AMD Radeon 860M vs NVIDIA GeForce RTX 5080 Mobile Comparison

Head-to-Head Benchmarks

The recorded data contains two direct comparisons between the AMD Radeon 860M and the NVIDIA GeForce RTX 5080 Mobile, and both are decisive victories for the NVIDIA part. In Geekbench OpenCL, the Radeon 860M scores 22,759, while the RTX 5080 Mobile reaches 166,986. That is a delta of 86.4% in favor of NVIDIA, meaning the RTX 5080 Mobile delivers roughly seven times the compute throughput in this workload. The gap is slightly narrower but still enormous in Geekbench Vulkan, where the AMD part scores 30,043 against NVIDIA's 169,754, a delta of 82.3%.

The average benchmark score tells the same story at a glance. The Radeon 860M averages 26,401 across its recorded tests, while the RTX 5080 Mobile averages 38,349. The NVIDIA part sits in the 81st percentile of all GPUs in the database, while the AMD part sits in the 72nd percentile. That 9-point percentile gap understates the raw performance difference because the RTX 5080 Mobile's benchmark suite includes far more demanding tests.

When placed against their nearest rivals, the two parts occupy different performance tiers entirely. The Radeon 860M's closest competitor is the NVIDIA GeForce MX550, which scores 26,421, a delta of only 0.1%. The RTX 5080 Mobile's nearest rival is the NVIDIA GeForce MX570 at 38,299, a delta of 0.1%. The AMD part also trades blows with the RTX 5060 (26,331, 0.3% ahead) and the RX 5700 XT 50th Anniversary (26,553, 0.6% behind). The RTX 5080 Mobile similarly sits within 1% of the RTX 4080 Mobile (38,135) and the MX570 A (38,691).

What the head-to-head numbers reveal is that these are not competing products in any practical sense. The RTX 5080 Mobile is a full desktop-class discrete GPU in mobile form, while the Radeon 860M is an integrated graphics solution. The Vulkan test shows the NVIDIA part scoring 5.6 times higher than the AMD part, and the OpenCL test shows a 7.3 times advantage. In both recorded benchmarks, the winner is the RTX 5080 Mobile, and the win is not marginal but categorical.

FAQ

Q: Which GPU wins in Geekbench OpenCL?

A: The NVIDIA GeForce RTX 5080 Mobile wins with a score of 166,986 against the AMD Radeon 860M's 22,759, a delta of 86.4% in NVIDIA's favor.

Q: How do their average benchmark scores compare?

A: The RTX 5080 Mobile averages 38,349 across all recorded benchmarks, while the Radeon 860M averages 26,401. The NVIDIA part also holds a higher percentile rank at 81 versus 72.

Q: What is the closest rival to each GPU?

A: The Radeon 860M's nearest rival is the NVIDIA GeForce MX550 at 26,421 (0.1% delta). The RTX 5080 Mobile's nearest rival is the NVIDIA GeForce MX570 at 38,299 (0.1% delta).

Q: Does the Radeon 860M win any recorded benchmark against the RTX 5080 Mobile?

A: No. In the two direct head-to-head tests, Geekbench OpenCL and Geekbench Vulkan, the RTX 5080 Mobile wins both. The Radeon 860M records zero wins.

Q: How much faster is the RTX 5080 Mobile in Vulkan?

A: The RTX 5080 Mobile scores 169,754 in Geekbench Vulkan versus 30,043 for the Radeon 860M, a delta of 82.3%. That is roughly 5.6 times higher.

Q: What is the RTX 5080 Mobile's percentile ranking?

A: The RTX 5080 Mobile sits in the 81st percentile of all GPUs in the database, compared to the Radeon 860M's 72nd percentile.

Where Each One Wins

The recorded data leaves no ambiguity about which part wins in raw compute. The RTX 5080 Mobile takes both head-to-head benchmarks, and it does so by margins that put the two products in entirely different performance classes. For any workload that stresses OpenCL or Vulkan compute, the NVIDIA part is the clear choice.

The Radeon 860M does not win any benchmark in the database. Its role is as a low-power integrated solution, and its benchmark scores reflect that positioning: 22,759 in OpenCL and 30,043 in Vulkan. These are respectable numbers for an IGP, placing it within 1% of the MX550 and the RX 5700 XT 50th Anniversary. But against the RTX 5080 Mobile, there is no contest.

