AMD Radeon HD 8690M vs NVIDIA T600 Comparison

Head-to-Head Benchmarks

The database contains a single shared benchmark result for these two GPUs, the Geekbench OpenCL test, and it delivers a decisive outcome. The NVIDIA T600 scores 27,875 points, while the AMD Radeon HD 8690M manages 6,137 points. That is a 354.2% advantage for the T600, meaning the NVIDIA part delivers roughly four and a half times the raw compute throughput of the AMD part in this workload. The delta is so large that it separates these cards into entirely different performance classes, despite both being end-of-life mobile or low-profile workstation parts.

Context from the nearest rivals reinforces how lopsided this matchup is. The T600's average benchmark score across all recorded tests is 7,035, and its closest competitor in the database is the NVIDIA GeForce GTX 680M at 7,023, a 0.2% gap. The GTX 970 sits 1.7% higher at 7,157. The AMD Radeon HD 8690M, by contrast, has an average benchmark score of 6,137, with its nearest rival being the Intel Iris Pro Graphics 6200 at 6,117, a 0.3% difference. The HD 8690M's OpenCL score is its only recorded benchmark, so its average equals that single result. In percentile terms, the T600 sits at the 39th percentile of all GPUs in the database, while the HD 8690M rests at the 36th percentile. The percentile gap is modest, but the raw performance gap is enormous, which suggests the HD 8690M's percentile is buoyed by the fact that many older or lower-end parts populate the database, while the T600's percentile reflects a crowded mid-range field.

It is importantly the T600 has nine recorded benchmark scores covering DirectX 9, 10, 11, 12, G2D, G3D, and compute workloads, whereas the HD 8690M only has the single OpenCL entry. This means the T600's average is derived from a much broader testing surface, while the HD 8690M's average is a single data point. The head-to-head comparison, however, is clean: one test, one winner, and the winner is not close.

Architecture Differences

The two GPUs come from different architectural lineages, and the data shows a generational chasm. The NVIDIA T600 uses the TU117 chip, built on the Turing architecture, fabricated by TSMC on a 12 nm process. It packs 4,700 million transistors into a 200 mm² die, yielding a transistor density of 23.5 million per square millimeter. The AMD Radeon HD 8690M uses the Sun chip, based on GCN 1.0 architecture, also fabricated by TSMC but on a 28 nm process. It contains 690 million transistors on a 56 mm² die, for a density of 12.3 million per square millimeter. The T600 therefore has nearly seven times the transistor count and more than three and a half times the die area, with roughly double the transistor density.

The compute configurations differ sharply. The T600 has 640 shading units, 40 texture mapping units, and 32 raster operation units. The HD 8690M has 320 shading units, 20 TMUs, and only 8 ROPs. The T600 doubles the shader count and TMU count, and quadruples the ROP count. Clock speeds tell a different story: the HD 8690M runs at a base clock of 925 MHz and a boost clock of 975 MHz, while the T600 runs at 735 MHz base and 1,335 MHz boost. The AMD part starts higher but has a much narrower boost range; the NVIDIA part boosts aggressively, ending up 36.9% higher than its base clock and 360 MHz above the AMD part's boost ceiling.

The resulting throughput figures are stark. The T600 delivers 42.72 GPixel/s of pixel rate and 53.40 GTexel/s of texture rate, while the HD 8690M delivers 7.800 GPixel/s and 19.50 GTexel/s. In floating-point compute, the T600 reaches 1.709 TFLOPS in FP32, with an FP16 rate of 3.418 TFLOPS at a 2:1 ratio. The HD 8690M reaches 624.0 GFLOPS in FP32 and has no recorded FP16 capability. The T600's FP32 output is nearly triple the AMD part's, and its pixel throughput is more than five times higher.

Memory architecture adds another layer of separation. The T600 has 4 GB of GDDR6 memory on a 128-bit bus, with a bandwidth of 160.0 GB/s. The HD 8690M has 2 GB of GDDR5 memory on a 64-bit bus, with a bandwidth of 32.00 GB/s. The T600 offers double the capacity, double the bus width, and five times the bandwidth. Memory clocks are listed at 1,250 MHz with 10 Gbps effective for the T600, versus 1,000 MHz with 4 Gbps effective for the HD 8690M.

The feature sets also diverge. The T600 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, while the HD 8690M supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The T600's Vulkan version is newer, and its DirectX feature level is higher. The T600 connects via PCIe 3.0 x16, while the HD 8690M uses PCIe 3.0 x8, halving the available bus bandwidth. The T600 has four mini-DisplayPort 1.4a outputs; the HD 8690M has no display outputs recorded in the database. The T600 is a single-slot card with no power connectors and a 40 W TDP, with a suggested PSU of 200 W. The HD 8690M has no TDP, slot width, power connector, or PSU data recorded.

