GeForce RTX 4090 vs GT 650M SLI
Aggregate performance score
We've compared GeForce GT 650M SLI with GeForce RTX 4090, including specs and performance data.
RTX 4090 outperforms GT 650M SLI by a whopping 1867% based on our aggregate benchmark results.
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 637 | 2 |
| Place by popularity | not in top-100 | 8 |
| Cost-effectiveness evaluation | no data | 18.86 |
| Power efficiency | no data | 15.22 |
| Architecture | Kepler (2012−2018) | Ada Lovelace (2022−2024) |
| GPU code name | N13E-GE | AD102 |
| Market segment | Laptop | Desktop |
| Release date | 22 March 2012 (12 years ago) | 20 September 2022 (2 years ago) |
| Launch price (MSRP) | no data | $1,599 |
Cost-effectiveness evaluation
The higher the performance-to-price ratio, the better. We use the manufacturer's recommended prices 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. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.
| Pipelines / CUDA cores | 768 | 16384 |
| Core clock speed | 790 MHz | 2235 MHz |
| Boost clock speed | 835 MHz | 2520 MHz |
| Number of transistors | no data | 76,300 million |
| Manufacturing process technology | 28 nm | 5 nm |
| Power consumption (TDP) | no data | 450 Watt |
| Texture fill rate | no data | 1,290 |
| Floating-point processing power | no data | 82.58 TFLOPS |
| ROPs | no data | 176 |
| TMUs | no data | 512 |
| Tensor Cores | no data | 512 |
| Ray Tracing Cores | no data | 128 |
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 | medium sized | no data |
| Interface | no data | PCIe 4.0 x16 |
| Length | no data | 304 mm |
| Width | no data | 3-slot |
| Supplementary power connectors | no data | 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 | GDDR5 | GDDR6X |
| Maximum RAM amount | 2x2 GB | 24 GB |
| Memory bus width | 2x 128 Bit | 384 Bit |
| Memory clock speed | 4000 MHz | 1313 MHz |
| Memory bandwidth | no data | 1.01 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 | no data | 1x HDMI 2.1, 3x DisplayPort 1.4a |
| HDMI | - | + |
Supported technologies
Supported technological solutions. This information will prove useful if you need some particular technology for your purposes.
| Optimus | + | - |
API and SDK compatibility
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 11 | 12 Ultimate (12_2) |
| Shader Model | no data | 6.8 |
| OpenGL | no data | 4.6 |
| OpenCL | no data | 3.0 |
| Vulkan | - | 1.3 |
| CUDA | + | 8.9 |
| 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.
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.
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:
| 900p | 79
−1862%
| 1550−1600
+1862%
|
| Full HD | 50
−418%
| 259
+418%
|
| 1440p | 10−12
−1880%
| 198
+1880%
|
| 4K | 7−8
−1929%
| 142
+1929%
|
Cost per frame, $
| 1080p | no data | 6.17 |
| 1440p | no data | 8.08 |
| 4K | no data | 11.26 |
FPS performance in popular games
Full HD
Low Preset
| Atomic Heart | 10−12
−2845%
|
324
+2845%
|
| Counter-Strike 2 | 10−12
−1827%
|
212
+1827%
|
| Cyberpunk 2077 | 10−11
−2170%
|
227
+2170%
|
Full HD
Medium Preset
| Atomic Heart | 10−12
−2309%
|
265
+2309%
|
| Battlefield 5 | 18−20
−937%
|
190−200
+937%
|
| Counter-Strike 2 | 10−12
−1855%
|
215
+1855%
|
| Cyberpunk 2077 | 10−11
−2140%
|
224
+2140%
|
| Far Cry 5 | 12−14
−1508%
|
209
+1508%
|
| Fortnite | 27−30
−979%
|
300−350
+979%
|
| Forza Horizon 4 | 21−24
−1464%
|
300−350
+1464%
|
| Forza Horizon 5 | 10−11
−2710%
|
281
+2710%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 18−20
−883%
|
170−180
+883%
|
| Valorant | 60−65
−1028%
|
650−700
+1028%
|
Full HD
High Preset
| Atomic Heart | 10−12
−2027%
|
234
+2027%
|
| Battlefield 5 | 18−20
