GeForce RTX 3070 vs Radeon RX 6900 XT
Aggregate performance score
We've compared Radeon RX 6900 XT and GeForce RTX 3070, covering specs and all relevant benchmarks.
RX 6900 XT outperforms RTX 3070 by a significant 20% based on our aggregate benchmark results.
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 32 | 58 |
| Place by popularity | not in top-100 | 34 |
| Cost-effectiveness evaluation | 31.78 | 51.86 |
| Power efficiency | 16.09 | 18.22 |
| Architecture | RDNA 2.0 (2020−2025) | Ampere (2020−2025) |
| GPU code name | Navi 21 | GA104 |
| Market segment | Desktop | Desktop |
| Release date | 28 October 2020 (4 years ago) | 1 September 2020 (4 years ago) |
| Launch price (MSRP) | $999 | $499 |
Cost-effectiveness evaluation
The higher the performance-to-price ratio, the better. We use the manufacturer's recommended prices for comparison.
RTX 3070 has 63% better value for money than RX 6900 XT.
Performance to price scatter graph
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.
| Pipelines / CUDA cores | 5120 | 5888 |
| Core clock speed | 1825 MHz | 1500 MHz |
| Boost clock speed | 2250 MHz | 1725 MHz |
| Number of transistors | 26,800 million | 17,400 million |
| Manufacturing process technology | 7 nm | 8 nm |
| Power consumption (TDP) | 300 Watt | 220 Watt |
| Texture fill rate | 720.0 | 317.4 |
| Floating-point processing power | 23.04 TFLOPS | 20.31 TFLOPS |
| ROPs | 128 | 96 |
| TMUs | 320 | 184 |
| Tensor Cores | no data | 184 |
| Ray Tracing Cores | 80 | 46 |
Form factor & compatibility
Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| Length | 267 mm | 242 mm |
| Width | 3-slot | 2-slot |
| Supplementary power connectors | 2x 8-pin | 1x 12-pin |
VRAM capacity and type
Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.
| Memory type | GDDR6 | GDDR6 |
| Maximum RAM amount | 16 GB | 8 GB |
| Memory bus width | 256 Bit | 256 Bit |
| Memory clock speed | 2000 MHz | 1750 MHz |
| Memory bandwidth | 512.0 GB/s | 448.0 GB/s |
| Shared memory | - | - |
| Resizable BAR | + | + |
Connectivity and outputs
Types and number of video connectors present on the reviewed GPUs. As a rule, data in this section is precise only for desktop reference ones (so-called Founders Edition for NVIDIA chips). OEM manufacturers may change the number and type of output ports, while for notebook cards availability of certain video outputs ports depends on the laptop model rather than on the card itself.
| Display Connectors | 1x HDMI, 2x DisplayPort, 1x USB Type-C | 1x HDMI, 3x DisplayPort |
| HDMI | + | + |
API and SDK compatibility
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
| Shader Model | 6.5 | 6.5 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 2.0 |
| Vulkan | 1.2 | 1.2 |
| CUDA | - | 8.5 |
| DLSS | - | + |
Synthetic benchmark performance
Non-gaming benchmark results comparison. The combined score is measured on a 0-100 point scale.
Combined synthetic benchmark score
This is our combined benchmark score.
Passmark
This is the most ubiquitous GPU benchmark. It gives the graphics card a thorough evaluation under various types of load, providing four separate benchmarks for Direct3D versions 9, 10, 11 and 12 (the last being done in 4K resolution if possible), and few more tests engaging DirectCompute capabilities.
3DMark 11 Performance GPU
3DMark 11 is an obsolete DirectX 11 benchmark by Futuremark. It used four tests based on two scenes, one being few submarines exploring the submerged wreck of a sunken ship, the other is an abandoned temple deep in the jungle. All the tests are heavy with volumetric lighting and tessellation, and despite being done in 1280x720 resolution, are relatively taxing. Discontinued in January 2020, 3DMark 11 is now superseded by Time Spy.
