AMD Radeon Pro WX 5100 vs NVIDIA GeForce GTX 1660 Ti Comparison
AMD Radeon Pro WX 5100
GeForce GTX 1660 Ti
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon Pro WX 5100 vs NVIDIA GeForce GTX 1660 Ti
Head-to-Head Benchmarks
The head-to-head comparison between the AMD Radeon Pro WX 5100 and the NVIDIA GeForce GTX 1660 Ti is dominated by NVIDIA, which takes 8 of the 9 shared benchmark tests. The only AMD victory comes in the Geekbench Vulkan test, where the WX 5100 scores 26,909 against the GTX 1660 Ti's 5,827, a staggering 361.8% advantage. This is the single largest delta in either direction across the entire comparison, indicating a fundamental strength in the AMD card's Vulkan execution path.
In the compute-oriented Geekbench OpenCL test, the GTX 1660 Ti posts 51,029 versus 24,217 for the WX 5100, a 52.5% deficit for the AMD card. This is the second-largest gap in the dataset, trailing only the Vulkan result. The OpenCL result aligns with the Passmark GPU compute score, where the GTX 1660 Ti records 5,226 against 2,053, a 60.7% difference in NVIDIA's favor.
The DirectX legacy tests show a consistent NVIDIA lead. In Passmark DirectX 9, the GTX 1660 Ti scores 190 versus 88, a 53.7% margin. DirectX 10 shows 66 versus 29, a 56.1% gap. DirectX 11 widens further: 103 versus 36, a 65% difference, the largest NVIDIA margin in the entire benchmark set. DirectX 12 narrows the gap somewhat, with 51 versus 26, a 49% deficit for the AMD card.
The overall graphics metric, Passmark G3D, strongly favors NVIDIA: 12,871 versus 5,496, a 57.3% difference. This is the headline performance indicator, and it is unflattering to the WX 5100. The 2D test is the closest of all nine, with the GTX 1660 Ti at 804 and the WX 5100 at 777, a mere 3.4% difference, suggesting that basic desktop rendering is nearly identical between the two.
The average benchmark score calculation reinforces the split. The WX 5100 has an average score of 8,863, while the GTX 1660 Ti averages 7,723. This inversion occurs because the GTX 1660 Ti's Vulkan score is so low that it drags down its average, despite winning most other tests. The AMD card's Vulkan result is so strong that it boosts its average above NVIDIA's, even though NVIDIA wins eight individual tests.
The Verdict
The data presents a clear choice for most users. The GTX 1660 Ti wins the majority of workload categories, including DirectX 9, 10, 11, and 12, GPU compute, and the overall G3D metric. For any application that relies on DirectX, OpenCL, or general GPU compute, the GTX 1660 Ti is the superior option, with margins ranging from 49% to 65% in the DirectX tests and 60.7% in compute.
The WX 5100's sole victory in Vulkan is substantial, but it is a single data point. The 361.8% lead in that test suggests the AMD card has a specialized strength, but it does not offset the NVIDIA card's dominance across eight other categories. The GTX 1660 Ti wins 8 of 9 head-to-head tests, a decisive ratio.
From a percentile standpoint, the WX 5100 sits at the 44th percentile of all GPUs, while the GTX 1660 Ti sits at the 41st percentile. These are close, but the percentile ranking considers all benchmark scores, including the Vulkan outlier. The GTX 1660 Ti's lower percentile reflects its weak Vulkan showing, not its overall DirectX or compute performance.
Users who prioritize Vulkan workloads should consider the WX 5100. Users who prioritize anything else, particularly DirectX gaming or compute tasks, should choose the GTX 1660 Ti. The data does not support the WX 5100 as a general-purpose performer; it is a specialist that happens to excel in one API.
Architecture Differences
The two cards come from different manufacturers and different architectural generations. The AMD Radeon Pro WX 5100 uses the GCN 4.0 architecture on a 14 nm process from GlobalFoundries. The NVIDIA GeForce GTX 1660 Ti uses the Turing architecture on a 12 nm process from TSMC. The process node difference is small but relevant: 14 nm versus 12 nm, with the latter being a denser, more modern fabrication.
Transistor counts differ notably. The WX 5100 packs 5,700 million transistors on a 232 mm² die, yielding a transistor density of 24.6 million per square millimeter. The GTX 1660 Ti contains 6,600 million transistors on a 284 mm² die, with a slightly lower density of 23.2 million per square millimeter. Despite the larger die, the NVIDIA card's density is marginally lower, indicating a different design philosophy in layout and power management.
The chip names also differ: Ellesmere for AMD, TU116 for NVIDIA. The AMD card is part of the Radeon Pro Polaris generation (WX x100), while the NVIDIA card belongs to the GeForce 16 generation. Both are end-of-life products, but their release dates differ: November 2016 for the WX 5100, February 2019 for the GTX 1660 Ti. This 27-month gap explains some of the architectural divergence.
Neither card has dedicated ray tracing cores or tensor cores, so both rely on traditional shader execution. The API support differs slightly: the WX 5100 supports DirectX 12 (12_0) and Vulkan 1.3, while the GTX 1660 Ti supports DirectX 12 (12_1) and Vulkan 1.4. The higher DirectX feature level and Vulkan version on the NVIDIA card are notable for compatibility with newer titles.
