AMD Radeon 680M vs NVIDIA GeForce GTX 1070 Ti Comparison

Head-to-Head Benchmarks

The benchmark data is unambiguous: the NVIDIA GeForce GTX 1070 Ti wins all three head-to-head tests, with margins that range from substantial to overwhelming. The largest gap appears in the Vulkan compute test, where the GTX 1070 Ti scores 61,006 against the AMD Radeon 680M's 21,965, a delta of -64% from the AMD part's perspective. In practical terms, that means the NVIDIA card delivers roughly 2.8 times the performance in this workload, a gap that would be immediately noticeable in any Vulkan-based application or game.

The OpenCL test tells a similar story, though with a slightly smaller margin. The GTX 1070 Ti posts 49,219 points, while the Radeon 680M manages 23,468, resulting in a -52.3% delta. This is still a decisive victory for NVIDIA, but the AMD integrated GPU is at least operating in the same general performance class here, rather than being completely outclassed. The most striking result, however, comes from the 3DMark Steel Nomad DX12 test. The GTX 1070 Ti scores 1,471, while the Radeon 680M trails far behind at 378, a -74.3% delta. This is the single largest performance gap in the entire comparison, and it highlights how far apart these two products are in raw DirectX 12 rasterization performance.

Looking at the aggregate benchmark averages reinforces this hierarchy. The Radeon 680M has an average benchmark score of 15,270 across all tested workloads, while the GTX 1070 Ti averages 14,277. This seems contradictory at first glance — the AMD part has a higher average despite losing every head-to-head test — but the explanation lies in the test suites. The Radeon 680M's average is heavily influenced by its strong Geekbench OpenCL and Vulkan scores, which are weighted differently in the aggregate calculation. The GTX 1070 Ti, by contrast, has a broader test profile that includes several PassMark legacy DX9, DX10, and DX11 tests, some of which pull its average down.

The percentile rankings tell a more nuanced story. The Radeon 680M sits at the 57th percentile of all GPUs, while the GTX 1070 Ti sits at the 56th percentile. These are nearly identical positions, suggesting that while the NVIDIA card wins the direct comparisons, the AMD integrated GPU is not far behind in the broader GPU landscape. The nearest rival data confirms this: the Radeon 680M's closest competitor is the NVIDIA GeForce GTX 580, which scores 15,283 (a -0.1% delta), while the GTX 1070 Ti's nearest rival is the Intel Iris Xe MAX Graphics at 14,315 (a -0.3% delta). Both cards are clustered with other mid-range parts, but the GTX 1070 Ti's wins are so decisive in the head-to-head that the overall picture remains clear.

Architecture Differences

The two GPUs represent fundamentally different design philosophies and generations. The AMD Radeon 680M is built on RDNA 2.0 architecture using a 6 nm TSMC process, packing 13,100 million transistors into a 208 mm² die. That works out to a transistor density of 63.0 million per square millimeter, a figure that reflects the modern, densely packed design. The chip itself is codenamed Rembrandt+ and belongs to the Navi II IGP generation, positioned as a successor to Vega II IGP and predecessor to Navi III IGP.

The NVIDIA GeForce GTX 1070 Ti, by contrast, uses the older Pascal architecture on a 16 nm TSMC process. It contains 7,200 million transistors on a much larger 314 mm² die, resulting in a transistor density of just 22.9 million per square millimeter. The chip is GP104, from the GeForce 10-series generation, succeeding the GeForce 900 series and preceding the GeForce 20 series. The manufacturing process gap is significant: 6 nm versus 16 nm represents two full process generations, and the density difference of nearly 3x explains how AMD fits 13.1 billion transistors into a smaller physical package.

The compute resources diverge sharply. The Radeon 680M has 768 shading units, 48 texture mapping units, and 32 ROPs, along with 12 ray tracing cores. The GTX 1070 Ti has 2,432 shading units, 152 TMUs, and 64 ROPs, with no ray tracing cores at all. The NVIDIA card's raw execution resources are roughly 3.2x larger in shading units and 3.2x larger in texture units, which directly explains its massive lead in texture rate (255.8 GTexel/s versus 105.6 GTexel/s) and pixel rate (107.7 GPixel/s versus 70.40 GPixel/s).

The floating-point throughput tells a similar story. The GTX 1070 Ti delivers 8.186 TFLOPS of FP32 performance, while the Radeon 680M manages 3.379 TFLOPS. The FP16 comparison is more complex: the AMD part delivers 6.758 TFLOPS at a 2:1 ratio, while the NVIDIA card delivers just 127.9 GFLOPS at a 1:64 ratio. This means the Radeon 680M is actually far superior in FP16 workloads, offering roughly 53x the FP16 throughput of the GTX 1070 Ti, a consequence of AMD's more modern architecture design.

Clock speeds also differ substantially. The Radeon 680M runs at a 2000 MHz base and 2200 MHz boost, while the GTX 1070 Ti runs at 1607 MHz base and 1683 MHz boost. Despite the lower clocks, the NVIDIA card's vastly larger execution units more than compensate. Memory configurations are completely different: the Radeon 680M uses system-shared memory with dependent bandwidth, while the GTX 1070 Ti features 8 GB of dedicated GDDR5 on a 256-bit bus, delivering 256.3 GB/s of bandwidth. The power envelope is equally divergent: the Radeon 680M draws just 50 W as an integrated GPU, while the GTX 1070 Ti consumes 180 W and requires a dual-slot cooler with a single 8-pin power connector and a 450 W suggested PSU.

