GeForce RTX 3090 vs 9650M GS
Aggregate performance score
We've compared GeForce 9650M GS with GeForce RTX 3090, including specs and performance data.
RTX 3090 outperforms 9650M GS by a whopping 9669% 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 | 1248 | 38 |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | no data | 19.11 |
| Power efficiency | 1.73 | 13.97 |
| Architecture | Tesla (2006−2010) | Ampere (2020−2025) |
| GPU code name | G84 | GA102 |
| Market segment | Laptop | Desktop |
| Release date | 13 March 2008 (18 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
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 | 32 | 10496 |
| Core clock speed | 625 MHz | 1395 MHz |
| Boost clock speed | no data | 1695 MHz |
| Number of transistors | 289 million | 28,300 million |
| Manufacturing process technology | 80 nm | 8 nm |
| Power consumption (TDP) | 29 Watt | 350 Watt |
| Texture fill rate | 10.00 | 556.0 |
| Floating-point processing power | 0.08 TFLOPS | 35.58 TFLOPS |
| ROPs | 8 | 112 |
| TMUs | 16 | 328 |
| Tensor Cores | no data | 328 |
| Ray Tracing Cores | no data | 82 |
| L1 Cache | no data | 10.3 MB |
| L2 Cache | 32 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 |
| Interface | PCIe 2.0 x16 | PCIe 4.0 x16 |
| Length | no data | 336 mm |
| Width | no data | 3-slot |
| Supplementary power connectors | None | 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 | GDDR3 | GDDR6X |
| Maximum RAM amount | 512 MB | 24 GB |
| Memory bus width | 128 Bit | 384 Bit |
| Memory clock speed | 800 MHz | 1219 MHz |
| Memory bandwidth | 25.6 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 | - | + |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 11.1 (10_0) | 12 Ultimate (12_2) |
| Shader Model | 4.0 | 6.5 |
| OpenGL | 3.3 | 4.6 |
| OpenCL | 1.1 | 2.0 |
| Vulkan | N/A | 1.2 |
| CUDA | 1.1 | 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 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:
| Full HD | 1−2
−18900%
| 190
+18900%
|
| 1440p | 1−2
−12100%
| 122
+12100%
|
| 4K | 0−1 | 82 |
Cost per frame, $
| 1080p | no data | 7.89 |
| 1440p | no data | 12.29 |
| 4K | no data | 18.28 |
FPS performance in popular games
Full HD
Low
| Cyberpunk 2077 | 1−2
−20800%
|
209
+20800%
|
Full HD
Medium
| Cyberpunk 2077 | 1−2
−17700%
|
178
+17700%
|
| Far Cry 5 | 0−1 | 208 |
| Forza Horizon 4 | 5−6
−4980%
|
254
+4980%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 8−9
−2075%
|
170−180
+2075%
|
| Valorant | 27−30
−1189%
|
350−400
+1189%
|
Full HD
High
| Counter-Strike: Global Offensive | 18−20
−1368%
|
270−280
+1368%
|
| Cyberpunk 2077 | 1−2
−15300%
|
154
+15300%
|
| Dota 2 | 12−14
−1708%
|
217
+1708%
|
| Far Cry 5 | 0−1 | 196 |
| Forza Horizon 4 | 5−6
−4840%
|
247
+4840%
|
| Metro Exodus | 0−1 | 176 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 8−9
−2075%
|
170−180
+2075%
|
| The Witcher 3: Wild Hunt | 6−7
−6050%
|
369
+6050%
|
| Valorant | 27−30
−1189%
|
350−400
+1189%
|
Full HD
Ultra
| Cyberpunk 2077 | 1−2
−13500%
|
136
+13500%
|
| Dota 2 | 12−14
−1675%
|
213
+1675%
|
| Far Cry 5 | 0−1 | 183 |
