AMD Radeon RX 6800M vs NVIDIA GeForce GTX 1060 5 GB Comparison
AMD Radeon RX 6800M
GeForce GTX 1060 5 GB
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon RX 6800M vs NVIDIA GeForce GTX 1060 5 GB
# FAQ
Q: What is the NVIDIA GeForce GTX 1060 5 GB based on?
A: The NVIDIA GeForce GTX 1060 5 GB is built on the GP106 chip and uses the Pascal architecture. It is manufactured on a 16 nm process at TSMC.
Q: What is the memory configuration of the GTX 1060 5 GB?
A: The card features 5 GB of GDDR5 memory on a 160-bit bus, delivering a bandwidth of 160.2 GB/s.
Q: How does the AMD Radeon RX 6800M compare in raw compute?
A: The RX 6800M is based on the Navi 22 chip with RDNA 2.0 architecture, built on a 7 nm process at TSMC. Its recorded FP32 performance is 12.24 TFLOPS, which is significantly higher than the GTX 1060's 4.375 TFLOPS.
Q: What are the key API and feature differences?
A: The GTX 1060 5 GB supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The RX 6800M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The AMD card also brings 40 ray tracing cores to the table, while the NVIDIA card has none.
Q: Which card has a higher memory bandwidth?
A: The AMD Radeon RX 6800M has a distinct advantage here. Its 192-bit bus and GDDR6 memory provide 384.0 GB/s of bandwidth, versus 160.2 GB/s for the GTX 1060 5 GB.
Q: What is the production status of these cards?
A: Both are listed as End-of-life / End-of-life in the database. The GTX 1060 5 GB was released on 2017-12-25, while the RX 6800M was released on 2021-05-30.
The Verdict
The benchmark data separates these two mobile GPUs cleanly. The NVIDIA GeForce GTX 1060 5 GB has an average benchmark score of 33694, placing it in the 78th percentile of all GPUs in the database. The AMD Radeon RX 6800M is a different class of hardware. Its average score is 28874, but that average is misleading: it wins both head-to-head OpenCL and Vulkan runs against the GTX 1060 by large margins. The recorded data shows the RX 6800M leads by 61.9% in OpenCL and 64.1% in Vulkan. Across all records, the RX 6800M sits in the 74th percentile, a lower percentile ranking than the GTX 1060, but the two cards are not direct peers. The GTX 1060 5 GB wins no head-to-head contest, and the RX 6800M wins both, which makes the pick straightforward: choose the AMD Radeon RX 6800M for workloads that can use its newer feature set and much higher throughput. The GeForce GTX 1060 5 GB remains the lower-power, legacy option with a smaller memory bus.
Head-to-Head Benchmarks
The recorded head-to-head results come from two tests: Geekbench OpenCL and Geekbench Vulkan. In both, the AMD Radeon RX 6800M is the winner.
In Geekbench OpenCL, the AMD Radeon RX 6800M scores 87621 against the NVIDIA GeForce GTX 1060 5 GB's 33397. That is a 61.9% advantage for the AMD part. The margin is even larger in the Vulkan workload. The RX 6800M posts 94766 versus 33991 for the GTX 1060, a 64.1% lead. The AMD card's performance advantage is consistent across both APIs, suggesting a substantial architectural throughput gap rather than a driver quirk.
Specification Differences
The two cards differ across several fundamental specification fields.
Memory configuration is a major separator. The GTX 1060 5 GB uses 5 GB of GDDR5 on a 160-bit bus, for 160.2 GB/s of bandwidth. The RX 6800M goes to 12 GB of GDDR6 on a 192-bit bus, which results in 384.0 GB/s. That is roughly 140% more memory bandwidth for the AMD part, a decisive difference in bandwidth-bound work.
The compute resources also tilt towards AMD. The RX 6800M has 2560 shading units, 160 texture mapping units, and 64 ROPs. The GTX 1060 has 1280 shading units, 80 TMUs, and 40 ROPs. The AMD has exactly double the shader and texture units, and 60% more ROPs.
Clock behavior is a further divergence. The GTX 1060 5 GB runs at a base clock of 1506 MHz and boosts to 1709 MHz. The RX 6800M has a much higher base clock of 2116 MHz and a boost clock of 2390 MHz. The AMD also has a separate "game" clock of 2300 MHz recorded in the data.
Architecturally, the RX 6800M is built on RDNA 2.0 with 40 ray tracing cores and a 7 nm process at TSMC with 17,200 million transistors. The GTX 1060 is Pascal, and TSMC 16 nm is a smaller process, but the 4,400 million transistors show a much smaller and older design. The RX 6800M has no direct predecessor; the GTX 1060 succeeds the GeForce 900 series, and is followed by the GeForce 20 series.
The API support also differs. The AMD card lists DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, while the NVIDIA lists DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The AMD is the only one with dedicated ray tracing cores in the data.
Where Each One Wins
Choose the AMD Radeon RX 6800M if:
- Your workload involves heavy compute or memory bandwidth. The RX 6800M's 12 GB GDDR6 frame buffer as well as its 384.0 GB/s of bandwidth are unmatched by the GTX 1060, and its FP32 of 12.24 TFLOPS is nearly three times as high.
- You want the latest API features and ray tracing. The RDNA 2 architecture has 40 ray tracing cores and supports DirectX 12 Ultimate.
- You are running Vulkan or OpenCL workloads: the head-to-head data shows a 61.9% to 64.1% lead.
Choose the NVIDIA GeForce GTX 1060 5 GB if:
- You need a smaller, lower-power legacy card. The GTX 1060 uses a 120 W TDP, while the RX 6800M is 145 W and it has no auxiliary power connector (the NVIDIA uses a 6-pin connector).
- You are in an older ecosystem (PCIe 3.0) as the RX 6800M is PCIe 4.0 and the GTX 1060 is PCIe 3.0 x16.
- You don't need the extra memory bandwidth and prefer the older GDDR5 memory type and lower 160-bit bus to the 192-bit GDDR6 setup.
The data reveals two different cards for two different purposes. The GTX 1060 5 GB is a more conservative Pascal-era part, strong in the 78th percentile of all GPUs, but overshadowed by the newer RDNA 2 part in the metrics that matter. The RX 6800M is the higher-performance option with more memory, much more bandwidth, and a more modern feature set. If the workload is the latest APIs, large texture data and ray tracing, the AMD card is the one to pick. For anything else, the NVIDIA is a proven, mature, and lower-power alternative.