AMD Radeon HD 8970M vs AMD Radeon Pro Vega 16 Comparison
AMD Radeon HD 8970M
Radeon Pro Vega 16
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon HD 8970M vs AMD Radeon Pro Vega 16
The AMD Radeon Pro Vega 16 and AMD Radeon HD 8970M are both end-of-life mobile graphics solutions, but they represent very different eras of AMD's GPU design. The Pro Vega 16 is a newer, power-efficient integrated part built on a modern process, while the HD 8970M is an older, higher-TDP discrete module. The benchmark data shows a clear split: the older card wins the single available head-to-head test, yet the newer card holds a higher overall percentile ranking, indicating a more complex relationship than a simple generational leap.
Head-to-Head Benchmarks
The only direct benchmark comparison available is the Geekbench OpenCL test, and the results are surprising given the age gap. The AMD Radeon HD 8970M scores 21,237 points, while the AMD Radeon Pro Vega 16 scores 18,268 points. This gives the HD 8970M a decisive 14% advantage in this specific compute workload. The deltaPct of -14 indicates that the Pro Vega 16 is 14% slower than its rival in this test.
This result is particularly noteworthy because the newer Pro Vega 16 has a higher theoretical FP32 throughput of 2.437 TFLOPS, compared to the HD 8970M's 2.304 TFLOPS. Despite having 5.8% more raw floating-point power, the Pro Vega 16 falls behind in the actual OpenCL benchmark. This suggests that the older GCN 1.0 architecture in the HD 8970M is better optimized for this particular workload, or that the newer card's drivers are less effective in this scenario.
However, the overall benchmark picture is more nuanced. The Pro Vega 16 has an average benchmark score of 23,250 across three tests (Geekbench Metal, OpenCL, and Vulkan), placing it in the 68th percentile of all GPUs. The HD 8970M, with only its single OpenCL score, averages 21,237 and sits in the 66th percentile. The Pro Vega 16's additional Vulkan score of 21,832 and Metal score of 29,650 show strong performance in those APIs, which the HD 8970M cannot match because it lacks those benchmark results. This means the HD 8970M wins the one common test, but the Pro Vega 16 demonstrates superior performance in the broader range of modern graphics APIs.
FAQ
Q: Which GPU has the higher average benchmark score?
A: The AMD Radeon Pro Vega 16 has a higher average benchmark score of 23,250, compared to the AMD Radeon HD 8970M's 21,237. This places the Pro Vega 16 in the 68th percentile of all GPUs, while the HD 8970M sits in the 66th percentile.
Q: Is the AMD Radeon HD 8970M faster than the Pro Vega 16 in OpenCL?
A: Yes, in the Geekbench OpenCL test, the HD 8970M scores 21,237, which is 14% higher than the Pro Vega 16's score of 18,268. This is the only direct head-to-head benchmark available, and the older card wins it.
Q: What is the memory bandwidth difference between these two cards?
A: The Pro Vega 16 uses 4 GB of HBM2 memory on a 1024-bit bus, delivering a bandwidth of 307.2 GB/s. The HD 8970M uses 4 GB of GDDR5 memory on a 256-bit bus, with a bandwidth of 153.6 GB/s. The Pro Vega 16 has exactly double the memory bandwidth of the HD 8970M.
Q: How do the power requirements compare?
A: The AMD Radeon Pro Vega 16 has a TDP of 75 W, while the AMD Radeon HD 8970M has a TDP of 100 W. The newer card consumes 25% less power, making it significantly more power-efficient per unit of performance.
Q: Which card supports newer graphics APIs?
A: The Pro Vega 16 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. The HD 8970M supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. Both support DirectX 12 and OpenGL 4.6, but the Pro Vega 16 has a newer DirectX feature level and a newer Vulkan version.
Q: What are the production statuses of these two GPUs?
A: Both the AMD Radeon Pro Vega 16 and the AMD Radeon HD 8970M are listed as end-of-life products. The Pro Vega 16 was released on November 13, 2018, while the HD 8970M was released on May 13, 2013.
Where Each One Wins
The AMD Radeon HD 8970M wins in raw OpenCL compute performance. Its score of 21,237 in Geekbench OpenCL is 14% higher than the Pro Vega 16's 18,268. This makes it the better choice for users who rely heavily on OpenCL-based applications, such as certain scientific computing tasks or older rendering workloads. Its higher shading unit count of 1280, compared to 1024 on the Pro Vega 16, likely contributes to this advantage in parallel compute tasks.
