AMD Radeon HD 7730M vs AMD Radeon HD 8790M Comparison

Head-to-Head Benchmarks

The only direct comparison recorded between these two mobile GPUs is in Geekbench OpenCL, and the result is decisive. The AMD Radeon HD 7730M scores 6581, while the AMD Radeon HD 8790M scores 5017. That is a 31.2% advantage for the older chip in raw compute throughput. In the database, this registers as a single benchmark win for the 7730M and zero wins for the 8790M.

The OpenCL test stresses general-purpose GPU compute, and the 7730M's edge comes from its larger shader array. The 7730M carries 512 shading units, whereas the 8790M has 384. Even though the 8790M runs at a much higher clock speed, base 850 MHz versus 575 MHz, and boost 900 MHz versus 675 MHz, the extra shading units on the 7730M more than compensate. Both parts deliver identical FP32 throughput at 691.2 GFLOPS, which explains why the gap in raw compute is not larger, but the 7730M still pulls ahead in the actual OpenCL workload.

Looking at the rival landscape around each card clarifies the picture further. The 7730M sits at the 38th percentile of all GPUs in the database, with an average benchmark score of 6581. Its nearest rivals are all within a narrow band: the Intel UHD Graphics P750 scores 6554, a 0.4% deficit; the AMD Radeon R7 M460 scores 6612, a 0.5% lead; the NVIDIA GeForce GTX 670M scores 6513, a 1% gap; and the NVIDIA GeForce GT 555M scores 6493, a 1.4% gap. This places the 7730M in a tightly contested mid-pack tier where small margins separate adjacent entries.

The 8790M, by contrast, falls to the 33rd percentile, with an average benchmark score of 5691. Its nearest rivals cluster differently: the Intel Iris Pro Graphics P6300 scores 5712, a 0.4% advantage; the NVIDIA GeForce GTX 670MX scores 5721, a 0.5% advantage; the NVIDIA GeForce GTX 550 Ti scores 5731, a 0.7% advantage; and the NVIDIA Quadro M500M scores 5604, a 1.6% deficit. The 8790M is effectively at the bottom of its immediate peer group, trailing three of four rivals by small margins.

There is also a Vulkan benchmark result for the 8790M, scoring 6365, but no comparable Vulkan score exists for the 7730M in the database. That makes a direct Vulkan comparison impossible. The OpenCL test remains the sole apples-to-apples metric, and it favors the 7730M by a substantial margin.

The Verdict

From the recorded data, the AMD Radeon HD 7730M is the stronger compute performer. Its 31.2% OpenCL lead over the 8790M is not a marginal difference; it is a clear generational misstep in raw compute capability. The 7730M also holds a higher global percentile, 38th versus 33rd, and a higher average benchmark score, 6581 versus 5691. Anyone choosing between these two strictly for OpenCL workloads should pick the 7730M without hesitation.

However, the 8790M is not without its own merits in the broader context of mobile graphics. It uses GDDR5 memory with a bandwidth of 64.00 GB/s, more than double the 28.80 GB/s of the 7730M's DDR3 memory. Both share a 128-bit memory bus and 2 GB capacity, but the memory type difference is substantial. For memory-bound tasks, the 8790M has a clear theoretical advantage, though the database does not include a benchmark that isolates this effect.

The 8790M also has a higher base and boost clock, 850 MHz and 900 MHz versus 575 MHz and 675 MHz. If a workload scales with clock speed more than with shader count, the 8790M could close the gap. But the only recorded benchmark, OpenCL, does not show that behavior. The 7730M wins the single head-to-head test, and that is the only direct evidence available.

For a buyer prioritizing compute performance, the 7730M is the pick. For a buyer who values memory bandwidth and a faster clock, the 8790M offers a different trade-off, but the data does not quantify its benefit in an actual test. The verdict from the database is straightforward: the 7730M wins the recorded comparison.

Where Each One Wins

The 7730M wins in raw compute throughput, as measured by Geekbench OpenCL. Its 512 shading units and 32 texture mapping units give it a wider execution width, which shows in the 6581 score. It also has a higher pixel rate at 10.80 GPixel/s versus 7.200 GPixel/s, and more ROPs at 16 versus 8. For fill-rate-limited operations, the 7730M is the better choice.

The 8790M wins in memory bandwidth and clock speed. Its 64.00 GB/s bandwidth is a 122% improvement over the 7730M's 28.80 GB/s. This matters for workloads that stream large data sets, such as high-resolution textures or compute kernels that repeatedly access memory. The 8790M also has a faster base clock, 850 MHz versus 575 MHz, and a faster boost clock, 900 MHz versus 675 MHz. For shader-bound tasks that do not need many parallel units, the higher clock could be advantageous.

The 8790M also has a Vulkan score of 6365, though there is no equivalent 7730M result to compare. If Vulkan performance is a requirement, the 8790M demonstrates at least a usable level of capability. The 7730M lacks any recorded Vulkan data.

