AMD Radeon HD 8850M vs NVIDIA GeForce GTX 980 Comparison

The NVIDIA GeForce GTX 980 is in a completely different performance class than the AMD Radeon HD 8850M, a verdict driven by overwhelming architectural and specification advantages. The single shared benchmark paints a stark picture: in Geekbench OpenCL, the GTX 980 scores 34,676 points against the HD 8850M’s 7,447 points, a 365.6% advantage. This is not a close contest; the data positions the GTX 980 as a high-end desktop GPU from a later generation, while the HD 8850M is an older mobile-class part with a much smaller silicon footprint.

Head-to-Head Benchmarks

The only benchmark where both GPUs appear is Geekbench OpenCL, and the result is a complete rout. The NVIDIA GeForce GTX 980 delivers a score of 34,676, while the AMD Radeon HD 8850M manages just 7,447. The delta is 365.6% in favor of the GTX 980, meaning the NVIDIA part is roughly four and a half times faster in this compute workload. This is not a marginal victory; it is a generational and market-segment chasm.

For context, the GTX 980’s average benchmark score across all tested workloads is 8,167, which places it in the 43rd percentile of all GPUs. Its nearest rivals include the NVIDIA GeForce GTX 950M (average score 8,135, delta 0.4%) and the AMD Radeon R9 M360 (average score 8,129, delta 0.5%). The HD 8850M, in contrast, has an average benchmark score of 7,447, placing it in the 40th percentile. Its nearest rivals are the Intel UHD Graphics 750 (average score 7,441, delta 0.1%) and the AMD Radeon R7 350 (average score 7,425, delta 0.3%). Notably, the HD 8850M is also within 0.3% of the NVIDIA GeForce GTX 1650’s average score of 7,472, but the delta is negative (-0.3%), meaning the HD 8850M actually trails that modern desktop part.

The GTX 980’s individual benchmark results further illustrate its breadth. It scores 11,095 in Passmark G3D, 4,753 in Passmark GPU Compute, and 164 in Passmark DirectX 9. Its 3DMark Steel Nomad DX12 score is 474, and Geekbench Vulkan reaches 22,543. The HD 8850M has no other recorded scores, so its OpenCL result is its only data point. This lack of additional data reinforces that the HD 8850M is not designed for the same demanding workloads; the GTX 980 is clearly a discrete desktop solution for gaming and compute, while the HD 8850M appears to be a lower-tier mobile part.

Architecture Differences

The two GPUs come from different architectural generations and are built for different purposes. The NVIDIA GeForce GTX 980 uses the GM204 chip, which is based on the Maxwell 2.0 architecture and belongs to the GeForce 900 generation. It is manufactured on a 28 nm process at TSMC, with a die size of 398 mm² and 5,200 million transistors, resulting in a transistor density of 13.1 million per mm². In contrast, the AMD Radeon HD 8850M uses the Venus chip, based on the older GCN 1.0 architecture and belonging to the Solar System (HD 8800M) generation. It is also on a 28 nm process at TSMC, but the die is only 123 mm² with 1,500 million transistors, yielding a lower density of 12.2 million per mm².

The core configurations diverge sharply. The GTX 980 has 2,048 shading units, 128 texture mapping units (TMUs), and 64 render output units (ROPs). The HD 8850M has just 640 shading units, 40 TMUs, and 16 ROPs. This represents a 3.2x difference in shading units, a 3.2x difference in TMUs, and a 4x difference in ROPs. Clock speeds also favor the NVIDIA part: the GTX 980 runs at a base of 1,127 MHz and a boost of 1,216 MHz, while the HD 8850M runs at a base of 575 MHz and a boost of 625 MHz. The memory subsystems are equally lopsided. The GTX 980 features 4 GB of GDDR5 on a 256-bit bus, delivering 224.4 GB/s of bandwidth. The HD 8850M has 2 GB of GDDR5 on a 128-bit bus, producing just 64.00 GB/s.

API support differs as well. The GTX 980 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The HD 8850M supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The GTX 980’s higher DirectX feature level and newer Vulkan version indicate better support for modern rendering techniques. The GTX 980 also has a defined power profile with a 165 W TDP, dual-slot cooler, and 2x 6-pin power connectors, while the HD 8850M has no TDP, slot width, or power connector data listed, consistent with a mobile part where those figures are not typically published.

FAQ

Q: Which GPU is faster in the shared Geekbench OpenCL test?

A: The NVIDIA GeForce GTX 980 is decisively faster, scoring 34,676 points versus the AMD Radeon HD 8850M’s 7,447 points, a 365.6% advantage.

Q: How do the two GPUs rank against their nearest rivals?

A: The GTX 980’s average score of 8,167 is 0.4% ahead of the GTX 950M and 0.5% ahead of the Radeon R9 M360. The HD 8850M’s average score of 7,447 is 0.1% ahead of the Intel UHD Graphics 750 and 0.3% ahead of the Radeon R7 350, but 0.3% behind the GTX 1650.

Q: What is the core configuration difference?

A: The GTX 980 has 2,048 shading units, 128 TMUs, and 64 ROPs, while the HD 8850M has 640 shading units, 40 TMUs, and 16 ROPs. The GTX 980 also runs at much higher clocks: 1,127 MHz base versus 575 MHz base.

Q: How does memory bandwidth compare?

A: The GTX 980 offers 224.4 GB/s of bandwidth from 4 GB of GDDR5 on a 256-bit bus. The HD 8850M offers 64.00 GB/s from 2 GB of GDDR5 on a 128-bit bus.

