AMD Radeon HD 6850 X2 vs AMD Radeon R5 M320 Comparison

Where Each One Wins

The benchmark database records only a single shared test between these two GPUs, the Geekbench OpenCL compute workload, and the AMD Radeon R5 M320 takes that test decisively. The R5 M320 scores 5051 points against the HD 6850 X2's 3977 points, a 27% advantage in raw compute throughput. This is the only head-to-head metric available, so the R5 M320 holds a clean sweep in the recorded data with one win and zero losses.

The Radeon R5 M320 also appears in a second benchmark, Geekbench Vulkan, where it scores 4262 points. The HD 6850 X2 has no recorded Vulkan result in the database, meaning the R5 M320 is the only one of the two with any measurable modern API compute performance. The R5 M320's average benchmark score across all recorded tests sits at 4657, placing it in the 27th percentile of all GPUs tracked by the database. The HD 6850 X2 has an average score of 3977, which puts it in the 24th percentile. The percentile gap is small, but the R5 M320 sits slightly higher in the overall distribution.

The HD 6850 X2 does not win any recorded benchmark category. Its only entry, the OpenCL score, is lower than the R5 M320's equivalent result. However, the HD 6850 X2 was built for a different era of graphics workloads, and its architectural strengths lie in areas the database does not capture in these two compute-focused tests. The R5 M320 wins the compute race in every measured category, while the HD 6850 X2's case rests entirely on specifications rather than recorded benchmark results.

Architecture Differences

The two GPUs come from fundamentally different architectural lineages. The Radeon R5 M320 uses the Jet chip built on GCN 1.0 architecture, manufactured on a 28 nm process at TSMC. The HD 6850 X2 uses the Barts chip built on TeraScale 2 architecture, manufactured on a 40 nm process, also at TSMC. The process node difference is significant: 28 nm versus 40 nm gives the newer chip a density advantage that shows up across the specification sheet.

Transistor counts tell that story clearly. The Jet chip packs 690 million transistors into a 56 mm² die, yielding a transistor density of 12.3 million per square millimeter. The Barts chip carries 1,700 million transistors across a 255 mm² die, for a density of just 6.7 million per square millimeter. The R5 M320's chip is both smaller and denser, a direct result of the newer manufacturing process.

The R5 M320 belongs to the Gem System generation, specifically the R5 M300 family, and lists Solar System as its predecessor and Polaris Mobile as its successor. The HD 6850 X2 belongs to the Northern Islands generation, the HD 6800 family, with Evergreen as its predecessor and Southern Islands as its successor. The R5 M320 was released on May 4, 2015, while the HD 6850 X2 came out on September 18, 2011, a gap of roughly four years that explains the architectural shift.

Memory architecture differs sharply. The R5 M320 uses 4 GB of DDR3 on a 64 bit bus, delivering 16.00 GB/s of bandwidth. The HD 6850 X2 uses 2 GB of GDDR5 on a 256 bit bus, delivering 134.4 GB/s of bandwidth. The HD 6850 X2 has a massive 8.4x bandwidth advantage, a figure that reflects its desktop dual-GPU heritage versus the R5 M320's integrated-class mobile positioning. The R5 M320's memory runs at 1000 MHz with 2 Gbps effective, while the HD 6850 X2's memory runs at 1050 MHz with 4.2 Gbps effective.

The HD 6850 X2 is a dual-slot card with two 8-pin power connectors and a 254 W TDP, with a suggested power supply of 600 W. The R5 M320 is an IGP, meaning integrated graphics, with no recorded TDP, no power connectors, and no suggested PSU. The slot width for the R5 M320 is listed as IGP, indicating it is not a discrete add-in card. The HD 6850 X2's display outputs are 2x DVI, 1x HDMI 1.3a, and 1x DisplayPort 1.1, while the R5 M320's display outputs are listed as portable device dependent, reflecting its mobile integration.

API support also differs. The R5 M320 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The HD 6850 X2 supports DirectX 11.2 (11_0) and OpenGL 4.4, with no Vulkan support recorded. This makes the R5 M320 the only one of the two capable of running Vulkan workloads at all.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon R5 M320 has an average benchmark score of 4657, while the AMD Radeon HD 6850 X2 has an average score of 3977.

Q: How much faster is the R5 M320 in the shared OpenCL benchmark?

A: The R5 M320 scores 5051 in Geekbench OpenCL, which is 27% higher than the HD 6850 X2's 3977.

Q: Does the HD 6850 X2 support Vulkan?

A: No. The HD 6850 X2 has no Vulkan support recorded in the database, while the R5 M320 supports Vulkan 1.2.170.

Q: Which GPU has more memory bandwidth?

A: The HD 6850 X2 has 134.4 GB/s of bandwidth from its 256 bit GDDR5 memory, far exceeding the R5 M320's 16.00 GB/s from its 64 bit DDR3 memory.

