AMD Radeon 780M
vs
NVIDIA GeForce GTX 1650

vs
AMD Radeon 780M vs NVIDIA GeForce GTX 1650 graphics card comparison

GPU Comparison Result

AMD Radeon 780M vs GeForce GTX 1650: How Close Has Integrated Graphics Come to an Old Discrete Card

The GeForce GTX 1650 was released in 2019 as an affordable graphics card for Full HD gaming. The Radeon 780M appeared nearly four years later within AMD’s mobile processors and operates without dedicated video memory. The GTX 1650 is still faster, but the best systems with the Radeon 780M trail it by just 15-20%.

This comparison refers specifically to the desktop GeForce GTX 1650, not the GTX 1650 Mobile. The mobile GTX 1650 is weaker, resulting in an even smaller gap with the Radeon 780M.

What is the Fundamental Difference

Radeon 780M GeForce GTX 1650
Type Integrated Discrete
Architecture RDNA 3 Turing
Compute Units 12 CU, 768 shaders 896 CUDA cores
Memory System DDR5/LPDDR5X 4 GB GDDR5/GDDR6
Memory Bus Depends on the system 128 bits
Typical Use Laptops, Mini-PCs Desktop PCs
Year of Release 2023 2019

The GTX 1650 has its own 4 GB of video memory and is powered independently from the processor. The Radeon 780M uses system memory and shares power with the CPU cores.

Therefore, the performance of the 780M significantly depends on the specific device. Fast LPDDR5X or DDR5, a high power limit, and efficient cooling can considerably enhance performance. The same iGPU in a slim ultrabook versus a more powerful laptop can yield noticeably different results.

Radeon 780M and GTX 1650 in Real Systems

For comparison, we’ll use 3DMark Time Spy Graphics. For Radeon 780M, results from specific laptops are used; for the GTX 1650, results from desktops with specific graphics cards.

Device / Configuration GPU Processor Time Spy Graphics
Tuxedo Pulse 14 Gen3 Radeon 780M Ryzen 7 7840HS 3027
Schenker XMG Evo 14 M24 Radeon 780M Ryzen 7 8845HS 2860
Ninkear A16 Pro Radeon 780M Ryzen 7 8845HS 2732
Framework Laptop 13.5 Radeon 780M Ryzen 7 7840U 2575
Acer Swift Edge 16 Radeon 780M Ryzen 7 7840U 2363
Gainward GeForce GTX 1650 4 GB GTX 1650 Ryzen 7 2700X 3492
KFA2 GTX 1650 EX GTX 1650 Ryzen 9 3900X 3512
KFA2 GTX 1650 EX GTX 1650 Ryzen 9 5900X 3538
ASUS TUF Gaming GTX 1650 GTX 1650 Ryzen 7 9800X3D 3622

The variance in scores for the Radeon 780M is significant: from 2363 points in the Acer Swift Edge 16 to 3027 in the Tuxedo Pulse 14 Gen3. The difference between these two laptops with the same iGPU approaches 30%.

The results for the desktop GTX 1650 are much more stable, generally falling between 3490 and 3620 points.

The best laptop with the Radeon 780M in the table falls behind the typical GTX 1650 by about 15-20%. In thinner laptops with stricter power limits, the gap can exceed 30%.

The average score for the Radeon 780M is around 2755 points compared to 3521 for the GTX 1650. On average, the GTX 1650 is about 28% faster.

What Happens in Games

Gaming tests confirm the results from Time Spy. The GTX 1650 is generally faster, but the extent of that advantage depends on the specific game and the system with the Radeon 780M.

At Full HD, the GTX 1650 often outperforms the Radeon 780M by about 20-30%, although in certain titles, the gap can be significantly larger.

For the Radeon 780M, memory and power limits are particularly crucial. Slow memory, low power limits for the APU, or weak cooling can significantly reduce FPS.

The GTX 1650 is more predictable in this regard. It has dedicated video memory, and the GPU's power consumption does not share resources with the CPU cores.

The Radeon 780M has not yet matched the GTX 1650 in gaming performance, but it has reached the level of older budget discrete graphics cards.

Why TFLOPS Can Be Misleading Here

The theoretical FP32 performance of the Radeon 780M is significantly higher than that of the GTX 1650, but these values cannot be directly compared.

RDNA 3 and Turing differ significantly in architecture, so a high TFLOPS figure does not automatically translate to a corresponding increase in FPS.

Thus, Time Spy and gaming tests are much more indicative in this context.

4 GB of GTX 1650 Has Become a Limitation

Dedicated video memory remains an advantage for the GTX 1650, but 4 GB is considered insufficient by modern standards. In new games, this amount quickly limits texture quality and other settings.

The Radeon 780M can allocate more system memory for graphics, but it cannot match the speed of dedicated VRAM. The CPU and GPU share the same memory subsystem, and the bandwidth remains one of the main limitations for integrated graphics.

Today, both are primarily suitable for 1080p resolutions and moderate graphics settings.

Is it Worth Buying a GTX 1650 in 2026?

New GTX 1650 graphics cards are still found for sale, but prices for remaining cards often do not correspond to their performance. Purchasing a new GTX 1650 for several hundred dollars does not make sense.

If a used GTX 1650 can be found for around $50-70, it may still be a cheap upgrade for an older office PC. Versions without additional power requirements are particularly interesting.

The Radeon 780M cannot be purchased separately. It is part of AMD's mobile APUs and is found in laptops and mini-PCs.

Conclusion

The GeForce GTX 1650 remains faster than the Radeon 780M, but the comparison is no longer very flattering for it.

In well-optimized systems, the Radeon 780M lags behind the desktop GTX 1650 by about 15-25%, although it is integrated into a mobile processor, lacks dedicated video memory, and shares the overall power budget with it.

