GeForce RTX 5090 vs RTX 3070
Aggregate performance score
We've compared GeForce RTX 3070 and GeForce RTX 5090, covering specs and all relevant benchmarks.
RTX 5090 outperforms RTX 3070 by an impressive 80% based on our aggregate benchmark results.
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 45 | 1 |
| Place by popularity | 40 | 59 |
| Cost-effectiveness evaluation | 57.81 | 11.23 |
| Power efficiency | 18.17 | 12.48 |
| Architecture | Ampere (2020−2024) | Blackwell 2.0 (2025) |
| GPU code name | GA104 | GB202 |
| Market segment | Desktop | Desktop |
| Release date | 1 September 2020 (4 years ago) | 30 January 2025 (recently) |
| Launch price (MSRP) | $499 | $1,999 |
Cost-effectiveness evaluation
The higher the performance-to-price ratio, the better. We use the manufacturer's recommended prices for comparison.
RTX 3070 has 415% better value for money than RTX 5090.
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 | 5888 | 21760 |
| Core clock speed | 1500 MHz | 2017 MHz |
| Boost clock speed | 1725 MHz | 2407 MHz |
| Number of transistors | 17,400 million | 92,200 million |
| Manufacturing process technology | 8 nm | 5 nm |
| Power consumption (TDP) | 220 Watt | 575 Watt |
| Texture fill rate | 317.4 | 1,637 |
| Floating-point processing power | 20.31 TFLOPS | 104.8 TFLOPS |
| ROPs | 96 | 176 |
| TMUs | 184 | 680 |
| Tensor Cores | 184 | 680 |
| Ray Tracing Cores | 46 | 170 |
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 5.0 x16 |
| Length | 242 mm | 304 mm |
| Width | 2-slot | 2-slot |
| Supplementary power connectors | 1x 12-pin | 1x 16-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 | GDDR7 |
| Maximum RAM amount | 8 GB | 32 GB |
| Memory bus width | 256 Bit | 512 Bit |
| Memory clock speed | 1750 MHz | 1750 MHz |
| Memory bandwidth | 448.0 GB/s | 1.79 TB/s |
| Shared memory | - | - |
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, 3x DisplayPort | 1x HDMI 2.1b, 3x DisplayPort 2.1b |
| 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 | 6.8 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.0 | 3.0 |
| Vulkan | 1.2 | 1.4 |
| CUDA | 8.5 | 10.1 |
| 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.
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 | 150
−38%
| 207
+38%
|
| 1440p | 98
−87.8%
| 184
+87.8%
|
| 4K | 64
−120%
| 141
+120%
|
Cost per frame, $
| 1080p | 3.33
+190%
| 9.66
−190%
|
| 1440p | 5.09
+113%
| 10.86
−113%
|
| 4K | 7.80
+81.8%
| 14.18
−81.8%
|
- RTX 3070 has 190% lower cost per frame in 1080p
- RTX 3070 has 113% lower cost per frame in 1440p
- RTX 3070 has 82% lower cost per frame in 4K
FPS performance in popular games
Full HD
Low Preset
| Counter-Strike 2 | 149
−43.6%
|
210−220
+43.6%
|
| Cyberpunk 2077 | 147
−76.9%
|
260−270
+76.9%
|
Full HD
Medium Preset
| Battlefield 5 | 110−120
+0%
|
110−120
+0%
|
| Counter-Strike 2 | 135
−58.5%
|
210−220
+58.5%
|
| Cyberpunk 2077 | 124
−77.4%
|
220−230
+77.4%
|
| Forza Horizon 4 | 311
−76.8%
|
550−600
+76.8%
|
| Forza Horizon 5 | 139
−131%
|
300−350
+131%
|
| Metro Exodus | 124
−59.7%
|
190−200
+59.7%
|
| Red Dead Redemption 2 | 118
−32.2%
|
150−160
+32.2%
|
| Valorant | 246
−236%
|
800−850
+236%
|
Full HD
High Preset
| Battlefield 5 | 110−120
+0%
|
110−120
+0%
|
| Counter-Strike 2 | 117
−82.9%
|
210−220
+82.9%
|
| Cyberpunk 2077 | 109
−74.3%
|
190−200
+74.3%
|
| Dota 2 | 137
−27%
|
170−180
+27%
|
| Far Cry 5 | 91
−74.7%
|
159
