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

AMD
RADEON

AMD Radeon Pro 570

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 2017
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
39,945
N/A
geekbench_opencl
27,702
33,397
geekbench_vulkan
31,974
33,991

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

# NVIDIA GeForce GTX 1060 5 GB vs AMD Radeon Pro 570

The NVIDIA GeForce GTX 1060 5 GB and AMD Radeon Pro 570 represent two distinct approaches to mid-range graphics in 2017, with the former built on TSMC's 16 nm Pascal architecture and the latter on GlobalFoundries' 14 nm GCN 4.0 architecture. Both cards land at the 78th percentile against all GPUs, yet their benchmark profiles diverge sharply: the GTX 1060 5 GB averages 33694 across OpenCL and Vulkan tests, while the Radeon Pro 570 averages 33207 across Metal, OpenCL, and Vulkan tests—a difference of roughly 1.5% in average score, but with far larger swings in individual workloads.

Where Each One Wins

The GTX 1060 5 GB wins the head-to-head comparison decisively in the two shared benchmark tests. In Geekbench OpenCL, the NVIDIA card scores 33397 against the Radeon Pro 570's 27702, a 20.6% advantage. This is the single largest margin between the two cards and reflects the GTX 1060's higher FP32 throughput of 4.375 TFLOPS versus 3.960 TFLOPS, as well as its superior pixel fill rate of 68.36 GPixel/s compared to 35.36 GPixel/s. The Vulkan test shows a narrower but still clear win: 33991 versus 31974, a 6.3% edge for the GTX 1060. These results indicate the NVIDIA card holds a consistent lead in compute-oriented and cross-platform graphics workloads.

The Radeon Pro 570's only benchmark win comes in Geekbench Metal, where it scores 39945—a test the GTX 1060 does not appear in. This Metal score is notably higher than the NVIDIA card's best result (33991 in Vulkan), suggesting AMD's card is particularly well-suited to Apple's Metal API environment. However, the Radeon Pro 570's OpenCL score of 27702 is its weakest result, trailing its own Vulkan score by 14.9%, which highlights inconsistency across different compute frameworks. The card's 4 GB of GDDR5 memory on a 256-bit bus delivers 217.0 GB/s bandwidth—35.5% more than the GTX 1060's 160.2 GB/s—yet this bandwidth advantage does not translate into OpenCL or Vulkan wins.

The Verdict

The data points to the NVIDIA GeForce GTX 1060 5 GB as the stronger all-around performer for users running OpenCL or Vulkan workloads. Its 20.6% OpenCL lead over the Radeon Pro 570 is substantial, and its 6.3% Vulkan advantage reinforces a pattern of consistent superiority in general-purpose compute. The GTX 1060 also achieves this with lower power draw: 120 W TDP versus 150 W, and it requires only a single 6-pin power connector while the Radeon Pro 570 uses no external connectors at all—though the latter's integrated form factor (IGP slot width) makes direct power comparison less meaningful.

The AMD Radeon Pro 570 is the better choice specifically for Metal-centric environments, where its 39945 score demonstrates a capability that the GTX 1060 cannot match in the available data. This card also offers more memory bandwidth (217.0 GB/s) and a wider 256-bit bus, which could benefit memory-intensive workloads, but its lower FP32 throughput and texture rate (123.8 GTexel/s versus 136.7 GTexel/s) hold it back in the tested compute scenarios. For users prioritizing OpenCL or Vulkan performance, the GTX 1060 5 GB is the clear winner; for those locked into Metal environments, the Radeon Pro 570's specialized strength justifies consideration.

Head-to-Head Benchmarks

The Geekbench OpenCL test delivers the most dramatic separation between these two cards. The GTX 1060 5 GB scores 33397, while the Radeon Pro 570 manages only 27702—a 20.6% gap that places the NVIDIA card in a different performance tier for this workload. This margin is larger than the difference between the GTX 1060's average score (33694) and its nearest rival, the NVIDIA RTX A5000 (33622, just 0.2% behind), which suggests the Radeon Pro 570's OpenCL performance is unusually weak relative to its other results. The GTX 1060's higher base clock of 1506 MHz and boost clock of 1709 MHz, combined with its 1280 shading units, likely drive this advantage despite having fewer shading units than the Radeon's 1792.

In Geekbench Vulkan, the gap narrows considerably. The GTX 1060 scores 33991, and the Radeon Pro 570 scores 31974, yielding a 6.3% delta. This closer margin indicates that Vulkan's lower-level API access allows the Radeon Pro 570 to better utilize its hardware, particularly its wider memory bus and higher bandwidth. Notably, the GTX 1060's Vulkan score is nearly identical to its OpenCL score (33991 versus 33397, a 1.8% difference), while the Radeon Pro 570 shows a 15.4% improvement from OpenCL to Vulkan (27702 to 31974). This pattern suggests AMD's architecture benefits more from explicit APIs, whereas NVIDIA maintains consistent performance across abstraction layers.

The Radeon Pro 570's Metal score of 39945 stands as its single strongest result, exceeding its Vulkan score by 24.9% and its OpenCL score by 44.2%. This massive spread across three APIs makes the card's performance highly workload-dependent. Meanwhile, the GTX 1060's two scores (33397 and 33991) differ by only 1.8%, indicating far greater stability across compute frameworks. For users with mixed workloads, the NVIDIA card's predictability is an advantage; for those exclusively in Metal environments, the AMD card's peak capability is compelling.

