AMD Radeon 680M vs NVIDIA GeForce GTX 1660 SUPER Comparison

Head-to-Head Benchmarks

The data is unambiguous: the NVIDIA GeForce GTX 1660 SUPER wins every recorded head-to-head benchmark against the AMD Radeon 680M. In the three shared tests, the GTX 1660 SUPER leads by margins ranging from roughly 55% to over 70%. The largest gap appears in the 3DMark Steel Nomad DX12 test, where the GTX 1660 SUPER scores 1278 against the Radeon 680M's 378, a delta of -70.4% from the AMD part's perspective. That is a decisive victory, placing the NVIDIA card in a completely different performance class for modern DirectX 12 workloads.

The compute-oriented benchmarks tell a similar story, though with slightly narrower margins. In Geekbench OpenCL, the GTX 1660 SUPER records 52490 points versus 23468 for the Radeon 680M, a delta of -55.3%. The Vulkan result follows the same pattern: 57102 for NVIDIA versus 21965 for AMD, a delta of -61.5%. Across all three tests, the NVIDIA card's advantage is consistent and substantial, with no single test where the AMD integrated part manages to close the gap to within a competitive range. The recorded data shows a clean sweep: winsA equals 0, winsB equals 3.

It is importantly the Radeon 680M still holds its own in the broader database context. Its average benchmark score of 15270 places it at the 57th percentile among all GPUs, which is actually higher than the GTX 1660 SUPER's 53rd percentile and its average score of 12986. This apparent contradiction stems from the fact that the Radeon 680M's average is buoyed by its strong Geekbench scores, while the head-to-head tests skew toward heavier 3D workloads. The Radeon 680M's nearest rivals include the NVIDIA GeForce RTX 2060 at 15290 (delta -0.1%), the NVIDIA GeForce GTX 580 at 15283 (delta -0.1%), and the AMD Radeon RX 7600 at 15171 (delta 0.7%). The GTX 1660 SUPER, meanwhile, sits near the AMD Radeon RX 580 at 12928 (delta 0.4%) and the NVIDIA GeForce RTX 3050 Ti Mobile at 12940 (delta 0.4%).

The Verdict

Based strictly on the recorded benchmark data, the NVIDIA GeForce GTX 1660 SUPER is the clear choice for anyone prioritizing raw 3D rendering performance, particularly in DirectX 12 titles. The 70.4% lead in 3DMark Steel Nomad is not a marginal advantage; it represents a fundamental gap in rasterization throughput that no driver optimization or power tuning on the AMD side can overcome. For gaming, content creation with GPU acceleration, or any workload that stresses the graphics pipeline, the GTX 1660 SUPER delivers results that are roughly 1.5 to 2.4 times faster across the measured tests.

However, the Radeon 680M is not without its merits. Its 57th percentile ranking versus the GTX 1660 SUPER's 53rd percentile, combined with a higher average benchmark score (15270 versus 12986), indicates that the AMD part performs better in certain compute-oriented scenarios relative to the broader GPU landscape. The Radeon 680M's nearest rivals include the RTX 2060 and GTX 580, both of which are desktop discrete cards, which speaks to the efficiency of AMD's integrated design. If the use case involves lightweight gaming, media playback, or general system acceleration where the GPU is not the primary bottleneck, the Radeon 680M's lower power envelope and integrated nature make it a sensible option.

The verdict is straightforward: the GTX 1660 SUPER wins on absolute performance, while the Radeon 680M wins on efficiency and versatility within its form factor. Neither card is a substitute for the other. The data does not support a recommendation for the Radeon 680M in any scenario where maximum frame rates or render times are the sole criterion. Conversely, the GTX 1660 SUPER's higher power draw and dual-slot footprint make it unsuitable for the ultra-compact or power-constrained systems where the Radeon 680M is designed to operate.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. The AMD Radeon 680M is built on the RDNA 2.0 architecture, manufactured on a 6 nm process at TSMC. It is a Rembrandt+ chip, part of the Navi II IGP generation, designed as an integrated graphics processor. The NVIDIA GeForce GTX 1660 SUPER uses the Turing architecture on a 12 nm process, also from TSMC, with the TU116 chip. The process node difference is significant: 6 nm versus 12 nm means the AMD part packs transistors more densely, at 63.0 million per square millimeter compared to 23.2 million for NVIDIA.