Where the RTX 5080 Mobile excels is in sustained high-throughput workloads. Its Passmark G3D score of 27,711 and GPU compute score of 12,134 reinforce the picture from the Geekbench tests. The AMD part has no equivalent Passmark scores in the database, so the comparison relies on the two Geekbench tests, both of which NVIDIA wins by over 80%.

For a portable device where power draw matters and integrated graphics is acceptable, the Radeon 860M is a functional option. For any application where GPU compute is the bottleneck, the RTX 5080 Mobile is the only rational pick based on the data.

Specification Differences

The two parts differ in nearly every measurable specification. The Radeon 860M uses 512 shading units, 32 texture mapping units, and 16 ROPs. The RTX 5080 Mobile uses 7,680 shading units, 240 TMUs, and 96 ROPs. That is 15 times more shading units and 7.5 times more TMUs.

Memory configuration is another major divergence. The Radeon 860M uses system-shared memory with a system-dependent bus width and bandwidth. The RTX 5080 Mobile has 16 GB of dedicated GDDR7 memory on a 256-bit bus, delivering 896.0 GB/s of bandwidth. The AMD part's memory bandwidth is listed as system dependent, meaning it scales with the host platform's memory configuration.

Clock speeds also differ significantly. The Radeon 860M has a base clock of 600 MHz and a boost clock of 3000 MHz. The RTX 5080 Mobile has a base clock of 975 MHz and a boost clock of 1500 MHz. Despite the lower boost clock, the NVIDIA part's much larger shader count produces dramatically higher throughput.

Pixel and texture rates reflect the same hierarchy. The Radeon 860M outputs 48.00 GPixel/s and 96.00 GTexel/s. The RTX 5080 Mobile outputs 144.0 GPixel/s and 360.0 GTexel/s. Floating-point performance follows: 3.072 TFLOPS FP32 for AMD versus 23.04 TFLOPS for NVIDIA, a 7.5 times difference. Both parts support FP16 at a 1:1 ratio with FP32.

The RTX 5080 Mobile also carries dedicated hardware that the AMD part lacks. It has 60 ray tracing cores and 240 tensor cores. The Radeon 860M has 8 RT cores and no tensor cores listed. The power envelope is equally lopsided: 15 W TDP for the AMD IGP versus 80 W TDP for the NVIDIA part.

The bus interface differs as well. The Radeon 860M connects via PCIe 4.0 x8, while the RTX 5080 Mobile uses PCIe 5.0 x16. Both are listed as IGP slot width with no power connectors, and both depend on the portable device for display outputs.

Architecture Differences

The architectural divide between these two GPUs is fundamental. The AMD Radeon 860M is built on RDNA 3.5, the latest iteration of AMD's graphics architecture, fabricated on a 4 nm process at TSMC. It belongs to the Navi III IGP generation and uses the Krackan Point chip. The RTX 5080 Mobile is built on Blackwell 2.0, NVIDIA's current architecture, on a 5 nm process also at TSMC, using the GB203 chip in the GeForce 50 Mobile generation.

Process technology favors AMD in density terms, but the NVIDIA chip is far larger and more complex. The GB203 contains 45,600 million transistors on a 378 mm² die, giving a transistor density of 120.6M per mm². The Radeon 860M's transistor count and die size are listed as unknown, so direct density comparison is not possible from the recorded data.

The RDNA 3.5 architecture in the AMD part is designed for efficient integrated graphics. Its 512 shading units and 8 RT cores are modest by any standard, but they operate within a 15 W power envelope. The architecture prioritizes power efficiency over raw throughput. The Blackwell 2.0 architecture in the RTX 5080 Mobile is designed for maximum performance in a mobile discrete GPU. Its 7,680 shading units, 60 RT cores, and 240 tensor cores require an 80 W budget and deliver 23.04 TFLOPS.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API coverage is identical. The release timeline shows the Radeon 860M arriving on 2025-02-28, with the RTX 5080 Mobile following on 2025-04-01. The AMD part's predecessor is listed as Navi II IGP, while the RTX 5080 Mobile's predecessor is GeForce 40 Mobile.

The transistor density figure for the NVIDIA chip, 120.6M per mm², indicates a dense, complex design. The AMD part's 4 nm process is technically smaller than NVIDIA's 5 nm node, but without transistor count data for the Krackan Point chip, the architectural comparison rests on the functional differences: core counts, memory subsystem, and dedicated compute units. The data shows two architectures aimed at different purposes, one for integration and efficiency, the other for high-end mobile compute.