AMD Radeon RX 560X vs NVIDIA GeForce GTX 1060 3 GB Comparison

AMD
RADEON

AMD Radeon RX 560X

CORE STATE Polaris 21
VRAM 4 GB
CLOCK SPEED 1275 MHz
TDP 75 W
BUS WIDTH 128 bit
ARCHITECTURE GCN 4.0
nm
PROCESS 14 nm
LAUNCH DATE 2018
VS
NVIDIA
GEFORCE

GeForce GTX 1060 3 GB

CORE STATE GP106
VRAM 3 GB
CLOCK SPEED 1708 MHz
TDP 120 W
BUS WIDTH 192 bit
ARCHITECTURE Pascal
nm
PROCESS 16 nm
LAUNCH DATE 2016

PERFORMANCE BENCHMARKS

geekbench_opencl
17,020
27,001
geekbench_vulkan
20,231
17,475
geekbench_metal
N/A
13,527

Analysis: AMD Radeon RX 560X vs NVIDIA GeForce GTX 1060 3 GB

The data presents a clear generational clash between NVIDIA’s Pascal-era GTX 1060 3 GB and AMD’s GCN 4.0-based RX 560X. While both cards sit near the 62nd-64th percentile of all GPUs, their performance profiles diverge sharply depending on the API workload, revealing that the choice between them hinges on which benchmark environment matters more to the user. The head-to-head results show a split decision, but the magnitude of the wins is far from symmetrical.

Head-to-Head Benchmarks

The GeForce GTX 1060 3 GB delivers a decisive victory in the Geekbench OpenCL test, posting a score of 27,001 against the RX 560X’s 17,020. That is a 58.6% advantage, a staggering margin that underscores the GTX 1060’s raw compute throughput advantage. This result aligns with the specification sheet: the GTX 1060 offers 3.935 TFLOPS of FP32 performance versus 2.611 TFLOPS for the RX 560X, and its 192-bit memory bus yields 192.2 GB/s of bandwidth compared to the RX 560X’s 128-bit bus and 112.0 GB/s. The OpenCL score is not close; it reflects a fundamental gap in shading power and memory subsystem capacity.

However, the Vulkan results flip the script. The AMD Radeon RX 560X wins the Geekbench Vulkan test with a score of 20,231 against the GTX 1060’s 17,475, a 13.6% margin in AMD’s favor. This is a remarkable outcome given the GTX 1060’s hardware advantages. The RX 560X’s GCN 4.0 architecture appears to handle Vulkan’s low-level draw call overhead and asynchronous compute features more efficiently, at least in this synthetic workload. The fact that AMD wins despite having 128 fewer shading units (1024 vs 1152) and less than half the ROP count (16 vs 48) suggests that driver optimization and architectural scheduling play an outsized role in Vulkan performance. The data also shows the GTX 1060’s FP16 throughput is crippled at 61.49 GFLOPS (1:64 ratio), while the RX 560X runs FP16 at 2.611 TFLOPS (1:1 ratio), which may contribute to AMD’s Vulkan showing if the test leverages half-precision paths.

Counting wins, each card takes one benchmark. The GTX 1060’s OpenCL win is nearly 4.3 times larger in percentage terms than the RX 560X’s Vulkan victory. This asymmetry matters: a 58.6% lead in one test versus a 13.6% lead in the other means the GTX 1060 has a higher ceiling for compute-heavy workloads, while the RX 560X offers a more modest but real advantage in Vulkan-specific scenarios.

FAQ

Q: Which card has the higher average benchmark score?

A: The NVIDIA GeForce GTX 1060 3 GB leads with an average benchmark score of 19,334, while the AMD Radeon RX 560X trails at 18,626. The GTX 1060 also holds a slightly higher percentile rank at 64 versus the RX 560X’s 62.

Q: How does the GTX 1060 compare to its nearest rivals?

A: The GTX 1060 sits 0.4% below the AMD Radeon RX 7900 XTX (19,410) and 0.8% above the NVIDIA TITAN Xp (19,177). It also edges out the NVIDIA GeForce GTX 780 (19,164) by 0.9% and the NVIDIA Tesla K20m (19,089) by 1.3%.

Q: What does the RX 560X’s nearest rival data tell us about its positioning?

A: The RX 560X is 0.3% below the AMD Radeon Pro 5700 XT (18,685) and 0.5% above the AMD FirePro D500 (18,533). It trails the NVIDIA GeForce RTX 2070 (18,789) by 0.9% and the NVIDIA Tesla K80 (18,866) by 1.3%, indicating it sits in a tight cluster of mid-range performers.

Q: Which card has better memory bandwidth?

A: The GTX 1060 has a 192-bit bus and 192.2 GB/s bandwidth, while the RX 560X has a 128-bit bus and 112.0 GB/s. The GTX 1060’s bandwidth is roughly 71.6% higher, which helps explain its large OpenCL margin.