The transistor counts reflect this density advantage. The Radeon 680M contains 13,100 million transistors on a 208 mm² die, while the GTX 1660 SUPER has 6,600 million transistors on a larger 284 mm² die. Despite the AMD chip having roughly twice the transistor count, it occupies a smaller physical area, which is a direct consequence of the more advanced manufacturing process. The die size difference also highlights the integrated nature of the Radeon 680M: it shares its package with a CPU, whereas the GTX 1660 SUPER is a standalone discrete GPU.

Feature support diverges as well. The Radeon 680M includes 12 ray tracing cores, while the GTX 1660 SUPER has none. This means the AMD part supports hardware-accelerated ray tracing, a feature absent from the NVIDIA card. However, the GTX 1660 SUPER's lack of RT cores does not necessarily hurt it in the recorded benchmarks, as none of the tests specifically target ray tracing performance. The Radeon 680M also supports DirectX 12 Ultimate (12_2), while the GTX 1660 SUPER is limited to DirectX 12 (12_1). Both cards support OpenGL 4.6 and Vulkan 1.4, so API compatibility is otherwise equal.

Memory architecture is another major differentiator. The Radeon 680M uses system shared memory, with bandwidth described as system dependent. The GTX 1660 SUPER has dedicated 6 GB of GDDR6 memory on a 192-bit bus, delivering 336.0 GB/s of bandwidth. This dedicated memory arrangement gives the NVIDIA card a substantial advantage in memory-intensive workloads, as it does not compete with the CPU for bandwidth. The Radeon 680M's memory performance is entirely dependent on the host system's RAM configuration, which is a variable that the database cannot fully control.

Specification Differences

The specification tables show clear differences across nearly every measurable field. Clock speeds favor the AMD part on paper: the Radeon 680M has a base clock of 2000 MHz and a boost clock of 2200 MHz, while the GTX 1660 SUPER runs at 1530 MHz base and 1785 MHz boost. However, the NVIDIA card compensates with a much larger execution resource pool. The GTX 1660 SUPER has 1408 shading units, 88 texture mapping units, and 48 raster output units, versus 768 shading units, 48 TMUs, and 32 ROPs for the Radeon 680M. This nearly 2:1 advantage in shading units explains why the NVIDIA part wins despite lower clocks.

Pixel and texture rates reflect this resource disparity. The GTX 1660 SUPER achieves 85.68 GPixel/s and 157.1 GTexel/s, while the Radeon 680M manages 70.40 GPixel/s and 105.6 GTexel/s. Floating point performance follows the same trend: the NVIDIA card delivers 5.027 TFLOPS FP32 and 10.05 TFLOPS FP16 (2:1), versus 3.379 TFLOPS FP32 and 6.758 TFLOPS FP16 (2:1) for the AMD part. The GTX 1660 SUPER is roughly 49% faster in FP32 and 49% faster in FP16 than the Radeon 680M.

Power and physical specifications differ dramatically. The Radeon 680M has a TDP of 50 W and is classified as an IGP with no power connectors. The GTX 1660 SUPER has a TDP of 125 W, requires a single 8-pin power connector, and has a suggested PSU rating of 300 W. The NVIDIA card is dual-slot with dimensions of 229 mm length, 111 mm height, and 35 mm width, while the Radeon 680M has no physical dimensions listed because it is integrated into a motherboard or laptop. The bus interface also differs: PCIe 4.0 x8 for AMD versus PCIe 3.0 x16 for NVIDIA. Display outputs are portable device dependent for the Radeon 680M, while the GTX 1660 SUPER offers 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a. The production status shows the Radeon 680M as active with a release date of January 2023, while the GTX 1660 SUPER is end-of-life with an October 2019 release date and a launch MSRP of 229 USD.

