AMD Radeon RX 5700 vs NVIDIA GeForce RTX 2070 Comparison

AMD
RADEON

AMD Radeon RX 5700

CORE STATE Navi 10
VRAM 8 GB
CLOCK SPEED 1725 MHz
TDP 180 W
BUS WIDTH 256 bit
ARCHITECTURE RDNA 1.0
nm
PROCESS 7 nm
LAUNCH DATE 2019
VS
NVIDIA
GEFORCE

GeForce RTX 2070

CORE STATE TU106
VRAM 8 GB
CLOCK SPEED 1620 MHz
TDP 175 W
BUS WIDTH 256 bit
ARCHITECTURE Turing
nm
PROCESS 12 nm
LAUNCH DATE 2018

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
1,862
1,569
geekbench_metal
86,079
N/A
geekbench_opencl
69,603
79,966
geekbench_vulkan
64,166
82,521
passmark_directx_10
87
111
passmark_directx_11
101
129
passmark_directx_12
54
60
passmark_directx_9
209
211
passmark_g2d
877
819
passmark_g3d
14,314
16,094
passmark_gpu_compute
6,517
6,411

Analysis: AMD Radeon RX 5700 vs NVIDIA GeForce RTX 2070

Head-to-Head Benchmarks

The head-to-head data shows a clear split between the two cards, with the NVIDIA GeForce RTX 2070 winning 7 of the 10 recorded tests, while the AMD Radeon RX 5700 takes 3 wins. However, the magnitude of the victories matters more than the raw count.

The AMD Radeon RX 5700's most decisive win comes in the 3DMark Steel Nomad DX12 test, where it scores 1862 against the RTX 2070's 1569, a 18.7% advantage. This is the largest delta in either direction across all benchmarks, and it signals that in modern DirectX 12 workloads, the RX 5700 holds a substantial edge. The RX 5700 also wins the Passmark G2D test, scoring 877 versus 819, a 7.1% lead, which points to better 2D and desktop composition performance. Its third win is in Passmark GPU Compute, where it scores 6517 against 6411, a slim 1.7% margin, indicating a slight advantage in raw compute throughput.

The NVIDIA GeForce RTX 2070 counters with several wins, but the margins are generally moderate. The largest is in Geekbench Vulkan, where the RTX 2070 scores 82521 versus the RX 5700's 64166, a 22.2% lead. This is the single biggest gap in either direction, and it suggests the RTX 2070 is significantly stronger in Vulkan API workloads. The RTX 2070 also wins Geekbench OpenCL, scoring 79966 against 69603, a 13% advantage. In Passmark DirectX 11, the RTX 2070 scores 129 versus 101, a 21.7% lead, and in Passmark DirectX 10 it scores 111 versus 87, a 21.6% lead. The Passmark DirectX 12 test shows a smaller gap, with the RTX 2070 at 60 versus 54, a 10% lead. The Passmark G3D test, which is often used as a proxy for overall gaming performance, favors the RTX 2070 with a score of 16094 against 14314, an 11.1% advantage. The closest contest is in Passmark DirectX 9, where the RTX 2070 wins by a hair, 211 versus 209, a 0.9% margin.

The data shows a pattern: the RX 5700 excels in the modern DirectX 12 Steel Nomad test and in compute-style workloads, while the RTX 2070 dominates in Vulkan, OpenCL, and the legacy DirectX 10 and 11 tests. The Passmark G3D result is particularly relevant for gaming, as it gives the RTX 2070 a clear lead. The average benchmark scores in the database confirm this: the RX 5700's average is 22170, while the RTX 2070's average is 18789, a gap driven by the RX 5700's strong Steel Nomad and compute results. However, the RTX 2070's percentile ranking of 63 against all GPUs is slightly below the RX 5700's 67, suggesting the RX 5700 sits in a higher overall tier despite losing the majority of head-to-head tests.

FAQ

Q: Which card wins the most head-to-head benchmarks?

A: The NVIDIA GeForce RTX 2070 wins 7 of the 10 recorded head-to-head tests, while the AMD Radeon RX 5700 wins 3.

Q: What is the largest performance gap in either direction?

A: The largest gap is in Geekbench Vulkan, where the RTX 2070 leads by 22.2%, scoring 82521 versus the RX 5700's 64166. The second-largest is in 3DMark Steel Nomad DX12, where the RX 5700 leads by 18.7%.

Q: How do the two cards compare in DirectX 12 performance?

A: The results are mixed. In 3DMark Steel Nomad DX12, the RX 5700 leads by 18.7%, but in Passmark DirectX 12, the RTX 2070 leads by 10%. This suggests the RX 5700 is stronger in newer, more demanding DX12 workloads, while the RTX 2070 holds a smaller edge in the Passmark DX12 test.

Q: Which card has a higher average benchmark score?

A: The AMD Radeon RX 5700 has a higher average benchmark score of 22170, compared to the RTX 2070's 18789. This is largely due to the RX 5700's strong performance in the Steel Nomad and compute tests.

Q: What is the difference in overall GPU percentile rankings?

