AMD Radeon RX 570X vs NVIDIA GeForce GTX 1660 Comparison

Head-to-Head Benchmarks

The recorded data shows a decisive sweep for the NVIDIA GeForce GTX 1660 across all three shared benchmark tests. The most striking disparity appears in PassMark G3D, where the GTX 1660 scores 11,646 against the RX 570X's 1,923. That is a delta of 83.5% in favor of NVIDIA, indicating a fundamentally different level of graphics throughput. The RX 570X's score places it in the 55th percentile of all GPUs in the database, while the GTX 1660 sits at the 51st percentile; despite the lower percentile ranking, the GTX 1660's raw G3D result is over six times higher.

In Geekbench OpenCL, the gap narrows but remains substantial. The GTX 1660 posts 47,850, while the RX 570X manages 38,939, a difference of 18.6%. This suggests that compute-oriented workloads favor NVIDIA's architecture, though the AMD card is not completely outclassed in this domain. The PassMark G2D test shows only a marginal advantage for NVIDIA: 776 versus 750, a 3.4% delta. Both cards handle 2D operations at a comparable level, which is expected given that 2D performance is rarely a bottleneck in modern systems.

Looking at the aggregate benchmark score, the RX 570X averages 13,871 across all recorded tests, while the GTX 1660 averages 11,680. This inversion is notable: the RX 570X has a higher average score, yet loses decisively in the head-to-head comparisons. The explanation lies in the benchmark composition. The RX 570X's nearest rivals include the NVIDIA RTX A2000 Mobile (average score 13,821, 0.4% lower) and the AMD Radeon 660M (13,812, 0.4% lower), which suggests its average is buoyed by tests that favor its specific feature set. The GTX 1660's nearest rivals are the AMD Radeon RX 7800 XT (11,627, 0.5% higher) and the AMD Radeon Pro 5500M (11,528, 1.3% lower), indicating a cluster of mid-range cards with similar aggregate performance.

The wins tally confirms the outcome: NVIDIA secures all three head-to-head victories, while AMD records zero. The largest single-test margin is the PassMark G3D result, where the GTX 1660 leads by 83.5%. This is not a close contest; it is a generational gap in 3D rendering capability. For users prioritizing gaming or GPU-accelerated 3D workloads, the data leaves no ambiguity.

The Verdict

The benchmark results point to a clear recommendation for the NVIDIA GeForce GTX 1660 for any workload that heavily depends on 3D graphics performance. The PassMark G3D score alone, with an 83.5% advantage, makes the RX 570X difficult to justify for gaming or rendering tasks. The GTX 1660 also leads in OpenCL compute by 18.6%, covering productivity and acceleration scenarios beyond pure graphics.

The RX 570X does hold a higher aggregate average score (13,871 versus 11,680), but this is misleading in a head-to-head context. Its average is lifted by tests where it performs comparably to the GTX 1660, such as PassMark G2D (within 3.4%), rather than by any dominant strength. The GTX 1660's percentile ranking (51st) is slightly lower than the RX 570X's (55th), yet the direct comparisons show NVIDIA winning every shared test. This suggests the RX 570X benefits from a broader set of benchmark entries in the database, not from superior performance against this specific rival.

For users with an existing RX 570X, the data does not indicate an urgent need to upgrade for 2D tasks or basic compute, but for 3D applications the gap is too large to ignore. For new purchases, the GTX 1660 is the only rational choice based on these measurements. The RX 570X's 8 GB memory capacity and wider 256-bit bus are not reflected in the benchmark outcomes; they do not translate into winning scores.

FAQ

Q: Which card wins in PassMark G3D, and by how much?

A: The NVIDIA GeForce GTX 1660 wins with a score of 11,646 against the AMD Radeon RX 570X's 1,923. This is an 83.5% advantage for NVIDIA.

Q: Is the AMD Radeon RX 570X competitive in any shared benchmark?

A: Yes, in PassMark G2D the RX 570X scores 750, which is only 3.4% behind the GTX 1660's 776. This is the closest margin across all head-to-head tests.

Q: What is the average benchmark score for each card?

A: The AMD Radeon RX 570X has an average benchmark score of 13,871, while the NVIDIA GeForce GTX 1660 averages 11,680 across all recorded tests.

Q: How does the GTX 1660 compare to its nearest rivals in the database?

A: The GTX 1660's nearest rival is the AMD Radeon RX 7800 XT with an average score of 11,627, which is 0.5% higher. The AMD Radeon Pro 5500M is 1.3% lower at 11,528.

Q: What is the percentile ranking of each GPU?

A: The AMD Radeon RX 570X ranks in the 55th percentile of all GPUs, while the NVIDIA GeForce GTX 1660 ranks in the 51st percentile.

Q: Which card has the higher OpenCL score?

A: The NVIDIA GeForce GTX 1660 scores 47,850 in Geekbench OpenCL, compared to the AMD Radeon RX 570X's 38,939, a difference of 18.6%.

