AMD Radeon 890M vs NVIDIA GeForce GTX 1070 Comparison

AMD
RADEON

AMD Radeon 890M

CORE STATE Strix Point
VRAM System Shared
CLOCK SPEED 2900 MHz
TDP 15 W
BUS WIDTH System Shared
ARCHITECTURE RDNA 3.5
nm
PROCESS 4 nm
LAUNCH DATE 2024
VS
NVIDIA
GEFORCE

GeForce GTX 1070

CORE STATE GP104
VRAM 8 GB
CLOCK SPEED 1683 MHz
TDP 150 W
BUS WIDTH 256 bit
ARCHITECTURE Pascal
nm
PROCESS 16 nm
LAUNCH DATE 2016

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
590
1,082
geekbench_opencl
37,254
44,700
geekbench_vulkan
40,808
22,121
passmark_directx_10
33
82
passmark_directx_11
73
100
passmark_directx_12
37
48
passmark_directx_9
97
197
passmark_g2d
979
846
passmark_g3d
8,076
13,498
passmark_gpu_compute
4,157
6,102
geekbench_metal
N/A
18,801

Analysis: AMD Radeon 890M vs NVIDIA GeForce GTX 1070

The NVIDIA GeForce GTX 1070 and AMD Radeon 890M represent two very different approaches to graphics processing: a dedicated desktop GPU from 2016 versus a modern integrated graphics processor from 2024. The benchmark data reveals a clear performance hierarchy, with the GTX 1070 winning 8 of 10 head-to-head tests, but the Radeon 890M shows surprising strength in specific workloads. The average benchmark scores are close—9780 for the GTX 1070 versus 9210 for the Radeon 890M—but the distribution of wins tells a more nuanced story. The GTX 1070's percentile ranking of 47 against all GPUs edges out the Radeon 890M's 45th percentile, yet the newer AMD part delivers a notable victory in Vulkan performance that reshapes the comparison.

Head-to-Head Benchmarks

The most decisive victory for the NVIDIA GeForce GTX 1070 comes in the 3DMark Steel Nomad DX12 test, where it scores 1082 against the AMD Radeon 890M's 590. That is an 83.4% advantage, the largest delta in the entire benchmark suite. This gap indicates that in modern DirectX 12 gaming workloads, the dedicated GTX 1070 is in a different performance class entirely, nearly doubling the frame throughput of the integrated Radeon 890M.

The DirectX legacy tests further cement NVIDIA's dominance. In PassMark DirectX 10, the GTX 1070 scores 82 versus 33 for the Radeon 890M, a 148.5% lead. The DirectX 9 test shows a similar pattern: 197 for the GTX 1070 versus 97 for the Radeon 890M, a 103.1% advantage. Even in DirectX 11, the GTX 1070's score of 100 beats the Radeon 890M's 73 by 37%. The DirectX 12 PassMark result is closer, with the GTX 1070 winning 48 to 37, a 29.7% margin, but still a clear NVIDIA victory.

The synthetic 3D performance metrics reinforce the GTX 1070's lead. In PassMark G3D, the GTX 1070 scores 13498 against 8076 for the Radeon 890M, a 67.1% advantage. The GPU compute test shows the GTX 1070 ahead at 6102 versus 4157, a 46.8% margin. OpenCL performance also favors NVIDIA, with the GTX 1070 scoring 44700 against 37254, a 20% lead.

However, the AMD Radeon 890M secures two notable wins. The most striking is in Geekbench Vulkan, where the Radeon 890M scores 40808 against the GTX 1070's 22121. That is a 45.8% advantage for AMD, a massive reversal of the overall trend. This result suggests that the RDNA 3.5 architecture's Vulkan implementation is exceptionally strong, potentially reflecting better driver optimization for modern API overhead. The Radeon 890M also wins the PassMark G2D test, scoring 979 versus 846, a 13.6% margin, indicating superior 2D desktop rendering performance.

Where Each One Wins

The NVIDIA GeForce GTX 1070 is the clear choice for DirectX-based gaming and compute workloads. Its wins span every DirectX version tested—9, 10, 11, and 12—with margins ranging from 29.7% to 148.5%. The 3DMark Steel Nomad DX12 result, showing an 83.4% lead, positions the GTX 1070 as the stronger option for modern game engines that rely heavily on DirectX 12 features. The GPU compute advantage of 46.8% further indicates that tasks like rendering, physics simulation, or any general-purpose GPU compute will run faster on the GTX 1070.

