AMD Radeon HD 6850 X2 vs NVIDIA GeForce GTX 460M Comparison

The Verdict

The benchmark data presents a clear, if somewhat counterintuitive, outcome. The NVIDIA GeForce GTX 460M is the decisive winner in the single available head-to-head metric, the Geekbench OpenCL test, posting a score of 4282 against the AMD Radeon HD 6850 X2's 3977. This represents a 7.7% performance advantage for the NVIDIA part. This is a significant result because it flips the expectation set by the raw specifications; the AMD card is a larger, hungrier, and numerically more powerful piece of hardware on paper, yet it loses in the actual compute benchmark.

Given this data, the choice is straightforward for anyone prioritizing raw compute performance as measured by this test: the GTX 460M is the superior option. Its higher benchmark score and better percentile ranking (25th vs 24th percentile) make it the more capable part for general compute workloads. However, the choice is not without context. The AMD card consumes a massive amount of power and requires a robust power supply, whereas the NVIDIA chip is a low-power mobile part. If the use case is a desktop system with a strong power supply and the need for multiple display outputs, the HD 6850 X2's feature set might be appealing, but its lower benchmark score makes it hard to recommend on performance alone. For a builder strictly looking at the data, the GTX 460M is the winner.

Architecture Differences

The two GPUs come from different architectural philosophies and eras. The NVIDIA GeForce GTX 460M is built on the Fermi architecture, using the GF106 chip, and belongs to the GeForce 400M generation. In contrast, the AMD Radeon HD 6850 X2 is based on the TeraScale 2 architecture, utilizing the Barts chip, and is part of the Northern Islands (HD 6800) generation. Both are manufactured by TSMC on a 40 nm process node, but the similarities end there.

The transistor counts and die sizes reveal stark differences in complexity. The AMD Barts chip packs 1,700 million transistors on a 255 mm² die, resulting in a transistor density of 6.7M / mm². The NVIDIA GF106 is a smaller chip, with 1,170 million transistors on a 238 mm² die, for a lower density of 4.9M / mm². This indicates that AMD's design is more complex and densely packed, yet the NVIDIA chip still manages to win the benchmark. The NVIDIA part is a mobile-focused "MXM Module," while the AMD card is a dual-slot desktop behemoth. The AMD card's architecture is fundamentally designed for desktop power and throughput, while NVIDIA's is for a more constrained mobile environment, making the GTX 460M's benchmark victory even more notable.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The NVIDIA GeForce GTX 460M has the higher score at 4282, compared to the AMD Radeon HD 6850 X2's 3977. This gives the NVIDIA part a 7.7% lead in this specific test.

Q: What are the power requirements for each card?

A: The AMD Radeon HD 6850 X2 has a TDP of 254 W and requires a 600 W power supply and two 8-pin power connectors. The NVIDIA GeForce GTX 460M, in contrast, has a TDP of only 50 W and requires no power connectors, drawing power directly from its MXM Module slot.

Q: How much memory does each card have and what is the bus width?

A: The NVIDIA GeForce GTX 460M comes with 1536 MB of GDDR5 memory on a 192-bit bus. The AMD Radeon HD 6850 X2 has 2 GB of GDDR5 memory on a wider 256-bit bus.

Q: Which card offers more display outputs?

A: The AMD Radeon HD 6850 X2 provides a fixed set of outputs including 2x DVI, 1x HDMI 1.3a, and 1x DisplayPort 1.1. The NVIDIA GeForce GTX 460M's display outputs are "Portable Device Dependent," meaning they are determined by the laptop manufacturer.

Q: What is the memory bandwidth difference?

A: The AMD Radeon HD 6850 X2 has a significantly higher memory bandwidth of 134.4 GB/s, while the NVIDIA GeForce GTX 460M has 60.00 GB/s. Despite this, the NVIDIA card still wins the compute benchmark.

Q: What are the pixel and texture rates for each card?

A: The AMD Radeon HD 6850 X2 has a pixel rate of 25.60 GPixel/s and a texture rate of 38.40 GTexel/s. The NVIDIA GeForce GTX 460M has a lower pixel rate of 5.400 GPixel/s and a texture rate of 21.60 GTexel/s.