FAQ

Q: Which card has the higher average benchmark score?

A: The NVIDIA GeForce GTX 1060 5 GB averages 33694 across its two benchmark results, while the AMD Radeon Pro 570 averages 33207 across three results—a difference of 487 points, or roughly 1.5% in favor of the NVIDIA card.

Q: How does the GTX 1060 5 GB compare to its closest rivals?

A: The GTX 1060 sits within 1% of several competitors: it trails the AMD Radeon HD 7950 by 0.8% (33951), trails the AMD Radeon RX 480 by 0.9% (33997), and trails the AMD Radeon RX 7700S by 0.5% (33849). It leads the NVIDIA RTX A5000 by 0.2% (33622).

Q: What is the biggest performance gap in the head-to-head tests?

A: The largest gap is in Geekbench OpenCL, where the GTX 1060 5 GB beats the Radeon Pro 570 by 20.6%—a difference of 5695 points (33397 versus 27702). The Vulkan gap is much smaller at 6.3%, or 2017 points.

Q: Does the Radeon Pro 570 win any benchmark against the GTX 1060?

A: The Radeon Pro 570 does not win any of the shared benchmarks. Its only notable result is in Geekbench Metal, where it scores 39945, but the GTX 1060 has no Metal score in the data for comparison.

Q: Which card has better memory specifications?

A: The Radeon Pro 570 offers 4 GB of GDDR5 memory on a 256-bit bus, delivering 217.0 GB/s bandwidth. The GTX 1060 5 GB has 5 GB of GDDR5 on a narrower 160-bit bus, yielding 160.2 GB/s. The AMD card provides 35.5% more bandwidth, while the NVIDIA card offers 1 GB more capacity.

Q: How do their compute capabilities compare?

A: The GTX 1060 delivers 4.375 TFLOPS of FP32 performance, exceeding the Radeon Pro 570's 3.960 TFLOPS by 10.5%. The NVIDIA card also has higher texture rate (136.7 GTexel/s versus 123.8 GTexel/s) and pixel rate (68.36 GPixel/s versus 35.36 GPixel/s). However, the Radeon Pro 570 offers 1:1 FP16 performance (3.960 TFLOPS), while the GTX 1060's FP16 is heavily reduced at 68.36 GFLOPS (1:64).

Architecture Differences

The two cards stem from fundamentally different architectural designs. The GTX 1060 5 GB uses NVIDIA's Pascal architecture on the GP106 chip, fabricated by TSMC at 16 nm. It packs 4,400 million transistors into a 200 mm² die, achieving a transistor density of 22.0M per mm². The Radeon Pro 570 uses AMD's GCN 4.0 architecture on the Ellesmere chip, built by GlobalFoundries at 14 nm, with 5,700 million transistors across 232 mm²—a density of 24.6M per mm². The AMD chip is 16% larger and contains 29.5% more transistors, yet the NVIDIA card still delivers higher FP32 throughput.

Core configuration differs significantly. The GTX 1060 has 1280 shading units, 80 texture mapping units, and 40 render output units. The Radeon Pro 570 counters with 1792 shading units, 112 TMUs, and 32 ROPs. AMD's advantage in shading units and TMUs (40% and 40% more, respectively) is offset by NVIDIA's higher clock speeds: the GTX 1060 runs at 1506 MHz base and 1709 MHz boost, while the Radeon Pro 570 operates at 1000 MHz base and 1105 MHz boost. This clock difference—50.6% higher base and 54.7% higher boost for NVIDIA—explains how fewer cores can produce more throughput.

Memory subsystems take opposite strategies. The GTX 1060 uses 5 GB of GDDR5 on a 160-bit bus, running at 2002 MHz (8 Gbps effective) for 160.2 GB/s bandwidth. The Radeon Pro 570 uses 4 GB of GDDR5 on a 256-bit bus, at 1695 MHz (6.8 Gbps effective), achieving 217.0 GB/s. AMD's 60% wider bus compensates for its 15.3% lower memory clock, resulting in 35.5% more bandwidth. The GTX 1060's extra gigabyte of capacity could benefit larger datasets, but the Radeon's bandwidth headroom suits streaming textures.

Power and form factor diverge sharply. The GTX 1060 is a dual-slot card, 250 mm long (9.8 inches), with a 120 W TDP, one 6-pin power connector, and a suggested 300 W PSU. The Radeon Pro 570 is an integrated GPU (IGP) with no power connectors, a 150 W TDP, and portable-device-dependent display outputs. The NVIDIA card offers 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a outputs, while the AMD card's outputs depend on the host device. API support also differs: the GTX 1060 supports DirectX 12 (12_1), while the Radeon Pro 570 supports DirectX 12 (12_0); both offer OpenGL 4.6, but the AMD card supports Vulkan 1.3 versus NVIDIA's Vulkan 1.4. The GTX 1060's successor is GeForce 20, while the Radeon Pro 570 lists no successor or predecessor, reflecting its specialized position in the Radeon Pro Mac 500 series.

DETAILED SPECIFICATIONS

SPECIFICATION
Pro 570
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 (500 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 570 Details View GeForce GTX 1060 5 GB Details