AMD Radeon Pro 570X vs NVIDIA GeForce GTX 1060 5 GB Comparison

AMD
RADEON

AMD Radeon Pro 570X

CORE STATE Ellesmere
VRAM 4 GB
CLOCK SPEED 1105 MHz
TDP 150 W
BUS WIDTH 256 bit
ARCHITECTURE GCN 4.0
nm
PROCESS 14 nm
LAUNCH DATE 2019
VS
NVIDIA
GEFORCE

GeForce GTX 1060 5 GB

CORE STATE GP106
VRAM 5 GB
CLOCK SPEED 1709 MHz
TDP 120 W
BUS WIDTH 160 bit
ARCHITECTURE Pascal
nm
PROCESS 16 nm
LAUNCH DATE 2017

PERFORMANCE BENCHMARKS

geekbench_metal
40,066
N/A
geekbench_opencl
27,604
33,397
geekbench_vulkan
28,859
33,991

Analysis: AMD Radeon Pro 570X vs NVIDIA GeForce GTX 1060 5 GB

The benchmark data presents a clear picture: the NVIDIA GeForce GTX 1060 5 GB and AMD Radeon Pro 570X are both end-of-life graphics processors, but they target different ecosystems and exhibit distinct performance profiles. The GTX 1060 5 GB, a Pascal-based part from the GeForce 10-series, holds a decisive lead in the two cross-platform compute tests recorded, while the Radeon Pro 570X, a GCN 4.0 part designed for Mac systems, counters with a strong showing in its native Metal API. The data suggests that while the NVIDIA card is the faster compute device in OpenCL and Vulkan, the AMD card’s advantage lies in its integration and API support within its intended platform.

Head-to-Head Benchmarks

The head-to-head results are unambiguous in favor of the NVIDIA GeForce GTX 1060 5 GB. In the Geekbench OpenCL test, the GTX 1060 5 GB scores 33,397, while the Radeon Pro 570X trails at 27,604. This translates to a 21% advantage for the NVIDIA card, a substantial margin that indicates significantly higher raw compute throughput in a cross-vendor, cross-platform API. The NVIDIA card’s superior FP32 performance of 4.375 TFLOPS, compared to the AMD card’s 3.960 TFLOPS, likely underpins this result, but the 21% delta in the actual test is larger than the 10% difference in theoretical peak throughput, suggesting other architectural efficiencies are at play in the OpenCL workload.

The Vulkan results echo the same trend, though with a slightly narrower gap. The GTX 1060 5 GB scores 33,991, against the Radeon Pro 570X’s 28,859, resulting in a 17.8% lead for the NVIDIA card. This consistency across two different low-level APIs reinforces the conclusion that the GTX 1060 5 GB is the more capable compute engine in this comparison. Notably, the NVIDIA card’s Vulkan score is nearly identical to its OpenCL score, showing a balanced performance profile across APIs, whereas the AMD card’s Vulkan score of 28,859 is actually higher than its OpenCL score of 27,604, hinting at a slight preference for Vulkan in its driver stack.

Despite these two decisive wins for the NVIDIA card, the Radeon Pro 570X is not without a counterpoint. The AMD card was benchmarked in Geekbench Metal, where it scores 40,066. This is a significantly higher score than either of its OpenCL or Vulkan results, and it also eclipses the NVIDIA card’s scores in those APIs. The data does not provide a head-to-head Metal comparison for the GTX 1060 5 GB, as the NVIDIA card lacks a Metal benchmark entry, but the Radeon Pro 570X’s Metal performance demonstrates that it excels in its native Apple ecosystem. This single data point, however, does not change the overall head-to-head record, which stands at 2 wins for the NVIDIA card and 0 for the AMD card.

The Verdict

From the benchmark data alone, the NVIDIA GeForce GTX 1060 5 GB is the superior compute performer for general-purpose and cross-platform workloads. Its 21% lead in OpenCL and 17.8% lead in Vulkan are decisive, and its average benchmark score of 33,694 places it at the 78th percentile of all GPUs, a respectable standing. The AMD Radeon Pro 570X, with an average score of 32,176 and a 76th percentile rank, is slightly behind in aggregate, but its 40,066 Metal score is a powerful outlier. This suggests that for users locked into a Mac environment where Metal is the primary API, the Radeon Pro 570X is likely the better choice, despite its lower scores in cross-platform tests. The data implies a clear split: choose the GTX 1060 5 GB for raw compute in OpenCL/Vulkan environments, and choose the Radeon Pro 570X for macOS-specific applications where Metal is the standard.