The feature sets reflect their respective eras. The Radeon 680M supports DirectX 12 Ultimate (12_2), while the GTX 1070 Ti is limited to DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The Radeon 680M connects via PCIe 4.0 x8, while the GTX 1070 Ti uses PCIe 3.0 x16. The NVIDIA card has physical display outputs (1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a), while the AMD IGP's outputs are portable-device dependent.

The Verdict

The data makes the hierarchy clear. The NVIDIA GeForce GTX 1070 Ti is the significantly faster GPU in every head-to-head benchmark, with leads ranging from 52.3% to 74.3%. Its advantage is most pronounced in DirectX 12 rasterization, where it more than triples the Radeon 680M's score in 3DMark Steel Nomad. For any user prioritizing raw gaming or compute performance, the GTX 1070 Ti is the unambiguous choice.

However, the Radeon 680M's position as an integrated GPU with a 50 W TDP changes the evaluation context. It delivers roughly half to one-third of the GTX 1070 Ti's performance while consuming less than a third of the power. The 680M's FP16 throughput is dramatically higher, and its modern RDNA 2.0 architecture includes ray tracing cores that the Pascal-based GTX 1070 Ti completely lacks. The Radeon 680M is also an active production product, while the GTX 1070 Ti is end-of-life, released in November 2017 at a launch MSRP of 399 USD.

The percentile data shows both cards sit within one point of each other in the global GPU distribution (57th versus 56th percentile), indicating that for the average user, the real-world difference may be less stark than the head-to-head numbers suggest. The Radeon 680M's higher average benchmark score (15,270 versus 14,277) further complicates a simple "one is better" narrative. The GTX 1070 Ti wins where it matters most for gaming, but the Radeon 680M holds its own in the broader benchmark landscape.

FAQ

Q: Which GPU wins in raw DirectX 12 performance?

A: The NVIDIA GeForce GTX 1070 Ti wins decisively, scoring 1,471 in 3DMark Steel Nomad DX12 compared to the AMD Radeon 680M's 378, a 74.3% advantage.

Q: Does the AMD Radeon 680M have any performance advantage at all?

A: In FP16 compute, yes. The Radeon 680M delivers 6.758 TFLOPS at a 2:1 ratio, while the GTX 1070 Ti manages only 127.9 GFLOPS at a 1:64 ratio — a difference of roughly 53x in favor of AMD.

Q: How do the power requirements compare?

A: The Radeon 680M is an integrated GPU with a 50 W TDP and no power connectors, while the GTX 1070 Ti is a dual-slot card with a 180 W TDP, a single 8-pin connector, and a 450 W suggested PSU.

Q: Which GPU supports newer DirectX features?

A: The Radeon 680M supports DirectX 12 Ultimate (12_2), while the GTX 1070 Ti is limited to DirectX 12 (12_1). The Radeon 680M also includes 12 ray tracing cores, which the GTX 1070 Ti lacks entirely.

Q: How do their overall benchmark averages compare?

A: The Radeon 680M has a higher average benchmark score of 15,270, compared to the GTX 1070 Ti's 14,277. Both cards sit at nearly the same percentile ranking (57th versus 56th).

Q: What is the memory configuration difference?

A: The Radeon 680M uses system-shared memory with dependent bandwidth, while the GTX 1070 Ti has 8 GB of dedicated GDDR5 on a 256-bit bus with 256.3 GB/s bandwidth.

Where Each One Wins

The NVIDIA GeForce GTX 1070 Ti is the clear winner in all three head-to-head benchmarks: 3DMark Steel Nomad DX12, Geekbench OpenCL, and Geekbench Vulkan. Its most dominant performance comes in the DirectX 12 test, where it achieves 1,471 points versus 378 for the Radeon 680M, a 74.3% margin. In Vulkan, the GTX 1070 Ti scores 61,006 against 21,965, a 64% lead. In OpenCL, it scores 49,219 against 23,468, a 52.3% lead. For gaming, content creation, or any workload that relies on traditional rasterization or general-purpose compute, the GTX 1070 Ti is the superior choice. Its 8 GB of dedicated GDDR5 memory with 256.3 GB/s bandwidth also gives it a substantial advantage in memory-heavy applications.

The AMD Radeon 680M has no head-to-head benchmark wins, but it does have advantages in specific architectural areas. Its FP16 throughput of 6.758 TFLOPS is vastly superior to the GTX 1070 Ti's 127.9 GFLOPS, making it better suited for workloads that leverage half-precision compute. Its 12 ray tracing cores provide hardware acceleration for ray-traced effects, a feature the GTX 1070 Ti cannot offer. The Radeon 680M's 6 nm process node and 50 W TDP make it dramatically more power-efficient, and its system-shared memory architecture means it requires no dedicated VRAM allocation. For users building a compact, low-power system where an integrated GPU is the only option, the Radeon 680M provides respectable performance in a fraction of the power envelope. Its higher average benchmark score (15,270 versus 14,277) and higher global percentile (57th versus 56th) also indicate that, outside the specific head-to-head tests, it holds its own against the aging GTX 1070 Ti in the broader GPU landscape.