A: The RX 5700 ranks in the 67th percentile against all GPUs, while the RTX 2070 ranks in the 63rd percentile.

Q: In which specific test does the RX 5700 win by the smallest margin?

A: The RX 5700 wins Passmark GPU Compute by just 1.7%, scoring 6517 versus 6411. The RTX 2070's smallest win is in Passmark DirectX 9, where it leads by 0.9%.

Architecture Differences

The two cards are built on fundamentally different architectures. The AMD Radeon RX 5700 uses the Navi 10 chip with RDNA 1.0 architecture, manufactured on a 7 nm process at TSMC. The NVIDIA GeForce RTX 2070 uses the TU106 chip with Turing architecture, also from TSMC but on a 12 nm process. This process difference is significant: the RX 5700's 7 nm node allows for a much smaller die size of 251 mm², while the RTX 2070's die is 445 mm². The transistor counts are similar, with the RX 5700 at 10,300 million and the RTX 2070 at 10,800 million, but the transistor density tells the story: the RX 5700 packs 41.0M transistors per mm², versus 24.3M for the RTX 2070.

The architectural feature sets diverge sharply. The RTX 2070 includes 36 ray tracing cores and 288 tensor cores, which are entirely absent from the RX 5700. This means the RTX 2070 has dedicated hardware for ray-traced effects and AI-accelerated workloads, while the RX 5700 relies on standard shader units for all tasks. Both cards have 2304 shading units, 144 texture mapping units, and 64 ROPs, so the base shader configuration is identical. However, the RTX 2070's pixel rate is 103.7 GPixel/s and its texture rate is 233.3 GTexel/s, while the RX 5700 achieves 110.4 GPixel/s and 248.4 GTexel/s, giving the RX 5700 slight advantages in both rasterization throughput metrics.

The floating-point performance also differs. The RX 5700 delivers 7.949 TFLOPS in FP32 and 15.90 TFLOPS in FP16 (2:1 ratio), while the RTX 2070 delivers 7.465 TFLOPS in FP32 and 14.93 TFLOPS in FP16 (2:1 ratio). The RX 5700's higher FP32 and FP16 figures align with its wins in the compute-oriented Passmark GPU Compute test. The RTX 2070 supports DirectX 12 Ultimate (12_2), while the RX 5700 is limited to DirectX 12 (12_1), a difference that matters for feature-level support in newer games. The RX 5700 uses a PCIe 4.0 x16 interface, while the RTX 2070 uses PCIe 3.0 x16, which gives the RX 5700 double the theoretical bandwidth to the host system.

Specification Differences

The specification sheets show several key differences beyond the architecture. The process node differs: the RX 5700 uses 7 nm while the RTX 2070 uses 12 nm. The die size is 251 mm² for the RX 5700 versus 445 mm² for the RTX 2070. The clock speeds differ, with the RX 5700 having a base clock of 1465 MHz and a boost clock of 1725 MHz, while the RTX 2070 has a base clock of 1410 MHz and a boost clock of 1620 MHz. The RX 5700 also has an explicit game clock of 1625 MHz, a specification the RTX 2070 does not list.

The thermal design power (TDP) is close, with the RX 5700 rated at 180 W and the RTX 2070 at 175 W. The power connectors differ: the RX 5700 requires 1x 6-pin plus 1x 8-pin, while the RTX 2070 requires only 1x 8-pin. Both cards suggest a 450 W power supply. The physical dimensions vary, with the RX 5700 being longer at 268 mm (10.6 inches) versus 229 mm (9 inches) for the RTX 2070. The heights are nearly identical at 111 mm (4.4 inches) for the RX 5700 and 113 mm (4.4 inches) for the RTX 2070, and the widths are 36 mm (1.4 inches) versus 35 mm (1.4 inches).

The display outputs differ as well. The RX 5700 offers 1x HDMI 2.0b and 3x DisplayPort 1.4a, while the RTX 2070 offers 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C. The memory configuration is identical: both have 8 GB of GDDR6 with a 256-bit bus and 448.0 GB/s bandwidth, and both run at 1750 MHz (14 Gbps effective). The release dates differ, with the RTX 2070 launching on 2018-10-16 and the RX 5700 on 2019-07-06. The RTX 2070's predecessor is GeForce 10 and its successor is GeForce 30, while the RX 5700's predecessor is Vega and its successor is Navi II.

Where Each One Wins

The benchmark results point to distinct use cases for each card. The AMD Radeon RX 5700 wins in scenarios that favor modern DirectX 12 implementations and raw compute throughput. Its 18.7% lead in 3DMark Steel Nomad DX12 is the standout result, indicating that newer games built on advanced DX12 features will likely run better on the RX 5700. The Passmark GPU Compute win, albeit narrow at 1.7%, reinforces the RX 5700's strength in compute-heavy tasks such as rendering or scientific workloads. The Passmark G2D win of 7.1% suggests the RX 5700 is also better suited for 2D applications and desktop environments.

