AMD Radeon 890M vs NVIDIA GeForce 945M 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 945M

CORE STATE GM107
VRAM 2 GB
CLOCK SPEED 1020 MHz
TDP 75 W
BUS WIDTH 128 bit
ARCHITECTURE Maxwell
nm
PROCESS 28 nm
LAUNCH DATE 2015

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
590
N/A
geekbench_opencl
37,254
8,099
geekbench_vulkan
40,808
N/A
passmark_directx_10
33
N/A
passmark_directx_11
73
N/A
passmark_directx_12
37
N/A
passmark_directx_9
97
N/A
passmark_g2d
979
N/A
passmark_g3d
8,076
N/A
passmark_gpu_compute
4,157
N/A

Analysis: AMD Radeon 890M vs NVIDIA GeForce 945M

Head-to-Head Benchmarks

The only directly comparable benchmark recorded in the database for both products is Geekbench OpenCL, and the result is decisive. The AMD Radeon 890M scores 37,254, while the NVIDIA GeForce 945M scores 8,099. That gives the AMD part a 360% advantage in this specific compute-oriented test. To frame it another way, the 890M delivers more than four and a half times the raw OpenCL throughput of the 945M.

Context from the nearest rivals list reinforces how lopsided this head-to-head is. The 890M's average benchmark score across all recorded tests is 9,210, which places it within 0.1% of the AMD Radeon Vega 8 (9,221) and 0.7% behind the NVIDIA GeForce GTX 960 (9,273). The 945M, by contrast, has an average score of 8,099, putting it 0.2% ahead of the NVIDIA GRID K2 (8,080) and 0.4% behind the NVIDIA GeForce GTX 950M (8,135). The gap between the two averages is roughly 1,100 points, or about 13.7% in favor of the 890M, but the OpenCL delta of 360% dwarfs that overall average difference. This suggests the 890M's architecture is disproportionately strong in compute workloads relative to its overall graphics standing.

The 890M also shows a wider benchmark spread in its favor when looking at individual tests. Its Passmark G3D score is 8,076, its Passmark G2D score is 979, and its Passmark GPU Compute score is 4,157. The 945M has no corresponding Passmark entries in the database, so no direct comparison is possible there. The 890M's 3DMark Steel Nomad DX12 score of 590 and its Geekbench Vulkan score of 40,808 further illustrate its compute and modern API capabilities, while the 945M lacks recorded scores for those tests entirely.

Where Each One Wins

The AMD Radeon 890M wins the only benchmark where both products have recorded results, so the use-case split is heavily one-sided. The 890M's Geekbench OpenCL score of 37,254 versus the 945M's 8,099 indicates that the 890M is the clear choice for OpenCL-accelerated applications, which includes many compute tasks, physics simulations, and certain productivity workloads that offload work to the GPU.

The 890M also holds advantages in areas where the 945M has no recorded data at all. Its Passmark G3D score of 8,076 and Passmark GPU Compute score of 4,157 suggest strong general graphics and compute performance relative to its average score of 9,210. The 3DMark Steel Nomad DX12 score of 590 and Geekbench Vulkan score of 40,808 point to modern API readiness, especially DirectX 12 Ultimate and Vulkan 1.4 support. For gaming and applications built around these newer APIs, the 890M has both the architectural support and the recorded scores to back it up.

The NVIDIA GeForce 945M, with only a single Geekbench OpenCL score in the database, has no benchmark wins in this comparison. Its strengths would have to be inferred from its specification profile rather than measured results. The 945M does have a dedicated 2 GB DDR3 memory configuration with a 128 bit bus and 28.80 GB/s bandwidth, which is a fixed allocation rather than system shared, but the recorded data does not show this translating into any benchmark victory over the 890M. The 945M sits at the 42nd percentile of all GPUs, while the 890M sits at the 45th, a modest but consistent edge for the AMD part.

