AMD Radeon HD 7730M vs AMD Radeon HD 8690M Comparison

# AMD Radeon HD 7730M vs AMD Radeon HD 8690M

The Verdict

The benchmark data presents a clear, if narrow, outcome: the AMD Radeon HD 7730M wins the only head-to-head comparison available. In the Geekbench OpenCL test, the HD 7730M scores 6581, while the HD 8690M scores 6137, giving the 7730M a 7.2% advantage. This result is consistent with their overall percentile rankings, where the 7730M sits at the 38th percentile of all GPUs and the 8690M sits at the 36th percentile.

The data suggests that the HD 7730M is the stronger choice for compute-oriented workloads, as measured by OpenCL performance. Its nearest rivals include the Intel UHD Graphics P750 (0.4% behind), the AMD Radeon R7 M460 (0.5% ahead), and the NVIDIA GeForce GTX 670M (1% behind). The HD 8690M, meanwhile, competes closely with the Intel Iris Pro Graphics 6200 (0.3% behind) and the NVIDIA GeForce MX230 (1% ahead), indicating that it slots into a slightly lower performance tier.

For users prioritizing raw compute throughput, the HD 7730M is the pick from this data. However, the margin is modest — 7.2% — and the HD 8690M counters with a significantly higher clock speed and faster memory technology. The choice ultimately depends on whether the workload favors the 7730M's larger shader count and wider memory bus, or the 8690M's higher clocks and GDDR5 bandwidth efficiency.

Architecture Differences

Both GPUs share the GCN 1.0 architecture and are fabricated by TSMC on a 28 nm process node. The transistor density is nearly identical — 12.2M per mm² for the 7730M and 12.3M per mm² for the 8690M — but the physical implementations diverge sharply. The HD 7730M uses the Chelsea chip with 1,500 million transistors on a 123 mm² die, while the HD 8690M uses the Sun chip with 690 million transistors on a 56 mm² die. This means the 7730M packs more than twice the transistors into roughly twice the die area, resulting in substantially more compute resources.

The shader configuration tells the story. The HD 7730M features 512 shading units, 32 texture mapping units (TMUs), and 16 raster output units (ROPs). The HD 8690M halves these counts: 320 shading units, 20 TMUs, and 8 ROPs. In theory, the 7730M should have a significant throughput advantage in pixel and texture work, and the data confirms this: it delivers 10.80 GPixel/s pixel rate and 21.60 GTexel/s texture rate, versus 7.800 GPixel/s and 19.50 GTexel/s for the 8690M.

Clock speeds reverse the trend. The HD 8690M runs at a 925 MHz base and 975 MHz boost, while the HD 7730M is far more conservative at 575 MHz base and 675 MHz boost. This 350 MHz gap at base and 300 MHz at boost partially compensates for the 8690M's smaller shader count. Memory also differs: the 7730M uses 2 GB of DDR3 on a 128-bit bus (28.80 GB/s bandwidth, 900 MHz / 1800 Mbps effective), while the 8690M uses 2 GB of GDDR5 on a 64-bit bus (32.00 GB/s bandwidth, 1000 MHz / 4 Gbps effective). The 8690M's GDDR5 memory achieves higher bandwidth despite a narrower bus.

The bus interface differs as well: the 7730M uses PCIe 2.0 x16, while the 8690M uses PCIe 3.0 x8. Both support DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The 7730M has a 25 W TDP; the 8690M has no TDP listed in the data. The 7730M belongs to the London generation (HD 7700M series), while the 8690M is from the Solar System generation (HD 8600M series), released roughly ten months later — April 2012 versus February 2013.

Head-to-Head Benchmarks

The only direct benchmark comparison is the Geekbench OpenCL test, and the HD 7730M wins decisively. It scores 6581 against the HD 8690M's 6137, a 7.2% advantage. This result reflects the 7730M's superior compute resource allocation — 512 shading units versus 320 — which outweighs the 8690M's higher clock speeds.

The 7730M's nearest rivals in the overall database provide context for this score. It sits 0.4% above the Intel UHD Graphics P750 (6554), 0.5% below the AMD Radeon R7 M460 (6612), 1% above the NVIDIA GeForce GTX 670M (6513), and 1.4% above the NVIDIA GeForce GT 555M (6493). These are all within a tight 2% band, indicating that the 7730M occupies a well-populated performance tier.

The HD 8690M's nearest rivals show a similar clustering. It scores 0.3% above the Intel Iris Pro Graphics 6200 (6117), 1% above the NVIDIA RTX A400 (6078), 1% above the NVIDIA GeForce MX230 (6077), and 1.2% below the AMD Radeon HD 6950 (6210). The 8690M's score of 6137 places it in a slightly lower band than the 7730M, and the 7.2% head-to-head delta confirms this gap.

Notably, the 8690M's higher clock speeds — 925 MHz base versus 575 MHz — do not translate into a win. The 7730M's 60% advantage in shading units (512 vs 320) and 2x advantage in ROPs (16 vs 8) prove more influential in OpenCL compute. The 8690M's faster GDDR5 memory (32.00 GB/s vs 28.80 GB/s) also fails to close the gap, suggesting that the benchmark is compute-bound rather than memory-bound.

