GeForce RTX 3090 vs GTX 670MX
Aggregate performance score
We've compared GeForce GTX 670MX with GeForce RTX 3090, including specs and performance data.
RTX 3090 outperforms 670MX by a whopping 1236% 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 | 700 | 38 |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | no data | 18.04 |
| Power efficiency | 4.91 | 14.06 |
| Architecture | Kepler (2012−2018) | Ampere (2020−2025) |
| GPU code name | GK104 | GA102 |
| Market segment | Laptop | Desktop |
| Release date | 1 October 2012 (13 years ago) | 1 September 2020 (5 years ago) |
| Launch price (MSRP) | no data | $1,499 |
Cost-effectiveness evaluation
The higher the ratio, the better. We use the manufacturer's recommended prices.
Performance to price scatter graph
Currently popular graphics cards are shown for comparison.
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Real power consumption of some graphics cards can well exceed their nominal TDP, especially if overclocked.
| Pipelines / CUDA cores | 960 | 10496 |
| Core clock speed | 600 MHz | 1395 MHz |
| Boost clock speed | no data | 1695 MHz |
| Number of transistors | 3,540 million | 28,300 million |
| Manufacturing process technology | 28 nm | 8 nm |
| Power consumption (TDP) | 75 Watt | 350 Watt |
| Texture fill rate | 48.08 | 556.0 |
| Floating-point processing power | 1.154 TFLOPS | 35.58 TFLOPS |
| ROPs | 24 | 112 |
| TMUs | 80 | 328 |
| Tensor Cores | no data | 328 |
| Ray Tracing Cores | no data | 82 |
| L1 Cache | 80 KB | 10.3 MB |
| L2 Cache | 384 KB | 6 MB |
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).
| Laptop size | large | no data |
| Bus support | PCI Express 2.0 | no data |
| Interface | PCIe 3.0 x16 | PCIe 4.0 x16 |
| Length | no data | 336 mm |
| Width | no data | 3-slot |
| Supplementary power connectors | no data | 1x 12-pin |
| SLI options | + | - |
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 | GDDR5 | GDDR6X |
| Maximum RAM amount | 3 GB | 24 GB |
| Memory bus width | 192 Bit | 384 Bit |
| Memory clock speed | 1400 MHz | 1219 MHz |
| Memory bandwidth | 67.2 GB/s | 936.2 GB/s |
| Shared memory | - | - |
| Resizable BAR | - | + |
Connectivity and outputs
This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.
| Display Connectors | No outputs | 1x HDMI, 3x DisplayPort |
| HDMI | + | + |
| HDCP | + | - |
| Maximum VGA resolution | Up to 2048x1536 | no data |
Supported technologies
Supported technological solutions. This information will prove useful if you need some particular technology for your purposes.
| Optimus | + | - |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 API | 12 Ultimate (12_2) |
| Shader Model | 5.1 | 6.5 |
| OpenGL | 4.5 | 4.6 |
| OpenCL | 1.1 | 2.0 |
| Vulkan | 1.1.126 | 1.2 |
| CUDA | + | 8.5 |
| DLSS | - | + |
Synthetic benchmarks
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 Vantage Performance
3DMark Vantage is an outdated DirectX 10 benchmark using 1280x1024 screen resolution. It taxes the graphics card with two scenes, one depicting a girl escaping some militarized base located within a sea cave, the other displaying a space fleet attack on a defenseless planet. It was discontinued in April 2017, and Time Spy benchmark is now recommended to be used instead.
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.
GeekBench 5 OpenCL
Geekbench 5 is a widespread graphics card benchmark combined from 11 different test scenarios. All these scenarios rely on direct usage of GPU's processing power, no 3D rendering is involved. This variation uses OpenCL API by Khronos Group.
GeekBench 5 Vulkan
Geekbench 5 is a widespread graphics card benchmark combined from 11 different test scenarios. All these scenarios rely on direct usage of GPU's processing power, no 3D rendering is involved. This variation uses Vulkan API by AMD & Khronos Group.
