AMD Radeon 760M

AMD Radeon 760M
AMD Radeon 760M graphics card review

AMD Radeon 760M: Tests in Laptops, Gaming, and Comparison with 680M and 780M

The AMD Radeon 760M is an integrated RDNA 3 graphics solution with 8 compute units and 512 shader processors. It is found in the Ryzen 5 family of Phoenix and Hawk Point processors and holds an intermediate position between the Radeon 740M and the more powerful 780M. The Radeon 760M is sufficiently fast for less demanding and many modern games, although the final performance greatly depends on memory, power limits, and the settings of the specific device.

Radeon 760M in Specific Devices

Device Processor RAM 3DMark Time Spy Graphics 3DMark Fire Strike Graphics
HP ZBook Firefly 14 G10 A Ryzen 5 PRO 7640HS 16 GB 1,997 6,121
Lenovo LOQ 15APH8 Ryzen 5 7640HS 16 GB 2,234 6,162
GMKtec NucBox M6 Ultra Ryzen 5 7640HS 32 GB 2,281 6,558
Acer Swift Edge 16 SFE16-44 Ryzen 5 8640U 16 GB 2,293 6,559
Lenovo Yoga 7 2-in-1 14AHP9 Ryzen 5 8640HS 16 GB 2,495 7,004

The variance clearly shows the main feature of integrated graphics. In Time Spy Graphics, the Lenovo Yoga 7 scores 2,495 points, while the HP ZBook Firefly scores 1,997. The difference is around 25%, even though both utilize the Radeon 760M.

The reason is not solely related to cooling. Power limits of the processor, frequencies, firmware settings, and memory bandwidth also influence the results. Therefore, it is better to assess the performance of the Radeon 760M based on the specific laptop instead of just relying on the GPU name.

The Lenovo LOQ in this table requires a separate note: the series is often equipped with discrete graphics. The result provided refers specifically to the integrated Radeon 760M.

Why Memory Matters

The Radeon 760M does not have its own GDDR memory; the graphics core utilizes system RAM. This is a key performance factor for iGPUs.

DDR5 or fast LPDDR5X provides the necessary bandwidth for the graphics. Thus, two laptops with the same processor might deliver different results even under similar power limits.

It is especially important to check the memory configuration in models with standard SO-DIMM. A slow or poorly optimized RAM subsystem can significantly reduce the performance of the integrated graphics.

Radeon 760M vs. Radeon 680M and 780M

Despite the transition from RDNA 2 to RDNA 3, the Radeon 760M did not become an unconditional replacement for the Radeon 680M. The older 680M has 12 compute units compared to the 8 in the 760M, placing their real-world performance very close to each other.

The average score of the Radeon 760M in 3DMark Time Spy Graphics is around 2,260 points, while the Radeon 680M scores approximately 2,214. The difference is too small to be considered significant: a specific laptop can easily change the standings.

The Radeon 780M is noticeably faster. With an average score of around 2,657 points, it has an advantage of about 15-20% over the 760M.

Thus, the transition from 680M to 760M does not appear to be a serious upgrade on its own. The jump from 760M to 780M is more noticeable, especially if gaming is among the primary use cases.

Radeon 760M Performance in Games

Gaming tests demonstrate the GPU's performance limits well. In less demanding and competitive titles, comfortable frame rates can be achieved at Full HD, but heavy AAA games quickly force a reduction in settings.

Game, 1920×1080 Low Medium High
Counter-Strike 2 ~105 FPS ~77 FPS ~39 FPS
F1 24 ~62 FPS ~54 FPS ~41 FPS
Ghost of Tsushima ~44 FPS ~38 FPS ~31 FPS
Cyberpunk 2077 ~31 FPS ~24 FPS ~20 FPS
Baldur's Gate 3 ~30 FPS ~24 FPS ~19 FPS

Counter-Strike 2 and F1 24 work adequately fast even at Full HD. For Cyberpunk 2077 and Baldur's Gate 3, you should expect around 30 FPS at low quality.

In demanding titles, it is wiser to use FSR or lower the resolution. For the Radeon 760M, this approach is more practical than trying to maintain native Full HD at all costs.

Hardware ray tracing is supported, but there is virtually no performance headroom for RT in modern demanding games.

Devices Featuring Radeon 760M

The Radeon 760M can be found in a very diverse range of computers. The Lenovo Yoga 7 2-in-1 utilizes it within a compact convertible with the Ryzen 5 8640HS, the Acer Swift Edge 16 features it in a slim laptop with the Ryzen 5 8640U, and the HP ZBook Firefly 14 includes it in a mobile work system.

There are mini PCs as well. For example, the GMKtec NucBox M6 Ultra combines the Ryzen 5 7640HS with 32 GB of DDR5 and scores 2,281 points in Time Spy Graphics. The form factor itself does not determine the speed of the Radeon 760M; the overall system configuration is what matters.

How Good Is the Radeon 760M

The Radeon 760M represents the minimum level of modern AMD iGPUs at which Full HD gaming becomes practically viable. Esports titles perform reliably, and many heavy games can be run at around 30-60 FPS after adjusting settings.

The Radeon 780M is faster and preferable for only a slight difference in price. However, paying extra just for the upgrade from 760M to 780M may not always be justified: memory, power limits, and the quality of the laptop might be more critical than the difference between the two iGPUs.

The Radeon 760M does not replace a discrete graphics card. However, for a laptop or mini PC without a separate GPU, it provides entry-level gaming graphics. For esports titles, Full HD is suitable without serious caveats, while intensive games usually require Low settings, FSR, or reduced resolution.

Basic

Label Name
AMD
Platform
Integrated
Launch Date
January 2023
Model Name
Radeon 760M
Generation
Navi III IGP
Base Clock
1500MHz
Boost Clock
2800MHz
Bus Interface
PCIe 4.0 x8
Transistors
25,390 million
RT Cores
6
Compute Units
8
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.
24
Foundry
TSMC
Process Size
4 nm
Architecture
RDNA 3.0

Memory Specifications

Memory Size
System Shared
Memory Type
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.
System Shared
Memory Clock
SystemShared
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.
System Dependent

Display and Media

Outputs
Portable Device Dependent

Theoretical Performance

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.
44.80 GPixel/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.
67.20 GTexel/s
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.
8.602 TFLOPS
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.
268.8 GFLOPS
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.
4.387 TFLOPS

Miscellaneous

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.
384
L1 Cache
128 KB per Array
L2 Cache
2MB
TDP
15W
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
OpenCL Version
2.1
OpenGL
4.6
DirectX
12 Ultimate (12_2)
Power Connectors
None
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.
16
Shader Model
6.7

Benchmarks

FP32 (float)
Score
4.387 TFLOPS
3DMark Time Spy
Score
2329
Blender
Score
191.62

Compared to Other GPU

FP32 (float) / TFLOPS
4.579 +4.4%
4.387
3DMark Time Spy
4682 +101%
3619 +55.4%
1526 -34.5%
371.8 -84%
Blender
1408.56 +635.1%
802 +318.5%
391 +104%
191.62
45.58 -76.2%