AMD Radeon RX 6700M vs NVIDIA GeForce RTX 2080 Comparison

The Verdict

The recorded data splits this comparison cleanly along workload lines. The AMD Radeon RX 6700M wins only one of the ten head-to-head benchmark comparisons, while the NVIDIA GeForce RTX 2080 takes the remaining nine. Based on the database, the RTX 2080 is the stronger overall performer, especially in legacy DirectX workloads and general compute. However, the RX 6700M holds a narrow but real lead in the modern 3DMark Steel Nomad DX12 test, which suggests it is the better choice for newer, DirectX 12-centric rendering paths.

For a laptop user prioritizing current-generation game engines and power efficiency, the RX 6700M makes sense. It delivers a 5.3% win in the DX12 Steel Nomad test while consuming 135 W versus the RTX 2080's 215 W. For desktop users running a mix of older titles, Vulkan applications, and compute-heavy tasks, the RTX 2080 is the clear pick, as it leads by double-digit percentages in eight of the ten recorded benchmarks. The RTX 2080 also carries a higher average benchmark score of 22895 against the RX 6700M's 25633, but note that this average includes different test suites, so the head-to-head deltas are more meaningful.

The percentile rankings reinforce this split. The RX 6700M sits at the 71st percentile of all GPUs, while the RTX 2080 sits at the 68th. That three-point gap is small, and the nearest rivals for each card include cards with nearly identical average scores. The RX 6700M's closest rival, the AMD Radeon Pro W5700, is only 0.4% behind it, while the RTX 2080's nearest rival, the Intel Arc B580, is 0.5% ahead. Neither card is dominant in its class.

Architecture Differences

The two GPUs come from different manufacturing nodes and design philosophies. The AMD Radeon RX 6700M uses the Navi 22 chip built on TSMC's 7 nm process, while the NVIDIA GeForce RTX 2080 uses the TU104 chip on TSMC's 12 nm node. The RX 6700M packs 17,200 million transistors into a 335 mm² die, giving a transistor density of 51.3 million per square millimeter. The RTX 2080 has 13,600 million transistors on a much larger 545 mm² die, resulting in a density of 25.0 million per square millimeter. The newer 7 nm process allows the AMD chip to be smaller and denser.

The architecture names reflect their generations. The RX 6700M is RDNA 2.0, part of the Radeon RX 6000 series, while the RTX 2080 is Turing, part of the GeForce 20-series. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature parity is not a differentiator.

Memory configurations differ significantly. The RX 6700M has 10 GB of GDDR6 on a 160-bit bus, delivering 320.0 GB/s of bandwidth. The RTX 2080 has 8 GB of GDDR6 on a 256-bit bus, delivering 448.0 GB/s. The NVIDIA card has a wider bus and higher bandwidth, but the AMD card has more capacity. Clock speeds also diverge: the RX 6700M boosts to 2400 MHz with a game clock of 2300 MHz, while the RTX 2080 boosts to 1710 MHz. The AMD card's memory runs at 16 Gbps effective, versus 14 Gbps for the NVIDIA card.

Compute unit counts tell a similar story. The RX 6700M has 2304 shading units, 144 texture mapping units, and 64 raster output units, plus 36 ray tracing cores. The RTX 2080 has 2944 shading units, 184 TMUs, and 64 ROPs, plus 46 RT cores and 368 tensor cores. The NVIDIA card has more shading units and dedicated tensor cores, which the AMD card lacks entirely. In raw throughput, the RX 6700M reaches 11.06 TFLOPS FP32 and 22.12 TFLOPS FP16, while the RTX 2080 reaches 10.07 TFLOPS FP32 and 20.14 TFLOPS FP16. The AMD card is slightly ahead in peak FP32 and FP16 rates despite having fewer shading units, thanks to its higher boost clock.

Physical design differs as well. The RX 6700M is an integrated GPU package with no slot width, no power connectors, and portable-device-dependent display outputs. The RTX 2080 is a dual-slot card measuring 267 mm in length, 116 mm in height, and 35 mm in width, requiring a 1x 6-pin plus 1x 8-pin power connection and a 550 W suggested power supply. The RTX 2080 also offers fixed display outputs: 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C.

FAQ

Q: Which GPU is faster in the modern 3DMark Steel Nomad DX12 benchmark?

A: The AMD Radeon RX 6700M wins this test with a score of 1845 against the RTX 2080's 1752, a 5.3% advantage.

Q: Which GPU has more memory?

A: The RX 6700M has 10 GB of GDDR6, while the RTX 2080 has 8 GB of GDDR6.

Q: Which GPU has higher memory bandwidth?

A: The RTX 2080 leads with 448.0 GB/s on a 256-bit bus, compared to the RX 6700M's 320.0 GB/s on a 160-bit bus.

Q: Does the RTX 2080 support tensor cores?

A: Yes, the RTX 2080 includes 368 tensor cores, a feature absent from the RX 6700M.

Q: Which GPU has a higher boost clock?

A: The RX 6700M boosts to 2400 MHz, whereas the RTX 2080 boosts to 1710 MHz.

Q: What is the power draw difference?

A: The RX 6700M is rated at 135 W, while the RTX 2080 is rated at 215 W.

