NVIDIA GeForce RTX 4070 vs NVIDIA GeForce RTX 3070 Ti

NVIDIA GeForce RTX 4070 $599
NVIDIA GeForce RTX 3070 Ti $600
5888 Shaders
12GB GDDR6X
2475MHz
6144 Shaders
8GB GDDR6X
1770MHz
Peak AI Performance
932.66 TOPS
INT4 Tensor Sparse
Peak AI Performance
695.99 TOPS
INT4 Tensor Sparse
FP32
29.15 TFLOPS
FP32
21.75 TFLOPS
FP16
29.15 TFLOPS
FP16
21.75 TFLOPS
Form Factor
PCIe Card
2.1-Slots
Form Factor
PCIe Card
2.1-Slots
TDP
200W
TDP
290W
Power Connectors
-
1x 16-Pin 12VHPWR
Power Connectors
1x 12-Pin
-

Peak AI Performance

  • 34% faster vs GeForce RTX 3070 Ti
  • 25% slower vs GeForce RTX 4070
GeForce RTX 4070 - 932.66 TOPS INT4 Tensor Sparse
x1.34
GeForce RTX 3070 Ti - 695.99 TOPS INT4 Tensor Sparse
x1

FP32

  • 34% faster vs GeForce RTX 3070 Ti
  • 25% slower vs GeForce RTX 4070
GeForce RTX 4070 - 29.15 TFLOPS FP32
x1.34
GeForce RTX 3070 Ti - 21.75 TFLOPS FP32
x1