Q: Is there a difference in power connector requirements?

A: Yes. The GTX 1060 requires a 1x 6-pin power connector and has a 120 W TDP, while the RX 560X has no power connectors and a 75 W TDP. The suggested PSU also differs: 300 W for the GTX 1060 versus 250 W for the RX 560X.

Q: Which card supports a newer Vulkan version?

A: The GTX 1060 supports Vulkan 1.4, while the RX 560X supports Vulkan 1.3. Despite the older API version, the RX 560X wins the Vulkan benchmark, suggesting implementation efficiency outweighs version number.

Architecture Differences

The two cards are built on fundamentally different architectures. The GTX 1060 uses NVIDIA’s Pascal architecture on a GP106 chip, fabricated on TSMC’s 16 nm process. The RX 560X employs AMD’s GCN 4.0 architecture on the Polaris 21 chip, built on GlobalFoundries’ 14 nm node. The transistor counts diverge significantly: NVIDIA packs 4,400 million transistors into a 200 mm² die, yielding a density of 22.0M / mm². AMD fits 3,000 million transistors into a smaller 123 mm² die, achieving a higher density of 24.4M / mm². This means AMD is more efficient at packing transistors, but NVIDIA uses its larger budget to deploy more functional units.

The shading unit counts tell a story of divergent design priorities. The GTX 1060 features 1,152 shading units, 72 TMUs, and 48 ROPs. The RX 560X counters with 1,024 shading units, 64 TMUs, but only 16 ROPs. That ROP disparity is enormous—the GTX 1060 has three times the raster operation units. This is reflected in pixel fill rates: the GTX 1060 achieves 81.98 GPixel/s versus the RX 560X’s 20.40 GPixel/s. For texture work, the GTX 1060’s 123.0 GTexel/s also outpaces the RX 560X’s 81.60 GTexel/s.

FP16 compute reveals a stark architectural philosophy difference. The GTX 1060’s FP16 throughput is a paltry 61.49 GFLOPS, explicitly noted as a 1:64 ratio relative to FP32. The RX 560X runs FP16 at 2.611 TFLOPS, a 1:1 ratio matching its FP32 output. This suggests AMD designed Polaris with more balanced half-precision capability, which could benefit workloads that use FP16 acceleration. The GTX 1060’s Pascal architecture clearly prioritized FP32 throughput at the expense of FP16, a choice that shows up in the Vulkan results where the RX 560X takes the win.

The memory architectures also differ beyond bandwidth. The GTX 1060 uses 3 GB of GDDR5 on a 192-bit bus, while the RX 560X uses 4 GB of GDDR5 on a 128-bit bus. The capacity advantage goes to AMD, but the bandwidth advantage belongs to NVIDIA. Both cards support DirectX 12, but the GTX 1060 implements version 12_1 while the RX 560X implements 12_0, indicating NVIDIA’s feature level is one tier higher. OpenGL support is identical at 4.6, but Vulkan versions differ as noted earlier.

Specification Differences

The most obvious differences lie in process node and foundry: the GTX 1060 uses TSMC’s 16 nm process, while the RX 560X uses GlobalFoundries’ 14 nm node. Clock speeds favor NVIDIA, with the GTX 1060 boosting to 1,708 MHz versus the RX 560X’s 1,275 MHz boost. Base clocks are 1,506 MHz and 1,175 MHz respectively. Memory clocks also differ: the GTX 1060 runs at 2,002 MHz (8 Gbps effective) versus the RX 560X’s 1,750 MHz (7 Gbps effective).

Memory capacity and bus width are clear differentiators: 3 GB on a 192-bit bus for NVIDIA versus 4 GB on a 128-bit bus for AMD. The resultant bandwidth figures are 192.2 GB/s and 112.0 GB/s. TDP is a major split: the GTX 1060 draws 120 W and needs a 6-pin connector, while the RX 560X draws only 75 W with no connector required. The suggested PSU reflects this, with 300 W versus 250 W. Physical dimensions differ too: the GTX 1060 measures 250 mm (9.8 inches) in length, while the RX 560X is a compact 170 mm (6.7 inches). Bus interface also differs, with PCIe 3.0 x16 on the GTX 1060 versus PCIe 3.0 x8 on the RX 560X.

Release dates are separated by roughly 20 months: the GTX 1060 launched on 2016-08-17, while the RX 560X arrived on 2018-04-10. Both are now end-of-life products. The GTX 1060 has a launch MSRP of 199 USD, while the RX 560X has no recorded launch MSRP in the data. The GTX 1060’s predecessor is listed as GeForce 900 and its successor as GeForce 20, while the RX 560X’s predecessor is Polaris and successor is Vega.

