AMD Radeon 860M vs NVIDIA GeForce RTX 2080 SUPER Comparison

The AMD Radeon 860M and the NVIDIA GeForce RTX 2080 SUPER occupy opposite ends of the hardware spectrum: one is a 15 W integrated processor designed for thin-and-light laptops, the other a 250 W discrete add-in card from a previous generation. The benchmark data shows a decisive performance gap, but the more interesting story lies in what each part represents for the systems they target. The RTX 2080 SUPER is the clear performance victor, while the 860M redefines what an iGPU can do, landing near the same overall percentile despite its power envelope.

Head-to-Head Benchmarks

The two shared benchmark tests tell a one-sided story. In Geekbench OpenCL, the RTX 2080 SUPER scores 99,226, which is 77.1% higher than the Radeon 860M’s 22,759. The Vulkan result is similarly lopsided: NVIDIA’s card posts 111,284 against AMD’s 30,043, a 73% deficit for the iGPU. These are not close calls. The discrete card’s advantage comes from having 6x the shading units (3072 vs 512), 6x the texture units (192 vs 32), and 4x the ROPs (64 vs 16), along with dedicated memory bandwidth.

However, context matters. The 860M’s average benchmark score of 26,401 places it at the 72nd percentile of all GPUs, which is higher than the RTX 2080 SUPER’s 69th percentile (average score 24,170). This is because the 860M’s scores are clustered in modern API tests, while the RTX 2080 SUPER’s average includes legacy DirectX 9, 10, 11, and 12 Passmark results that drag its aggregate down. In the head-to-head Vulkan test, the 860M actually performs better relative to its own average than the RTX 2080 SUPER does, suggesting the RDNA 3.5 architecture handles modern APIs more efficiently per unit of compute.

The deltaPct figures in the rivals list reinforce this. The 860M sits within 1.1% of the NVIDIA RTX A4000 and the AMD RX 5700 XT 50th Anniversary, both of which are far more power-hungry discrete cards. The RTX 2080 SUPER, meanwhile, is statistically tied with the GTX 780 Ti (0.3% slower) and GTX 1630 (0.4% slower), indicating that its raw compute advantage has aged poorly against newer architectures.

Architecture Differences

The fundamental divide is process node and design philosophy. The Radeon 860M uses a 4 nm TSMC process, while the RTX 2080 SUPER uses 12 nm TSMC. This explains the massive efficiency gap: the 860M carries a 15 W TDP, the RTX 2080 SUPER a 250 W TDP. AMD’s chip integrates 512 shading units, 32 TMUs, 16 ROPs, and 8 ray tracing cores into a Krackan Point IGP with RDNA 3.5 architecture. NVIDIA’s TU104 die packs 3,072 shading units, 192 TMUs, 64 ROPs, 48 RT cores, and 384 tensor cores into a Turing architecture part.

Memory is a stark differentiator. The 860M uses system shared memory with bandwidth described as “System Dependent,” meaning performance scales with the laptop’s RAM configuration. The RTX 2080 SUPER has 8 GB of GDDR6 on a 256-bit bus, delivering 495.9 GB/s. That fixed bandwidth is a major reason the discrete card wins so decisively in compute-heavy tasks.

The 860M’s FP32 throughput is 3.072 TFLOPS, and its FP16 rate is identical at 3.072 TFLOPS (1:1 ratio). The RTX 2080 SUPER pushes 11.15 TFLOPS FP32 and 22.30 TFLOPS FP16 (2:1 ratio), showing NVIDIA’s Turing architecture favored mixed-precision compute for AI workloads via tensor cores. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, but the RTX 2080 SUPER’s 384 tensor cores give it a feature the 860M lacks entirely.

The 860M has no power connectors and is an IGP, while the RTX 2080 SUPER requires a dual-slot cooler, 1x 6-pin + 1x 8-pin power connectors, and a suggested 600 W PSU. The 860M uses PCIe 4.0 x8; the RTX 2080 SUPER uses PCIe 3.0 x16. The 860M is still in active production; the RTX 2080 SUPER is end-of-life.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The AMD Radeon 860M has an average benchmark score of 26,401, which is higher than the RTX 2080 SUPER’s 24,170. However, the RTX 2080 SUPER wins both head-to-head tests by large margins.

Q: Why does the 860M have a higher percentile rank despite losing head-to-head?

A: The 860M sits at the 72nd percentile of all GPUs, while the RTX 2080 SUPER is at the 69th. The RTX 2080 SUPER’s average is dragged down by low Passmark DirectX 9, 10, and 12 scores (227, 140, and 75 respectively), while the 860M only has modern Geekbench results in its average.

Q: What is the power consumption difference?

A: The 860M has a 15 W TDP, while the RTX 2080 SUPER has a 250 W TDP. The RTX 2080 SUPER also requires a 600 W suggested PSU and two power connectors, whereas the 860M uses none.

Q: Can the 860M handle ray tracing?

A: Yes, the 860M has 8 ray tracing cores, though the RTX 2080 SUPER has 48. Both support DirectX 12 Ultimate, which includes ray tracing features.

Q: How do their memory subsystems compare?

