AMD Radeon HD 8690M vs NVIDIA GeForce RTX 5070 Ti SUPER Comparison
AMD Radeon HD 8690M
GeForce RTX 5070 Ti SUPER
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon HD 8690M vs NVIDIA GeForce RTX 5070 Ti SUPER
The NVIDIA GeForce RTX 5070 Ti SUPER and the AMD Radeon HD 8690M represent two extremes of the GPU timeline, separated by over a decade of silicon evolution. The data shows a complete mismatch in nearly every measurable specification, from process technology to memory bandwidth. However, the benchmark database places both cards at the 36th percentile among all GPUs, which is a statistical anomaly driven by the different workloads each card was tested under, rather than an indicator of comparable real-world performance.
Where Each One Wins
The NVIDIA GeForce RTX 5070 Ti SUPER is designed for modern, high-resolution gaming and compute workloads. Its benchmark result comes from the 3DMark Steel Nomad DX12 test, where it scores 6269.5 points. This is a DirectX 12 Ultimate workload that stresses ray tracing, mesh shaders, and other contemporary rendering features. The card’s 70 RT cores and 280 tensor cores are purpose-built for these tasks, and its 16 GB of GDDR7 memory on a 256-bit bus provides 896.0 GB/s of bandwidth, a figure that allows it to feed its 8960 shading units without bottlenecking.
The AMD Radeon HD 8690M is a legacy mobile part from 2013, built for entry-level laptop graphics. Its sole benchmark entry is Geekbench OpenCL, scoring 6137 points. This test measures general-purpose compute throughput, which is a different metric than gaming rasterization. The HD 8690M wins in the niche of legacy compatibility and low-power mobile operation, where its 28 nm process and 690 million transistors require far less power than the RTX 5070 Ti SUPER’s 350 W TDP. However, the data does not include any gaming benchmarks for the HD 8690M, so its victory in this comparison is confined to a single synthetic compute test.
In terms of production status, the RTX 5070 Ti SUPER is listed as Active, while the HD 8690M is End-of-life. This means the NVIDIA card is the only one you can currently purchase new, making it the default choice for anyone building a system today. The HD 8690M’s nearest rivals include the Intel Iris Pro Graphics 6200 and NVIDIA GeForce MX230, both of which are also older or low-end parts, reinforcing its position as a legacy product.
Architecture Differences
The architectural gap between these two GPUs is vast. The RTX 5070 Ti SUPER is built on the Blackwell 2.0 architecture using TSMC’s 5 nm process, packing 45,600 million transistors into a 378 mm² die. This yields a transistor density of 120.6 million per square millimeter. In contrast, the HD 8690M uses the GCN 1.0 architecture on a 28 nm process, with just 690 million transistors on a 56 mm² die, resulting in a density of 12.3 million per square millimeter. The RTX 5070 Ti SUPER has roughly 66 times more transistors, which explains its massive lead in compute resources.
The memory subsystems are equally divergent. The RTX 5070 Ti SUPER features 16 GB of GDDR7 memory with a 256-bit bus width, delivering 896.0 GB/s of bandwidth. The HD 8690M has 2 GB of GDDR5 on a 64-bit bus, providing only 32.00 GB/s. This 28x difference in memory bandwidth is critical for modern games and high-resolution textures, where the older card would simply run out of data to process.
Feature support also separates the two. The RTX 5070 Ti SUPER supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6, along with hardware ray tracing via its 70 RT cores. The HD 8690M only supports DirectX 12 (11_1), Vulkan 1.2.170, and OpenGL 4.6, with no equivalent RT or tensor core hardware. The NVIDIA card also uses a PCIe 5.0 x16 interface, while the AMD card is limited to PCIe 3.0 x8, halving the available bandwidth for data transfers from the CPU.
Head-to-Head Benchmarks
Direct comparison is difficult because the two cards were tested using different benchmark suites. The RTX 5070 Ti SUPER has a single entry in 3DMark Steel Nomad DX12, scoring 6269.5. The HD 8690M has a single entry in Geekbench OpenCL, scoring 6137. Neither card appears in the other’s benchmark suite, so there is no overlapping data point for a true apples-to-apples comparison.