FAQ

Q: Which GPU wins in DirectX 12 gaming performance?

A: The NVIDIA GeForce GTX 1660 SUPER wins decisively. In the 3DMark Steel Nomad DX12 test, it scores 1278 versus 378 for the AMD Radeon 680M, a delta of -70.4% from the AMD part's perspective.

Q: Does the AMD Radeon 680M support ray tracing?

A: Yes, the Radeon 680M includes 12 ray tracing cores as part of its RDNA 2.0 architecture. The NVIDIA GeForce GTX 1660 SUPER has no ray tracing cores.

Q: How do the two compare in compute benchmarks?

A: The GTX 1660 SUPER leads in Geekbench OpenCL with 52490 points versus 23468, and in Geekbench Vulkan with 57102 versus 21965. The deltas are -55.3% and -61.5% respectively.

Q: What is the memory configuration difference?

A: The GTX 1660 SUPER has 6 GB of dedicated GDDR6 memory on a 192-bit bus with 336.0 GB/s bandwidth. The Radeon 680M uses system shared memory with bandwidth described as system dependent.

Q: Which GPU has a higher percentile ranking?

A: The Radeon 680M is at the 57th percentile among all GPUs, while the GTX 1660 SUPER is at the 53rd percentile. The Radeon 680M also has a higher average benchmark score of 15270 versus 12986.

Q: What are the power requirements for each?

A: The Radeon 680M has a 50 W TDP and no power connectors, while the GTX 1660 SUPER has a 125 W TDP, requires one 8-pin power connector, and has a suggested PSU of 300 W.

Where Each One Wins

The NVIDIA GeForce GTX 1660 SUPER wins in every scenario where absolute graphics performance is the primary requirement. The 3DMark Steel Nomad result, with a 70.4% advantage, indicates that modern DirectX 12 games will run significantly faster on the NVIDIA card. The Geekbench OpenCL and Vulkan results, showing 55.3% and 61.5% leads respectively, extend this advantage to compute-heavy applications such as video encoding, 3D rendering, and GPU-accelerated scientific workloads. The dedicated 6 GB of GDDR6 memory with 336.0 GB/s bandwidth means the GTX 1660 SUPER will not suffer from memory contention in multi-tasking scenarios, making it the better choice for a dedicated gaming desktop or workstation.

The AMD Radeon 680M wins in scenarios where its integrated nature and efficiency are the deciding factors. Its 50 W TDP is less than half of the GTX 1660 SUPER's 125 W, and it requires no power connectors or additional cooling beyond what a laptop or compact desktop already provides. For thin-and-light laptops, mini PCs, or all-in-one systems where a discrete GPU is physically impossible, the Radeon 680M is the only option of the two. Its higher percentile ranking (57th versus 53rd) and higher average benchmark score (15270 versus 12986) suggest that in the broader GPU ecosystem, it punches above its weight class relative to its power envelope. The 12 ray tracing cores provide a feature that the GTX 1660 SUPER lacks entirely, which could matter for future titles that support hardware RT on integrated graphics.

The recorded data also shows that the Radeon 680M's nearest rivals include the RTX 2060 and GTX 580, both desktop discrete cards, which is remarkable for an integrated processor. However, the GTX 1660 SUPER's nearest rivals include the RX 580 and RTX 3050 Ti Mobile, indicating it is competitive with mid-range discrete options. For users who already have a system with a Radeon 680M, the data does not suggest an urgent need to upgrade if the workload is light. For users building a new system with a discrete GPU budget, the GTX 1660 SUPER's benchmark wins are too large to ignore. The choice ultimately comes down to form factor and power constraints, not performance, where the NVIDIA card is the clear winner.