+74.7%
|
| Fortnite | 220−230
−34.5%
|
300−350
+34.5%
|
| Forza Horizon 4 | 256
−75.8%
|
450−500
+75.8%
|
| Forza Horizon 5 | 144
−123%
|
300−350
+123%
|
| Grand Theft Auto V | 134
−79.1%
|
240−250
+79.1%
|
| Metro Exodus | 107
−85%
|
190−200
+85%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 210−220
+0%
|
210−220
+0%
|
| Red Dead Redemption 2 | 98
−59.2%
|
150−160
+59.2%
|
| The Witcher 3: Wild Hunt | 170−180
+0%
|
170−180
+0%
|
| Valorant | 159
−419%
|
800−850
+419%
|
| World of Tanks | 270−280
+0%
|
270−280
+0%
|
Full HD
Ultra Preset
| Battlefield 5 | 110−120
+0%
|
110−120
+0%
|
| Counter-Strike 2 | 105
−92.4%
|
202
+92.4%
|
| Cyberpunk 2077 | 91
−75.8%
|
160−170
+75.8%
|
| Dota 2 | 125
−76%
|
220−230
+76%
|
| Far Cry 5 | 110−120
−76.3%
|
200−210
+76.3%
|
| Forza Horizon 4 | 223
−79.4%
|
400−450
+79.4%
|
| Forza Horizon 5 | 118
−172%
|
300−350
+172%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 210−220
+0%
|
210−220
+0%
|
| Valorant | 237
−249%
|
800−850
+249%
|
1440p
High Preset
| Counter-Strike 2 | 35−40
−423%
|
183
+423%
|
| Dota 2 | 98
−72.4%
|
160−170
+72.4%
|
| Grand Theft Auto V | 95
−78.9%
|
170−180
+78.9%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
−71.4%
|
300−310
+71.4%
|
| Red Dead Redemption 2 | 63
−47.6%
|
90−95
+47.6%
|
| World of Tanks | 350−400
−35.1%
|
500−550
+35.1%
|
1440p
Ultra Preset
| Battlefield 5 | 85−90
−1.2%
|
85−90
+1.2%
|
| Cyberpunk 2077 | 58
−72.4%
|
100−105
+72.4%
|
| Far Cry 5 | 160−170
+0%
|
160−170
+0%
|
| Forza Horizon 4 | 166
−74.7%
|
290−300
+74.7%
|
| Forza Horizon 5 | 98
−144%
|
230−240
+144%
|
| Metro Exodus | 101
−65.3%
|
160−170
+65.3%
|
| The Witcher 3: Wild Hunt | 110−120
−76.5%
|
210−220
+76.5%
|
| Valorant | 208
−175%
|
550−600
+175%
|
4K
High Preset
| Counter-Strike 2 | 30−35
−461%
|
174
+461%
|
| Dota 2 | 117
−59.8%
|
180−190
+59.8%
|
| Grand Theft Auto V | 115
−73.9%
|
200−210
+73.9%
|
| Metro Exodus | 49
−241%
|
167
+241%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 200−210
−1.5%
|
200−210
+1.5%
|
| Red Dead Redemption 2 | 43
−48.8%
|
60−65
+48.8%
|
| The Witcher 3: Wild Hunt | 117
−79.5%
|
210−220
+79.5%
|
4K
Ultra Preset
| Battlefield 5 | 70−75
−23%
|
90−95
+23%
|
| Counter-Strike 2 | 16
−244%
|
55
+244%
|
| Cyberpunk 2077 | 28
−78.6%
|
50−55
+78.6%
|
| Dota 2 | 125
−76%
|
220−230
+76%
|
| Far Cry 5 | 100−110
+0%
|
100−110
+0%
|
| Fortnite | 95−100
+0%
|
95−100
+0%
|
| Forza Horizon 4 | 98
−73.5%
|
170−180
+73.5%
|
| Forza Horizon 5 | 57
−182%
|
160−170
+182%
|
| Valorant | 116
−214%
|
350−400
+214%
|
This is how RTX 3070 and RTX 5090 compete in popular games:
- RTX 5090 is 38% faster in 1080p
- RTX 5090 is 88% faster in 1440p
- RTX 5090 is 120% 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 RTX 5090 is 461% faster.
All in all, in popular games:
- RTX 5090 is ahead in 35 tests (78%)
- there's a draw in 10 tests (22%)
Pros & cons summary
| Performance score | 55.70 | 100.00 |
| Recency | 1 September 2020 | 30 January 2025 |
| Maximum RAM amount | 8 GB | 32 GB |
| Chip lithography | 8 nm | 5 nm |
| Power consumption (TDP) | 220 Watt | 575 Watt |
RTX 3070 has 161.4% lower power consumption.
RTX 5090, on the other hand, has a 79.5% higher aggregate performance score, an age advantage of 4 years, a 300% higher maximum VRAM amount, and a 60% more advanced lithography process.
The GeForce RTX 5090 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.
