AMD Radeon HD 6850 X2 vs NVIDIA GeForce GTX 560M Comparison

Head-to-Head Benchmarks

The recorded data contains a single direct comparison between these two mobile and desktop graphics solutions: the Geekbench OpenCL benchmark. In this test, the NVIDIA GeForce GTX 560M scores 4855, while the AMD Radeon HD 6850 X2 scores 3977. The delta percentage is -18.1%, meaning the NVIDIA part leads by a substantial margin in this compute workload. This is a decisive win for the GeForce GTX 560M, and it is the only head-to-head measurement available in the database.

The magnitude of this difference is notable. An 18.1% gap in OpenCL performance places the GTX 560M in a clearly higher performance tier for this specific test. For context, the AMD Radeon HD 6850 X2 sits at the 24th percentile among all GPUs in the database. Its nearest rivals are the NVIDIA Quadro K2000 (average score 3964, a 0.3% difference), the NVIDIA GeForce 830M (3957, 0.5%), the AMD Radeon R5 M420 (3956, 0.5%), and the NVIDIA GeForce GT 745M (3953, 0.6%). These are all essentially tied with the Radeon, within a fraction of a percent of each other. The Radeon HD 6850 X2 is thus clustered tightly with entry-level and older workstation parts in this metric.

Meanwhile, the GTX 560M rests at the 28th percentile. Its nearest rivals include the NVIDIA GeForce 940MX (4844, 0.2% difference), the AMD Radeon R6 M255DX (4867, -0.2%), the NVIDIA GeForce GTS 450 (4893, -0.8%), and notably the NVIDIA GeForce RTX 5060 Ti 8 GB (4901, -0.9%). The presence of a modern RTX 50-series card in that list is surprising. It suggests that in this particular OpenCL workload, the older GTX 560M performs nearly on par with a much newer product, at least according to the recorded scores. This is a strong indicator that the GTX 560M's compute capability is respectable for its era.

The data does not show any benchmark where the AMD Radeon HD 6850 X2 comes out ahead. The wins tally is 0 for the AMD part and 1 for the NVIDIA part. This is a one-sided comparison in the available metrics. The question then becomes whether the architecture and specification differences can explain this outcome, and whether any use case might favor the AMD card despite its lower OpenCL score.

Architecture Differences

The two GPUs come from different architectural philosophies. The AMD Radeon HD 6850 X2 is built on TeraScale 2, using the Barts chip, and belongs to the Northern Islands generation (HD 6800). The NVIDIA GeForce GTX 560M uses the Fermi 2.0 architecture, with the GF116 chip, from the GeForce 500M generation. Both are fabricated by TSMC on the same 40 nm process node, so the manufacturing process itself is not a differentiator.

The transistor counts diverge significantly. The AMD chip packs 1,700 million transistors on a 255 mm² die, yielding a transistor density of 6.7 million per square millimeter. The NVIDIA chip contains 1,170 million transistors on a 238 mm² die, for a density of 4.9 million per square millimeter. The AMD die is larger and more complex in raw transistor terms, yet it still loses in the OpenCL benchmark. This suggests architectural efficiency differences rather than a simple transistor count advantage.

The memory subsystems are very different. The AMD card uses 2 GB of GDDR5 on a 256-bit bus, delivering 134.4 GB/s of bandwidth. The NVIDIA card has 1536 MB (1.5 GB) of GDDR5 on a 192-bit bus, with 60.00 GB/s of bandwidth. The AMD card has more than double the memory bandwidth, which should theoretically help in memory-intensive workloads. The memory clock speeds reflect this: the AMD memory runs at 1050 MHz (4.2 Gbps effective), while the NVIDIA memory runs at 625 MHz (2.5 Gbps effective). Despite this massive bandwidth advantage, the AMD card still trails in OpenCL.