3DMark Fire Strike Graphics
Fire Strike is a DirectX 11 benchmark for gaming PCs. It features two separate tests displaying a fight between a humanoid and a fiery creature made of lava. Using 1920x1080 resolution, Fire Strike shows off some realistic graphics and is quite taxing on hardware.
Gaming performance
Let's see how good the compared graphics cards are for gaming. Particular gaming benchmark results are measured in FPS.
Average FPS across all PC games
Here are the average frames per second in a large set of popular games across different resolutions:
| Full HD | 196
+32.4%
| 148
−32.4%
|
| 1440p | 134
+35.4%
| 99
−35.4%
|
| 4K | 83
+31.7%
| 63
−31.7%
|
Cost per frame, $
| 1080p | 5.10
−51.2%
| 3.37
+51.2%
|
| 1440p | 7.46
−47.9%
| 5.04
+47.9%
|
| 4K | 12.04
−52%
| 7.92
+52%
|
- RTX 3070 has 51% lower cost per frame in 1080p
- RTX 3070 has 48% lower cost per frame in 1440p
- RTX 3070 has 52% lower cost per frame in 4K
FPS performance in popular games
Full HD
Low Preset
| Counter-Strike 2 | 300−350
+11.2%
|
270−280
−11.2%
|
| Cyberpunk 2077 | 160−170
+9.5%
|
147
−9.5%
|
| Sons of the Forest | 120−130
+5.2%
|
115
−5.2%
|
Full HD
Medium Preset
| Battlefield 5 | 195
+30.9%
|
149
−30.9%
|
| Counter-Strike 2 | 300−350
−7.5%
|
330
+7.5%
|
| Cyberpunk 2077 | 160−170
+15.8%
|
139
−15.8%
|
| Far Cry 5 | 190−200
+25.3%
|
154
−25.3%
|
| Fortnite | 300−350
+27.4%
|
230−240
−27.4%
|
| Forza Horizon 4 | 283
+37.4%
|
200−210
−37.4%
|
| Forza Horizon 5 | 190−200
+20.1%
|
159
−20.1%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
+0%
|
170−180
+0%
|
| Sons of the Forest | 120−130
+16.3%
|
104
−16.3%
|
| Valorant | 350−400
+23.9%
|
290−300
−23.9%
|
Full HD
High Preset
| Battlefield 5 | 196
+48.5%
|
132
−48.5%
|
| Counter-Strike 2 | 300−350
+19.5%
|
257
−19.5%
|
| Counter-Strike: Global Offensive | 270−280
+0%
|
270−280
+0%
|
| Cyberpunk 2077 | 160−170
+27.8%
|
126
−27.8%
|
| Dota 2 | 170−180
+27.8%
|
133
−27.8%
|
| Far Cry 5 | 190−200
+30.4%
|
148
−30.4%
|
| Fortnite | 300−350
+27.4%
|
230−240
−27.4%
|
| Forza Horizon 4 | 279
+35.4%
|
200−210
−35.4%
|
| Forza Horizon 5 | 190−200
+29.1%
|
148
−29.1%
|
| Grand Theft Auto V | 160−170
+20.9%
|
139
−20.9%
|
| Metro Exodus | 164
+36.7%
|
120
−36.7%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
+0%
|
170−180
+0%
|
| Sons of the Forest | 120−130
+17.5%
|
103
−17.5%
|
| The Witcher 3: Wild Hunt | 323
+40.4%
|
230
−40.4%
|
| Valorant | 350−400
+23.9%
|
290−300
−23.9%
|
Full HD
Ultra Preset
| Battlefield 5 | 197
+65.5%
|
119
−65.5%
|
| Cyberpunk 2077 | 160−170
+57.8%
|
102
−57.8%
|
| Dota 2 | 170−180
+36%
|
125
−36%
|
| Far Cry 5 | 190−200
+36.9%
|
141
−36.9%
|
| Forza Horizon 4 | 248
+20.4%
|
200−210
−20.4%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
+0%
|
170−180
+0%
|
| Sons of the Forest | 120−130
+18.6%
|
102