Architecture Differences

The two GPUs are built on fundamentally different architectures from different process nodes. The NVIDIA GeForce GTX 1060 5 GB uses the GP106 chip, based on the Pascal architecture, and is manufactured on a 16 nm process by TSMC. In contrast, the AMD Radeon Pro 570X uses the Ellesmere chip, based on the GCN 4.0 architecture, and is manufactured on a 14 nm process by GlobalFoundries. These differences have significant implications for the physical design and feature set. The NVIDIA chip integrates 4,400 million transistors on a 200 mm² die, yielding a transistor density of 22.0M per mm². The AMD chip is larger in both respects, with 5,700 million transistors on a 232 mm² die, giving it a higher density of 24.6M per mm². This suggests that AMD’s 14 nm process allowed for a denser packing of transistors, but the Pascal architecture appears to extract more compute performance from its smaller transistor count in the tested workloads.

The functional unit configuration also differs considerably. The GTX 1060 5 GB has 1,280 shading units, 80 texture mapping units (TMUs), and 40 render output units (ROPs). The Radeon Pro 570X counters with more shading units at 1,792, more TMUs at 112, but fewer ROPs at 32. This configuration explains some of the performance differences: the AMD card has more parallel shader cores, but the NVIDIA card’s higher clock speeds and ROP count contribute to its higher pixel rate of 68.36 GPixel/s, versus the AMD card’s 35.36 GPixel/s. The texture rate is closer, with the NVIDIA card at 136.7 GTexel/s and the AMD card at 123.8 GTexel/s. The FP16 performance is a stark differentiator: the Radeon Pro 570X offers 3.960 TFLOPS at FP16 with a 1:1 ratio to FP32, while the GTX 1060 5 GB is heavily limited to 68.36 GFLOPS at FP16, a 1:64 ratio. This indicates that for workloads leveraging half-precision, the AMD card is vastly superior.

Specification Differences

The specification differences between the two cards are numerous and impactful, starting with memory configuration. The NVIDIA GeForce GTX 1060 5 GB features 5 GB of GDDR5 memory on a 160-bit bus, delivering a bandwidth of 160.2 GB/s. The AMD Radeon Pro 570X has 4 GB of GDDR5 memory on a wider 256-bit bus, resulting in a higher bandwidth of 217.0 GB/s. This is a notable trade-off: the NVIDIA card has more capacity, but the AMD card has more bandwidth, which can be critical for certain high-resolution textures or compute workloads. Clock speeds also diverge significantly. The GTX 1060 5 GB runs at a base clock of 1506 MHz and a boost clock of 1709 MHz, while the Radeon Pro 570X operates at a much lower 1000 MHz base and 1105 MHz boost. The memory clocks also differ, with the NVIDIA card running at 2002 MHz (8 Gbps effective) and the AMD card at 1695 MHz (6.8 Gbps effective). The higher clocks of the NVIDIA card are a primary reason for its compute performance lead, compensating for its narrower memory bus.

Power and physical specifications further separate the two. The GTX 1060 5 GB has a TDP of 120 W, requires a single 6-pin power connector, and is a dual-slot card measuring 250 mm (9.8 inches) in length. The Radeon Pro 570X has a higher TDP of 150 W but is an IGP (integrated graphics processor) with no power connectors and dimensions that are listed as portable device dependent. This indicates the AMD card is designed for integration into laptops or all-in-one systems, rather than as a standalone add-in board. The display outputs also reflect this: the NVIDIA card offers 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a, while the AMD card’s outputs are listed as "Portable Device Dependent." In terms of API support, the NVIDIA card supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, while the AMD card supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3. This gives the NVIDIA card a slight edge in API version support, particularly with Vulkan 1.4 versus 1.3.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce GTX 1060 5 GB has an average benchmark score of 33,694, placing it at the 78th percentile of all GPUs. The AMD Radeon Pro 570X has a lower average score of 32,176, at the 76th percentile.

Q: How does the AMD Radeon Pro 570X perform in the Metal API?

A: The Radeon Pro 570X scores 40,066 in Geekbench Metal, which is its highest benchmark score across all tested APIs. This suggests strong performance in Apple’s native graphics API.

Q: What is the memory bandwidth difference between the two cards?

A: The AMD Radeon Pro 570X has a higher memory bandwidth of 217.0 GB/s, thanks to its 256-bit bus, while the NVIDIA GeForce GTX 1060 5 GB has 160.2 GB/s from its 160-bit bus.

Q: Which GPU has more shading units?

A: The AMD Radeon Pro 570X has 1,792 shading units, while the NVIDIA GeForce GTX 1060 5 GB has 1,280 shading units. Despite having more, the AMD card’s lower clock speeds result in lower FP32 performance.

Q: What is the difference in FP16 performance?