Where Each One Wins

Given the single shared benchmark, the use-case split is heavily skewed. The NVIDIA T600 wins the only direct comparison, and by a margin that makes it the clear choice for any compute-oriented workload that can leverage OpenCL. The recorded data shows the T600 also has a broad benchmark profile: its Passmark G3D score is 6,479, its G2D score is 756, and its GPU compute score is 2,402. It posts DirectX scores of 114 in DirectX 9, 49 in DirectX 11, 32 in DirectX 10, and 25 in DirectX 12. These numbers indicate a GPU that handles legacy DirectX workloads reasonably well, with a strong G3D result relative to its compute score.

The AMD Radeon HD 8690M has no recorded wins in the database, and its only benchmark result is the OpenCL score of 6,137. This means the data cannot support any scenario where the HD 8690M outperforms the T600 in a measured test. The HD 8690M's architecture, with GCN 1.0 and a 28 nm process, is from an earlier generation, and its smaller memory bus and lower bandwidth suggest it would struggle in bandwidth-sensitive tasks. Its pixel rate of 7.800 GPixel/s and texture rate of 19.50 GTexel/s are both far below the T600's figures, so any fill-rate-bound workload would favor the NVIDIA part.

The T600's advantages in shading units, ROPs, memory bandwidth, and FP32 throughput point to it being better suited for modern workloads such as GPU-accelerated compute, video encoding, and higher-resolution rendering. The HD 8690M, with its lower clock boost and smaller memory footprint, would be more appropriate for basic display output or very light 3D acceleration, but the database does not include benchmarks that capture such usage. The data simply does not support any win for the AMD part, so the practical conclusion is that the T600 is the superior option across every measured dimension.

Specification Differences

The two cards differ in nearly every recorded specification. The T600 uses a 12 nm process, the HD 8690M uses 28 nm. The T600 has 4,700 million transistors, the HD 8690M has 690 million. Die size is 200 mm² versus 56 mm². Transistor density is 23.5M per mm² versus 12.3M per mm². The T600's base clock is 735 MHz versus 925 MHz for the HD 8690M, but its boost clock is 1,335 MHz versus 975 MHz. Memory clock is 1,250 MHz with 10 Gbps effective for the T600, versus 1,000 MHz with 4 Gbps effective for the HD 8690M.

Memory capacity is 4 GB versus 2 GB. Memory type is GDDR6 versus GDDR5. Bus width is 128 bit versus 64 bit. Bandwidth is 160.0 GB/s versus 32.00 GB/s. Shading units are 640 versus 320. TMUs are 40 versus 20. ROPs are 32 versus 8. Pixel rate is 42.72 GPixel/s versus 7.800 GPixel/s. Texture rate is 53.40 GTexel/s versus 19.50 GTexel/s. FP32 compute is 1.709 TFLOPS versus 624.0 GFLOPS. The T600 has an FP16 rate of 3.418 TFLOPS; the HD 8690M has none recorded.

The TDP is 40 W for the T600; the HD 8690M has no TDP recorded. The T600 is single-slot with no power connectors and a suggested PSU of 200 W; the HD 8690M has no such data. Bus interface is PCIe 3.0 x16 for the T600 versus PCIe 3.0 x8 for the HD 8690M. Display outputs are 4x mini-DisplayPort 1.4a for the T600, none recorded for the AMD part. DirectX support is 12 (12_1) for the T600 versus 12 (11_1) for the HD 8690M. Vulkan support is 1.4 for the T600 versus 1.2.170 for the HD 8690M. OpenGL is 4.6 for both. Release dates differ: the T600 launched in April 2021, the HD 8690M in February 2013. Both are end-of-life. The T600's predecessor is Quadro Volta and its successor is Workstation Ampere; the HD 8690M's predecessor is London and its successor is Gem System.

FAQ

Q: How much faster is the NVIDIA T600 than the AMD Radeon HD 8690M in OpenCL?

A: The T600 scores 27,875 in Geekbench OpenCL, while the HD 8690M scores 6,137. That is a 354.2% delta, making the T600 roughly four and a half times faster.

Q: What are the memory bandwidth figures for each card?

A: The T600 has 160.0 GB/s of bandwidth with 4 GB of GDDR6 on a 128-bit bus. The HD 8690M has 32.00 GB/s with 2 GB of GDDR5 on a 64-bit bus.

Q: Which card has more shading units?

A: The NVIDIA T600 has 640 shading units, while the AMD Radeon HD 8690M has 320.

Q: Do both cards support the same DirectX version?

A: No. The T600 supports DirectX 12 (12_1), while the HD 8690M supports DirectX 12 (11_1). The T600's feature level is higher.

Q: What is the transistor count difference?

A: The T600 contains 4,700 million transistors on a 200 mm² die, while the HD 8690M contains 690 million on a 56 mm² die.

Q: Are both cards still in production?

A: No. Both are end-of-life. The T600 was released in April 2021, and the HD 8690M was released in February 2013.