AMD FirePro W5170M vs NVIDIA GeForce GTX 950A Comparison

Head-to-Head Benchmarks

The only common benchmark recorded for both mobile workstation parts is Geekbench OpenCL, and the result is decisive. The NVIDIA GeForce GTX 950A scores 10,273 points, while the AMD FirePro W5170M scores 8,140 points. That gives the NVIDIA part a 26.2% advantage in compute throughput, a substantial margin for two GPUs aimed at similar portable workstations.

Context from the database clarifies how significant that gap is. The GTX 950A sits just 0.9% behind the AMD Radeon RX 6500M, a much newer discrete mobile part, and lands within 2% of the AMD Radeon RX 550X. It also edges out the AMD Radeon R9 M375 by 2%. In short, the GTX 950A competes with a generation of GPUs that came years after its 2015 release, at least in raw OpenCL compute.

The FirePro W5170M, by contrast, posts an average benchmark score of 8,595 when including its Vulkan result, but its OpenCL score alone is 8,140. Its nearest rivals cluster tightly around that figure: the Intel Arc A380 is 0.4% ahead, the AMD Radeon HD 8870M is 1.6% behind, and the NVIDIA GeForce MX330 is 1.6% behind. The NVIDIA Quadro P2200 leads it by 1.1%. The FirePro is therefore within a narrow band of mid-range mobile GPUs, but the GTX 950A is simply in a higher performance tier. The data records one head-to-head win for the GTX 950A and zero for the FirePro, so the compute hierarchy is unambiguous.

Architecture Differences

The two GPUs come from different architectural lineages, which explains much of the performance delta. The NVIDIA GeForce GTX 950A uses the GM107 chip built on the Maxwell architecture, while the AMD FirePro W5170M uses the Tropo chip based on GCN 1.0. Both are fabricated on a 28 nm process at TSMC, so the manufacturing node is not a differentiator.

Transistor counts and die sizes do differ. The GTX 950A packs 1,870 million transistors into a 148 mm² die, yielding a transistor density of 12.6 million per mm². The FirePro W5170M carries 1,500 million transistors on a smaller 123 mm² die, for a density of 12.2 million per mm². NVIDIA's part is both larger and denser, and that extra silicon translates into more compute resources at similar power envelopes.

Clock speeds favor NVIDIA as well. The GTX 950A runs at a 993 MHz base clock with a 1,124 MHz boost, while the FirePro W5170M runs at 900 MHz base and 925 MHz boost. The NVIDIA part therefore sustains a higher operating frequency under load, compounding its architectural efficiency advantage. Memory clocks also diverge sharply: the GTX 950A runs its DDR3 at 1001 MHz (2 Gbps effective), while the FirePro runs its GDDR5 at 1125 MHz (4.5 Gbps effective). That difference in memory technology is significant, but the GTX 950A still wins compute benchmarks by a wide margin, suggesting the core architecture does the heavy lifting.

Both GPUs share identical shading unit counts at 640, the same 40 texture mapping units, and the same 16 raster output units. The GTX 950A's higher clocks push its pixel rate to 17.98 GPixel/s and texture rate to 44.96 GTexel/s, versus 14.80 GPixel/s and 37.00 GTexel/s for the FirePro. FP32 compute also favors NVIDIA: 1,438.7 GFLOPS versus 1,184.0 GFLOPS. The GTX 950A is simply a faster implementation of similar resource counts.

API support differs in notable ways. The GTX 950A lists DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. The FirePro W5170M lists DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The NVIDIA part has a newer Vulkan revision, while AMD claims slightly higher DirectX feature level. For legacy compatibility, the GTX 950A's predecessor is the GeForce 800A, while the FirePro's predecessor is the FirePro Mobility. The FirePro's successor is the Radeon Pro Mobile, whereas the GTX 950A has no recorded successor.

The Verdict

For users prioritizing raw compute performance, the data points unequivocally to the NVIDIA GeForce GTX 950A. Its 26.2% OpenCL lead over the AMD FirePro W5170M is the defining metric, and its position in the database places it within 1% of much newer GPUs like the Radeon RX 6500M. The GTX 950A also holds advantages in clock speed, pixel throughput, texture rate, and FP32 compute, all from the same 28 nm process.