Q: Which GPU has better API support for modern games?

A: The GTX 980 supports DirectX 12 (12_1) and Vulkan 1.4, while the HD 8850M supports DirectX 12 (11_1) and Vulkan 1.2.170. The GTX 980’s higher feature level is advantageous.

Q: Are these GPUs from the same era?

A: No. The GTX 980 was released on 2014-09-18, while the HD 8850M was released on 2013-03-31. The GTX 980 is also from a newer generation (GeForce 900 vs. Solar System).

Specification Differences

• Chip: GM204 (NVIDIA) vs. Venus (AMD)
• Architecture: Maxwell 2.0 vs. GCN 1.0
• Generation: GeForce 900 vs. Solar System (HD 8800M)
• Transistors: 5,200 million vs. 1,500 million
• Die Size: 398 mm² vs. 123 mm²
• Transistor Density: 13.1M / mm² vs. 12.2M / mm²
• Base Clock: 1,127 MHz vs. 575 MHz
• Boost Clock: 1,216 MHz vs. 625 MHz
• Memory Clock: 1753 MHz (7 Gbps effective) vs. 1000 MHz (4 Gbps effective)
• Memory Size: 4 GB vs. 2 GB
• Memory Bus Width: 256 bit vs. 128 bit
• Memory Bandwidth: 224.4 GB/s vs. 64.00 GB/s
• Shading Units: 2048 vs. 640
• TMUs: 128 vs. 40
• ROPs: 64 vs. 16
• Pixel Rate: 77.82 GPixel/s vs. 10.00 GPixel/s
• Texture Rate: 155.6 GTexel/s vs. 25.00 GTexel/s
• FP32 Performance: 4.981 TFLOPS vs. 800.0 GFLOPS
• TDP: 165 W vs. (not listed)
• Slot Width: Dual-slot vs. (not listed)
• Power Connectors: 2x 6-pin vs. (not listed)
• Suggested PSU: 450 W vs. (not listed)
• Display Outputs: 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.2 vs. (not listed)
• DirectX Support: 12 (12_1) vs. 12 (11_1)
• Vulkan Support: 1.4 vs. 1.2.170
• Release Date: 2014-09-18 vs. 2013-03-31
• Predecessor: GeForce 700 vs. London
• Successor: GeForce 10 vs. Gem System
• Dimensions: 267 mm length, 111 mm height, 40 mm width vs. (not listed)

Where Each One Wins

The NVIDIA GeForce GTX 980 wins every measurable category. It is faster in the sole head-to-head benchmark, has more shading units, TMUs, ROPs, higher clocks, more memory, wider memory bus, and higher bandwidth. Its pixel rate of 77.82 GPixel/s and texture rate of 155.6 GTexel/s are dramatically higher than the HD 8850M’s 10.00 GPixel/s and 25.00 GTexel/s. The GTX 980’s FP32 compute of 4.981 TFLOPS dwarfs the HD 8850M’s 800.0 GFLOPS. The GTX 980 is also the only one with a published TDP (165 W), slot width (dual-slot), power connectors (2x 6-pin), and display outputs (1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.2), signifying a fully-featured desktop card.

The AMD Radeon HD 8850M has no single benchmark win. Its only advantage is in physical footprint: the die is 123 mm² versus 398 mm², and it has a lower transistor count (1,500 million vs. 5,200 million). This suggests it is a more power-efficient or compact design, but no TDP is listed to confirm. The HD 8850M also has a lower transistor density (12.2M / mm² vs. 13.1M / mm²), indicating less efficient packing. In terms of performance, the HD 8850M’s only recorded score (7,447 in OpenCL) puts it in the 40th percentile, while the GTX 980’s average score (8,167) puts it in the 43rd percentile. These percentile values, however, are based on different benchmarks and should not be compared directly.

The Verdict

The data is unambiguous: the NVIDIA GeForce GTX 980 is the superior GPU for any task requiring significant graphics or compute performance. It is 365.6% faster in the single shared benchmark, which aligns with its 3.2x more shading units, 4x more ROPs, and 3.5x higher memory bandwidth. The GTX 980 is a desktop-class card from a newer generation (Maxwell 2.0) with a modern API feature set (DirectX 12_1, Vulkan 1.4), making it suitable for gaming and compute workloads. Its average benchmark score of 8,167 places it in the 43rd percentile, alongside the GTX 950M and R9 M360, which are also mid-range mobile parts. This suggests the GTX 980 is a strong mid-range performer even by modern standards.

The AMD Radeon HD 8850M is a legacy mobile GPU from 2013, based on the older GCN 1.0 architecture. Its single benchmark score of 7,447 places it in the 40th percentile, near the Intel UHD Graphics 750 and Radeon R7 350. It is not competitive with the GTX 980 in any way, and its lack of additional benchmark data indicates it is not intended for demanding applications. The HD 8850M’s smaller die (123 mm²) and lower transistor count (1,500 million) suggest it is designed for power-constrained laptops, but without a TDP figure, the data cannot confirm efficiency claims.

For anyone choosing between these two, the GTX 980 is the only rational choice for performance. It wins every benchmark, has more of every hardware resource, and offers better software support. The HD 8850M is only relevant in a legacy laptop context where it is the pre-installed option, but even then, the data shows it is far behind. The verdict is simple: the GTX 980 is a dedicated, high-performance desktop GPU, while the HD 8850M is a low-end mobile part from an older era.