Specification Differences

The two GPUs differ across nearly every measured specification. The RX 6700M uses a 7 nm process from TSMC, while the RTX 2080 uses 12 nm. The transistor counts are 17,200 million versus 13,600 million. The die sizes are 335 mm² versus 545 mm². The base clocks are 1489 MHz versus 1515 MHz, and the boost clocks are 2400 MHz versus 1710 MHz. The RX 6700M has a game clock of 2300 MHz, which the RTX 2080 does not list.

Memory differs in size, type, bus width, and bandwidth. The RX 6700M offers 10 GB of GDDR6, a 160-bit bus, and 320.0 GB/s bandwidth. The RTX 2080 offers 8 GB of GDDR6, a 256-bit bus, and 448.0 GB/s bandwidth. Memory clocks are 2000 MHz (16 Gbps effective) for the AMD card and 1750 MHz (14 Gbps effective) for the NVIDIA card.

Compute resources differ. The RX 6700M has 2304 shading units, 144 TMUs, and 64 ROPs, with 36 RT cores and no tensor cores. The RTX 2080 has 2944 shading units, 184 TMUs, and 64 ROPs, with 46 RT cores and 368 tensor cores. Pixel rates are 153.6 GPixel/s for the AMD card and 109.4 GPixel/s for the NVIDIA card. Texture rates are 345.6 GTexel/s versus 314.6 GTexel/s. FP32 throughput is 11.06 TFLOPS versus 10.07 TFLOPS, and FP16 throughput is 22.12 TFLOPS versus 20.14 TFLOPS.

Power and physical specs also diverge. The RX 6700M has a 135 W TDP, is an IGP form factor, has no power connectors, and uses PCIe 4.0 x16. The RTX 2080 has a 215 W TDP, is dual-slot, requires 1x 6-pin plus 1x 8-pin connectors, and uses PCIe 3.0 x16. The RTX 2080 has fixed display outputs and listed dimensions, while the RX 6700M's outputs are portable-device dependent and its dimensions are unlisted.

Head-to-Head Benchmarks

The head-to-head data shows one AMD win and nine NVIDIA wins. The RX 6700M's sole victory comes in the 3dmark_3dmark_steel_nomad_dx12 test, scoring 1845 against the RTX 2080's 1752, a 5.3% margin. This is a meaningful result for modern DirectX 12 workloads, where the AMD architecture's higher clock speed compensates for its narrower memory bus.

The RTX 2080 dominates the remaining tests. In geekbench_opencl, it scores 91313 versus 77666, a 14.9% lead. In geekbench_vulkan, it scores 107797 versus 90816, a 15.8% lead. The Vulkan gap is notable because both cards support Vulkan 1.4, yet the NVIDIA card's shading unit count and bandwidth advantage show through.

PassMark results are heavily NVIDIA-favored. In DirectX 10, the RTX 2080 scores 136 versus 92, a 32.4% lead. In DirectX 11, it scores 158 versus 129, an 18.4% lead. In DirectX 12, it scores 72 versus 65, a 9.7% lead. In DirectX 9, it scores 223 versus 155, a 30.5% lead. The largest delta is in the 2D test: passmark_g2d shows 907 versus 555, a 38.8% lead for NVIDIA. In passmark_g3d, the RTX 2080 scores 18720 versus 13536, a 27.7% lead. In passmark_gpu_compute, it scores 7872 versus 5191, a 34.1% lead.

The pattern is consistent: the RTX 2080 wins every legacy and compute test by margins ranging from 9.7% to 38.8%, while the RX 6700M wins only the single modern DX12 test. The RX 6700M's higher pixel rate (153.6 GPixel/s versus 109.4 GPixel/s) and texture rate (345.6 GTexel/s versus 314.6 GTexel/s) do not translate into wins outside of Steel Nomad.

Where Each One Wins

The AMD Radeon RX 6700M wins in scenarios that stress modern DirectX 12 rendering. Its 5.3% advantage in the Steel Nomad DX12 test indicates it handles current-generation game engines well. It also offers 10 GB of memory, which is useful for high-resolution textures, and it does so with a much lower power draw of 135 W versus 215 W. For laptops, the RX 6700M's IGP form factor and lack of external power connectors make it the only viable choice in this comparison, as the RTX 2080 is a dual-slot desktop card.

The NVIDIA GeForce RTX 2080 wins in nearly every other category. Its Vulkan performance is 15.8% ahead, making it the better pick for Vulkan-based titles and applications. Its OpenCL compute performance is 14.9% ahead, and its GPU compute score is 34.1% ahead, so it excels in general-purpose compute workloads. Its DirectX 9 and DirectX 10 leads of 30.5% and 32.4% respectively make it the stronger card for older game libraries. The 38.8% lead in 2D performance suggests faster desktop compositing and 2D application rendering. The RTX 2080 also has tensor cores, which the RX 6700M lacks entirely, making it the only option here for tensor-accelerated workloads.

The RTX 2080's higher memory bandwidth of 448.0 GB/s versus 320.0 GB/s likely contributes to its wins in bandwidth-sensitive tests, though the RX 6700M's higher FP32 throughput of 11.06 TFLOPS versus 10.07 TFLOPS did not produce a compute win. The RTX 2080's larger shading unit count of 2944 versus 2304, and its 368 tensor cores, give it a structural advantage in parallel workloads.

In summary, the RX 6700M is the modern DX12 and efficiency pick, while the RTX 2080 is the legacy, Vulkan, and compute pick. The database shows a 1-to-9 split in favor of NVIDIA, but the single AMD win is in the test most representative of future game engines.