Specification Differences

The two cards differ across nearly every core specification. The AMD Radeon RX 570X uses the Polaris 20 chip on a 14 nm process from GlobalFoundries, while the NVIDIA GeForce GTX 1660 uses the TU116 chip on a 12 nm process from TSMC. The RX 570X packs 5,700 million transistors on a 232 mm² die, while the GTX 1660 has 6,600 million transistors on a 284 mm² die. Transistor density is slightly higher on the AMD part at 24.6M per mm² versus 23.2M per mm² for NVIDIA.

Clock speeds favor NVIDIA substantially. The GTX 1660 has a base clock of 1530 MHz and a boost clock of 1785 MHz, compared to the RX 570X's 1168 MHz base and 1244 MHz boost. Memory configuration also differs: the RX 570X has 8 GB of GDDR5 on a 256-bit bus, delivering 224.0 GB/s of bandwidth, while the GTX 1660 has 6 GB of GDDR5 on a 192-bit bus, providing 192.1 GB/s. The RX 570X's memory clocks at 1750 MHz (7 Gbps effective), while the GTX 1660 runs at 2001 MHz (8 Gbps effective).

Compute unit counts diverge as well. The RX 570X has 2048 shading units, 128 texture mapping units, and 32 raster operation units. The GTX 1660 has 1408 shading units, 88 TMUs, and 48 ROPs. Pixel and texture rates reflect these differences: the GTX 1660 achieves 85.68 GPixel/s versus the RX 570X's 39.81 GPixel/s, while texture rates are nearly identical at 159.2 GTexel/s for AMD and 157.1 GTexel/s for NVIDIA. Floating-point performance is close: the RX 570X delivers 5.095 TFLOPS in FP32, while the GTX 1660 delivers 5.027 TFLOPS. In FP16, the RX 570X maintains a 1:1 ratio at 5.095 TFLOPS, while the GTX 1660 doubles to 10.05 TFLOPS with a 2:1 ratio.

Power requirements differ significantly. The RX 570X has a TDP of 150 W with a suggested PSU of 450 W and a single 6-pin connector. The GTX 1660 has a TDP of 120 W, a suggested PSU of 300 W, and a single 8-pin connector. Physical dimensions vary: the RX 570X is 241 mm long (9.5 inches), while the GTX 1660 is 229 mm long (9 inches) with a height of 111 mm (4.4 inches) and width of 35 mm (1.4 inches). Both are dual-slot cards with PCIe 3.0 x16 interfaces.

Display outputs differ slightly. The RX 570X offers 1x DVI, 1x HDMI 2.0b, and 3x DisplayPort 1.4a. The GTX 1660 offers 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a. API support also varies: the RX 570X supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3, while the GTX 1660 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4.

Architecture Differences

The architectural divide is fundamental. The AMD Radeon RX 570X is built on GCN 4.0, the fourth iteration of AMD's Graphics Core Next design, implemented in the Polaris generation. The NVIDIA GeForce GTX 1660 uses the Turing architecture, NVIDIA's newer design that also underpins the GeForce 20-series. The process nodes reflect this gap: AMD uses a mature 14 nm process from GlobalFoundries, while NVIDIA uses a denser 12 nm process from TSMC.

The RX 570X's GCN 4.0 architecture is characterized by a high count of shading units (2048) arranged in a wide configuration, paired with a 256-bit memory bus. The GTX 1660's Turing architecture uses fewer shading units (1408) but achieves higher clock speeds and a higher pixel rate (85.68 GPixel/s versus 39.81 GPixel/s). This indicates a different design philosophy: AMD favors raw compute width, while NVIDIA optimizes for per-clock efficiency and rasterization throughput.

Neither card includes dedicated ray tracing cores or tensor cores, as both are null in the database. The RX 570X's FP16 performance is identical to FP32 at a 1:1 ratio, meaning it does not accelerate half-precision workloads. The GTX 1660 doubles its FP16 throughput to 10.05 TFLOPS with a 2:1 ratio, providing a clear advantage for workloads that leverage half-precision arithmetic.

The RX 570X is the predecessor to AMD's Vega architecture, while the GTX 1660 is the successor to the GeForce 10-series and the predecessor to the GeForce 20-series. The RX 570X was released on April 10, 2018, while the GTX 1660 arrived on March 13, 2019. The GTX 1660 launched with an MSRP of 219 USD, a fact that can be stated once. Both cards are end-of-life in production status.

The Vulkan API support differs: the RX 570X supports Vulkan 1.3, while the GTX 1660 supports Vulkan 1.4. DirectX support also differs, with the RX 570X at feature level 12_0 and the GTX 1660 at 12_1. These differences may matter for specific applications that rely on newer API features, though the benchmark data shows the GTX 1660's performance advantage is the dominant factor in real-world comparisons.