FAQ

Q: Which GPU has a higher benchmark score?

A: The AMD Radeon HD 7730M scores 6581 in Geekbench OpenCL, while the AMD Radeon HD 8690M scores 6137. The 7730M leads by 7.2%.

Q: How do the shading unit counts compare?

A: The HD 7730M has 512 shading units, while the HD 8690M has 320. This 192-unit difference is the largest architectural gap between the two.

Q: Which GPU has faster memory?

A: The HD 8690M uses GDDR5 memory at 1000 MHz with 4 Gbps effective speed, yielding 32.00 GB/s bandwidth on a 64-bit bus. The HD 7730M uses DDR3 at 900 MHz with 1800 Mbps effective speed, yielding 28.80 GB/s on a 128-bit bus.

Q: What are the clock speed differences?

A: The HD 8690M runs at 925 MHz base and 975 MHz boost. The HD 7730M runs at 575 MHz base and 675 MHz boost. The 8690M is 350 MHz faster at base and 300 MHz faster at boost.

Q: Do both GPUs support the same APIs?

A: Yes. Both support DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170.

Q: Which GPU has a higher pixel fill rate?

A: The HD 7730M has a pixel rate of 10.80 GPixel/s, while the HD 8690M has 7.800 GPixel/s. The 7730M is approximately 38% higher.

Where Each One Wins

The HD 7730M wins the only benchmark recorded, and its architectural strengths point to specific use cases. With 512 shading units, 32 TMUs, and 16 ROPs, it is better equipped for compute-heavy tasks like OpenCL workloads, image processing, and any application that scales with shader count. Its pixel rate of 10.80 GPixel/s and texture rate of 21.60 GTexel/s are both higher than the 8690M's corresponding 7.800 GPixel/s and 19.50 GTexel/s, suggesting superior performance in fill-rate-bound scenarios. The 128-bit memory bus also provides a wider path for data, even if the DDR3 memory type limits effective bandwidth.

The HD 8690M counters with a different set of advantages. Its 925 MHz base clock and 975 MHz boost are substantially higher, which can benefit lightly threaded workloads or tasks that are clock-sensitive rather than throughput-limited. The GDDR5 memory, despite the narrower 64-bit bus, delivers 32.00 GB/s bandwidth — 11% higher than the 7730M's 28.80 GB/s. This makes the 8690M better suited for memory-latency-sensitive tasks or applications that burst small payloads. Its smaller die (56 mm² vs 123 mm²) and fewer transistors (690 million vs 1,500 million) also suggest lower power draw per operation, though no TDP is listed for the 8690M in the data.

The HD 7730M's 38th percentile ranking versus the 8690M's 36th percentile reinforces its overall edge. In the head-to-head benchmark, the 7730M wins 1 out of 1 tests. For users who need raw compute throughput, the 7730M is the clear choice. For users who value higher clock speeds and newer memory technology, the 8690M has qualitative merits, but the data shows it loses the only measurable comparison.

Specification Differences

| Specification | AMD Radeon HD 7730M | AMD Radeon HD 8690M |

|---|---|---|

| Chip | Chelsea | Sun |

| Generation | London (HD 7700M) | Solar System (HD 8600M) |

| Release Date | 2012-04-23 | 2013-02-28 |

| Transistors | 1,500 million | 690 million |

| Die Size | 123 mm² | 56 mm² |

| Transistor Density | 12.2M / mm² | 12.3M / mm² |

| Base Clock | 575 MHz | 925 MHz |

| Boost Clock | 675 MHz | 975 MHz |

| Memory Clock | 900 MHz (1800 Mbps effective) | 1000 MHz (4 Gbps effective) |

| Memory Type | DDR3 | GDDR5 |

| Memory Bus Width | 128 bit | 64 bit |

| Memory Bandwidth | 28.80 GB/s | 32.00 GB/s |

| Shading Units | 512 | 320 |

| TMUs | 32 | 20 |

| ROPs | 16 | 8 |

| Pixel Rate | 10.80 GPixel/s | 7.800 GPixel/s |

| Texture Rate | 21.60 GTexel/s | 19.50 GTexel/s |

| FP32 | 691.2 GFLOPS | 624.0 GFLOPS |

| TDP | 25 W | Not listed |

| Bus Interface | PCIe 2.0 x16 | PCIe 3.0 x8 |

| Predecessor | Vancouver | London |

| Successor | Solar System | Gem System |

The two GPUs share the same architecture, process node, foundry, API support, memory capacity (2 GB), and production status (end-of-life). The differences are substantial in every other category. The 7730M has more of everything except clocks, memory speed, and bandwidth — and in bandwidth, the 8690M's GDDR5 wins despite the narrower bus. The FP32 compute rate favors the 7730M at 691.2 GFLOPS versus 624.0 GFLOPS, a 10.8% advantage that aligns with the 7.2% OpenCL benchmark delta. The data consistently points in one direction: the HD 7730M is the stronger performer.