring DirectX 12 gaming performance.

The GeForce RTX 3070 is equipped with 8 GB of VRAM, which is currently the sweet spot for smooth 1440p gaming and modern texture quality.

The GeForce RTX 3070 has a Total Graphics Power (TGP) of 220 W. This value dictates your power supply (PSU) requirements, so ensure your system can handle this draw effectively.

The GeForce RTX 3070 operates at a base clock speed of 1500 MHz. This is the minimum clock frequency the graphics card runs at under standard workloads before boosting.

The GeForce RTX 3070 features a boost clock speed of 1725 MHz. This clock speed represents the maximum frequency the graphics card can dynamically reach during intense gaming or rendering workloads.

The GeForce RTX 3070 is built on the Ampere architecture. This foundational design determines the GPU's power efficiency and support for modern features like hardware Ray Tracing and AI upscaling (DLSS/FSR).

With a performance score of 70%, the GeForce RTX 3070 is ranked as a high-performance graphics card in our extensive GPU database.

The GeForce RTX 3070 uses the PCIe 4.0 x16 interface, ensuring high-speed data transfer and minimal bottlenecking between the graphics card and the motherboard.

The GeForce RTX 3070 features 5,888 shading units (cores), which are directly responsible for rendering complex visual effects and lighting in modern games.

The GeForce RTX 3070 contains approximately 17 billion transistors, packed into its silicon to deliver massive parallel processing capabilities.

The GeForce RTX 3070 is manufactured on a 8 nm node. A smaller process node generally translates to better energy efficiency and improved thermal management.