AMD Radeon HD 8970M vs NVIDIA GeForce RTX 3080 Mobile Comparison

Head-to-Head Benchmarks

The only directly comparable benchmark between these two mobile GPUs is Geekbench OpenCL, and the result is decisively lopsided. The NVIDIA GeForce RTX 3080 Mobile scores 104,831 points, while the AMD Radeon HD 8970M scores 21,237 points. That translates to a 393.6% advantage for the NVIDIA part, meaning the RTX 3080 Mobile delivers nearly five times the OpenCL compute performance of the older AMD chip.

This single head-to-head data point tells the story of two eras. The RTX 3080 Mobile sits in the 69th percentile of all GPUs in the database, while the HD 8970M sits in the 66th percentile. Despite the huge raw score gap, both rank surprisingly close in overall percentile, which reflects how the database's broader benchmark suite weights different workloads. The RTX 3080 Mobile's average benchmark score of 23,628 places it just 0.5% behind the NVIDIA GeForce GTX TITAN Z (23,736) and 0.5% ahead of the NVIDIA GeForce RTX 3070 Ti Mobile (23,518). It also trails the AMD Radeon RX 9070 by 1% (23,877) and leads the AMD Radeon AI PRO R9700 by 1.3% (23,315).

For the HD 8970M, its average score of 21,237 puts it 0.4% ahead of the AMD Radeon RX Vega M GL (21,153), 1% ahead of the NVIDIA GeForce RTX 5050 (21,035), and 0.7% behind the NVIDIA RTX A4000 Mobile (21,379). The closest rival is the NVIDIA Quadro RTX 5000 at 21,629, which beats the HD 8970M by 1.8%. These narrow margins suggest the HD 8970M, despite its age, still occupies a competitive niche among lower-tier modern GPUs in the database's aggregate scoring.

Looking beyond the single head-to-head, the RTX 3080 Mobile's benchmark suite reveals substantial compute versatility. It posts 104,066 in Geekbench Vulkan, 16,321 in Passmark G3D, 7,276 in Passmark GPU Compute, and 637 in Passmark G2D. Its DirectX scores range from 72 (Passmark DirectX 12) to 170 (Passmark DirectX 9), with 120 in DirectX 10 and 146 in DirectX 11. It also manages 2,644 in 3DMark Steel Nomad DX12. The HD 8970M, by contrast, only has a single recorded benchmark in the database, the Geekbench OpenCL score of 21,237, so no additional comparison points exist for DirectX, Vulkan, or compute workloads.

The wins tally is 1-0 in favor of the RTX 3080 Mobile. This is not a close contest in any metric where both parts have data. The delta of 393.6% dwarfs the typical single-digit percentage differences seen among the nearest rivals of either card.

Architecture Differences

The architectural gulf between these two GPUs spans nearly a decade of silicon evolution. The NVIDIA GeForce RTX 3080 Mobile uses the GA104 chip built on Samsung's 8 nm process, packing 17,400 million transistors into a 392 mm² die for a transistor density of 44.4 million per mm². The AMD Radeon HD 8970M uses the Neptune chip on TSMC's 28 nm process, with 2,800 million transistors on a 212 mm² die, giving a density of 13.2 million per mm². The RTX 3080 Mobile's transistor count is roughly six times higher, and its density is more than three times greater.

The RTX 3080 Mobile belongs to the Ampere architecture, part of the GeForce 30 Mobile generation, released in January 2021. The HD 8970M is GCN 1.0 architecture, part of the Solar System (HD 8900M) generation, released in May 2013. The predecessor of the RTX 3080 Mobile is the GeForce 20 Mobile series, while the HD 8970M's predecessor is London, and its successor is Gem System.

Core configuration differs massively. The RTX 3080 Mobile has 6,144 shading units, 192 texture mapping units, 96 raster output units, 48 ray tracing cores, and 192 tensor cores. The HD 8970M has 1,280 shading units, 80 TMUs, and 32 ROPs, with no ray tracing or tensor cores at all. The RTX 3080 Mobile's shading unit count is 4.8 times higher, and its TMU and ROP counts are 2.4 and 3 times higher respectively.

Clock speeds also favor the newer card. The RTX 3080 Mobile runs at 1110 MHz base and 1545 MHz boost, while the HD 8970M runs at 850 MHz base and 900 MHz boost. The RTX 3080 Mobile's boost clock is 645 MHz higher. Memory clocks are similarly divergent: the RTX 3080 Mobile uses 1750 MHz (14 Gbps effective) GDDR6, while the HD 8970M uses 1200 MHz (4.8 Gbps effective) GDDR5. Memory capacity is 8 GB versus 4 GB, both on a 256-bit bus. Bandwidth jumps from 153.6 GB/s on the HD 8970M to 448.0 GB/s on the RTX 3080 Mobile, a 2.9 times improvement.

Pixel and texture rates reflect the raw throughput advantage. The RTX 3080 Mobile achieves 148.3 GPixel/s and 296.6 GTexel/s, compared to 28.80 GPixel/s and 72.00 GTexel/s for the HD 8970M. FP32 compute is 18.98 TFLOPS for the NVIDIA part versus 2.304 TFLOPS for the AMD part, an 8.2 times gap. The RTX 3080 Mobile also supports FP16 at 18.98 TFLOPS (1:1), while the HD 8970M has no recorded FP16 capability.

