NVIDIA GeForce RTX 4070 Ti vs NVIDIA GeForce RTX 3090 Ti

Comparative analysis of NVIDIA GeForce RTX 4070 Ti and NVIDIA GeForce RTX 3090 Ti videocards for all known characteristics in the following categories: Essentials, Technical info, Video outputs and ports, Compatibility, dimensions and requirements, API support, Memory. Benchmark videocards performance analysis: PassMark - G2D Mark, PassMark - G3D Mark, 3DMark Fire Strike - Graphics Score, Geekbench - OpenCL, CompuBench 1.5 Desktop - Face Detection (mPixels/s), CompuBench 1.5 Desktop - T-Rex (Frames/s), CompuBench 1.5 Desktop - Video Composition (Frames/s), CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s).

NVIDIA GeForce RTX 4070 Ti

NVIDIA GeForce RTX 3090 Ti

Differences

Reasons to consider the NVIDIA GeForce RTX 4070 Ti

Videocard is newer: launch date 11 month(s) later
Around 48% higher core clock speed: 2310 MHz vs 1560 MHz
Around 40% higher boost clock speed: 2610 MHz vs 1860 MHz
A newer manufacturing process allows for a more powerful, yet cooler running videocard: 4 nm vs 8 nm
Around 58% lower typical power consumption: 285 Watt vs 450 Watt
Around 7% better performance in PassMark - G3D Mark: 31655 vs 29502
3.6x better performance in CompuBench 1.5 Desktop - Face Detection (mPixels/s): 939.363 vs 259.78
Around 7% better performance in CompuBench 1.5 Desktop - T-Rex (Frames/s): 67.426 vs 63.049
Around 96% better performance in CompuBench 1.5 Desktop - Video Composition (Frames/s): 348.259 vs 177.665

Specifications (specs)
Launch date	3 Jan 2023 vs 27 Jan 2022
Core clock speed	2310 MHz vs 1560 MHz
Boost clock speed	2610 MHz vs 1860 MHz
Texture fill rate	626.4 GTexel/s vs 625.0 GTexel/s
Manufacturing process technology	4 nm vs 8 nm
Thermal Design Power (TDP)	285 Watt vs 450 Watt
Benchmarks
PassMark - G3D Mark	31655 vs 29502
CompuBench 1.5 Desktop - Face Detection (mPixels/s)	939.363 vs 259.78
CompuBench 1.5 Desktop - T-Rex (Frames/s)	67.426 vs 63.049
CompuBench 1.5 Desktop - Video Composition (Frames/s)	348.259 vs 177.665

Reasons to consider the NVIDIA GeForce RTX 3090 Ti

Around 40% higher pipelines: 10752 vs 7680
2x more maximum memory size: 24 GB vs 12 GB
Around 2% better performance in PassMark - G2D Mark: 1227 vs 1198
Around 14% better performance in 3DMark Fire Strike - Graphics Score: 5741 vs 5024
Around 4% better performance in CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s): 3026.082 vs 2895.895

Specifications (specs)
Pipelines	10752 vs 7680
Maximum memory size	24 GB vs 12 GB
Benchmarks
PassMark - G2D Mark	1227 vs 1198
3DMark Fire Strike - Graphics Score	5741 vs 5024
Geekbench - OpenCL	206084 vs 205069
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)	3026.082 vs 2895.895

Compare benchmarks

GPU 1: NVIDIA GeForce RTX 4070 Ti
GPU 2: NVIDIA GeForce RTX 3090 Ti

PassMark - G2D Mark

GPU 1

GPU 2

1198

1227

PassMark - G3D Mark

GPU 1

GPU 2

31655

29502

3DMark Fire Strike - Graphics Score

GPU 1

GPU 2

5024

5741

Geekbench - OpenCL

GPU 1

GPU 2

205069

206084

CompuBench 1.5 Desktop - Face Detection (mPixels/s)

GPU 1

GPU 2

939.363

259.78

CompuBench 1.5 Desktop - T-Rex (Frames/s)

GPU 1

GPU 2

67.426

63.049

CompuBench 1.5 Desktop - Video Composition (Frames/s)

GPU 1

GPU 2

348.259

177.665

CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)

GPU 1

GPU 2

2895.895

3026.082

Name	NVIDIA GeForce RTX 4070 Ti	NVIDIA GeForce RTX 3090 Ti
PassMark - G2D Mark	1198	1227
PassMark - G3D Mark	31655	29502
3DMark Fire Strike - Graphics Score	5024	5741
Geekbench - OpenCL	205069	206084
CompuBench 1.5 Desktop - Face Detection (mPixels/s)	939.363	259.78
CompuBench 1.5 Desktop - T-Rex (Frames/s)	67.426	63.049
CompuBench 1.5 Desktop - Video Composition (Frames/s)	348.259	177.665
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)	2895.895	3026.082

Compare specifications (specs)

	NVIDIA GeForce RTX 4070 Ti	NVIDIA GeForce RTX 3090 Ti
Essentials
Architecture	Ada Lovelace	Ampere
Code name	AD104	GA102
Launch date	3 Jan 2023	27 Jan 2022
Place in performance rating	8	25
Technical info
Boost clock speed	2610 MHz	1860 MHz
Core clock speed	2310 MHz	1560 MHz
Manufacturing process technology	4 nm	8 nm
Peak Double Precision (FP64) Performance	626.4 GFLOPS (1:64)	625.0 GFLOPS (1:64)
Peak Half Precision (FP16) Performance	40.09 TFLOPS (1:1)	40.00 TFLOPS (1:1)
Peak Single Precision (FP32) Performance	40.09 TFLOPS	40.00 TFLOPS
Pipelines	7680	10752
Pixel fill rate	208.8 GPixel/s	208.3 GPixel/s
Texture fill rate	626.4 GTexel/s	625.0 GTexel/s
Thermal Design Power (TDP)	285 Watt	450 Watt
Transistor count	35800 million	28300 million
Video outputs and ports
Display Connectors	1x HDMI 2.1, 3x DisplayPort 1.4a	1x HDMI 2.1, 3x DisplayPort 1.4a
Compatibility, dimensions and requirements
Form factor	Dual-slot	Triple-slot
Height	42 mm, 1.7 inches	61 mm, 2.4 inches
Interface	PCIe 4.0 x16	PCIe 4.0 x16
Length	285 mm, 11.2 inches	336 mm, 13.2 inches
Recommended system power (PSU)	600 Watt	850 Watt
Supplementary power connectors	1x 16-pin	1x 16-pin
Width	112 mm, 4.4 inches	140 mm, 5.5 inches
API support
DirectX	12 Ultimate (12_2)	12 Ultimate (12_2)
OpenCL	3.0	3.0
OpenGL	4.6	4.6
Shader Model	6.7	6.7
Vulkan
Memory
Maximum RAM amount	12 GB	24 GB
Memory bandwidth	504.2 GB/s	1,008 GB/s
Memory bus width	192 bit	384 bit
Memory clock speed	1313 MHz, 21 Gbps effective	1313 MHz, 21 Gbps effective
Memory type	GDDR6X	GDDR6X