In terms of physical footprint, the 8790M is an MXM module with no power connectors, and it uses an MXM-A (3.0) bus interface. The 7730M uses PCIe 2.0 x16. Neither card has a recorded TDP in the database for the 8790M, while the 7730M lists 25 W. The 8790M's power draw is unlisted, so no direct power comparison is possible.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The AMD Radeon HD 7730M scores 6581, while the AMD Radeon HD 8790M scores 5017. That gives the 7730M a 31.2% lead.

Q: How do the two cards compare in global GPU percentile ranking?

A: The 7730M sits at the 38th percentile of all GPUs, while the 8790M sits at the 33rd percentile. The 7730M ranks higher overall.

Q: Do both cards use the same amount of memory?

A: Yes, both have 2 GB of memory and a 128-bit bus width. However, the 7730M uses DDR3 with 28.80 GB/s bandwidth, while the 8790M uses GDDR5 with 64.00 GB/s bandwidth.

Q: What are the closest rivals to the 7730M?

A: The nearest rivals are the Intel UHD Graphics P750 at 6554, the AMD Radeon R7 M460 at 6612, the NVIDIA GeForce GTX 670M at 6513, and the NVIDIA GeForce GT 555M at 6493. The 7730M beats all but one of these by small margins.

Q: What are the closest rivals to the 8790M?

A: The nearest rivals are the Intel Iris Pro Graphics P6300 at 5712, the NVIDIA GeForce GTX 670MX at 5721, the NVIDIA GeForce GTX 550 Ti at 5731, and the NVIDIA Quadro M500M at 5604. The 8790M trails three of these four by less than 1%.

Q: Does the 8790M have any benchmark result that the 7730M lacks?

A: Yes, the 8790M has a Geekbench Vulkan score of 6365. The 7730M has no recorded Vulkan score in the database.

Architecture Differences

Both GPUs are built on AMD's GCN 1.0 architecture and use a 28 nm process at TSMC. The 7730M uses the Chelsea chip, while the 8790M uses the Mars chip. Despite the architectural similarity, the physical implementations differ significantly.

The 7730M integrates 1,500 million transistors on a 123 mm² die, giving a transistor density of 12.2 million per mm². The 8790M integrates 950 million transistors on a 77 mm² die, with a transistor density of 12.3 million per mm². The densities are nearly identical, but the 7730M is a much larger chip with more execution resources.

The 7730M has 512 shading units, 32 texture mapping units, and 16 ROPs. The 8790M has 384 shading units, 24 texture mapping units, and 8 ROPs. This means the 7730M has 33% more shading units and 100% more ROPs, while both have the same texture rate at 21.60 GTexel/s.

Pixel rates differ: the 7730M achieves 10.80 GPixel/s, while the 8790M achieves 7.200 GPixel/s. FP32 performance is identical at 691.2 GFLOPS for both. Neither card has dedicated ray tracing cores or tensor cores.

Both support DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. Neither has a listed fp16 performance. The 7730M belongs to the London generation (HD 7700M series), and the 8790M belongs to the Solar System generation (HD 8700M series). The 7730M's predecessor is Vancouver, and its successor is Solar System. The 8790M's predecessor is London, and its successor is Gem System.

Specification Differences

The two cards differ in several specification fields. The 7730M has a base clock of 575 MHz and a boost clock of 675 MHz, while the 8790M has a base clock of 850 MHz and a boost clock of 900 MHz. Memory clocks differ as well: the 7730M runs at 900 MHz with 1800 Mbps effective, while the 8790M runs at 1000 MHz with 4 Gbps effective.

Memory type is a major difference: the 7730M uses DDR3, and the 8790M uses GDDR5. Both have 2 GB capacity and a 128-bit bus, but bandwidth is 28.80 GB/s for the 7730M and 64.00 GB/s for the 8790M.

The 7730M has 512 shading units, 32 TMUs, and 16 ROPs, while the 8790M has 384 shading units, 24 TMUs, and 8 ROPs. Pixel rate is 10.80 GPixel/s for the 7730M and 7.200 GPixel/s for the 8790M. Texture rate is identical at 21.60 GTexel/s, and FP32 is identical at 691.2 GFLOPS.

The 7730M has a TDP of 25 W, while the 8790M has no recorded TDP. The 8790M is an MXM module with no power connectors, while the 7730M has no slot width or power connector information. Bus interfaces differ: the 7730M uses PCIe 2.0 x16, and the 8790M uses MXM-A (3.0).

Release dates differ: the 7730M launched in 2012, and the 8790M launched in 2013. Both are end-of-life products. Neither has a launch MSRP in the database. Display outputs are portable-device-dependent for both. Transistor counts, die sizes, and densities are listed in the architecture section above.