The AMD Radeon 890M's wins are more specialized but significant. The Vulkan victory is substantial—a 45.8% lead—meaning that applications built on Vulkan, such as certain game engines or emulators, will perform considerably better on the integrated AMD part. The G2D win, while smaller at 13.6%, suggests that everyday desktop usage, 2D applications, and possibly older or lighter workloads will feel more responsive on the Radeon 890M. Given that the Radeon 890M is an integrated processor with a 15 W TDP, its ability to beat a 150 W dedicated GPU in these specific tests is noteworthy.

For pure performance across the board, the GTX 1070 wins 8 out of 10 tests. But the Radeon 890M's architecture brings modern features that the 2016 Pascal part lacks. The Radeon 890M supports DirectX 12 Ultimate (12_2) versus the GTX 1070's DirectX 12 (12_1), and it includes 16 ray tracing cores. These architectural differences do not show up in the benchmark scores provided, but they matter for future software compatibility.

The Verdict

Based strictly on the benchmark data, the NVIDIA GeForce GTX 1070 is the superior performer for the vast majority of workloads. It leads in average benchmark score, 9780 versus 9210, and holds a higher percentile ranking at 47 compared to 45. For any user prioritizing raw frame rates in DirectX-based games or compute-heavy tasks, the GTX 1070 is the clear choice, offering leads of 67.1% in G3D and 46.8% in GPU compute.

The AMD Radeon 890M is not without merit. Its 45.8% Vulkan advantage is substantial and suggests that for Vulkan-centric applications, it is the better option. The G2D win also indicates smoother 2D desktop experiences. However, these two wins do not offset the GTX 1070's dominance elsewhere. The data shows that the GTX 1070 is more than 80% faster in the most demanding 3DMark test and more than double the speed in DirectX 9 and DirectX 10 legacy tests.

The choice ultimately depends on the use case. For a desktop system with a dedicated GPU slot and power budget, the GTX 1070 offers vastly superior gaming and compute performance. For a mobile or low-power integrated system, the Radeon 890M provides competitive performance in specific APIs while consuming 135 W less power. The GTX 1070's launch MSRP was 379 USD. Users needing maximum performance in DirectX titles should choose the GTX 1070; users prioritizing Vulkan performance or 2D responsiveness in a power-constrained form factor should consider the Radeon 890M.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce GTX 1070 has an average benchmark score of 9780, while the AMD Radeon 890M scores 9210, giving NVIDIA a lead of about 6.2%.

Q: How much faster is the GTX 1070 in 3DMark Steel Nomad DX12?

A: The GTX 1070 scores 1082 versus the Radeon 890M's 590, which is an 83.4% advantage for NVIDIA.

Q: In which test does the AMD Radeon 890M beat the GTX 1070 by the largest margin?

A: The Radeon 890M wins Geekbench Vulkan by 45.8%, scoring 40808 against the GTX 1070's 22121.

Q: What are the TDP figures for each GPU?

A: The NVIDIA GeForce GTX 1070 has a TDP of 150 W, while the AMD Radeon 890M has a TDP of 15 W, a 135 W difference.

Q: Which GPU has a higher percentile ranking among all GPUs?

A: The GTX 1070 ranks in the 47th percentile, while the Radeon 890M ranks in the 45th percentile.

Q: How do the two compare in PassMark G2D performance?

A: The Radeon 890M wins the G2D test with a score of 979 versus 846 for the GTX 1070, a 13.6% margin for AMD.

Architecture Differences

The two GPUs are built on fundamentally different architectures and process nodes. The NVIDIA GeForce GTX 1070 uses the GP104 chip on the Pascal architecture, manufactured on a 16 nm process at TSMC. It contains 7,200 million transistors on a 314 mm² die, yielding a transistor density of 22.9 million per mm². In contrast, the AMD Radeon 890M uses the Strix Point chip on the RDNA 3.5 architecture, built on a 4 nm process, also at TSMC. It packs 34,000 million transistors into a 233 mm² die, achieving a much higher density of 145.9 million per mm².