FP16

  • 34% faster vs GeForce RTX 3070 Ti
  • 25% slower vs GeForce RTX 4070
GeForce RTX 4070 - 29.15 TFLOPS FP16
x1.34
GeForce RTX 3070 Ti - 21.75 TFLOPS FP16
x1
  • 17% slower vs GeForce RTX 3070 Ti
  • 21% faster vs GeForce RTX 4070
GeForce RTX 4070 - 504GB/s
x1
GeForce RTX 3070 Ti - 608GB/s
x1.21
  • 31% lower vs GeForce RTX 3070 Ti
  • 45% higher vs GeForce RTX 4070
GeForce RTX 4070 - 200W
x1
GeForce RTX 3070 Ti - 290W
x1.45
  • 4% faster vs GeForce RTX 3070 Ti
  • 4% slower vs GeForce RTX 4070
GeForce RTX 4070 - GB6 OpenCL 137,755
x1.04
GeForce RTX 3070 Ti - GB6 OpenCL 132,100
x1
GB5 OpenCL N/A
0%
GB5 OpenCL 153,225
50%
GB5 CUDA N/A
0%
GB5 CUDA 166,375
47%
GB5 Vulkan N/A
0%
GB5 Vulkan 95,755
47%
OCT 2020.1 N/A
0%
OCT 2020.1 455
60%
Manufacturer
NVIDIA
Manufacturer
NVIDIA
Chip Designer
NVIDIA
Chip Designer
NVIDIA
Architecture
Ada Lovelace
Architecture
Ampere
Family
GeForce 40
Family
GeForce 30
Codename
NV184
AD104
Variant
AD104-250
Codename
NV174
GA104
Variant
GA104-400-A1
Market Segment
Desktop
Market Segment
Desktop
Release Date
4/13/2023
Release Date
6/10/2021
Foundry
TSMC
Foundry
Samsung
Fabrication Node
4N
Fabrication Node
8N
Die Size
295 mm²
Die Size
393 mm²
Transistor Count
35.8 Billion
Transistor Count
17.4 Billion
Transistor Density
121.56M/mm²
Transistor Density
44.33M/mm²
Form
PCIe Card
Form
PCIe Card
Shading Units
5888 Shaders
Shading Units
6144 Shaders
Texture Mapping Units
184 TMUs
Texture Mapping Units
192 TMUs
Render Output Units
64 ROPs
Render Output Units
96 ROPs
Tensor Cores
184 T-Cores
Tensor Cores
192 T-Cores
Ray-Tracing Cores
46 RT-Cores
Ray-Tracing Cores
48 RT-Cores
Streaming Multiprocessors
46 SMs
Streaming Multiprocessors
48 SMs
Graphics Processing Clusters
4 GPCs
Graphics Processing Clusters
6 GPCs
1920MHz Base
2475MHz
1575MHz Base
1770MHz
Peak AI Performance
932.66 TOPS
INT4 Tensor Sparse
Peak AI Performance
695.99 TOPS
INT4 Tensor Sparse
FP8
233.17 TFLOPS Tensor (FP16 Accumulate)
466.33 TFLOPS Tensor (FP16 Accumulate) Sparse
116.58 TFLOPS Tensor (FP32 Accumulate)
233.16 TFLOPS Tensor (FP32 Accumulate) Sparse
-
-
-
-
-
FP16
29.15 TFLOPS
116.58 TFLOPS Tensor (FP16 Accumulate)
233.16 TFLOPS Tensor (FP16 Accumulate) Sparse
58.29 TFLOPS Tensor (FP32 Accumulate)
116.58 TFLOPS Tensor (FP32 Accumulate) Sparse
FP16
21.75 TFLOPS
87 TFLOPS Tensor (FP16 Accumulate)
174 TFLOPS Tensor (FP16 Accumulate) Sparse
43.5 TFLOPS Tensor (FP32 Accumulate)
87 TFLOPS Tensor (FP32 Accumulate) Sparse
FP32
29.15 TFLOPS
FP32
21.75 TFLOPS
FP64
460 GFLOPS
FP64
340 GFLOPS
BF16
29.15 TFLOPS
58.29 TFLOPS Tensor
116.58 TFLOPS Tensor Sparse
BF16
21.75 TFLOPS
43.5 TFLOPS Tensor
87 TFLOPS Tensor Sparse
TF32
29.15 TFLOPS Tensor
58.29 TFLOPS Tensor Sparse
TF32
21.75 TFLOPS Tensor
43.5 TFLOPS Tensor Sparse
INT4
466.33 TOPS Tensor
932.66 TOPS Tensor Sparse
INT4
348 TOPS Tensor
695.99 TOPS Tensor Sparse
INT8
233.17 TOPS Tensor
466.33 TOPS Tensor Sparse
INT8
174 TOPS Tensor
348 TOPS Tensor Sparse
INT32
14.57 TOPS
INT32
10.88 TOPS
Ray Tracing
67.4 TOPS
Ray Tracing
42.5 TOPS
Pixel Fillrate
158.4 GPixel/s
Pixel Fillrate
169.92 GPixel/s
Texture Fillrate
455.4 GTexel/s
Texture Fillrate
339.84 GTexel/s
L1
64KB/SM Tex
128KB/SM
L1
64KB/SM Tex
128KB/SM
L2
36MB Shared
L2
4MB Shared
12GB
GDDR6X
8GB
GDDR6X
Bus Width
192Bit
Bus Width
256Bit
Clock
1313MHz
Transfer Rate
21GT/s
Bandwidth
504GB/s
Clock
1188MHz
Transfer Rate
19GT/s
Bandwidth
608GB/s
TDP
200W
TDP
290W
-
-
Temp
93°C Max
3x DisplayPort 1.4
1x HDMI 2.1
3x DisplayPort 1.4
1x HDMI 2.1
Max Resolution
7680x4320
Max Resolution
7680x4320
Max Resolution Refresh Rate
60Hz
Max Resolution Refresh Rate
60Hz
Variable Refresh Rate
G-Sync
FreeSync
Variable Refresh Rate
G-Sync
FreeSync
Display Stream Compression (DSC)
Supported
Display Stream Compression (DSC)
Supported
Multi Monitor Support
4
Multi Monitor Support
4
Content Protection
HDCP 2.3
Content Protection
HDCP 2.3
Model
2x NVENC 8
Model
NVENC 7
Codec
AVC (H.264)
HEVC (H.265)
AV1
Codec
AVC (H.264)
HEVC (H.265)
-
Model
NVDEC 5
Model
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
Direct X
12
Direct 3D
12_3
Direct X
12
Direct 3D
12_2
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
OpenGL
4.6
OpenCL
3.0
Vulkan
1.2
Shader Model
6.7
CUDA
8.9
PureVideo HD
VP12
VDPAU
Feature Set L
Shader Model
6.6
CUDA
8.6
PureVideo HD
VP11
VDPAU
Feature Set K
2x Fans
2x Fans
Power Connectors
-
1x 16-Pin 12VHPWR
Power Connectors
1x 12-Pin
-
Slots Required
2.1
PCIe Version
4.0
PCIe Lanes
16
Slots Required
2.1
PCIe Version
4.0
PCIe Lanes
16
Height
112 mm (4.41 in)
Width
244 mm (9.61 in)
Depth
42 mm (1.65 in)
Height
112 mm (4.41 in)
Width
267 mm (10.51 in)
Depth
43 mm (1.69 in)
Change Comparison