However, the nearest rival data provides context. The RTX 5070 Ti SUPER’s closest competitor is the NVIDIA GeForce RTX 4070 Ti SUPER AD102, which scores 6270, a delta of 0%. This indicates the RTX 5070 Ti SUPER is essentially performance-equivalent to that card in the Steel Nomad test. The AMD FirePro W600 scores 6223, putting the RTX 5070 Ti SUPER 0.8% ahead. The Quadro K620 scores 6282, which is 0.2% faster, and the Radeon R7 M350 scores 6327, which is 0.9% faster. These deltas are all within a single percentage point, suggesting the RTX 5070 Ti SUPER sits in a tightly packed performance band for this specific workload.
For the HD 8690M, its nearest rivals include the Intel Iris Pro Graphics 6200 at 6117 (0.3% slower), the NVIDIA RTX A400 at 6078 (1% slower), and the NVIDIA GeForce MX230 at 6077 (1% slower). The AMD Radeon HD 6950 scores 6210, which is 1.2% faster than the HD 8690M. These results show the HD 8690M is competitive with other low-end and legacy parts, but the absolute scores are in a completely different performance tier than the RTX 5070 Ti SUPER’s results.
The raw compute specifications tell the real story. The RTX 5070 Ti SUPER delivers 43.94 TFLOPS of FP32 performance, while the HD 8690M manages 624.0 GFLOPS, a 70x difference. Pixel fill rates are 235.4 GPixel/s versus 7.800 GPixel/s, and texture rates are 686.6 GTexel/s versus 19.50 GTexel/s. Even the memory bandwidth difference of 896.0 GB/s versus 32.00 GB/s is a 28x gap. These numbers indicate that in any workload that scales with compute or memory throughput, the RTX 5070 Ti SUPER would outperform the HD 8690M by orders of magnitude.
FAQ
Q: Which card has a higher benchmark score?
A: The RTX 5070 Ti SUPER scores 6269.5 in 3DMark Steel Nomad DX12, while the HD 8690M scores 6137 in Geekbench OpenCL. The difference is only 2.2%, but the tests are entirely different workloads, so this is not a meaningful comparison.
Q: Does the AMD Radeon HD 8690M support ray tracing?
A: No. The HD 8690M has no RT cores listed in its specifications. The RTX 5070 Ti SUPER has 70 RT cores, which enable hardware-accelerated ray tracing.
Q: What is the memory bandwidth difference?
A: The RTX 5070 Ti SUPER has 896.0 GB/s of bandwidth from its 16 GB GDDR7 memory on a 256-bit bus. The HD 8690M has 32.00 GB/s from its 2 GB GDDR5 memory on a 64-bit bus.
Q: Are these cards in the same performance percentile?
A: Both are listed at the 36th percentile among all GPUs, but this is due to the differing benchmark workloads. The RTX 5070 Ti SUPER’s percentile is based on a modern DX12 test, while the HD 8690M’s is based on a compute test.
Q: Which card has a larger die and more transistors?
A: The RTX 5070 Ti SUPER has a 378 mm² die with 45,600 million transistors. The HD 8690M has a 56 mm² die with 690 million transistors.
Q: Is the HD 8690M still in production?
A: No. The HD 8690M is listed as End-of-life with a release date of February 2013. The RTX 5070 Ti SUPER is listed as Active with a release date of December 2025.
The Verdict
The data makes the choice clear for any modern use case. The RTX 5070 Ti SUPER is the only viable option for gaming, content creation, or any compute-heavy workload. Its 43.94 TFLOPS of FP32 performance, 896.0 GB/s memory bandwidth, and 16 GB of VRAM place it in a performance tier that the HD 8690M cannot approach. The HD 8690M’s 624.0 GFLOPS and 32.00 GB/s bandwidth are insufficient for even entry-level modern games, and its lack of RT cores means it cannot handle ray-traced titles at all.
The HD 8690M’s only theoretical advantage is power consumption, though the FACT PACK does not list a TDP for it. Its 28 nm process and 690 million transistors suggest it draws far less than the RTX 5070 Ti SUPER’s 350 W TDP, but this is not quantified. For anyone with a legacy laptop from the HD 8600M era, the card might still function for basic 2D tasks or very old games, but the production status is End-of-life, meaning driver support and availability are limited.
The RTX 5070 Ti SUPER’s nearest rival, the RTX 4070 Ti SUPER AD102, scores within 0% of it, indicating that NVIDIA has positioned this card as a direct successor in performance. The 5 nm process and Blackwell 2.0 architecture provide modern feature support like DirectX 12 Ultimate and Vulkan 1.4, which the HD 8690M lacks. For any buyer, the RTX 5070 Ti SUPER is the data-backed choice; the HD 8690M is a historical artifact with no place in a current system.