AMD Radeon Pro 555X vs NVIDIA GeForce GTX 660 Ti Comparison

Head-to-Head Benchmarks

The recorded benchmark data presents a clear picture in the two directly comparable workloads. In Geekbench OpenCL, the NVIDIA GeForce GTX 660 Ti scores 15,113 against the AMD Radeon Pro 555X's 12,628, a decisive 19.7% advantage. The Vulkan result follows a similar pattern: 15,012 for the NVIDIA part versus 12,855 for the AMD part, yielding a 16.8% lead. These are substantial margins in both compute and graphics API workloads, indicating the GTX 660 Ti holds a meaningful performance tier above the Radeon Pro 555X.

The average benchmark score across all recorded tests reinforces this separation. The GTX 660 Ti posts an average of 15,063, while the Radeon Pro 555X averages 13,321. That is a difference of approximately 13% in favor of the NVIDIA card. Notably, the Radeon Pro 555X does have a Geekbench Metal score of 14,479, a test the GTX 660 Ti does not have recorded. This Metal result is its strongest showing, but it does not close the gap in the shared workloads. The data shows two wins for the GTX 660 Ti and zero for the Radeon Pro 555X in their head-to-head comparisons.

Looking at the broader competitive landscape, each card sits in a different neighborhood. The GTX 660 Ti's average score places it just 0.1% behind the AMD Radeon Pro 560X (15,082) and 0.7% behind the AMD Radeon RX 7600 (15,171). It is also within 0.9% of the NVIDIA GeForce RTX 3050 OEM (15,199). These are tight margins, meaning the GTX 660 Ti is essentially at parity with a cluster of much newer GPUs in these synthetic benchmarks. On the other side, the Radeon Pro 555X's average score is nearly identical to the NVIDIA GeForce GTX 480 (13,300), trailing by just 0.2%. It is also close to the AMD FirePro M6100 (13,354) and the AMD Radeon RX 5500M (13,356), each within 0.3% or less. The picture is one of two cards competing in different performance brackets, with the GTX 660 Ti clearly occupying the higher tier.

The percentile rankings confirm this split. The GTX 660 Ti sits at the 57th percentile among all GPUs, while the Radeon Pro 555X is at the 54th percentile. This is a modest overall ranking difference, but the raw score delta in the shared tests is what matters most for direct comparisons.

The Verdict

The evidence from the database is unambiguous for users who need raw compute or graphics performance in cross-platform workloads. The NVIDIA GeForce GTX 660 Ti is the faster card in every shared benchmark. In OpenCL, it is 19.7% ahead, and in Vulkan, it is 16.8% ahead. Anyone choosing between these two strictly on measured performance should select the GTX 660 Ti.

However, context matters. The Radeon Pro 555X is designed for a different environment entirely. It is an integrated-class part for portable systems, drawing 75 W with no power connectors, whereas the GTX 660 Ti is a dual-slot, 150 W card requiring two 6-pin connectors and a 450 W power supply. The Radeon Pro 555X also has a Metal score of 14,479, which is notably higher than its OpenCL score of 12,628. For users in Apple's ecosystem where Metal is the primary API, the Radeon Pro 555X may be the only option that fits the hardware constraints, and its Metal performance is respectable.

The data does not suggest the Radeon Pro 555X is a better performer in absolute terms. It is not. But it serves a specific niche: low-power, portable, and Mac-compatible systems. The GTX 660 Ti is a desktop card from 2012, end-of-life, and requires a full PCIe 3.0 x16 slot. The Radeon Pro 555X, released in 2018, uses a PCIe 3.0 x8 interface and is described as "Portable Device Dependent" for display outputs.

For a desktop builder with the power budget and slot space, the GTX 660 Ti is the clear choice based on benchmark results. For a laptop or all-in-one Mac user, the Radeon Pro 555X is the only one of these two that is physically compatible, and its Metal performance of 14,479 is its saving grace.

Where Each One Wins

The GTX 660 Ti wins in every measured cross-platform workload. Its OpenCL score of 15,113 surpasses the Radeon Pro 555X's 12,628 by a wide margin. Its Vulkan score of 15,012 beats the 12,855 posted by the AMD card. In terms of raw throughput, the GTX 660 Ti also has higher specifications: 2.634 TFLOPS FP32 versus 1,393.2 GFLOPS for the Radeon Pro 555X, a 112 TMU count versus 48, and 24 ROPs versus 16. The NVIDIA card's memory bandwidth of 144.2 GB/s is also higher than the AMD card's 94.08 GB/s, despite the GTX 660 Ti having a narrower 192-bit bus compared to the 128-bit bus on the Radeon Pro 555X. The GTX 660 Ti's 2 GB of GDDR5 is smaller than the 4 GB on the Radeon Pro 555X, but in the recorded benchmarks, capacity did not translate into a performance win.

