GPU DB

Back to All GPUs

GPU Comparison

Side-by-side comparison of specifications and performance

Performance Comparison

NVIDIA GeForce RTX 4070 SUPER

85%
85%

NVIDIA GeForce RTX 3090 Ti

117%
117%
NVIDIA GeForce RTX 3090 Ti is approximately 32% faster

Technical Specifications

SpecificationNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
Graphics ProcessorDifferent
AD104
GA102
CoresDifferent
7168
10752
TMUsDifferent
224
336
ROPsDifferent
80
112
Memory SizeDifferent
12 GB
24 GB
Memory Type
GDDR6X
GDDR6X
Bus WidthDifferent
192 bit
384 bit

Graphics Processor

FeatureNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
GPU NameDifferent
AD104
GA102
GPU VariantDifferent
AD104-350-A1
GA102-350-A1
ArchitectureDifferent
Ada Lovelace
Ampere
FoundryDifferent
TSMC
Samsung
Process SizeDifferent
5 nm
8 nm
TransistorsDifferent
35,800 million
28,300 million
DensityDifferent
121.8M / mm²
45.1M / mm²
Die SizeDifferent
294 mm²
628 mm²
Chip PackageDifferent
N/A
BGA-3328

Graphics Card

FeatureNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
Release DateDifferent
Jan 8th, 2024
Jan 27th, 2022
AvailabilityDifferent
Jan 17th, 2024
Mar 29th, 2022
GenerationDifferent
GeForce 40
GeForce 30
PredecessorDifferent
GeForce 30
GeForce 20
SuccessorDifferent
GeForce 50
GeForce 40
Production
Active
Active
Launch PriceDifferent
599 USD
1,999 USD
Current PriceDifferent
Amazon/Newegg
N/A
Bus Interface
PCIe 4.0 x16
PCIe 4.0 x16
ReviewsDifferent
42 in our database
31 in our database

Clock Speeds

FeatureNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
Base ClockDifferent
1980 MHz
1560 MHz
Boost ClockDifferent
2475 MHz
1860 MHz
Memory Clock
1313 MHz21 Gbps effective
1313 MHz21 Gbps effective

Memory

FeatureNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
Memory SizeDifferent
12 GB
24 GB
Memory Type
GDDR6X
GDDR6X
Memory BusDifferent
192 bit
384 bit
BandwidthDifferent
504.2 GB/s
1.01 TB/s

Render Config

FeatureNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
Shading UnitsDifferent
7168
10752
TMUsDifferent
224
336
ROPsDifferent
80
112
SM CountDifferent
56
84
Tensor CoresDifferent
224
336
RT CoresDifferent
56
84
L1 Cache
128 KB (per SM)
128 KB (per SM)
L2 CacheDifferent
48 MB
6 MB

Theoretical Performance

FeatureNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
Pixel RateDifferent
198.0 GPixel/s
208.3 GPixel/s
Texture RateDifferent
554.4 GTexel/s
625.0 GTexel/s
FP16 (half)Different
35.48 TFLOPS (1:1)
40.00 TFLOPS (1:1)
FP32 (float)Different
35.48 TFLOPS
40.00 TFLOPS
FP64 (double)Different
554.4 GFLOPS (1:64)
625.0 GFLOPS (1:64)

Board Design

FeatureNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
Slot WidthDifferent
Dual-slot
Triple-slot
LengthDifferent
267 mm10.5 inches
336 mm13.2 inches
HeightDifferent
112 mm4.4 inches
140 mm5.5 inches
WidthDifferent
42 mm1.7 inches
61 mm2.4 inches
TDPDifferent
220 W
450 W
Suggested PSUDifferent
550 W
850 W
Outputs
1x HDMI 2.13x DisplayPort 1.4a
1x HDMI 2.13x DisplayPort 1.4a
Power Connectors
1x 16-pin
1x 16-pin
Board NumberDifferent
PG141 SKU 335
PG136 SKU 5

Graphics Features

FeatureNVIDIA GeForce RTX 4070 SUPERNVIDIA GeForce RTX 3090 Ti
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
OpenCL
3.0
3.0
Vulkan
1.3
1.3
CUDADifferent
8.9
8.6
Shader Model
6.8
6.8

Comparison Summary

7
Specifications Compared
8
Feature Categories
54
Detailed Features

More Comparisons with NVIDIA GeForce RTX 4070 SUPER

Compare NVIDIA GeForce RTX 4070 SUPER with other GPUs of similar performance

More Comparisons with NVIDIA GeForce RTX 3090 Ti

Compare NVIDIA GeForce RTX 3090 Ti with other GPUs of similar performance