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

A: The R5 M320's Jet chip has a density of 12.3M transistors per square millimeter, while the HD 6850 X2's Barts chip has a density of 6.7M per square millimeter.

Q: Which GPU has the higher shading unit count?

A: The HD 6850 X2 has 960 shading units, compared to the R5 M320's 320 shading units.

Specification Differences

The R5 M320 and HD 6850 X2 diverge across nearly every hardware specification. The R5 M320 uses the Jet chip with GCN 1.0 architecture, while the HD 6850 X2 uses the Barts chip with TeraScale 2 architecture. The process node is 28 nm for the R5 M320 and 40 nm for the HD 6850 X2. Transistor count is 690 million for the R5 M320 versus 1,700 million for the HD 6850 X2, with die sizes of 56 mm² and 255 mm² respectively.

The R5 M320 has no base or boost clock recorded, while the HD 6850 X2 also has no base or boost clock recorded. Memory clocks differ slightly: the R5 M320 runs at 1000 MHz with 2 Gbps effective, the HD 6850 X2 at 1050 MHz with 4.2 Gbps effective. Memory capacity is 4 GB DDR3 for the R5 M320 versus 2 GB GDDR5 for the HD 6850 X2. The bus width is 64 bit for the R5 M320 and 256 bit for the HD 6850 X2. Bandwidth is 16.00 GB/s versus 134.4 GB/s.

The R5 M320 has 320 shading units, 20 texture units, and 8 ROPs, while the HD 6850 X2 has 960 shading units, 48 texture units, and 32 ROPs. Pixel rate is 6.840 GPixel/s for the R5 M320 and 25.60 GPixel/s for the HD 6850 X2. Texture rate is 17.10 GTexel/s versus 38.40 GTexel/s. FP32 compute is 547.2 GFLOPS for the R5 M320 and 1.536 TFLOPS for the HD 6850 X2. The HD 6850 X2 has a 254 W TDP, while the R5 M320 has no TDP recorded and is classified as an IGP. The HD 6850 X2 is dual-slot with 2x 8-pin power connectors and a 600 W suggested PSU, none of which apply to the R5 M320. The bus interface is PCIe 3.0 x8 for the R5 M320 and PCIe 2.0 x16 for the HD 6850 X2.

Head-to-Head Benchmarks

The only direct comparison in the database is Geekbench OpenCL, and the result is unambiguous. The AMD Radeon R5 M320 posts 5051 points, while the AMD Radeon HD 6850 X2 posts 3977 points. That is a 27% delta in favor of the R5 M320, a substantial margin in a compute workload that stresses raw shader throughput and memory efficiency.

The R5 M320's 27% lead in OpenCL is notable given the HD 6850 X2's hardware advantages on paper. The HD 6850 X2 has three times the shading units, 960 versus 320, and more than eight times the memory bandwidth, 134.4 GB/s versus 16.00 GB/s. Yet the R5 M320 still wins the compute test comfortably. This suggests the GCN 1.0 architecture extracts far more useful compute throughput per resource than TeraScale 2, and the newer 28 nm process likely contributes to better instruction efficiency.

The R5 M320 also has a recorded Vulkan score of 4262, which the HD 6850 X2 cannot match because it has no Vulkan support at all. The HD 6850 X2's DirectX support tops out at 11.2 (11_0), while the R5 M320 supports DirectX 12 (11_1). In any modern API workload, the R5 M320 is the only contender.

The average benchmark scores reinforce the single-test result. The R5 M320 averages 4657 across its recorded benchmarks, which places it at the 27th percentile of all GPUs. The HD 6850 X2 averages 3977, at the 24th percentile. The R5 M320's nearest rivals in the database include the AMD Radeon RX 9060 XT 16 GB at an identical average score of 4657, the NVIDIA Quadro P400 at 4684 with a 0.6% delta, the NVIDIA GeForce GTX 970M at 4628 with a 0.6% delta, and the NVIDIA Quadro M3000M at 4621 with a 0.8% delta. The HD 6850 X2's nearest rivals include the NVIDIA Quadro K2000 at 3964 with a 0.3% delta, the NVIDIA GeForce 830M at 3957 with a 0.5% delta, the AMD Radeon R5 M420 at 3956 with a 0.5% delta, and the NVIDIA GeForce GT 745M at 3953 with a 0.6% delta.

The raw compute numbers align with the benchmark outcomes. The R5 M320 delivers 547.2 GFLOPS of FP32 performance, while the HD 6850 X2 delivers 1.536 TFLOPS, nearly three times as much on paper. The HD 6850 X2's pixel rate of 25.60 GPixel/s and texture rate of 38.40 GTexel/s also dwarf the R5 M320's 6.840 GPixel/s and 17.10 GTexel/s. Yet in the actual recorded OpenCL test, the R5 M320 wins by 27%. The database shows that theoretical throughput does not translate directly into compute benchmark success, and the R5 M320's architectural efficiency carries the day.