The NVIDIA GeForce RTX 2070 wins in Vulkan and OpenCL, with leads of 22.2% and 13% respectively, making it the better choice for workloads that leverage these APIs. The Vulkan advantage is particularly relevant for Linux gaming and for games that use Vulkan as their primary rendering path. The RTX 2070 also dominates the legacy DirectX 10 and 11 tests, with leads of 21.6% and 21.7%, which matters for older game titles that have not been updated to newer APIs. The Passmark G3D win of 11.1% is the most direct evidence of overall 3D gaming superiority, and the Passmark DirectX 12 win of 10% adds to this picture. The RTX 2070's ray tracing cores also provide a qualitative advantage for games that support ray-traced effects, though the database's benchmark scores do not directly measure this capability.

The Verdict

The data describes two cards with complementary strengths. The AMD Radeon RX 5700 is the pick for users focused on modern DirectX 12 titles and compute-heavy applications. Its 18.7% lead in 3DMark Steel Nomad DX12 is a strong signal for future-proofing in newer games, and its higher average benchmark score of 22170 against 18789 places it in a higher overall tier, reflected in its 67th percentile versus the RTX 2070's 63rd. The RX 5700 also has the advantage of a 7 nm process, a smaller die, a PCIe 4.0 interface, and higher pixel and texture rates.

The NVIDIA GeForce RTX 2070 is the better choice for users who prioritize Vulkan performance, legacy DirectX compatibility, and overall 3D gaming as measured by Passmark G3D. Its 22.2% Vulkan lead and 11.1% G3D lead are significant, and it also offers dedicated ray tracing and tensor cores, which the RX 5700 lacks. The RTX 2070 wins 7 of 10 head-to-head tests, making it the more versatile card across a broader range of benchmarks.

For a user deciding between the two, the choice hinges on the workload. If the primary use is modern DX12 gaming and compute, the RX 5700's Steel Nomad and compute wins make it the data-driven recommendation. If the use case spans Vulkan, older DirectX titles, and general 3D gaming, the RTX 2070's multiple wins and feature set make it the safer pick. The RTX 2070's launch MSRP was 499 USD, while the RX 5700's launch MSRP was 349 USD, a difference that may factor into any decision. Both cards are end-of-life products, but the recorded data shows they remain competitive in different arenas.

DETAILED SPECIFICATIONS

SPECIFICATION
RX 5700
RTX 2070
Core Specs
Shading Units
2,304
2,304 0.0%
Shaders
2,304
2,304 0.0%
TMUs
144
144 0.0%
ROPs
64
64 0.0%
Compute Units
36
SM Count
36
Clocks
Base Clock
1465 MHz
1410 MHz
Boost Clock
1725 MHz
1620 MHz
Game Clock
1625 MHz
Memory Clock
1750 MHz 14 Gbps effective
1750 MHz 14 Gbps effective
Memory
Memory Size
8 GB
8 GB
VRAM (MB)
8,192
8,192 0.0%
Memory Type
GDDR6
GDDR6
Memory Bus
256 bit
256 bit
Bandwidth
448.0 GB/s
448.0 GB/s
Cache
L1 Cache
64 KB (per SM)
L2 Cache
4 MB
4 MB
Performance
Pixel Rate
110.4 GPixel/s
103.7 GPixel/s
Texture Rate
248.4 GTexel/s
233.3 GTexel/s
FP32 (TFLOPS)
7.949 TFLOPS
7.465 TFLOPS
FP64 (TFLOPS)
496.8 GFLOPS (1:16)
233.3 GFLOPS (1:32)
FP16 (TFLOPS)
15.90 TFLOPS (2:1)
14.93 TFLOPS (2:1)
AI/RT
RT Cores
36
Tensor Cores
288
Power
TDP
180 W
175 W
TDP (W)
180
175 -2.8%
Suggested PSU
450 W
450 W
Power Connectors
1x 6-pin + 1x 8-pin
1x 8-pin
Architecture
Architecture
RDNA 1.0
Turing
GPU Name
Navi 10
TU106
Generation
Navi (RX 5000)
GeForce 20
Process Size
7 nm
12 nm
Transistors
10,300 million
10,800 million
Die Size
251 mm²
445 mm²
Foundry
TSMC
TSMC
Density
41.0M / mm²
24.3M / mm²
API Support
DirectX
12 (12_1)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
2.1
3.0
CUDA
7.5
Shader Model
6.8
6.8
Physical
Slot Width
Dual-slot
Dual-slot
Length
268 mm 10.6 inches
229 mm 9 inches
Height
111 mm 4.4 inches
113 mm 4.4 inches
Outputs
1x HDMI 2.0b3x DisplayPort 1.4a
1x DVI1x HDMI 2.02x DisplayPort 1.4a1x USB Type-C
Bus Interface
PCIe 4.0 x16
PCIe 3.0 x16
Other
Launch Price
349 USD
499 USD
Production
End-of-life
End-of-life
Predecessor
Vega
GeForce 10
Successor
Navi II
GeForce 30
View Radeon RX 5700 Details View GeForce RTX 2070 Details