The Verdict

The benchmark data suggests the GTX 1060 3 GB is the stronger all-around performer. Its average benchmark score of 19,334 beats the RX 560X’s 18,626, and its OpenCL victory by 58.6% is far more decisive than the RX 560X’s 13.6% Vulkan win. The GTX 1060 also holds a higher percentile rank (64 vs 62). Users who prioritize compute-heavy OpenCL workloads should choose the GTX 1060 without hesitation. The card’s higher clock speeds, larger ROP count, and superior memory bandwidth translate directly into the dominant OpenCL result.

However, the RX 560X is not without merit. Its Vulkan win indicates that for applications built around that API, AMD’s architecture has a tangible edge. The RX 560X also offers more VRAM (4 GB vs 3 GB), a lower TDP (75 W vs 120 W), and requires no external power connector, making it a more flexible option for smaller builds. The data shows it can outperform the GTX 1060 in specific Vulkan scenarios, so users whose workloads are Vulkan-centric may prefer AMD. The GTX 1060’s FP16 limitation (1:64 ratio) could also be a deciding factor for compute tasks that leverage half-precision, where the RX 560X’s 1:1 ratio gives it a theoretical advantage.

Where Each One Wins

The GTX 1060 3 GB wins in OpenCL compute, as evidenced by the 58.6% score gap. It also wins on raw specifications: higher FP32 throughput (3.935 vs 2.611 TFLOPS), higher pixel rate (81.98 vs 20.40 GPixel/s), higher texture rate (123.0 vs 81.60 GTexel/s), and more bandwidth (192.2 vs 112.0 GB/s). For users running OpenCL-based applications, whether for rendering, physics simulation, or data processing, the GTX 1060 is the clear choice. Its higher average score (19,334 vs 18,626) also makes it the safer bet for mixed workloads.

The RX 560X wins in Vulkan, posting a 13.6% higher score. It also wins on power efficiency, with a 75 W TDP versus 120 W, and on VRAM capacity with 4 GB versus 3 GB. Its FP16 capability at 1:1 ratio is a significant advantage for half-precision workloads. The RX 560X’s smaller physical footprint (170 mm vs 250 mm) and lack of a power connector make it more suitable for compact or legacy systems with limited PSU headroom. Users targeting Vulkan-first applications—such as modern game engines that use Vulkan as a primary backend—may find the RX 560X delivers better performance per watt, even if its overall compute ceiling is lower.

DETAILED SPECIFICATIONS

SPECIFICATION
RX 560X
GTX 1060 3 GB
Core Specs
Shading Units
1,024
1,152 +12.5%
Shaders
1,024
1,152 +12.5%
TMUs
64
72 +12.5%
ROPs
16
48 +200.0%
Compute Units
16
—
SM Count
—
9
Clocks
Base Clock
1175 MHz
1506 MHz
Boost Clock
1275 MHz
1708 MHz
Memory Clock
1750 MHz 7 Gbps effective
2002 MHz 8 Gbps effective
Memory
Memory Size
4 GB
3 GB
VRAM (MB)
4,096
3,072 -25.0%
Memory Type
GDDR5
GDDR5
Memory Bus
128 bit
192 bit
Bandwidth
112.0 GB/s
192.2 GB/s
Cache
L1 Cache
16 KB (per CU)
48 KB (per SM)
L2 Cache
1024 KB
1536 KB
Performance
Pixel Rate
20.40 GPixel/s
81.98 GPixel/s
Texture Rate
81.60 GTexel/s
123.0 GTexel/s
FP32 (TFLOPS)
2.611 TFLOPS
3.935 TFLOPS
FP64 (TFLOPS)
163.2 GFLOPS (1:16)
123.0 GFLOPS (1:32)
FP16 (TFLOPS)
2.611 TFLOPS (1:1)
61.49 GFLOPS (1:64)
Power
TDP
75 W
120 W
TDP (W)
75
120 +60.0%
Suggested PSU
250 W
300 W
Power Connectors
None
1x 6-pin
Architecture
Architecture
GCN 4.0
Pascal
GPU Name
Polaris 21
GP106
Generation
Polaris (RX 500X)
GeForce 10
Process Size
14 nm
16 nm
Transistors
3,000 million
4,400 million
Die Size
123 mm²
200 mm²
Foundry
GlobalFoundries
TSMC
Density
24.4M / 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
Dual-slot
Dual-slot
Length
170 mm 6.7 inches
250 mm 9.8 inches
Outputs
1x DVI1x HDMI 2.0b1x DisplayPort 1.4a
1x DVI1x HDMI 2.03x DisplayPort 1.4a
Bus Interface
PCIe 3.0 x8
PCIe 3.0 x16
Other
Launch Price
—
199 USD
Production
End-of-life
End-of-life
Predecessor
Polaris
GeForce 900
Successor
Vega
GeForce 20
View Radeon RX 560X Details View GeForce GTX 1060 3 GB Details