The GTX 1070 features 1920 shading units, 120 texture mapping units, and 64 ROPs. Its pixel rate is 107.7 GPixel/s, and its texture rate is 202.0 GTexel/s. The FP32 performance is 6.463 TFLOPS. The Radeon 890M has 1024 shading units, 64 TMUs, and 32 ROPs, along with 16 ray tracing cores. Its pixel rate is 92.80 GPixel/s, and its texture rate is 185.6 GTexel/s. FP32 performance is 5.939 TFLOPS. Notably, the Radeon 890M achieves FP16 performance equal to its FP32 at 5.939 TFLOPS (1:1 ratio), while the GTX 1070's FP16 performance is only 101.0 GFLOPS (1:64 ratio).

Memory configurations differ dramatically. The GTX 1070 uses 8 GB of GDDR5 memory on a 256-bit bus, providing 256.3 GB/s of bandwidth. The Radeon 890M uses system shared memory, with bus width and bandwidth being system dependent. Clock speeds also differ, with the GTX 1070 running at 1506 MHz base and 1683 MHz boost, while the Radeon 890M runs at 400 MHz base and 2900 MHz boost.

API support shows the generational gap. The GTX 1070 supports DirectX 12 (12_1), while the Radeon 890M supports DirectX 12 Ultimate (12_2). Both support OpenGL 4.6 and Vulkan 1.4. The GTX 1070 uses a PCIe 3.0 x16 interface, while the Radeon 890M uses PCIe 4.0 x8. The GTX 1070 is a dual-slot card requiring a 1x 8-pin power connector and a 450 W suggested PSU, while the Radeon 890M is an IGP with no power connectors. The GTX 1070 is end-of-life, released in 2016, while the Radeon 890M is active, released in 2024.

DETAILED SPECIFICATIONS

SPECIFICATION
890M
GTX 1070
Core Specs
Shading Units
1,024
1,920 +87.5%
Shaders
1,024
1,920 +87.5%
TMUs
64
120 +87.5%
ROPs
32
64 +100.0%
Compute Units
16
SM Count
15
Clocks
Base Clock
400 MHz
1506 MHz
Boost Clock
2900 MHz
1683 MHz
Memory Clock
System Shared
2002 MHz 8 Gbps effective
Memory
Memory Size
System Shared
8 GB
VRAM (MB)
8,192
Memory Type
System Shared
GDDR5
Memory Bus
System Shared
256 bit
Bandwidth
System Dependent
256.3 GB/s
Cache
L1 Cache
128 KB per Array
48 KB (per SM)
L2 Cache
2 MB
2 MB
L0 Cache
32 KB per WGP
Performance
Pixel Rate
92.80 GPixel/s
107.7 GPixel/s
Texture Rate
185.6 GTexel/s
202.0 GTexel/s
FP32 (TFLOPS)
5.939 TFLOPS
6.463 TFLOPS
FP64 (TFLOPS)
371.2 GFLOPS (1:16)
202.0 GFLOPS (1:32)
FP16 (TFLOPS)
5.939 TFLOPS (1:1)
101.0 GFLOPS (1:64)
AI/RT
RT Cores
16
Power
TDP
15 W
150 W
TDP (W)
15
150 +900.0%
Suggested PSU
450 W
Power Connectors
None
1x 8-pin
Architecture
Architecture
RDNA 3.5
Pascal
GPU Name
Strix Point
GP104
Generation
Navi III IGP (Strix Point Mobile)
GeForce 10
Process Size
4 nm
16 nm
Transistors
34,000 million
7,200 million
Die Size
233 mm²
314 mm²
Foundry
TSMC
TSMC
Density
145.9M / mm²
22.9M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 (12_1)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
2.1
3.0
CUDA
6.1
Shader Model
6.8
6.8
Physical
Slot Width
IGP
Dual-slot
Length
267 mm 10.5 inches
Height
112 mm 4.4 inches
Outputs
Portable Device Dependent
1x DVI1x HDMI 2.03x DisplayPort 1.4a
Bus Interface
PCIe 4.0 x8
PCIe 3.0 x16
Other
Launch Price
379 USD
Production
Active
End-of-life
Predecessor
Navi II IGP
GeForce 900
Successor
GeForce 20
View Radeon 890M Details View GeForce GTX 1070 Details