−937%
|
190−200
+937%
|
| Counter-Strike 2 | 10−12
−1709%
|
199
+1709%
|
| Counter-Strike: Global Offensive | 139
−100%
|
270−280
+100%
|
| Cyberpunk 2077 | 10−11
−2050%
|
215
+2050%
|
| Dota 2 | 40−45
−517%
|
253
+517%
|
| Far Cry 5 | 12−14
−1446%
|
201
+1446%
|
| Fortnite | 27−30
−979%
|
300−350
+979%
|
| Forza Horizon 4 | 21−24
−1464%
|
300−350
+1464%
|
| Forza Horizon 5 | 10−11
−2650%
|
275
+2650%
|
| Grand Theft Auto V | 16−18
−988%
|
174
+988%
|
| Metro Exodus | 9−10
−2444%
|
229
+2444%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 18−20
−883%
|
170−180
+883%
|
| The Witcher 3: Wild Hunt | 12−14
−4354%
|
579
+4354%
|
| Valorant | 60−65
−1028%
|
650−700
+1028%
|
Full HD
Ultra Preset
| Battlefield 5 | 18−20
−937%
|
190−200
+937%
|
| Counter-Strike 2 | 10−12
−1582%
|
185
+1582%
|
| Cyberpunk 2077 | 10−11
−2010%
|
211
+2010%
|
| Dota 2 | 40−45
−446%
|
224
+446%
|
| Far Cry 5 | 12−14
−1338%
|
187
+1338%
|
| Forza Horizon 4 | 21−24
−1464%
|
300−350
+1464%
|
| Forza Horizon 5 | 10−11
−1800%
|
190−200
+1800%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 18−20
−883%
|
170−180
+883%
|
| The Witcher 3: Wild Hunt | 12−14
−2246%
|
305
+2246%
|
| Valorant | 60−65
−1033%
|
680
+1033%
|
Full HD
Epic Preset
| Fortnite | 27−30
−979%
|
300−350
+979%
|
1440p
High Preset
| Counter-Strike 2 | 7−8
−2471%
|
180−190
+2471%
|
| Counter-Strike: Global Offensive | 35−40
−1333%
|
500−550
+1333%
|
| Grand Theft Auto V | 5−6
−3140%
|
162
+3140%
|
| Metro Exodus | 3−4
−5900%
|
180
+5900%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 30−35
−430%
|
170−180
+430%
|
| Valorant | 50−55
−815%
|
450−500
+815%
|
1440p
Ultra Preset
| Battlefield 5 | 4−5
−4800%
|
190−200
+4800%
|
| Cyberpunk 2077 | 4−5
−3875%
|
159
+3875%
|
| Far Cry 5 | 9−10
−1978%
|
187
+1978%
|
| Forza Horizon 4 | 10−12
−2682%
|
300−350
+2682%
|
| Forza Horizon 5 | 7−8
−1757%
|
130−140
+1757%
|
| The Witcher 3: Wild Hunt | 7−8
−3600%
|
259
+3600%
|
1440p
Epic Preset
| Fortnite | 9−10
−1578%
|
150−160
+1578%
|
4K
High Preset
| Atomic Heart | 4−5
−2450%
|
102
+2450%
|
| Grand Theft Auto V | 16−18
−1069%
|
187
+1069%
|
| The Witcher 3: Wild Hunt | 1−2
−27900%
|
280
+27900%
|
| Valorant | 24−27
−1283%
|
300−350
+1283%
|
4K
Ultra Preset
| Battlefield 5 | 2−3
−6700%
|
130−140
+6700%
|
| Cyberpunk 2077 | 1−2
−8000%
|
81
+8000%
|
| Dota 2 | 16−18
−1319%
|
227
+1319%
|
| Far Cry 5 | 5−6
−3300%
|
170
+3300%
|
| Forza Horizon 4 | 6−7
−4983%
|
300−350
+4983%
|
| Forza Horizon 5 | 2−3
−1650%
|
35−40
+1650%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 5−6
−1820%
|
95−100
+1820%
|
4K
Epic Preset
| Fortnite | 5−6
−1480%
|
75−80
+1480%
|
4K
High Preset
| Counter-Strike 2 | 130
+0%
|
130
+0%
|
| Metro Exodus | 137
+0%
|
137
+0%
|
4K
Ultra Preset
| Counter-Strike 2 | 38
+0%
|
38
+0%
|
This is how GT 650M SLI and RTX 4090 compete in popular games:
- RTX 4090 is 1862% faster in 900p
- RTX 4090 is 418% faster in 1080p
- RTX 4090 is 1880% faster in 1440p
- RTX 4090 is 1929% 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 4090 is 27900% faster.
All in all, in popular games:
- RTX 4090 is ahead in 61 test (95%)
- there's a draw in 3 tests (5%)
Pros & cons summary
| Performance score | 5.07 | 99.75 |
| Recency | 22 March 2012 | 20 September 2022 |
| Chip lithography | 28 nm | 5 nm |
RTX 4090 has a 1867.5% higher aggregate performance score, an age advantage of 10 years, and a 460% more advanced lithography process.
The GeForce RTX 4090 is our recommended choice as it beats the GeForce GT 650M SLI in performance tests.
Be aware that GeForce GT 650M SLI is a notebook card while GeForce RTX 4090 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.