The Radeon Pro 555X wins in efficiency and form factor. It draws 75 W versus 150 W, it is an IGP (integrated graphics processor) rather than a dual-slot add-in card, and it requires no external power connectors. Its transistor density is also much higher: 24.4M transistors per square millimeter versus 12.0M for the GTX 660 Ti, due to the 14 nm process node compared to 28 nm. The AMD card also supports newer API features, including DirectX 12 (12_0) versus the GTX 660 Ti's DirectX 12 (11_0), and Vulkan 1.3 versus Vulkan 1.2.175. Additionally, the Radeon Pro 555X has 4 GB of VRAM, which is double the GTX 660 Ti's capacity, a factor that could matter in texture-heavy workloads not represented in the recorded tests.

FAQ

Q: Which GPU is faster in OpenCL?

A: The NVIDIA GeForce GTX 660 Ti scores 15,113 in Geekbench OpenCL, which is 19.7% higher than the AMD Radeon Pro 555X's 12,628.

Q: How do the two cards compare in Vulkan performance?

A: The GTX 660 Ti records a Geekbench Vulkan score of 15,012, while the Radeon Pro 555X scores 12,855. The NVIDIA card leads by 16.8%.

Q: Does the AMD Radeon Pro 555X have any benchmark where it outperforms the GTX 660 Ti?

A: In the recorded head-to-head tests, no. The GTX 660 Ti wins both shared workloads. However, the Radeon Pro 555X has a Geekbench Metal score of 14,479, which is its highest recorded score, but no Metal result is available for the GTX 660 Ti.

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

A: The GTX 660 Ti has an average benchmark score of 15,063, placing it at the 57th percentile. The Radeon Pro 555X averages 13,321, at the 54th percentile.

Q: Which card is more power-efficient?

A: The AMD Radeon Pro 555X is rated at 75 W TDP and requires no power connectors, while the NVIDIA GeForce GTX 660 Ti is rated at 150 W and needs two 6-pin connectors plus a 450 W power supply.

Q: What are the memory specifications of each card?

A: The GTX 660 Ti has 2 GB of GDDR5 on a 192-bit bus with 144.2 GB/s bandwidth. The Radeon Pro 555X has 4 GB of GDDR5 on a 128-bit bus with 94.08 GB/s bandwidth.

Architecture Differences

The two GPUs come from fundamentally different design eras and philosophies. The NVIDIA GeForce GTX 660 Ti uses the GK104 chip, based on the Kepler architecture, manufactured on a 28 nm process at TSMC. It integrates 3,540 million transistors on a 294 mm² die, resulting in a transistor density of 12.0 million per square millimeter. The chip includes 1,344 shading units, 112 texture mapping units, and 24 raster operators. Its clock speeds are 915 MHz base and 980 MHz boost, with memory running at 1502 MHz (6 Gbps effective). The memory subsystem is 2 GB of GDDR5 on a 192-bit bus, delivering 144.2 GB/s. The card's FP32 throughput is 2.634 TFLOPS, and it has no recorded FP16 capability. It supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The GTX 660 Ti is a dual-slot card, 241 mm long, with two DVI outputs, one HDMI 1.4a, and one DisplayPort 1.2.

The AMD Radeon Pro 555X is built on the Polaris 21 chip, using the GCN 4.0 architecture, fabricated on a 14 nm process at GlobalFoundries. It contains 3,000 million transistors on a much smaller 123 mm² die, giving it a transistor density of 24.4 million per square millimeter, more than double that of the NVIDIA chip. The GPU has 768 shading units, 48 TMUs, and 16 ROPs. Clock speeds are not recorded for base or boost, but memory runs at 1470 MHz (5.9 Gbps effective). The memory configuration is 4 GB of GDDR5 on a 128-bit bus with 94.08 GB/s bandwidth. FP32 performance is 1,393.2 GFLOPS, and it has matching FP16 throughput at a 1:1 ratio, a feature the GTX 660 Ti does not list. The Radeon Pro 555X supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3, the latter being a newer version than what the GTX 660 Ti supports.

The interface and power delivery differ sharply. The GTX 660 Ti uses a PCIe 3.0 x16 connection, while the Radeon Pro 555X uses PCIe 3.0 x8. The NVIDIA card is a standalone dual-slot expansion card with a 150 W TDP, two 6-pin power connectors, and a suggested 450 W power supply. The AMD card is an IGP with a 75 W TDP and no power connectors, designed to be integrated into a portable device. Its display outputs are dependent on the host device, unlike the fixed DVI, HDMI, and DisplayPort outputs on the GTX 660 Ti. The release timeline also reflects this divergence: the GTX 660 Ti launched in August 2012 with a launch MSRP of 299 USD, while the Radeon Pro 555X arrived in July 2018. Both are now end-of-life products.