Specification Differences

The specification sheets for these two GPUs are wildly different, reflecting their divergent target platforms. The most glaring difference is in power consumption: the NVIDIA card is rated at 50 W TDP with no power connectors, while the AMD card is rated at 254 W TDP and requires 2x 8-pin connectors and a 600 W suggested PSU. The AMD card has a higher transistor count (1,700 million vs 1,170 million) and a larger die (255 mm² vs 238 mm²).

Memory configurations also differ significantly. The AMD card features 2 GB of memory on a 256-bit bus, yielding 134.4 GB/s of bandwidth. The NVIDIA card has 1536 MB on a 192-bit bus, providing 60.00 GB/s. The compute resources are also lopsided: the AMD card has 960 shading units, 48 TMUs, and 32 ROPs, while the NVIDIA card has 192 shading units, 32 TMUs, and 24 ROPs. The AMD card also has much higher theoretical throughput rates, including 1.536 TFLOPS of FP32 compute and a pixel rate of 25.60 GPixel/s, versus the NVIDIA card's 518.4 GFLOPS and 5.400 GPixel/s.

The form factors are incompatible. The NVIDIA part is an MXM Module for laptops, while the AMD part is a Dual-slot desktop card. The AMD card supports a fixed set of display outputs (2x DVI, 1x HDMI 1.3a, 1x DisplayPort 1.1), while the NVIDIA card's outputs are "Portable Device Dependent." The API support also shows minor differences, with the NVIDIA card supporting DirectX 12 (11_0) and OpenGL 4.6, while the AMD card supports DirectX 11.2 (11_0) and OpenGL 4.4. Neither card supports Vulkan.

Head-to-Head Benchmarks

The only direct comparison available is the Geekbench OpenCL test, and it yields a decisive result. The NVIDIA GeForce GTX 460M scores 4282, while the AMD Radeon HD 6850 X2 scores 3977. This is a 7.7% win for the NVIDIA card. This is a remarkable outcome given the hardware specifications. The AMD card has 5 times the shading units (960 vs 192), more than double the memory bandwidth (134.4 GB/s vs 60.00 GB/s), and almost three times the FP32 compute throughput (1.536 TFLOPS vs 518.4 GFLOPS). Yet, in this real-world compute test, the NVIDIA Fermi architecture proves to be more efficient and effective at executing the workload.

This result is further contextualized by the percentile rankings. The GTX 460M sits in the 25th percentile of all GPUs, while the HD 6850 X2 sits in the 24th percentile. The nearest rivals for the NVIDIA card include the AMD FirePro W2100 (score 4295, delta -0.3%) and the NVIDIA Quadro K3000M (score 4241, delta 1%). The AMD card's nearest rivals include the NVIDIA Quadro K2000 (score 3964, delta 0.3%) and the NVIDIA GeForce 830M (score 3957, delta 0.5%). This shows that the GTX 460M is competitive with professional workstation cards, while the HD 6850 X2 is closer to entry-level mobile parts, despite its desktop-class power draw.

Where Each One Wins

Based strictly on the benchmark data, the NVIDIA GeForce GTX 460M wins in the only category that matters for direct comparison: compute performance. Its 4282 Geekbench OpenCL score is superior, and it holds a higher percentile rank. This makes it the clear choice for any application where OpenCL compute performance is the primary bottleneck. Its extremely low 50 W TDP is also a massive advantage, making it suitable for systems with limited power delivery or thermal headroom, such as the laptops it was designed for.

The AMD Radeon HD 6850 X2 does not win any benchmark head-to-head comparisons in this data. However, its specification sheet points to potential advantages in other scenarios that are not tested here. Its 2 GB of memory and 256-bit bus provide significantly more memory bandwidth, which could be beneficial for certain tasks. Its support for 2x DVI, 1x HDMI 1.3a, and 1x DisplayPort 1.1 outputs makes it a more flexible option for connecting multiple desktop monitors. It also has a much higher theoretical pixel rate (25.60 GPixel/s) and texture rate (38.40 GTexel/s), suggesting it could excel in fill-rate-bound workloads, despite its lower overall compute score. Ultimately, the data shows the NVIDIA card is the winner in the available test, while the AMD card's advantages are purely theoretical based on its unverified specifications.