AMD Radeon Pro 455 vs NVIDIA GeForce GTX 1660 Comparison

The AMD Radeon Pro 455 and NVIDIA GeForce GTX 1660 occupy different corners of the GPU landscape, with the former designed for professional mobile workstations and the latter for desktop gaming. The benchmark data shows a decisive performance gap, with the GTX 1660 winning both head-to-head tests by wide margins, yet the Radeon Pro 455 maintains relevance in specific contexts due to its power efficiency and form factor. This analysis breaks down the data to clarify where each card stands.

Where Each One Wins

The data shows a clear split in use-case suitability. The NVIDIA GeForce GTX 1660 dominates in raw compute and graphics workloads, winning both available head-to-head benchmarks. In Geekbench OpenCL, the GTX 1660 scores 47,850 against the Radeon Pro 455's 10,336, a delta of -78.4% for the AMD card. Similarly, in Geekbench Vulkan, the GTX 1660 posts 50,137 versus 12,240, with the AMD card trailing by -75.6%. These are not marginal victories; they represent a fundamental performance tier difference.

The AMD Radeon Pro 455, however, wins on power efficiency and physical footprint. Its 35 W TDP is a fraction of the GTX 1660's 120 W, and its MXM Module slot width makes it suitable for portable systems where the dual-slot, 229 mm GTX 1660 cannot fit. The Radeon Pro 455 also carries no power connectors, whereas the GTX 1660 requires a single 8-pin connector and a 300 W suggested PSU. For applications prioritizing battery life, thermal headroom, or space constraints, the AMD card holds an advantage despite its lower performance. The GTX 1660's 6 GB memory versus 2 GB also favors it in memory-heavy tasks, but the Radeon Pro 455's lower power draw could be decisive in mobile professional environments.

Architecture Differences

The architecture gap is substantial. The Radeon Pro 455 uses the Baffin chip on GCN 4.0, built on a 14 nm process at GlobalFoundries. The GTX 1660 uses the TU116 chip on Turing, fabricated on a 12 nm process at TSMC. This node difference contributes to the GTX 1660's higher clock speeds and transistor density, though the numbers tell a nuanced story: the GTX 1660 packs 6,600 million transistors on a 284 mm² die (23.2M / mm²), while the Radeon Pro 455 has 3,000 million on 123 mm² (24.4M / mm²). Interestingly, the AMD chip has slightly higher transistor density, but the NVIDIA chip benefits from a larger die and nearly double the transistor count.

Core configurations differ dramatically. The GTX 1660 features 1,408 shading units, 88 TMUs, and 48 ROPs, versus the Radeon Pro 455's 768 shading units, 48 TMUs, and 16 ROPs. This translates to a pixel rate of 85.68 GPixel/s for the GTX 1660 against 13.68 GPixel/s for the AMD card, and a texture rate of 157.1 GTexel/s versus 41.04 GTexel/s. FP32 compute is 5.027 TFLOPS for the GTX 1660, compared to 1,313.3 GFLOPS for the Radeon Pro 455. The GTX 1660 also supports FP16 at a 2:1 ratio (10.05 TFLOPS), while the AMD card runs FP16 at 1:1 (1,313.3 GFLOPS), indicating a different compute philosophy.

API support also differs: the GTX 1660 supports DirectX 12 (12_1) and Vulkan 1.4, while the Radeon Pro 455 supports DirectX 12 (12_0) and Vulkan 1.3. Both support OpenGL 4.6. Neither card has ray tracing or tensor cores, as those are absent from both specifications.

The Verdict

From the data, the NVIDIA GeForce GTX 1660 is the clear choice for anyone needing raw performance. It outperforms the Radeon Pro 455 by roughly 3.6x in OpenCL and 4.1x in Vulkan, based on the head-to-head deltas. Its 6 GB memory, wider 192-bit bus, and higher bandwidth (192.1 GB/s versus 81.28 GB/s) make it suitable for modern gaming and compute workloads. The GTX 1660's percentile rank of 51 versus the Radeon Pro 455's 52 is nearly identical, but its average benchmark score of 11,680 is lower than the AMD card's 12,831 — a curious inversion explained by the different benchmark suites each card was tested with. The Radeon Pro 455's scores come exclusively from Geekbench Metal, OpenCL, and Vulkan, while the GTX 1660 includes PassMark tests, which may skew averages.

The AMD Radeon Pro 455's niche is clear: it is for systems where power draw, heat, and space are paramount. At 35 W with no power connectors and an MXM module form factor, it fits into portable workstations. Its performance, while far below the GTX 1660, is still competitive with its immediate rivals — it sits within 0.5% of the GeForce GTX 670 and within 0.3% of the AMD Radeon 740M, according to the nearestRivals data. Users who prioritize battery life and portability over frame rates should consider the Radeon Pro 455. Those who need performance should choose the GTX 1660 without hesitation, as the benchmark deltas are too large to ignore.

FAQ

Q: How much faster is the NVIDIA GeForce GTX 1660 in OpenCL?

A: The GTX 1660 scores 47,850 in Geekbench OpenCL, while the Radeon Pro 455 scores 10,336. The delta is -78.4%, meaning the GTX 1660 is approximately 3.6 times faster.

Q: What is the power consumption difference?