−18.6%
|
| The Witcher 3: Wild Hunt | 164
+35.5%
|
121
−35.5%
|
| Valorant | 411
+73.4%
|
237
−73.4%
|
Full HD
Epic Preset
| Fortnite | 300−350
+27.4%
|
230−240
−27.4%
|
1440p
High Preset
| Counter-Strike 2 | 190−200
+16.8%
|
167
−16.8%
|
| Counter-Strike: Global Offensive | 500−550
+29.4%
|
350−400
−29.4%
|
| Grand Theft Auto V | 130−140
+38.8%
|
98
−38.8%
|
| Metro Exodus | 102
+36%
|
75
−36%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
+0%
|
170−180
+0%
|
| Valorant | 400−450
+33.2%
|
300−350
−33.2%
|
1440p
Ultra Preset
| Battlefield 5 | 196
+90.3%
|
103
−90.3%
|
| Cyberpunk 2077 | 90−95
+48.4%
|
62
−48.4%
|
| Far Cry 5 | 160−170
+32%
|
125
−32%
|
| Forza Horizon 4 | 231
+36.7%
|
160−170
−36.7%
|
| Sons of the Forest | 100−110
+24.4%
|
82
−24.4%
|
| The Witcher 3: Wild Hunt | 150−160
+31.9%
|
110−120
−31.9%
|
1440p
Epic Preset
| Fortnite | 150−160
+0.7%
|
150−160
−0.7%
|
4K
High Preset
| Counter-Strike 2 | 85−90
+105%
|
43
−105%
|
| Grand Theft Auto V | 150−160
+33.3%
|
117
−33.3%
|
| Metro Exodus | 67
+36.7%
|
49
−36.7%
|
| The Witcher 3: Wild Hunt | 122
+35.6%
|
90
−35.6%
|
| Valorant | 300−350
+7.5%
|
300−350
−7.5%
|
4K
Ultra Preset
| Battlefield 5 | 134
+91.4%
|
70
−91.4%
|
| Counter-Strike 2 | 85−90
+27.5%
|
65−70
−27.5%
|
| Cyberpunk 2077 | 40−45
+46.7%
|
30
−46.7%
|
| Dota 2 | 150−160
+27.2%
|
125
−27.2%
|
| Far Cry 5 | 100−110
+48.6%
|
70
−48.6%
|
| Forza Horizon 4 | 162
+35%
|
120−130
−35%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 95−100
+4.3%
|
90−95
−4.3%
|
| Sons of the Forest | 70−75
+41.2%
|
51
−41.2%
|
4K
Epic Preset
| Fortnite | 75−80
+1.3%
|
75−80
−1.3%
|
This is how RX 6900 XT and RTX 3070 compete in popular games:
- RX 6900 XT is 32% faster in 1080p
- RX 6900 XT is 35% faster in 1440p
- RX 6900 XT is 32% faster in 4K
Here's the range of performance differences observed across popular games:
- in Counter-Strike 2, with 4K resolution and the High Preset, the RX 6900 XT is 105% faster.
- in Counter-Strike 2, with 1080p resolution and the Medium Preset, the RTX 3070 is 7% faster.
All in all, in popular games:
- RX 6900 XT is ahead in 59 tests (91%)
- RTX 3070 is ahead in 1 test (2%)
- there's a draw in 5 tests (8%)
Pros & cons summary
| Performance score | 61.93 | 51.45 |
| Recency | 28 October 2020 | 1 September 2020 |
| Maximum RAM amount | 16 GB | 8 GB |
| Chip lithography | 7 nm | 8 nm |
| Power consumption (TDP) | 300 Watt | 220 Watt |
RX 6900 XT has a 20.4% higher aggregate performance score, an age advantage of 1 month, a 100% higher maximum VRAM amount, and a 14.3% more advanced lithography process.
RTX 3070, on the other hand, has 36.4% lower power consumption.
The Radeon RX 6900 XT is our recommended choice as it beats the GeForce RTX 3070 in performance tests.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
