GeekRadar / The comparison desk

COMPARE.

GPU

NVIDIA

NVIDIA GeForce RTX 4070 10 GB

Graphics processor

Video memory: 10 GB^*
Memory bus: 160 bit^*
Rated memory bandwidth: 420.20 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Ada Lovelace^*
Process node: 5 nm^*
GPU cores: 7,168^*
Graphics interface: PCIe 4.0 x16^*
Reported chip die area: 294 mm²^*
Transistors: 35,800 million^*
GPU L2 cache: 48 MB^*
Graphics processor: AD104^*
Texture mapping units: 224^*
Raster output units: 80^*
Streaming multiprocessors: 56^*
Tensor cores: 224^*
Ray tracing cores: 56^*
Chip foundry: TSMC^*
Card or module format: Dual-slot^*
Declared GPU TDP: 220 W^*
Suggested system PSU: 550 W^*
External power connectors: 1x 16-pin^*
DirectX / feature level: 12 Ultimate (12_2)^*
OpenGL: 4.6^*
OpenCL: 3.0^*
Vulkan: 1.4^*
CUDA compute capability: 8.9^*
Shader model: 6.9^*

Memory & storage

Memory options: 10^*
Video memory: 10 GB^*
Video memory type: GDDR6X^*
Memory bus: 160 bit^*
Rated memory bandwidth: 420.20 GB/s^*

Power specifications

Thermal design power: 220 W^*

System configuration

Detailed specifications

Specifications 1 · 30 fields

proc transistors: 35,800 million^*
proc die size: 294^* mm²
cores: 7,168^*
tensor cores: 224^*
rt cores: 56^*
sm count: 56^*
base clock: 2,355^*
boost clock: 2,520^*
memory clock: 1,313^*
ram: 10^*
name: NVIDIA GeForce RTX 4070 10 GB^*
ram type: GDDR6X^*
bus width: 160^*
bus interface: PCIe 4.0 x16^*
ram bandwidth: 420.2^* GB/s
fp16 performance: 36.13^*
fp32 performance: 36.13^*
fp64 performance: 0.5645^*
tdp: 220^* W
suggested psu: 550^* W
slot width: Dual-slot^*
power connectors: 1x 16-pin^*
api cuda: 8.9^*
is retail board: No^*
gpi value: 13.72^*
vendor: nvd^*
manufacturer: NVIDIA^*
arch name: Ada Lovelace^*
proc foundry: TSMC^*
proc process size: 5^* nm

Specifications 2 · 40 fields

CUDA compute capability

8.9^*

DirectX / feature level

12 Ultimate (12_2)^*

OpenCL

3.0^*

OpenGL

4.6^*

Shader model

6.9^*

Vulkan

1.4^*

Reported theoretical BF16 peak

72.25^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Graphics interface

PCIe 4.0 x16^*

Memory bus

160^* bit

Reported shader processors

7,168^*

Reported theoretical FP16 peak

36.13^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP32 peak

36.13^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP64 peak

0.5645^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Declared L1 cache configuration

128 KB (per SM)^*

Declared GPU L2 cache

48 MB^*

Reported display outputs

1x HDMI 2.1, 3x DisplayPort 1.4a^*

Theoretical pixel fill rate

201.6 GPixel/s^*

External power connectors

1x 16-pin^*

Reported chip die area

294 mm²^*

Chip foundry

TSMC^*

Process node declaration

5 nm^*

Transistors

35.8 billion^*

Graphics processor

AD104^*

Reported video memory

10^* GB

Rated memory bandwidth

420.2 GB/s^*

Video memory type

GDDR6X^*

Raster output units

80^*

Ray tracing cores

56^*

Reported shading units

7,168^*

Card or module format

Dual-slot^*

Streaming multiprocessors

56^*

Suggested system PSU

550 W^*

Declared GPU TDP

220 W^*

Tensor cores

224^*

Theoretical texture fill rate

564.5 GTexel/s^*

Reported theoretical TF32 peak

36.13^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Texture mapping units

224^*

GPU cores

7,168^*

Video memory

10^* GB

Video memory type

GDDR6X^*

* Provisional data