Power envelopes are closer than the performance gap might suggest. The RTX 3080 Mobile is rated at 115 W TDP, while the HD 8970M is rated at 100 W TDP. Both are portable-device dependent for display outputs. The RTX 3080 Mobile uses PCIe 4.0 x16, while the HD 8970M uses PCIe 3.0 x16. The HD 8970M is an MXM Module form factor, while the RTX 3080 Mobile has no specified slot width and uses no external power connectors.

API support also separates them. The RTX 3080 Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The HD 8970M supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The RTX 3080 Mobile's DirectX 12 Ultimate feature set includes hardware ray tracing and variable rate shading, none of which exist on the older GCN part.

The Verdict

The data is unambiguous. The NVIDIA GeForce RTX 3080 Mobile wins the only direct benchmark comparison by 393.6%, and it is the sole winner in the head-to-head tally. For any workload measured by Geekbench OpenCL, the RTX 3080 Mobile is the clear choice. Its 104,831 score versus 21,237 for the HD 8970M means the NVIDIA part completes compute tasks roughly five times faster.

The HD 8970M cannot compete in raw performance, but its 66th percentile ranking shows it still holds ground in the aggregate database. Its average score of 21,237 sits within 1.8% of the NVIDIA Quadro RTX 5000 and within 1% of the NVIDIA GeForce RTX 5050. For users constrained to the HD 8970M's era or platform, it remains a functional GPU, but it is not a modern performer.

The RTX 3080 Mobile, with a 69th percentile ranking and an average score 11.3% higher than the HD 8970M's single benchmark, is the superior part for any modern application. Its nearest rivals are all within 1.3% of its average score, indicating it sits in a competitive cluster of mid-range GPUs. The HD 8970M's nearest rivals are similarly tight, but at a much lower absolute performance level.

Choose the RTX 3080 Mobile if you need current-generation features, high compute throughput, or broad benchmark coverage. Choose the HD 8970M only if your system requires its specific MXM form factor or if you are working within a legacy platform that cannot accommodate a modern mobile GPU. From the recorded data, there is no scenario where the HD 8970M outperforms the RTX 3080 Mobile.

FAQ

Q: How much faster is the NVIDIA GeForce RTX 3080 Mobile than the AMD Radeon HD 8970M in Geekbench OpenCL?

A: The RTX 3080 Mobile scores 104,831 versus 21,237 for the HD 8970M, a 393.6% advantage.

Q: What are the memory specifications of each GPU?

A: The RTX 3080 Mobile has 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth. The HD 8970M has 4 GB of GDDR5 on a 256-bit bus with 153.6 GB/s bandwidth.

Q: Do both GPUs support ray tracing?

A: No. The RTX 3080 Mobile has 48 ray tracing cores, while the HD 8970M has none.

Q: What is the TDP of each card?

A: The RTX 3080 Mobile is rated at 115 W, and the HD 8970M is rated at 100 W.

Q: Which GPU has higher FP32 compute performance?

A: The RTX 3080 Mobile delivers 18.98 TFLOPS, while the HD 8970M delivers 2.304 TFLOPS.

Q: How do their percentile rankings compare?

A: The RTX 3080 Mobile is in the 69th percentile of all GPUs, while the HD 8970M is in the 66th percentile.

Where Each One Wins

The RTX 3080 Mobile wins in every measurable category where both have data. Its Geekbench OpenCL score is 393.6% higher, making it the default choice for compute-heavy workloads like OpenCL-accelerated rendering, physics simulation, or data processing. Its 8 GB memory capacity and 448.0 GB/s bandwidth also make it better suited for large datasets or high-resolution textures, compared to the HD 8970M's 4 GB and 153.6 GB/s.

The RTX 3080 Mobile's 48 ray tracing cores and 192 tensor cores give it capabilities the HD 8970M simply lacks. Any application that uses DirectX 12 Ultimate features, ray tracing, or AI-accelerated tensor operations will only run on the NVIDIA part. Its Vulkan 1.4 support also exceeds the HD 8970M's Vulkan 1.2.170.

The HD 8970M's only potential advantage is its lower TDP at 100 W versus 115 W, which could matter in thermally constrained chassis. However, the performance difference is so large that this power savings is not a meaningful trade-off. The HD 8970M's MXM Module form factor may also fit specific legacy laptops, but the RTX 3080 Mobile's portable-device-dependent outputs mean it is designed for modern mobile systems.

For gaming, the RTX 3080 Mobile's Passmark G3D score of 16,321 and DirectX 11 score of 146 indicate strong legacy DirectX performance, while its DirectX 12 score of 72 and 3DMark Steel Nomad score of 2,644 show current-gen capability. The HD 8970M has no recorded gaming benchmarks to compare, so its gaming performance cannot be quantified from the database.

In summary, the RTX 3080 Mobile wins all compute, memory bandwidth, feature-set, and API-support categories. The HD 8970M wins only in lower power draw and legacy form factor compatibility. For any user choosing between these two based on the recorded data, the RTX 3080 Mobile is the unequivocal recommendation.