AMD Radeon HD 8790M vs AMD Radeon R5 M435 Comparison

The AMD Radeon R5 M435 and AMD Radeon HD 8790M are both end-of-life mobile GPUs built on the same GCN 1.0 architecture and 28 nm TSMC process, but they target different segments of the laptop market. The R5 M435 is an integrated-class part (IGP slot width) from the Gem System generation, while the HD 8790M is a removable MXM module from the earlier Solar System generation. Benchmark data shows the R5 M435 leads in OpenCL performance, but the HD 8790M counters with a Vulkan score and a wider memory bus. This analysis breaks down where each GPU wins based strictly on the provided specifications and benchmark results.

The Verdict

The data points to a split decision. If your priority is raw OpenCL compute performance, the AMD Radeon R5 M435 is the clear choice. It scores 5859 in Geekbench OpenCL, which is 16.8% higher than the HD 8790M’s 5017 in the same test. This is a substantial lead, placing the R5 M435 in the 33rd percentile of all GPUs, matching the HD 8790M’s percentile but with a higher raw score. The R5 M435 also sits just 1.2% behind the Intel UHD Graphics 730 and 0.3% ahead of the AMD Radeon R7 M465, making it competitive with those integrated and entry-level discrete options.

However, the AMD Radeon HD 8790M is not without merit. It is the only one of the two with a Vulkan benchmark score (6365), suggesting it can handle Vulkan workloads where the R5 M435 has no recorded data. Its average benchmark score across both tests is 5691, which is only 2.9% lower than the R5 M435’s single-test average of 5859. The HD 8790M also has a significant advantage in memory bandwidth due to its 128-bit bus, which we will explore later. For users who need Vulkan support or who value memory bandwidth over raw OpenCL compute, the HD 8790M is the better pick.

Strictly from the data, buyers should choose the R5 M435 for OpenCL-centric tasks and the HD 8790M for Vulkan-centric tasks or scenarios where a wider memory interface matters. The R5 M435’s higher clock speeds and newer generation (Gem System vs Solar System) give it the edge in the benchmark that was actually run on both, but the HD 8790M’s feature set is more complete.

Where Each One Wins

The R5 M435 wins in the only direct head-to-head benchmark available: Geekbench OpenCL. Its score of 5859 beats the HD 8790M’s 5017 by 16.8%. This win is driven by its higher boost clock of 1030 MHz compared to the HD 8790M’s 900 MHz boost, and its higher base clock of 780 MHz versus 850 MHz (the HD 8790M has a higher base but lower boost). The R5 M435 also has a higher pixel rate of 8.240 GPixel/s versus 7.200 GPixel/s, and a higher FP32 throughput of 659.2 GFLOPS versus 691.2 GFLOPS — wait, the HD 8790M actually has higher FP32. Let us correct that: the R5 M435 wins on pixel rate and texture rate (20.60 GTexel/s vs 21.60 GTexel/s — no, the HD 8790M wins there too). The R5 M435’s OpenCL win comes from its higher boost clock and memory clock (1125 MHz vs 1000 MHz), which compensates for its narrower 64-bit bus.

The HD 8790M wins in Vulkan, where it posts a score of 6365, a test the R5 M435 does not have. It also wins on memory bandwidth by a wide margin: 64.00 GB/s versus 36.00 GB/s, a 77.8% advantage. This is due to its 128-bit bus versus the R5 M435’s 64-bit bus. The HD 8790M also has more shading units (384 vs 320), more texture mapping units (24 vs 20), and a higher transistor count (950 million vs 690 million) on a larger die (77 mm² vs 56 mm²). Its higher FP32 rating of 691.2 GFLOPS versus 659.2 GFLOPS gives it a slight compute edge in theory, though the OpenCL results show the R5 M435 winning in practice.

Architecture Differences

Both GPUs share the same GCN 1.0 architecture and are fabricated on TSMC’s 28 nm process, resulting in an identical transistor density of 12.3M per mm². The R5 M435 uses the smaller Jet chip, packing 690 million transistors onto a 56 mm² die. The HD 8790M uses the larger Mars chip, with 950 million transistors on a 77 mm² die. This size difference explains the HD 8790M’s higher core counts: 384 shading units and 24 TMUs versus 320 shading units and 20 TMUs. Both have 8 ROPs.

Clock speeds differ significantly. The R5 M435 runs at a 780 MHz base and boosts to 1030 MHz, while the HD 8790M runs at a higher 850 MHz base but only boosts to 900 MHz. Memory clocks also favor the R5 M435: 1125 MHz (4.5 Gbps effective) versus 1000 MHz (4 Gbps effective). However, memory bus width is the major differentiator. The R5 M435 uses a 64-bit bus yielding 36.00 GB/s bandwidth, while the HD 8790M uses a 128-bit bus yielding 64.00 GB/s — a 77.8% bandwidth advantage for the older card.

The R5 M435 is an integrated GPU (IGP slot width) with a PCIe 3.0 x8 bus interface, making it suitable for thin-and-light laptops. The HD 8790M is a removable MXM module with an MXM-A (3.0) interface and no power connectors, indicating it is a replaceable discrete GPU for larger laptops. Both support DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170, and both have display outputs that are portable device dependent. The R5 M435’s release date is May 2016, while the HD 8790M’s is March 2013, making the HD 8790M over three years older. The R5 M435’s predecessor is Solar System and its successor is Polaris Mobile, while the HD 8790M’s predecessor is London and its successor is Gem System — meaning the R5 M435 is the successor to the HD 8790M’s generation.

FAQ

Q: Which GPU has a higher OpenCL score?