The compute unit counts also favor AMD on paper. The Radeon HD 6850 X2 has 960 shading units, 48 texture mapping units (TMUs), and 32 render output units (ROPs). The GTX 560M has only 192 shading units, 32 TMUs, and 24 ROPs. The AMD card has five times the shading units. The pixel rate is 25.60 GPixel/s for AMD versus 6.200 GPixel/s for NVIDIA. The texture rate is 38.40 GTexel/s versus 24.80 GTexel/s. The FP32 performance is 1.536 TFLOPS for AMD versus 595.2 GFLOPS for NVIDIA. By every raw specification, the AMD part should be dramatically faster, yet the benchmark shows the reverse.

This points to a fundamental difference in how the architectures execute compute workloads. TeraScale 2 was designed with a VLIW (Very Long Instruction Word) approach, which can struggle with certain types of parallel workloads that are not perfectly suited to its instruction packing. Fermi 2.0, by contrast, uses a more conventional scalar design that is often better at general-purpose compute tasks like those in OpenCL. The API support also differs: AMD lists DirectX 11.2 (11_0) and OpenGL 4.4, while NVIDIA lists DirectX 12 (11_0) and OpenGL 4.6. The NVIDIA card has a higher OpenGL version, which may contribute to its performance in some workloads. Neither card has Vulkan support recorded.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The NVIDIA GeForce GTX 560M scores 4855, which is 18.1% higher than the AMD Radeon HD 6850 X2's score of 3977. The NVIDIA card wins the only head-to-head benchmark recorded.

Q: Does the AMD Radeon HD 6850 X2 have more memory bandwidth?

A: Yes. The AMD card has 134.4 GB/s of bandwidth from a 256-bit bus and 2 GB of GDDR5 memory. The NVIDIA card has 60.00 GB/s from a 192-bit bus and 1536 MB of memory. The AMD bandwidth advantage is more than double.

Q: How do the shading unit counts compare?

A: The AMD Radeon HD 6850 X2 has 960 shading units, while the NVIDIA GeForce GTX 560M has 192. The AMD card also has 48 TMUs versus 32, and 32 ROPs versus 24. The AMD part leads in every raw compute resource count.

Q: What is the transistor density difference between the two chips?

A: The AMD Barts chip has 1,700 million transistors on a 255 mm² die, giving a density of 6.7 million per square millimeter. The NVIDIA GF116 chip has 1,170 million transistors on a 238 mm² die, for 4.9 million per square millimeter. Both are on TSMC's 40 nm process.

Q: Which GPU has a higher percentile ranking among all GPUs?

A: The NVIDIA GeForce GTX 560M is at the 28th percentile, while the AMD Radeon HD 6850 X2 is at the 24th percentile. The NVIDIA card ranks higher, though both are in the lower quartile of the database.

Q: What are the power connector requirements for each card?

A: The AMD Radeon HD 6850 X2 requires two 8-pin power connectors and has a 254 W TDP. The NVIDIA GeForce GTX 560M has no power connectors and a 75 W TDP, as it is an MXM module designed for portable devices.

Specification Differences

The two cards differ in nearly every measurable specification. The process node is the same (40 nm, TSMC), but the transistor counts differ: 1,700 million for AMD versus 1,170 million for NVIDIA. The die size is 255 mm² for AMD versus 238 mm² for NVIDIA. Transistor density is 6.7M/mm² for AMD versus 4.9M/mm² for NVIDIA.

Memory configuration is a major split. AMD uses 2 GB of GDDR5 on a 256-bit bus with 134.4 GB/s bandwidth. NVIDIA uses 1536 MB of GDDR5 on a 192-bit bus with 60.00 GB/s. The memory clocks are 1050 MHz (4.2 Gbps effective) for AMD versus 625 MHz (2.5 Gbps effective) for NVIDIA. The AMD card has a larger frame buffer and much higher bandwidth.

Compute resources are also heavily skewed toward AMD. The AMD card has 960 shading units, 48 TMUs, and 32 ROPs, yielding a pixel rate of 25.60 GPixel/s, a texture rate of 38.40 GTexel/s, and FP32 performance of 1.536 TFLOPS. The NVIDIA card has 192 shading units, 32 TMUs, and 24 ROPs, with a pixel rate of 6.200 GPixel/s, a texture rate of 24.80 GTexel/s, and FP32 of 595.2 GFLOPS.