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 | 40
−368%
| 187
+368%
|
| 1440p | 8−9
−1400%
| 120
+1400%
|
| 4K | 6−7
−1250%
| 81
+1250%
|
Cost per frame, $
| 1080p | no data | 8.02 |
| 1440p | no data | 12.49 |
| 4K | no data | 18.51 |
FPS performance in popular games
Full HD
Low
| Counter-Strike 2 | 21−24
−1562%
|
349
+1562%
|
| Cyberpunk 2077 | 10−11
−1990%
|
209
+1990%
|
| Palworld | 12−14
−1892%
|
239
+1892%
|
| Resident Evil 4 Remake | 7−8
−3757%
|
270
+3757%
|
Full HD
Medium
| Battlefield 5 | 20−22
−760%
|
172
+760%
|
| Counter-Strike 2 | 21−24
−1552%
|
347
+1552%
|
| Cyberpunk 2077 | 10−11
−1680%
|
178
+1680%
|
| Far Cry 5 | 14−16
−1386%
|
208
+1386%
|
| Fortnite | 27−30
−979%
|
300−350
+979%
|
| Forza Horizon 4 | 21−24
−1055%
|
254
+1055%
|
| Palworld | 12−14
−1842%
|
233
+1842%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 18−20
−811%
|
170−180
+811%
|
| Valorant | 60−65
−502%
|
350−400
+502%
|
Full HD
High
| Battlefield 5 | 20−22
−690%
|
158
+690%
|
| Counter-Strike 2 | 21−24
−1371%
|
309
+1371%
|
| Counter-Strike: Global Offensive | 96
−190%
|
270−280
+190%
|
| Cyberpunk 2077 | 10−11
−1440%
|
154
+1440%
|
| Dota 2 | 40−45
−429%
|
217
+429%
|
| Far Cry 5 | 14−16
−1300%
|
196
+1300%
|
| Fortnite | 27−30
−979%
|
300−350
+979%
|
| Forza Horizon 4 | 21−24
−1023%
|
247
+1023%
|
| Grand Theft Auto V | 16−18
−969%
|
171
+969%
|
| Metro Exodus | 9−10
−1856%
|
176
+1856%
|
| Palworld | 12−14
−1633%
|
208
+1633%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 18−20
−811%
|
170−180
+811%
|
| The Witcher 3: Wild Hunt | 14−16
−2536%
|
369
+2536%
|
| Valorant | 60−65
−502%
|
350−400
+502%
|
Full HD
Ultra
| Battlefield 5 | 20−22
−630%
|
146
+630%
|
| Cyberpunk 2077 | 10−11
−1260%
|
136
+1260%
|
| Dota 2 | 40−45
−420%
|
213
+420%
|
| Far Cry 5 | 14−16
−1207%
|
183
+1207%
|
| Forza Horizon 4 | 21−24
−886%
|
217
+886%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 18−20
−811%
|
170−180
+811%
|
| The Witcher 3: Wild Hunt | 14−16
−1200%
|
182
+1200%
|
| Valorant | 60−65
−393%
|
296
+393%
|
Full HD
Epic
| Fortnite | 27−30
−979%
|
300−350
+979%
|
| Palworld | 12−14
−1208%
|
157
+1208%
|
1440p
High
| Counter-Strike 2 | 9−10
−2467%
|
231
+2467%
|
| Counter-Strike: Global Offensive | 35−40
−1254%
|
500−550
+1254%
|
| Grand Theft Auto V | 3−4
−4900%
|
150
+4900%
|
| Metro Exodus | 3−4
−3733%
|
115
+3733%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 35−40
−400%
|
170−180
+400%
|
| Valorant | 45−50
−819%
|
400−450
+819%
|
1440p
Ultra
| Battlefield 5 | 4−5
−3150%
|
130
+3150%
|
| Cyberpunk 2077 | 3−4
−3000%
|
93
+3000%
|
| Far Cry 5 | 9−10
−1800%
|
171
+1800%
|
| Forza Horizon 4 | 10−12
−1691%
|
197
+1691%
|
| The Witcher 3: Wild Hunt | 6−7
−2450%
|
153
+2450%
|
1440p
Epic
| Fortnite | 9−10
−1578%
|
150−160
+1578%
|
| Palworld | 6−7
−1900%
|
120
+1900%
|
4K
High
| Grand Theft Auto V | 16−18
−1038%
|
182
+1038%
|
| Palworld | 0−1 | 96 |
| The Witcher 3: Wild Hunt | 1−2
−15300%
|
154
+15300%
|
| Valorant | 21−24
−1419%
|
300−350
+1419%
|
4K
Ultra
| Battlefield 5 | 2−3
−5550%
|
113
+5550%
|
| Cyberpunk 2077 | 1−2
−4500%
|
46
+4500%
|
| Dota 2 | 16−18
−1163%
|
202
+1163%
|
| Far Cry 5 | 4−5
−2600%
|
108
+2600%
|
| Forza Horizon 4 | 6−7
−2450%
|
153
+2450%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 5−6
−1820%
|
95−100
+1820%
|
4K
Epic
| Fortnite | 5−6
−1480%
|
75−80
+1480%
|
| Palworld | 0−1 | 72 |
4K
High
| Counter-Strike 2 | 59
+0%
|
59
+0%
|
| Metro Exodus | 76
+0%
|
76
+0%
|
4K
Ultra
| Counter-Strike 2 | 85−90
+0%
|
85−90
+0%
|
This is how GTX 670MX and RTX 3090 compete in popular games:
- RTX 3090 is 368% faster in 1080p
- RTX 3090 is 1400% faster in 1440p
- RTX 3090 is 1250% faster in 4K
Here's the range of performance differences observed across popular games:
- in The Witcher 3: Wild Hunt, with 4K resolution and the High Preset, the RTX 3090 is 15300% faster.
All in all, in popular games:
- RTX 3090 performs better in 60 tests (95%)
- there's a draw in 3 tests (5%)
Pros & cons summary
| Performance score | 4.76 | 63.61 |
| Recency | 1 October 2012 | 1 September 2020 |
| Maximum RAM amount | 3 GB | 24 GB |
| Chip lithography | 28 nm | 8 nm |
| Power consumption (TDP) | 75 Watt | 350 Watt |
GTX 670MX has 367% lower power consumption.
RTX 3090, on the other hand, has a 1236% higher aggregate performance score, an age advantage of 7 years, a 700% higher maximum VRAM amount, and a 250% more advanced lithography process.
The GeForce RTX 3090 is our recommended choice as it beats the GeForce GTX 670MX in performance tests.
Be aware that GeForce GTX 670MX is a notebook graphics card while GeForce RTX 3090 is a desktop one.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