A: The AMD Radeon Pro 570X offers 3.960 TFLOPS of FP16 performance (1:1 ratio with FP32), while the NVIDIA GeForce GTX 1060 5 GB is limited to 68.36 GFLOPS (1:64 ratio). The AMD card is vastly superior for FP16 workloads.

Q: Which card has a higher pixel fill rate?

A: The NVIDIA GeForce GTX 1060 5 GB has a pixel rate of 68.36 GPixel/s, which is nearly double the AMD Radeon Pro 570X’s 35.36 GPixel/s, due to its higher ROP count and clock speeds.

Where Each One Wins

The NVIDIA GeForce GTX 1060 5 GB wins in raw compute performance across open-standard APIs. Its 21% OpenCL lead and 17.8% Vulkan lead over the Radeon Pro 570X make it the clear choice for users running compute tasks in these environments. Its higher pixel rate of 68.36 GPixel/s and texture rate of 136.7 GTexel/s also suggest advantages in traditional rasterization workloads, particularly at higher resolutions where pixel throughput is critical. Furthermore, its support for Vulkan 1.4 and DirectX 12 (12_1) gives it a more modern feature set for cross-platform gaming and compute. The card’s dual-slot design and dedicated display outputs also make it suitable for desktop systems where multiple monitors are required.

The AMD Radeon Pro 570X wins decisively in the Apple ecosystem. Its Geekbench Metal score of 40,066 is the highest single benchmark result in this comparison, indicating that it is optimized for macOS and iOS development environments. For users running Metal-based applications, such as professional creative software or machine learning frameworks on a Mac, the Radeon Pro 570X is the only card here with a native Metal path. Its higher bandwidth of 217.0 GB/s could also benefit memory-intensive workloads, and its 1:1 FP16 performance of 3.960 TFLOPS makes it a better option for compute tasks that rely on half-precision arithmetic, such as certain AI inference workloads. The card’s IGP form factor and lack of power connectors also make it the only viable option for portable or integrated systems, as it is designed to be soldered onto a board rather than installed in a PCIe slot. In summary, the GTX 1060 5 GB wins on general compute and desktop graphics, while the Radeon Pro 570X wins on Apple-specific platforms and FP16-heavy workloads.

DETAILED SPECIFICATIONS

SPECIFICATION
Pro 570X
GTX 1060 5 GB
Core Specs
Shading Units
1,792
1,280 -28.6%
Shaders
1,792
1,280 -28.6%
TMUs
112
80 -28.6%
ROPs
32
40 +25.0%
Compute Units
28
—
SM Count
—
10
Clocks
Base Clock
1000 MHz
1506 MHz
Boost Clock
1105 MHz
1709 MHz
Memory Clock
1695 MHz 6.8 Gbps effective
2002 MHz 8 Gbps effective
Memory
Memory Size
4 GB
5 GB
VRAM (MB)
4,096
5,120 +25.0%
Memory Type
GDDR5
GDDR5
Memory Bus
256 bit
160 bit
Bandwidth
217.0 GB/s
160.2 GB/s
Cache
L1 Cache
16 KB (per CU)
48 KB (per SM)
L2 Cache
2 MB
1280 KB
Performance
Pixel Rate
35.36 GPixel/s
68.36 GPixel/s
Texture Rate
123.8 GTexel/s
136.7 GTexel/s
FP32 (TFLOPS)
3.960 TFLOPS
4.375 TFLOPS
FP64 (TFLOPS)
247.5 GFLOPS (1:16)
136.7 GFLOPS (1:32)
FP16 (TFLOPS)
3.960 TFLOPS (1:1)
68.36 GFLOPS (1:64)
Power
TDP
150 W
120 W
TDP (W)
150
120 -20.0%
Suggested PSU
—
300 W
Power Connectors
None
1x 6-pin
Architecture
Architecture
GCN 4.0
Pascal
GPU Name
Ellesmere
GP106
Generation
Radeon Pro Mac (500X Series)
GeForce 10
Process Size
14 nm
16 nm
Transistors
5,700 million
4,400 million
Die Size
232 mm²
200 mm²
Foundry
GlobalFoundries
TSMC
Density
24.6M / mm²
22.0M / mm²
API Support
DirectX
12 (12_0)
12 (12_1)
OpenGL
4.6
4.6
Vulkan
1.3
1.4
OpenCL
2.1
3.0
CUDA
—
6.1
Shader Model
6.7
6.8
Physical
Slot Width
IGP
Dual-slot
Length
—
250 mm 9.8 inches
Outputs
Portable Device Dependent
1x DVI1x HDMI 2.03x DisplayPort 1.4a
Bus Interface
PCIe 3.0 x16
PCIe 3.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
—
GeForce 900
Successor
—
GeForce 20
View Radeon Pro 570X Details View GeForce GTX 1060 5 GB Details