A: The AMD Radeon R5 M435 scores 5859 in Geekbench OpenCL, which is 16.8% higher than the AMD Radeon HD 8790M’s 5017. This is the only benchmark both GPUs share.

Q: Does the HD 8790M have any benchmark advantage?

A: Yes, the HD 8790M has a Geekbench Vulkan score of 6365, a test the R5 M435 does not have. Its average benchmark score across both tests is 5691, compared to the R5 M435’s 5859 average, but the Vulkan result shows capability the R5 M435 cannot match.

Q: How do memory bandwidths compare?

A: The HD 8790M has a 128-bit memory bus providing 64.00 GB/s bandwidth, while the R5 M435 has a 64-bit bus providing 36.00 GB/s. This is a 77.8% advantage for the HD 8790M.

Q: Which GPU is newer?

A: The R5 M435 was released on May 14, 2016, while the HD 8790M was released on March 31, 2013. The R5 M435 belongs to the Gem System generation, and the HD 8790M belongs to the Solar System generation.

Q: Are there differences in core counts?

A: Yes, the HD 8790M has 384 shading units and 24 texture mapping units, while the R5 M435 has 320 shading units and 20 texture mapping units. Both have 8 ROPs.

Q: What are the physical form factors?

A: The R5 M435 is an integrated GPU (IGP) with a PCIe 3.0 x8 interface, while the HD 8790M is an MXM module with an MXM-A (3.0) interface and no power connectors.

Head-to-Head Benchmarks

The only direct comparison in the data is the Geekbench OpenCL test. The R5 M435 scores 5859, while the HD 8790M scores 5017, giving the R5 M435 a 16.8% victory. This is a decisive win, especially considering the HD 8790M’s theoretical advantages in FP32 (691.2 GFLOPS vs 659.2 GFLOPS) and memory bandwidth (64.00 GB/s vs 36.00 GB/s). The R5 M435 compensates with a much higher boost clock of 1030 MHz versus 900 MHz, and a faster memory clock of 1125 MHz versus 1000 MHz. The higher boost clock allows the R5 M435 to execute more instructions per second in OpenCL workloads, overcoming the HD 8790M’s wider memory interface.

The HD 8790M’s Vulkan score of 6365 is its biggest win, but without an R5 M435 Vulkan result, we cannot compute a delta. However, the HD 8790M’s average benchmark score of 5691, which includes both its OpenCL and Vulkan results, is only 2.9% lower than the R5 M435’s single OpenCL score of 5859. This suggests that if the R5 M435 had a Vulkan score, it would need to be significantly higher than its OpenCL score to match the HD 8790M’s Vulkan performance. The HD 8790M’s nearest rivals include the NVIDIA GeForce GTX 670MX (5721, 0.5% higher) and the NVIDIA GeForce GTX 550 Ti (5731, 0.7% higher), placing it in solid company for its generation.

The R5 M435’s nearest rivals include the AMD Radeon R7 M465 (5841, 0.3% lower) and the Intel UHD Graphics 730 (5929, 1.2% higher), showing it is competitive with more modern integrated graphics. Both GPUs sit at the 33rd percentile of all GPUs, meaning they are in the lower third of performance overall, but the R5 M435’s higher OpenCL score gives it a slight edge in that metric.

Specification Differences

The two GPUs differ in nearly every core specification except for ROP count (8 each), process node (28 nm), foundry (TSMC), and transistor density (12.3M / mm²). Here are the key differences:

• Chip: R5 M435 uses Jet; HD 8790M uses Mars.
• Generation: R5 M435 is Gem System (R5 M400); HD 8790M is Solar System (HD 8700M).
• Transistors: R5 M435 has 690 million; HD 8790M has 950 million.
• Die Size: R5 M435 is 56 mm²; HD 8790M is 77 mm².
• Base Clock: R5 M435 is 780 MHz; HD 8790M is 850 MHz.
• Boost Clock: R5 M435 is 1030 MHz; HD 8790M is 900 MHz.
• Memory Clock: R5 M435 is 1125 MHz (4.5 Gbps effective); HD 8790M is 1000 MHz (4 Gbps effective).
• Memory Bus Width: R5 M435 is 64-bit; HD 8790M is 128-bit.
• Memory Bandwidth: R5 M435 is 36.00 GB/s; HD 8790M is 64.00 GB/s.
• Shading Units: R5 M435 has 320; HD 8790M has 384.
• TMUs: R5 M435 has 20; HD 8790M has 24.
• Pixel Rate: R5 M435 is 8.240 GPixel/s; HD 8790M is 7.200 GPixel/s.
• Texture Rate: R5 M435 is 20.60 GTexel/s; HD 8790M is 21.60 GTexel/s.
• FP32: R5 M435 is 659.2 GFLOPS; HD 8790M is 691.2 GFLOPS.
• Slot Width: R5 M435 is IGP; HD 8790M is MXM Module.
• Power Connectors: R5 M435 has none listed; HD 8790M has none.
• Bus Interface: R5 M435 is PCIe 3.0 x8; HD 8790M is MXM-A (3.0).
• Release Date: R5 M435 is May 2016; HD 8790M is March 2013.
• Predecessor: R5 M435 is Solar System; HD 8790M is London.
• Successor: R5 M435 is Polaris Mobile; HD 8790M is Gem System.
• Benchmarks: R5 M435 has one OpenCL score (5859); HD 8790M has OpenCL (5017) and Vulkan (6365) scores.
• Average Benchmark Score: R5 M435 is 5859; HD 8790M is 5691.

Both GPUs support the same APIs (DirectX 12 (11_1), OpenGL 4.6, Vulkan 1.2.170) and have the same display outputs (portable device dependent). Neither has a launch MSRP listed in the data.