GPU Comparison

Head-to-Head Benchmarks

The AMD Radeon 890M and AMD FirePro W5170M share only two benchmark results in the data, but those two results tell a decisive story. In Geekbench OpenCL, the Radeon 890M scores 37,254 against the FirePro W5170M's 8,140. That is a 357.7% delta, meaning the 890M delivers nearly four and a half times the compute throughput in this test. The Geekbench Vulkan result is similarly lopsided: 40,808 for the 890M versus 9,050 for the W5170M, a 350.9% advantage. There is no benchmark in the shared set where the FirePro W5170M wins; the head-to-head tally stands at 2 wins for the Radeon 890M and 0 for the FirePro W5170M.

Looking at the broader benchmark picture, the Radeon 890M's average benchmark score is 9,210, placing it at the 45th percentile of all GPUs. The FirePro W5170M's average score is 8,595, at the 44th percentile. The difference in average score is only 615 points, a gap of roughly 7%, but the raw Geekbench deltas are enormous, the average score includes many other tests where the FirePro has no data, so the comparison is not apples-to-apples across the full suite. The Radeon 890M's nearest rivals include the AMD Radeon Vega 8 (average score 9,221, delta -0.1%), the NVIDIA GeForce GTX 960 (9,273, -0.7%), and the NVIDIA GeForce GTX 465 (9,294, -0.9%). The 890M sits essentially level with Vega 8 and within 1% of those older discrete cards. The FirePro W5170M's nearest rivals are the Intel Arc A380 (8,558, delta 0.4%), the NVIDIA Quadro P2200 (8,686, -1.1%), and the AMD Radeon HD 8870M (8,462, 1.6%). The FirePro is competitive with that group, within about 1.6% either way, but it is a full tier below the 890M in raw compute.

The Radeon 890M also has a much richer benchmark profile. It has results for Passmark DirectX 9 (97), DirectX 10 (33), DirectX 11 (73), DirectX 12 (37), G2D (979), G3D (8,076), and GPU Compute (4,157). The FirePro W5170M has no Passmark entries at all. This means the 890M's average score is built on ten tests, while the FirePro's is built on just two, so the 890M's percentile ranking is more robust.

FAQ

Q: How much faster is the Radeon 890M than the FirePro W5170M in OpenCL?

A: The Radeon 890M scores 37,254 in Geekbench OpenCL versus 8,140 for the FirePro W5170M. That is a 357.7% delta, meaning the 890M is roughly 4.5x faster in this workload.

Q: Does the FirePro W5170M win any benchmark against the Radeon 890M?

A: No. In the only two shared tests, Geekbench OpenCL and Geekbench Vulkan, the Radeon 890M wins both. The head-to-head record is 2 wins for the 890M and 0 for the W5170M.

Q: Which GPU has the higher average benchmark score?

A: The Radeon 890M has an average benchmark score of 9,210, compared to 8,595 for the FirePro W5170M. The 890M also sits at the 45th percentile of all GPUs, one point higher than the W5170M's 44th percentile.

Q: How does the Radeon 890M compare to its nearest rivals?

A: The 890M is nearly identical to the AMD Radeon Vega 8, which scores 9,221 (a -0.1% delta). It is also within 1% of the NVIDIA GeForce GTX 960 (9,273, -0.7%) and the NVIDIA GeForce GTX 465 (9,294, -0.9%).

Q: What benchmarks does the FirePro W5170M have data for?

A: Only two: Geekbench OpenCL (8,140) and Geekbench Vulkan (9,050). It has no Passmark results, no 3DMark results, and no other compute benchmarks in the data.

Q: Is the FirePro W5170M competitive with modern integrated graphics?

A: Based on its nearest rivals, the W5170M is comparable to the Intel Arc A380 (8,558, delta 0.4%) and the NVIDIA GeForce MX330 (8,458, 1.6%). But it is far behind the Radeon 890M, which beats it by over 350% in both shared tests.

Architecture Differences

The two GPUs come from entirely different architectural eras. The Radeon 890M uses the RDNA 3.5 architecture on a 4 nm TSMC process, with a chip codenamed Strix Point. The FirePro W5170M uses GCN 1.0 on a 28 nm TSMC process, with a chip codenamed Tropo. The process node difference alone, 4 nm versus 28 nm, explains much of the efficiency and performance gap, though the data does not include direct power consumption figures for the W5170M.

Transistor counts are starkly different: the Radeon 890M packs 34,000 million transistors on a 233 mm² die, giving a transistor density of 145.9 million per square millimeter. The FirePro W5170M has 1,500 million transistors on a 123 mm² die, with a density of only 12.2 million per square millimeter. That is a density difference of over 12x, reflecting the jump from a mature planar process to a modern FinFET node.

The Radeon 890M is an integrated GPU (IGP) with "System Shared" memory, meaning it has no dedicated VRAM and relies on the host system's memory bandwidth. The FirePro W5170M is a discrete mobile GPU in an MXM module form factor, with 2 GB of GDDR5 memory on a 128-bit bus delivering 72.00 GB/s of bandwidth. The 890M's memory bandwidth is listed as "System Dependent," so no direct comparison is possible from the data.

Feature support also diverges. The Radeon 890M has 16 dedicated ray tracing cores, while the FirePro W5170M has none (null in the data). The 890M supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the W5170M tops out at DirectX 12 (11_1) and Vulkan 1.2.170. Both support OpenGL 4.6, but the 890M's API level is a full generation ahead.

