AMD Radeon RX 6950 XT vs NVIDIA GeForce RTX 4070

Comparative analysis of AMD Radeon RX 6950 XT and NVIDIA GeForce RTX 4070 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).

AMD Radeon RX 6950 XT

NVIDIA GeForce RTX 4070

Differences

Reasons to consider the AMD Radeon RX 6950 XT

Around 54% higher texture fill rate: 739.2 GTexel/s vs 480.2 GTexel/s
Around 33% higher maximum memory size: 16 GB vs 12 GB
Around 71% higher memory clock speed: 2250 MHz, 18 Gbps effective vs 1313 MHz, 21 Gbps effective
Around 4% better performance in PassMark - G3D Mark: 28130 vs 26966
Around 10% better performance in 3DMark Fire Strike - Graphics Score: 4233 vs 3853
Around 4% better performance in Geekbench - OpenCL: 174317 vs 167932

Specifications (specs)
Texture fill rate	739.2 GTexel/s vs 480.2 GTexel/s
Maximum memory size	16 GB vs 12 GB
Memory clock speed	2250 MHz, 18 Gbps effective vs 1313 MHz, 21 Gbps effective
Benchmarks
PassMark - G3D Mark	28130 vs 26966
3DMark Fire Strike - Graphics Score	4233 vs 3853
Geekbench - OpenCL	174317 vs 167932

Reasons to consider the NVIDIA GeForce RTX 4070

Videocard is newer: launch date 2 year(s) 9 month(s) later
Around 24% higher core clock speed: 2310 MHz vs 1860 MHz
Around 13% higher boost clock speed: 2610 MHz vs 2310 MHz
Around 15% higher pipelines: 5888 vs 5120
A newer manufacturing process allows for a more powerful, yet cooler running videocard: 4 nm vs 7 nm
Around 18% lower typical power consumption: 285 Watt vs 335 Watt
Around 11% better performance in PassMark - G2D Mark: 1157 vs 1047

Specifications (specs)
Launch date	2023 vs 10 May 2022
Core clock speed	2310 MHz vs 1860 MHz
Boost clock speed	2610 MHz vs 2310 MHz
Pipelines	5888 vs 5120
Manufacturing process technology	4 nm vs 7 nm
Thermal Design Power (TDP)	285 Watt vs 335 Watt
Benchmarks
PassMark - G2D Mark	1157 vs 1047

Compare benchmarks

GPU 1: AMD Radeon RX 6950 XT
GPU 2: NVIDIA GeForce RTX 4070

PassMark - G2D Mark

GPU 1

GPU 2

1047

1157

PassMark - G3D Mark

GPU 1

GPU 2

28130

26966

3DMark Fire Strike - Graphics Score

GPU 1

GPU 2

4233

3853

Geekbench - OpenCL

GPU 1

GPU 2

174317

167932

Name	AMD Radeon RX 6950 XT	NVIDIA GeForce RTX 4070
PassMark - G2D Mark	1047	1157
PassMark - G3D Mark	28130	26966
3DMark Fire Strike - Graphics Score	4233	3853
Geekbench - OpenCL	174317	167932
CompuBench 1.5 Desktop - Face Detection (mPixels/s)		417.133
CompuBench 1.5 Desktop - T-Rex (Frames/s)		55.476
CompuBench 1.5 Desktop - Video Composition (Frames/s)		245.639
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)		2459.317

Compare specifications (specs)

	AMD Radeon RX 6950 XT	NVIDIA GeForce RTX 4070
Essentials
Architecture	RDNA 2.0	Ada Lovelace
Code name	Navi 21	AD104
Launch date	10 May 2022	2023
Place in performance rating	33	32
Technical info
Boost clock speed	2310 MHz	2610 MHz
Compute units	80
Core clock speed	1860 MHz	2310 MHz
Manufacturing process technology	7 nm	4 nm
Peak Double Precision (FP64) Performance	1,478 GFLOPS (1:16)	480.2 GFLOPS (1:64)
Peak Half Precision (FP16) Performance	47.31 TFLOPS (2:1)	30.74 TFLOPS (1:1)
Peak Single Precision (FP32) Performance	23.65 TFLOPS	30.74 TFLOPS
Pipelines	5120	5888
Pixel fill rate	295.7 GPixel/s	167.0 GPixel/s
Texture fill rate	739.2 GTexel/s	480.2 GTexel/s
Thermal Design Power (TDP)	335 Watt	285 Watt
Transistor count	26800 million	35800 million
Video outputs and ports
Display Connectors	1x HDMI 2.1, 2x DisplayPort 1.4a	1x HDMI 2.1, 3x DisplayPort 1.4a
Compatibility, dimensions and requirements
Form factor	Triple-slot	Dual-slot
Height	50 mm, 2 inches	61 mm, 2.4 inches
Interface	PCIe 4.0 x16	PCIe 4.0 x16
Length	267 mm, 10.5 inches	336 mm, 13.2 inches
Recommended system power (PSU)	700 Watt	600 Watt
Supplementary power connectors	2x 8-pin	1x 12-pin
Width	120 mm, 4.7 inches	140 mm, 5.5 inches
API support
DirectX	12 Ultimate (12_2)	12 Ultimate (12_2)
OpenCL	2.1	3.0
OpenGL	4.6	4.6
Shader Model	6.7	6.7
Vulkan
Memory
Maximum RAM amount	16 GB	12 GB
Memory bandwidth	576.0 GB/s	504.2 GB/s
Memory bus width	256 bit	192 bit
Memory clock speed	2250 MHz, 18 Gbps effective	1313 MHz, 21 Gbps effective
Memory type	GDDR6	GDDR6X