A: The 860M uses system shared memory with system-dependent bandwidth. The RTX 2080 SUPER has 8 GB of GDDR6 on a 256-bit bus with 495.9 GB/s bandwidth.

Q: Which GPU is newer?

A: The 860M was released on 2025-02-28, while the RTX 2080 SUPER was released on 2019-07-22. The 860M is still in active production; the RTX 2080 SUPER is end-of-life.

Specification Differences

The key differences are in core configuration, memory, power, and physical design.

• Process Node: 4 nm (860M) vs 12 nm (RTX 2080 SUPER)
• Transistors: Unknown (860M) vs 13,600 million (RTX 2080 SUPER)
• Die Size: Unknown (860M) vs 545 mm² (RTX 2080 SUPER)
• Transistor Density: Not listed (860M) vs 25.0M / mm² (RTX 2080 SUPER)
• Base Clock: 600 MHz (860M) vs 1650 MHz (RTX 2080 SUPER)
• Boost Clock: 3000 MHz (860M) vs 1815 MHz (RTX 2080 SUPER)
• Memory: System Shared (860M) vs 8 GB GDDR6 (RTX 2080 SUPER)
• Memory Bus: System Shared (860M) vs 256 bit (RTX 2080 SUPER)
• Memory Bandwidth: System Dependent (860M) vs 495.9 GB/s (RTX 2080 SUPER)
• Shading Units: 512 (860M) vs 3072 (RTX 2080 SUPER)
• TMUs: 32 (860M) vs 192 (RTX 2080 SUPER)
• ROPs: 16 (860M) vs 64 (RTX 2080 SUPER)
• RT Cores: 8 (860M) vs 48 (RTX 2080 SUPER)
• Tensor Cores: None (860M) vs 384 (RTX 2080 SUPER)
• Pixel Rate: 48.00 GPixel/s (860M) vs 116.2 GPixel/s (RTX 2080 SUPER)
• Texture Rate: 96.00 GTexel/s (860M) vs 348.5 GTexel/s (RTX 2080 SUPER)
• FP32: 3.072 TFLOPS (860M) vs 11.15 TFLOPS (RTX 2080 SUPER)
• FP16: 3.072 TFLOPS (1:1) (860M) vs 22.30 TFLOPS (2:1) (RTX 2080 SUPER)
• TDP: 15 W (860M) vs 250 W (RTX 2080 SUPER)
• Slot Width: IGP (860M) vs Dual-slot (RTX 2080 SUPER)
• Power Connectors: None (860M) vs 1x 6-pin + 1x 8-pin (RTX 2080 SUPER)
• Bus Interface: PCIe 4.0 x8 (860M) vs PCIe 3.0 x16 (RTX 2080 SUPER)
• Release Date: 2025-02-28 (860M) vs 2019-07-22 (RTX 2080 SUPER)
• Launch MSRP: Not available (860M) vs 699 USD (RTX 2080 SUPER)

Where Each One Wins

The RTX 2080 SUPER wins every benchmark where the two are compared directly. Its 77.1% lead in OpenCL and 73% lead in Vulkan are decisive, driven by 6x the shading units and dedicated GDDR6 memory. It is the clear choice for any workload that demands raw compute throughput, such as video rendering, 3D modeling, or high-refresh gaming at resolutions above 1080p. Its 11.15 TFLOPS FP32 and 22.30 TFLOPS FP16 also make it suitable for compute tasks that leverage NVIDIA’s tensor cores, which the 860M lacks entirely.

The 860M wins in efficiency and system integration. It draws 15 W compared to 250 W, requires no power connectors, and fits into any laptop chassis as an IGP. Its 3.072 TFLOPS FP32 and 8 RT cores are remarkable for a chip that shares system memory, and its 72nd percentile ranking proves that it outperforms many older discrete GPUs. It is the better choice for ultraportable systems where battery life and thermals matter more than frame rates. The 860M’s identical 1:1 FP16 rate also suggests it handles modern shader workloads more predictably than the Turing card’s 2:1 ratio.

The Verdict

The data is unambiguous: if raw performance is the only criterion, the RTX 2080 SUPER wins by a massive margin. Its 99,226 OpenCL and 111,284 Vulkan scores dwarf the 860M’s 22,759 and 30,043. Anyone building a desktop or a thick gaming laptop with access to a 600 W PSU should choose the RTX 2080 SUPER, provided they can accept its 250 W power draw and dual-slot footprint.

But the 860M’s performance is remarkable for what it is. It achieves a 72nd percentile ranking on a 15 W TDP with system shared memory, beating the RTX 2080 SUPER’s 69th percentile overall. The 860M is the right choice for a thin-and-light laptop that needs competent 1080p gaming or GPU acceleration without a discrete card. It offers modern features like ray tracing cores and PCIe 4.0, and it is still in active production. The RTX 2080 SUPER is end-of-life, and its 12 nm process and 699 USD launch MSRP reflect an earlier era of GPU design. The 860M represents the future of integrated graphics; the RTX 2080 SUPER represents the past of discrete performance. Pick the 860M for portability and efficiency, or the RTX 2080 SUPER for sheer compute power — the benchmarks make the trade-off clear.