A: The Radeon Pro 455 has a TDP of 35 W, while the GTX 1660 has a TDP of 120 W. The GTX 1660 also requires a 1x 8-pin power connector and a 300 W suggested PSU, whereas the Radeon Pro 455 uses no power connectors.

Q: Which card has more memory?

A: The GTX 1660 has 6 GB of GDDR5 memory on a 192-bit bus, providing 192.1 GB/s bandwidth. The Radeon Pro 455 has 2 GB of GDDR5 on a 128-bit bus, providing 81.28 GB/s bandwidth.

Q: Do either of these cards support ray tracing?

A: No. Both the Radeon Pro 455 and GTX 1660 have null values for RT cores and tensor cores in their specifications.

Q: How do these cards compare to their nearest rivals?

A: The Radeon Pro 455's average score of 12,831 is nearly identical to the GeForce GTX 590 (12,830, 0% delta) and the AMD FirePro W5100 (12,847, -0.1% delta). The GTX 1660's average of 11,680 is within 1.8% of the NVIDIA Tesla K20c (11,479) and within 0.5% of the AMD Radeon RX 7800 XT (11,627).

Q: What are the physical size differences?

A: The GTX 1660 is a dual-slot card measuring 229 mm in length, 111 mm in height, and 35 mm in width. The Radeon Pro 455 uses an MXM Module form factor with no listed dimensions, and its display outputs are described as "Portable Device Dependent."

Head-to-Head Benchmarks

The head-to-head data shows a one-sided contest. In Geekbench OpenCL, the GTX 1660's score of 47,850 dwarfs the Radeon Pro 455's 10,336, a difference of -78.4% from the AMD card's perspective. This is the larger of the two deltas, indicating that OpenCL compute workloads heavily favor the NVIDIA architecture. The GTX 1660's 1,408 shading units and 88 TMUs, combined with its 5.027 TFLOPS FP32 throughput, provide a massive computational advantage over the Radeon Pro 455's 768 shading units and 1,313.3 GFLOPS.

In Geekbench Vulkan, the margin narrows slightly but remains decisive. The GTX 1660 scores 50,137 versus the Radeon Pro 455's 12,240, a -75.6% delta. This suggests the GTX 1660's Vulkan 1.4 support and Turing architecture extract more performance from the API than the Radeon Pro 455's Vulkan 1.3 implementation. The GTX 1660's higher memory bandwidth (192.1 GB/s versus 81.28 GB/s) likely contributes to this advantage in memory-intensive Vulkan workloads.

Notably, the Radeon Pro 455 has a Geekbench Metal score of 15,916, which has no direct counterpart in the GTX 1660's benchmark suite. This Metal score is the AMD card's strongest result, suggesting its GCN architecture is well-optimized for Apple's API. However, this does not offset the OpenCL and Vulkan deficits. The GTX 1660 also has a PassMark G3D score of 11,646 and a PassMark GPU Compute score of 4,963, while the Radeon Pro 455 has no PassMark results. The GTX 1660's PassMark DirectX scores range from 49 (DirectX 12) to 177 (DirectX 9), with the DirectX 11 score at 79 and DirectX 10 at 61. These numbers provide a broader picture of the GTX 1660's legacy API performance, which the Radeon Pro 455 lacks.

Specification Differences

The two cards differ across nearly every measurable specification. The Radeon Pro 455 uses a 14 nm process at GlobalFoundries, while the GTX 1660 uses 12 nm at TSMC. Transistor counts are 3,000 million versus 6,600 million, with die sizes of 123 mm² and 284 mm² respectively. The GTX 1660 has a base clock of 1530 MHz and boost clock of 1785 MHz, while the Radeon Pro 455 has no listed base or boost clocks. Memory clocks differ: the Radeon Pro 455 runs at 1270 MHz (5.1 Gbps effective), while the GTX 1660 runs at 2001 MHz (8 Gbps effective).

Memory specifications diverge sharply: 2 GB versus 6 GB GDDR5, 128-bit versus 192-bit bus, and 81.28 GB/s versus 192.1 GB/s bandwidth. Core counts favor the GTX 1660 across the board: 1,408 versus 768 shading units, 88 versus 48 TMUs, and 48 versus 16 ROPs. Pixel and texture rates are 85.68 GPixel/s and 157.1 GTexel/s for the GTX 1660, against 13.68 GPixel/s and 41.04 GTexel/s for the Radeon Pro 455. FP32 is 5.027 TFLOPS versus 1,313.3 GFLOPS, and FP16 is 10.05 TFLOPS (2:1) versus 1,313.3 GFLOPS (1:1).

Power and physical specs also differ: TDP is 120 W versus 35 W, slot width is dual-slot versus MXM Module, and power connectors are 1x 8-pin versus none. The GTX 1660 has a 300 W suggested PSU, while the Radeon Pro 455 has none listed. Bus interfaces are PCIe 3.0 x16 versus PCIe 3.0 x8. Display outputs are 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a for the GTX 1660, versus "Portable Device Dependent" for the Radeon Pro 455. Release dates are March 2019 for the GTX 1660 and October 2016 for the Radeon Pro 455. The GTX 1660 has a launch MSRP of 219 USD, while the Radeon Pro 455 has no listed MSRP. Both cards are end-of-life.