The Radeon 890M is the newer product by a wide margin: it was released in July 2024, while the FirePro W5170M dates to August 2014, a ten-year gap. The 890M's predecessor is the Navi II IGP, and it remains in active production. The W5170M is end-of-life, with the Radeon Pro Mobile as its successor.

Specification Differences

| Specification | AMD Radeon 890M | AMD FirePro W5170M |

|---|---|---|

| Architecture | RDNA 3.5 | GCN 1.0 |

| Process Node | 4 nm | 28 nm |

| Transistors | 34,000 million | 1,500 million |

| Die Size | 233 mm² | 123 mm² |

| Base Clock | 400 MHz | 900 MHz |

| Boost Clock | 2900 MHz | 925 MHz |

| Memory Type | System Shared | GDDR5 |

| Memory Size | System Shared | 2 GB |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 72.00 GB/s |

| Shading Units | 1024 | 640 |

| TMUs | 64 | 40 |

| ROPs | 32 | 16 |

| Ray Tracing Cores | 16 | None |

| Pixel Rate | 92.80 GPixel/s | 14.80 GPixel/s |

| Texture Rate | 185.6 GTexel/s | 37.00 GTexel/s |

| FP32 Performance | 5.939 TFLOPS | 1,184.0 GFLOPS |

| FP16 Performance | 5.939 TFLOPS (1:1) | Not listed |

| TDP | 15 W | Not listed |

| Slot Width | IGP | MXM Module |

| Bus Interface | PCIe 4.0 x8 | MXM-A (3.0) |

| DirectX Support | 12 Ultimate (12_2) | 12 (11_1) |

| Vulkan Support | 1.4 | 1.2.170 |

| Production Status | Active | End-of-life |

| Release Date | July 2024 | August 2014 |

The clock speed comparison is interesting: the FirePro W5170M runs at a much higher base clock (900 MHz versus 400 MHz), but the Radeon 890M's boost clock of 2900 MHz is over 3x higher than the W5170M's 925 MHz boost. The 890M's shading unit count (1024) is 60% higher than the W5170M's 640, and its texture rate of 185.6 GTexel/s is exactly 5x the W5170M's 37.00 GTexel/s. Pixel rate tells a similar story: 92.80 GPixel/s versus 14.80 GPixel/s.

The Verdict

The data is unambiguous: the AMD Radeon 890M is the superior GPU in every measurable way. It wins both head-to-head benchmarks by margins of over 350%, has a higher average benchmark score (9,210 versus 8,595), and offers more than 4x the FP32 compute throughput (5.939 TFLOPS versus 1,184.0 GFLOPS). The 890M also brings modern features that the FirePro W5170M simply cannot match: ray tracing cores, DirectX 12 Ultimate support, and a 4 nm process.

The FirePro W5170M is a decade-old, end-of-life product. Its only advantages in the data are its dedicated 2 GB GDDR5 memory with 72.00 GB/s bandwidth (versus system-shared memory), and its higher base clock of 900 MHz. But those advantages do not translate into benchmark wins, the 890M's system-shared memory and massive boost clock more than compensate.

Anyone choosing between these two should pick the Radeon 890M without hesitation, assuming the host platform supports it. The 890M is an IGP, so it requires a Strix Point mobile processor; the W5170M is an MXM module, so it requires a compatible laptop with an MXM slot. For modern workloads, gaming, compute, or content creation, the 890M is categorically faster. The W5170M is only relevant if you are maintaining legacy hardware that specifically requires an MXM-A (3.0) module, and even then, it will struggle with modern API demands.

Where Each One Wins

AMD Radeon 890M, Compute and modern workloads. The 890M wins both shared benchmarks by a landslide: 357.7% in OpenCL and 350.9% in Vulkan. Its FP32 performance of 5.939 TFLOPS is roughly 5x the W5170M's 1,184.0 GFLOPS. The 890M also supports DirectX 12 Ultimate and Vulkan 1.4, making it future-proof for games and applications that rely on newer APIs. Its 16 ray tracing cores give it a feature the W5170M lacks entirely. With a boost clock of 2900 MHz, the 890M can sustain high throughput in bursty workloads, and its 45th percentile ranking puts it roughly on par with older discrete cards like the GTX 960.

AMD FirePro W5170M, Legacy compatibility and dedicated memory. The W5170M does not win any benchmark, but it does have two structural advantages. First, its MXM-A (3.0) form factor means it can be swapped into older laptops that require a discrete MXM module; the 890M is an IGP and cannot be retrofitted. Second, its 2 GB of dedicated GDDR5 memory with 72.00 GB/s bandwidth is not dependent on system RAM, which can be an advantage in scenarios where the host system has slow or limited shared memory. Its higher base clock of 900 MHz (versus 400 MHz) suggests better sustained low-load efficiency, though the data does not include power figures to confirm this. The W5170M also has a longer history of driver maturity in professional FirePro software, though that is not quantified in the benchmark data.

In practical terms, the Radeon 890M is the right choice for any new build or modern laptop, while the FirePro W5170M is only a stopgap for keeping old hardware running. The 890M's 10-year newer release date, active production status, and overwhelming benchmark lead make the decision straightforward for anyone not constrained by legacy MXM slots.