The AMD FirePro W5170M does have one qualitative advantage: its GDDR5 memory provides 72.00 GB/s of bandwidth versus 32.03 GB/s for the GTX 950A's DDR3. For memory-bound workloads, that could matter, but the benchmark data shows the GTX 950A still wins the only recorded compute test by a wide margin. The FirePro also carries a newer DirectX feature level (12_1 versus 11_0), which may matter for specific application compatibility.

Given the end-of-life status of both parts, the decision rests on what a buyer values. The GTX 950A is the stronger compute performer and sits at a higher percentile (48th versus 44th) among all GPUs in the database. The FirePro W5170M offers faster memory and a newer DirectX revision but trails in every recorded compute metric. For general-purpose OpenCL acceleration, the GTX 950A is the clear choice. For applications that specifically require GDDR5 bandwidth or DirectX 12_1 features, the FirePro warrants consideration, but the benchmark evidence favors NVIDIA.

Specification Differences

The following fields differ between the two parts:

• Chip: GM107 (NVIDIA) versus Tropo (AMD)
• Architecture: Maxwell versus GCN 1.0
• Generation: GeForce 900A versus FirePro Mobile (Wx100M)
• Transistors: 1,870 million versus 1,500 million
• Die Size: 148 mm² versus 123 mm²
• Transistor Density: 12.6M / mm² versus 12.2M / mm²
• Base Clock: 993 MHz versus 900 MHz
• Boost Clock: 1124 MHz versus 925 MHz
• Memory Clock: 1001 MHz (2 Gbps effective) versus 1125 MHz (4.5 Gbps effective)
• Memory Type: DDR3 versus GDDR5
• Memory Bandwidth: 32.03 GB/s versus 72.00 GB/s
• Pixel Rate: 17.98 GPixel/s versus 14.80 GPixel/s
• Texture Rate: 44.96 GTexel/s versus 37.00 GTexel/s
• FP32 Compute: 1,438.7 GFLOPS versus 1,184.0 GFLOPS
• TDP: 75 W versus null (not recorded)
• Bus Interface: MXM-B (3.0) versus MXM-A (3.0)
• DirectX Support: 12 (11_0) versus 12 (11_1)
• Vulkan Support: 1.4 versus 1.2.170
• Release Date: 2015-03-12 versus 2014-08-24
• Predecessor: GeForce 800A versus FirePro Mobility
• Successor: null versus Radeon Pro Mobile

Identical fields include 2 GB memory size, 128-bit bus width, 640 shading units, 40 TMUs, 16 ROPs, 28 nm process node, TSMC foundry, MXM Module slot width, no power connectors, portable-device-dependent display outputs, and end-of-life production status.

FAQ

Q: Which GPU has a higher OpenCL score?

A: The NVIDIA GeForce GTX 950A scores 10,273 points versus 8,140 for the AMD FirePro W5170M, a 26.2% advantage.

Q: Do the two GPUs have the same number of shading units?

A: Yes, both have 640 shading units, along with 40 texture mapping units and 16 raster output units.

Q: Which card provides higher memory bandwidth?

A: The AMD FirePro W5170M provides 72.00 GB/s using GDDR5, while the NVIDIA GeForce GTX 950A provides 32.03 GB/s using DDR3.

Q: What are the boost clock speeds?

A: The NVIDIA GeForce GTX 950A boosts to 1124 MHz, while the AMD FirePro W5170M boosts to 925 MHz.

Q: Which GPU supports a newer Vulkan version?

A: The NVIDIA GeForce GTX 950A supports Vulkan 1.4, whereas the AMD FirePro W5170M supports Vulkan 1.2.170.

Q: Were both GPUs produced on the same manufacturing process?

A: Yes, both are built on a 28 nm process at TSMC, though the GTX 950A has a larger die at 148 mm² versus 123 mm² for the FirePro.