Architecture Differences

The two GPUs come from different architectural generations and manufacturing processes. The AMD Radeon 890M uses the Strix Point chip built on RDNA 3.5 architecture at TSMC's 4 nm process. The NVIDIA GeForce 945M uses the GM107 chip built on Maxwell architecture at TSMC's 28 nm process. That process gap is enormous: 4 nm versus 28 nm, which helps explain the transistor density disparity. The 890M packs 34,000 million transistors on a 233 mm² die, for a density of 145.9 million transistors per mm². The 945M has 1,870 million transistors on a 148 mm² die, or 12.6 million per mm². The 890M has roughly 18 times the transistor density of the 945M.

The 890M's RDNA 3.5 architecture brings hardware ray tracing support with 16 RT cores, which the Maxwell-based 945M lacks entirely. The 890M also supports DirectX 12 Ultimate (12_2), while the 945M only reaches DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4, so those API levels match. The 890M has 1,024 shading units, 64 texture mapping units, and 32 render output units. The 945M has 640 shading units, 40 TMUs, and 16 ROPs. The 890M roughly doubles the 945M in TMUs and ROPs, and leads by 384 shading units.

Clock speeds tell a different story. The 945M has a base clock of 928 MHz and a boost clock of 1,020 MHz, while the 890M has a base clock of 400 MHz and a boost clock of 2,900 MHz. The 890M's boost clock is nearly three times higher than the 945M's, which combined with the larger shader count drives its 5.939 TFLOPS FP32 throughput versus the 945M's 1,305.6 GFLOPS. Pixel rate follows suit: 92.80 GPixel/s for the 890M versus 16.32 GPixel/s for the 945M. Texture rate is 185.6 GTexel/s versus 40.80 GTexel/s.

Memory architecture also diverges sharply. The 890M uses system shared memory with system dependent bandwidth, while the 945M has 2 GB of dedicated DDR3 on a 128 bit bus delivering 28.80 GB/s. The 890M's memory clock is listed as "System Shared," meaning it borrows from the host system's RAM. The 945M's memory runs at 900 MHz with 1800 Mbps effective. The 890M has no power connectors and is an IGP, while the 945M is an MXM module with a bus interface of MXM-B (3.0). The 890M uses PCIe 4.0 x8. The 890M's TDP is 15 W, while the 945M's is 75 W, a fivefold difference in power draw that makes the 890M's performance lead even more notable.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon 890M has an average benchmark score of 9,210, while the NVIDIA GeForce 945M has an average of 8,099. The 890M also sits at the 45th percentile of all GPUs, compared to the 945M's 42nd percentile.

Q: How much faster is the AMD Radeon 890M in OpenCL?

A: The 890M scores 37,254 in Geekbench OpenCL versus 8,099 for the 945M, a 360% advantage for the AMD part.

Q: Does the NVIDIA GeForce 945M support ray tracing?

A: No. The 945M has no RT cores listed, while the AMD Radeon 890M has 16 RT cores. The 890M also supports DirectX 12 Ultimate (12_2), while the 945M only supports DirectX 12 (11_0).

Q: What are the memory differences between the two?

A: The 890M uses system shared memory with system dependent bandwidth, while the 945M has 2 GB of dedicated DDR3 on a 128 bit bus with 28.80 GB/s bandwidth.

Q: Which GPU is more power efficient?

A: The 890M has a TDP of 15 W, while the 945M has a TDP of 75 W. Despite the much lower power draw, the 890M delivers significantly higher compute throughput in the recorded benchmarks.

Q: Are both GPUs still in production?

A: No. The 890M is listed as Active, while the 945M is End-of-life. The 945M's predecessor is the GeForce 800M and its successor is GeForce 10 Mobile. The 890M's predecessor is Navi II IGP.