| Forza Horizon 4 | 5−6
−4240%
|
217
+4240%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 8−9
−2075%
|
170−180
+2075%
|
| The Witcher 3: Wild Hunt | 6−7
−2933%
|
182
+2933%
|
| Valorant | 27−30
−957%
|
296
+957%
|
1440p
High
| Counter-Strike 2 | 3−4
−7600%
|
231
+7600%
|
| Counter-Strike: Global Offensive | 3−4
−16567%
|
500−550
+16567%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 7−8
−2400%
|
170−180
+2400%
|
1440p
Ultra
| Forza Horizon 4 | 2−3
−9750%
|
197
+9750%
|
| The Witcher 3: Wild Hunt | 2−3
−7550%
|
153
+7550%
|
1440p
Epic
| Fortnite | 1−2
−15000%
|
150−160
+15000%
|
4K
High
| Grand Theft Auto V | 14−16
−1200%
|
182
+1200%
|
| Valorant | 3−4
−10833%
|
300−350
+10833%
|
4K
Ultra
| PLAYERUNKNOWN'S BATTLEGROUNDS | 2−3
−4700%
|
95−100
+4700%
|
4K
Epic
| Fortnite | 2−3
−3850%
|
75−80
+3850%
|
Full HD
Low
| Counter-Strike 2 | 349
+0%
|
349
+0%
|
| Resident Evil 4 Remake | 270
+0%
|
270
+0%
|
Full HD
Medium
| Battlefield 5 | 172
+0%
|
172
+0%
|
| Counter-Strike 2 | 347
+0%
|
347
+0%
|
| Fortnite | 300−350
+0%
|
300−350
+0%
|
| Forza Horizon 5 | 210
+0%
|
210
+0%
|
Full HD
High
| Battlefield 5 | 158
+0%
|
158
+0%
|
| Counter-Strike 2 | 309
+0%
|
309
+0%
|
| Fortnite | 300−350
+0%
|
300−350
+0%
|
| Forza Horizon 5 | 195
+0%
|
195
+0%
|
| Grand Theft Auto V | 171
+0%
|
171
+0%
|
Full HD
Ultra
| Battlefield 5 | 146
+0%
|
146
+0%
|
Full HD
Epic
| Fortnite | 300−350
+0%
|
300−350
+0%
|
1440p
High
| Grand Theft Auto V | 150
+0%
|
150
+0%
|
| Metro Exodus | 115
+0%
|
115
+0%
|
| Valorant | 400−450
+0%
|
400−450
+0%
|
1440p
Ultra
| Battlefield 5 | 130
+0%
|
130
+0%
|
| Cyberpunk 2077 | 93
+0%
|
93
+0%
|
| Far Cry 5 | 171
+0%
|
171
+0%
|
4K
High
| Counter-Strike 2 | 59
+0%
|
59
+0%
|
| Metro Exodus | 76
+0%
|
76
+0%
|
| The Witcher 3: Wild Hunt | 154
+0%
|
154
+0%
|
4K
Ultra
| Battlefield 5 | 113
+0%
|
113
+0%
|
| Counter-Strike 2 | 85−90
+0%
|
85−90
+0%
|
| Cyberpunk 2077 | 46
+0%
|
46
+0%
|
| Dota 2 | 202
+0%
|
202
+0%
|
| Far Cry 5 | 108
+0%
|
108
+0%
|
| Forza Horizon 4 | 153
+0%
|
153
+0%
|
This is how 9650M GS and RTX 3090 compete in popular games:
- RTX 3090 is 18900% faster in 1080p
- RTX 3090 is 12100% faster in 1440p
Here's the range of performance differences observed across popular games:
- in Cyberpunk 2077, with 1080p resolution and the Low Preset, the RTX 3090 is 20800% faster.
All in all, in popular games:
- RTX 3090 performs better in 28 tests (50%)
- there's a draw in 28 tests (50%)
Pros & cons summary
| Performance score | 0.65 | 63.50 |
| Recency | 13 March 2008 | 1 September 2020 |
| Maximum RAM amount | 512 MB | 24 GB |
| Chip lithography | 80 nm | 8 nm |
| Power consumption (TDP) | 29 Watt | 350 Watt |
9650M GS has 1107% lower power consumption.
RTX 3090, on the other hand, has a 9669% higher aggregate performance score, an age advantage of 12 years, a 4700% higher maximum VRAM amount, and a 900% more advanced lithography process.
The GeForce RTX 3090 is our recommended choice as it beats the GeForce 9650M GS in performance tests.
Be aware that GeForce 9650M GS 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.