In thinner laptops, the gap may increase to 30% or more, making it difficult to estimate the performance of a specific laptop based solely on the Radeon 780M designation.

If the GTX 1650 is already part of a computer, switching to a system with a Radeon 780M for the sole purpose of graphics would be a step backward.

However, for a laptop or mini-PC, the situation is much more interesting. In 2019, such performance required a separate 75-watt graphics card. Now, integrated graphics within a mobile processor have come very close to that level.

Advantages

  • Higher Boost Clock: 2900MHz (2900MHz vs 1665MHz)
  • Newer Launch Date: January 2023 (January 2023 vs April 2019)
  • Larger Memory Size: 4GB (System Shared vs 4GB)
  • Higher Bandwidth: 128.1 GB/s (System Dependent vs 128.1 GB/s)
  • More Shading Units: 896 (768 vs 896)

Basic

AMD
Label Name
NVIDIA
January 2023
Launch Date
April 2019
Integrated
Platform
Desktop
Radeon 780M
Model Name
GeForce GTX 1650
Navi III IGP
Generation
GeForce 16
1500MHz
Base Clock
1485MHz
2900MHz
Boost Clock
1665MHz
PCIe 4.0 x8
Bus Interface
PCIe 3.0 x16
25,390 million
Transistors
4,700 million
12
RT Cores
-
12
Compute Units
-
48
TMUs
?
Texture Mapping Units (TMUs) serve as components of the GPU, which are capable of rotating, scaling, and distorting binary images, and then placing them as textures onto any plane of a given 3D model. This process is called texture mapping.
56
TSMC
Foundry
TSMC
4 nm
Process Size
12 nm
RDNA 3.0
Architecture
Turing

Memory Specifications

System Shared
Memory Size
4GB
System Shared
Memory Type
GDDR5
System Shared
Memory Bus
?
The memory bus width refers to the number of bits of data that the video memory can transfer within a single clock cycle. The larger the bus width, the greater the amount of data that can be transmitted instantaneously, making it one of the crucial parameters of video memory. The memory bandwidth is calculated as: Memory Bandwidth = Memory Frequency x Memory Bus Width / 8. Therefore, when the memory frequencies are similar, the memory bus width will determine the size of the memory bandwidth.
128bit
SystemShared
Memory Clock
2001MHz
System Dependent
Bandwidth
?
Memory bandwidth refers to the data transfer rate between the graphics chip and the video memory. It is measured in bytes per second, and the formula to calculate it is: memory bandwidth = working frequency × memory bus width / 8 bits.
128.1 GB/s

Display and Media

Portable Device Dependent
Outputs
1x DVI
1x HDMI 2.0
1x DisplayPort 1.4a

Theoretical Performance

92.80 GPixel/s
Pixel Rate
?
Pixel fill rate refers to the number of pixels a graphics processing unit (GPU) can render per second, measured in MPixels/s (million pixels per second) or GPixels/s (billion pixels per second). It is the most commonly used metric to evaluate the pixel processing performance of a graphics card.
53.28 GPixel/s
139.2 GTexel/s
Texture Rate
?
Texture fill rate refers to the number of texture map elements (texels) that a GPU can map to pixels in a single second.
93.24 GTexel/s
17.82 TFLOPS
FP16 (half)
?
An important metric for measuring GPU performance is floating-point computing capability. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable. Single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks, while double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy.
5.967 TFLOPS
556.8 GFLOPS
FP64 (double)
?
An important metric for measuring GPU performance is floating-point computing capability. Double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy, while single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable.
93.24 GFLOPS
8.731 TFLOPS
FP32 (float)
?
An important metric for measuring GPU performance is floating-point computing capability. Single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks, while double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable.
3.044 TFLOPS

Miscellaneous

-
SM Count
?
Multiple Streaming Processors (SPs), along with other resources, form a Streaming Multiprocessor (SM), which is also referred to as a GPU's major core. These additional resources include components such as warp schedulers, registers, and shared memory. The SM can be considered the heart of the GPU, similar to a CPU core, with registers and shared memory being scarce resources within the SM.
14
768
Shading Units
?
The most fundamental processing unit is the Streaming Processor (SP), where specific instructions and tasks are executed. GPUs perform parallel computing, which means multiple SPs work simultaneously to process tasks.
896
128 KB per Array
L1 Cache
64 KB (per SM)
2MB
L2 Cache
1024KB
15W
TDP
75W
1.3
Vulkan Version
?
Vulkan is a cross-platform graphics and compute API by Khronos Group, offering high performance and low CPU overhead. It lets developers control the GPU directly, reduces rendering overhead, and supports multi-threading and multi-core processors.
1.3
2.1
OpenCL Version
3.0
4.6
OpenGL
4.6
-
CUDA
7.5
12 Ultimate (12_2)
DirectX
12 (12_1)
None
Power Connectors
None
32
ROPs
?
The Raster Operations Pipeline (ROPs) is primarily responsible for handling lighting and reflection calculations in games, as well as managing effects like anti-aliasing (AA), high resolution, smoke, and fire. The more demanding the anti-aliasing and lighting effects in a game, the higher the performance requirements for the ROPs; otherwise, it may result in a sharp drop in frame rate.
32
6.7
Shader Model
6.6
-
Suggested PSU
250W

Benchmarks

FP32 (float) / TFLOPS
Radeon 780M
8.731 +187%
GeForce GTX 1650
3.044
3DMark Steel Nomad
Radeon 780M
497 +54%
GeForce GTX 1650
323
3DMark Time Spy
Radeon 780M
2755
GeForce GTX 1650
3521 +28%
Blender
Radeon 780M
281.09
GeForce GTX 1650
430.53 +53%