Specification Differences

The two GPUs differ across nearly every specification category. The 890M uses a 4 nm process at TSMC, while the 945M uses 28 nm at the same foundry. The 890M has 34,000 million transistors on a 233 mm² die, while the 945M has 1,870 million on 148 mm². Transistor density is 145.9M per mm² versus 12.6M per mm². The 890M's base clock is 400 MHz and boost is 2,900 MHz, while the 945M's base is 928 MHz and boost is 1,020 MHz. The 890M has 1,024 shading units, 64 TMUs, 32 ROPs, and 16 RT cores. The 945M has 640 shading units, 40 TMUs, and 16 ROPs, with no RT cores. Pixel rate is 92.80 GPixel/s versus 16.32 GPixel/s. Texture rate is 185.6 GTexel/s versus 40.80 GTexel/s. FP32 throughput is 5.939 TFLOPS versus 1,305.6 GFLOPS. The 890M supports FP16 at 5.939 TFLOPS (1:1), while the 945M has no listed FP16 capability. TDP is 15 W versus 75 W. The 890M is an IGP with PCIe 4.0 x8, while the 945M is an MXM module with MXM-B (3.0). The 890M supports DirectX 12 Ultimate (12_2), while the 945M supports DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4. The 890M was released on 2024-07-14, while the 945M was released on 2015-10-26. The 890M is Active, the 945M is End-of-life.

The Verdict

The recorded data points to a clear winner for compute-oriented workloads. The AMD Radeon 890M beats the NVIDIA GeForce 945M by 360% in Geekbench OpenCL, the only benchmark where both have scores. The 890M also has a higher average benchmark score (9,210 versus 8,099), a higher percentile ranking (45th versus 42nd), and a broader set of recorded benchmark results across modern APIs like DirectX 12 Ultimate, Vulkan, and 3DMark Steel Nomad.

For users prioritizing raw compute performance, modern API support, ray tracing capability, and power efficiency, the 890M is the data-backed choice. Its 15 W TDP versus the 945M's 75 W, combined with its higher performance, makes the power efficiency argument one-sided. The 945M's only advantage in the specifications is its dedicated 2 GB DDR3 memory with 28.80 GB/s bandwidth, which provides a fixed memory allocation rather than relying on system shared memory. However, no recorded benchmark shows this translating into a performance win over the 890M.

The 945M is also an end-of-life product released in 2015, while the 890M is active and recent. For anyone choosing between these two based on the database records, the AMD Radeon 890M is the superior option across every measured dimension.

DETAILED SPECIFICATIONS

SPECIFICATION
890M
945M
Core Specs
Shading Units
1,024
640 -37.5%
Shaders
1,024
640 -37.5%
TMUs
64
40 -37.5%
ROPs
32
16 -50.0%
Compute Units
16
Clocks
Base Clock
400 MHz
928 MHz
Boost Clock
2900 MHz
1020 MHz
Memory Clock
System Shared
900 MHz 1800 Mbps effective
Memory
Memory Size
System Shared
2 GB
VRAM (MB)
2,048
Memory Type
System Shared
DDR3
Memory Bus
System Shared
128 bit
Bandwidth
System Dependent
28.80 GB/s
Cache
L1 Cache
128 KB per Array
64 KB (per SMM)
L2 Cache
2 MB
2 MB
L0 Cache
32 KB per WGP
Performance
Pixel Rate
92.80 GPixel/s
16.32 GPixel/s
Texture Rate
185.6 GTexel/s
40.80 GTexel/s
FP32 (TFLOPS)
5.939 TFLOPS
1,305.6 GFLOPS
FP64 (TFLOPS)
371.2 GFLOPS (1:16)
40.80 GFLOPS (1:32)
FP16 (TFLOPS)
5.939 TFLOPS (1:1)
AI/RT
RT Cores
16
Power
TDP
15 W
75 W
TDP (W)
15
75 +400.0%
Power Connectors
None
None
Architecture
Architecture
RDNA 3.5
Maxwell
GPU Name
Strix Point
GM107
Generation
Navi III IGP (Strix Point Mobile)
GeForce 900M
Process Size
4 nm
28 nm
Transistors
34,000 million
1,870 million
Die Size
233 mm²
148 mm²
Foundry
TSMC
TSMC
Density
145.9M / mm²
12.6M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
2.1
3.0
CUDA
5.0
Shader Model
6.8
6.7 (5.1)
Physical
Slot Width
IGP
MXM Module
Outputs
Portable Device Dependent
Portable Device Dependent
Bus Interface
PCIe 4.0 x8
MXM-B (3.0)
Other
Production
Active
End-of-life
Predecessor
Navi II IGP
GeForce 800M
Successor
GeForce 10 Mobile
View Radeon 890M Details View GeForce 945M Details