Specification Differences
The most significant specification gap is memory bandwidth. The WX 5100 uses 8 GB of GDDR5 on a 256-bit bus, delivering 160.0 GB/s. The GTX 1660 Ti uses 6 GB of GDDR6 on a 192-bit bus, delivering 288.0 GB/s. Despite the narrower bus, the GDDR6 memory runs at 12 Gbps effective versus 5 Gbps effective, resulting in an 80% bandwidth advantage for NVIDIA.
Clock speeds are dramatically different. The WX 5100 has a base clock of 713 MHz and a boost clock of 1086 MHz. The GTX 1660 Ti has a base clock of 1500 MHz and a boost clock of 1770 MHz. The NVIDIA card's boost clock is 63% higher than the AMD card's boost clock, contributing to its performance lead in most tests.
Shader and texture hardware differ in configuration. The WX 5100 has 1,792 shading units, 112 texture mapping units, and 32 ROPs. The GTX 1660 Ti has 1,536 shading units, 96 TMUs, and 48 ROPs. Despite having fewer shading units and TMUs, the NVIDIA card achieves higher pixel and texture rates: 84.96 GPixel/s versus 34.75 GPixel/s, and 169.9 GTexel/s versus 121.6 GTexel/s.
Floating-point performance favors NVIDIA in FP32: 5.437 TFLOPS versus 3.892 TFLOPS. The FP16 comparison is even more lopsided, with the GTX 1660 Ti delivering 10.87 TFLOPS (2:1 ratio) against the WX 5100's 3.892 TFLOPS (1:1 ratio). This means the NVIDIA card has a 2:1 advantage in FP16 workloads, which is significant for compute tasks that use reduced precision.
Power and physical specifications differ considerably. The WX 5100 has a TDP of 75 W, is single-slot, and requires no power connectors. The GTX 1660 Ti has a TDP of 120 W, is dual-slot, and requires a single 8-pin power connector. The suggested PSU rating is 250 W for AMD and 300 W for NVIDIA. The WX 5100 is also shorter at 173 mm versus 229 mm, though both are about 4.4 inches tall. The NVIDIA card adds 35 mm of width, making it bulkier.
Display outputs differ: the WX 5100 offers four DisplayPort 1.4a connections, while the GTX 1660 Ti offers one DVI, one HDMI 2.0, and one DisplayPort 1.4a. The AMD card's four DisplayPort outputs are clearly aimed at multi-display professional setups.
FAQ
Q: Which card wins more benchmark tests?
A: The NVIDIA GeForce GTX 1660 Ti wins 8 of the 9 head-to-head tests. The AMD Radeon Pro WX 5100 wins only the Geekbench Vulkan test.
Q: How large is the AMD card's Vulkan advantage?
A: The WX 5100 scores 26,909 in Geekbench Vulkan versus 5,827 for the GTX 1660 Ti, a 361.8% lead. This is the largest margin in either direction across all tests.
Q: What is the memory bandwidth difference?
A: The GTX 1660 Ti delivers 288.0 GB/s over a 192-bit GDDR6 interface, while the WX 5100 delivers 160.0 GB/s over a 256-bit GDDR5 interface. NVIDIA has an 80% bandwidth advantage.
Q: Which card has higher FP32 performance?
A: The GTX 1660 Ti achieves 5.437 TFLOPS in FP32, compared to 3.892 TFLOPS for the WX 5100. The NVIDIA card is approximately 40% faster in this metric.
Q: Do either of these cards support ray tracing?
A: No. Both cards have null entries for RT cores and tensor cores, meaning neither supports dedicated ray tracing hardware.
Q: What are the power requirements?
A: The WX 5100 has a 75 W TDP and needs no power connectors, with a 250 W suggested PSU. The GTX 1660 Ti has a 120 W TDP, requires one 8-pin connector, and suggests a 300 W PSU.
Where Each One Wins
The GTX 1660 Ti wins in every DirectX generation tested: DirectX 9, 10, 11, and 12. The margins are consistent, ranging from 49% in DirectX 12 to 65% in DirectX 11. This makes it the clear choice for gaming and any application built on Microsoft's graphics API. The compute advantage is equally decisive, with a 60.7% lead in Passmark GPU compute and a 52.5% lead in Geekbench OpenCL.
The GTX 1660 Ti also wins the overall Passmark G3D score by 57.3%, cementing its status as the superior general-purpose graphics card. Its 2D score is only 3.4% higher, so desktop productivity is essentially a tie, but the 3D and compute workloads are not close.
The WX 5100 has exactly one winning scenario: Geekbench Vulkan. The 361.8% margin is extraordinary, but it is an isolated result. No other test in the dataset shows a comparable Vulkan advantage, and the AMD card loses every other benchmark. Users running Vulkan-specific workloads, such as certain cross-platform games or Vulkan-based compute applications, may find the WX 5100 compelling.
The WX 5100 also offers practical advantages not captured in benchmarks: four DisplayPort outputs versus one each of DVI, HDMI, and DisplayPort on the NVIDIA card. For multi-monitor professional setups, the AMD card's display configuration is more flexible. The WX 5100 is also single-slot and requires no power connectors, making it easier to install in compact or densely populated systems.
The GTX 1660 Ti's higher boost clock, superior FP32 and FP16 throughput, and larger memory bandwidth explain its benchmark dominance. The WX 5100's lower power draw and smaller footprint are its compensating strengths. In raw performance terms, the data is unambiguous: the GTX 1660 Ti wins 8 of 9 tests, and the WX 5100 wins exactly one. Choose the NVIDIA card for performance, the AMD card for Vulkan specialization or multi-display output needs.