The AMD Radeon Pro Vega 16 wins in overall versatility and modern API support. Its average benchmark score of 23,250 is 9.5% higher than the HD 8970M's 21,237, driven by strong Metal (29,650) and Vulkan (21,832) results. The 68th percentile ranking versus 66th percentile also reflects this broader strength. The Pro Vega 16's superior memory bandwidth of 307.2 GB/s, double that of the HD 8970M, makes it better suited for memory-intensive modern workloads. Its lower TDP of 75 W also makes it a more practical choice for thin-and-light laptops where thermal management is critical.
Specification Differences
The two GPUs differ substantially in their core specifications. The Pro Vega 16 has 1024 shading units, 64 texture mapping units (TMUs), and 32 render output units (ROPs). The HD 8970M has more, with 1280 shading units and 80 TMUs, but also has 32 ROPs. This gives the HD 8970M a 25% higher shading unit count and a 25% higher TMU count.
Clock speeds also differ. The Pro Vega 16 runs at a base clock of 815 MHz and a boost clock of 1190 MHz. The HD 8970M has a higher base clock of 850 MHz but a lower boost clock of 900 MHz. The Pro Vega 16's higher boost clock allows it to reach a pixel rate of 38.08 GPixel/s, compared to the HD 8970M's 28.80 GPixel/s. In terms of texture fill rate, the Pro Vega 16 achieves 76.16 GTexel/s, which is slightly higher than the HD 8970M's 72.00 GTexel/s, despite the older card's larger TMU count.
Memory configurations are a major differentiator. The Pro Vega 16 uses 4 GB of HBM2 on a 1024-bit bus, achieving a bandwidth of 307.2 GB/s at a memory clock of 1200 MHz (2.4 Gbps effective). The HD 8970M uses 4 GB of GDDR5 on a 256-bit bus, with a bandwidth of 153.6 GB/s at a memory clock of 1200 MHz (4.8 Gbps effective). The Pro Vega 16's bandwidth advantage is significant.
Finally, the form factor and power differ. The Pro Vega 16 is an integrated graphics processor (IGP) with a 75 W TDP, while the HD 8970M is an MXM module with a 100 W TDP. Both use a PCIe 3.0 x16 bus interface and have portable-device-dependent display outputs.
Architecture Differences
The architectural gap between these two GPUs is vast. The Pro Vega 16 is built on the GCN 5.0 architecture, using the Vega 12 chip, and is manufactured on a 14 nm process at GlobalFoundries. The HD 8970M uses the older GCN 1.0 architecture with the Neptune chip, manufactured on a 28 nm process at TSMC. This process node difference is a major factor in the power efficiency disparity, with the 14 nm part running at 75 W versus the 28 nm part's 100 W.
The HD 8970M does have a listed transistor count of 2,800 million on a die size of 212 mm², giving a transistor density of 13.2M per mm². The Pro Vega 16 does not have a listed transistor count or die size in the data. The HD 8970M also has a clear predecessor (London) and successor (Gem System), while the Pro Vega 16's series and codename fields are null.
In terms of compute capabilities, the Pro Vega 16 supports FP16 with a 2:1 ratio, achieving 4.874 TFLOPS, while the HD 8970M has no listed FP16 support. The Pro Vega 16's FP32 output is 2.437 TFLOPS, slightly higher than the HD 8970M's 2.304 TFLOPS. The Pro Vega 16 also supports Vulkan 1.3 and DirectX 12 (12_1), while the HD 8970M only reaches Vulkan 1.2.170 and DirectX 12 (11_1), reflecting the newer architecture's more modern feature set.
The Verdict
The data presents a clear choice based on workload priorities. The AMD Radeon HD 8970M is the pick for users who require maximum OpenCL compute performance. Its 14% lead in the Geekbench OpenCL test, combined with its higher shading unit count, makes it the stronger option for OpenCL-centric applications. It is also the only choice for compatibility with older software ecosystems that may rely on its specific GCN 1.0 feature set.
The AMD Radeon Pro Vega 16 is the better overall GPU for modern usage. Its higher average benchmark score of 23,250 (versus 21,237) and its 68th percentile ranking demonstrate superior all-around performance, particularly in Metal and Vulkan workloads. The double memory bandwidth of 307.2 GB/s versus 153.6 GB/s makes it significantly more capable for high-resolution textures and large data sets. Its lower TDP of 75 W versus 100 W also makes it the more practical choice for portable devices.
For a user building or upgrading a laptop, the choice hinges on the specific application ecosystem. If the primary use case is OpenCL-heavy compute, the HD 8970M offers a concrete performance advantage. For everything else — especially modern games, Metal-based applications, or Vulkan titles — the Pro Vega 16 delivers better results with less power consumption. The Pro Vega 16 represents the forward-looking choice, while the HD 8970M is a specialized tool for a specific workload.