Power and physical characteristics differ substantially. The AMD card has a TDP of 254 W, is dual-slot, requires two 8-pin connectors, and suggests a 600 W PSU. The NVIDIA card has a TDP of 75 W, is an MXM module, requires no connectors, and has no suggested PSU. The bus interfaces are PCIe 2.0 x16 for AMD versus MXM-B (3.0) for NVIDIA. Display outputs are 2x DVI, 1x HDMI 1.3a, and 1x DisplayPort 1.1 for AMD, while NVIDIA is listed as "Portable Device Dependent."

API support also differs. AMD lists DirectX 11.2 (11_0) and OpenGL 4.4. NVIDIA lists DirectX 12 (11_0) and OpenGL 4.6. Neither has Vulkan support. The release dates are close: AMD on September 18, 2011, and NVIDIA on May 29, 2011. Both are end-of-life. The AMD predecessor is Evergreen and successor is Southern Islands. The NVIDIA predecessor is GeForce 400M and successor is GeForce 600M.

Where Each One Wins

Based strictly on the recorded data, the NVIDIA GeForce GTX 560M wins the only benchmark available. It is 18.1% ahead in OpenCL compute performance. This makes it the better choice for any workload that relies on general-purpose compute, such as OpenCL-accelerated applications. Its higher percentile ranking (28th versus 24th) reinforces this position. The GTX 560M also has a much lower power draw (75 W versus 254 W) and no power connector requirement, making it suitable for portable devices where power delivery is limited. Its MXM form factor is specifically designed for laptops and mobile workstations.

The AMD Radeon HD 6850 X2 does not win any benchmark in the database. However, its specification sheet suggests strengths that the benchmark data does not capture. It has more than double the memory bandwidth (134.4 GB/s versus 60.00 GB/s), which could benefit workloads that are memory-bandwidth-bound rather than compute-bound. It also has 2 GB of memory versus 1536 MB, which may help with larger textures or datasets. The raw compute resources (960 shading units, 48 TMUs, 32 ROPs) are far higher than the NVIDIA card, and the pixel and texture rates are several times higher. For traditional rasterization tasks that are not well-represented in OpenCL compute, the AMD card's architecture might perform differently, though the data does not confirm this.

The AMD card is a dual-slot desktop solution with PCIe 2.0 x16 and multiple display outputs (2x DVI, HDMI, DisplayPort). This makes it a fixed desktop component. The NVIDIA card is an MXM module with portable-device-dependent outputs. The choice between them, from a use-case perspective, is largely about form factor: the AMD card belongs in a desktop tower, while the NVIDIA card belongs in a laptop.

The Verdict

The benchmark data is unambiguous: the NVIDIA GeForce GTX 560M outperforms the AMD Radeon HD 6850 X2 in OpenCL compute by 18.1%. The GTX 560M scores 4855 versus 3977, and it holds a higher percentile ranking (28th versus 24th). For any user whose primary concern is compute performance, the NVIDIA card is the clear pick from the recorded measurements.

The AMD card's massive specification advantages (more shading units, more memory bandwidth, higher pixel and texture rates) do not translate into a benchmark win. This is a cautionary tale about relying on paper specifications alone. The TeraScale 2 architecture appears to be less efficient at executing the specific OpenCL workload tested, despite having 1.536 TFLOPS of FP32 performance versus 595.2 GFLOPS for NVIDIA. The Fermi 2.0 architecture, with far fewer resources, achieves a higher score, indicating better instruction-level efficiency for this workload.

The power difference is also striking. The AMD card requires 254 W and two 8-pin connectors, while the NVIDIA card draws 75 W with no connectors. For portable or power-constrained systems, the GTX 560M is the only viable option. For desktop users with a 600 W PSU, the AMD card could be installed, but the data suggests it would still lose the compute benchmark.

The verdict is straightforward: pick the NVIDIA GeForce GTX 560M if you value compute performance, lower power draw, and mobile compatibility. The AMD Radeon HD 6850 X2 offers superior raw specifications in memory bandwidth and shading units, but none of those advantages show up in the actual test results. The data does not support choosing the AMD card for any measured workload. The GTX 560M is the winner in this comparison, and the margin is substantial.