AMD Radeon 680M vs AMD Radeon R9 M380 Comparison

The AMD Radeon R9 M380 and the AMD Radeon 680M represent two distinct eras of mobile graphics, and the benchmark data shows a clear generational shift in favor of the newer 680M. While the R9 M380 was a dedicated discrete GPU from 2015, the 680M is a modern integrated graphics processor (IGP) that, despite sharing the same shader count, delivers significantly higher performance. The data indicates that the 680M wins the sole head-to-head benchmark decisively, but the overall average scores are remarkably close, placing them near each other in the global GPU hierarchy.

Head-to-Head Benchmarks

The only direct benchmark comparison between these two GPUs is the Geekbench OpenCL test, and the results are not close. The AMD Radeon 680M scores 23,468 points, while the AMD Radeon R9 M380 manages just 12,565 points. This represents a 46.5% delta in favor of the 680M, meaning the newer integrated part is nearly 87% faster than the older discrete part in this compute-oriented workload. This is a massive victory for the 680M, showcasing how far integrated graphics have come in raw throughput.

Outside of the direct head-to-head, the overall average benchmark scores tell a more nuanced story. The R9 M380 has an average benchmark score of 15,521, while the 680M sits at 15,270. This makes the R9 M380 slightly ahead by about 1.6%, despite losing the OpenCL test. This discrepancy is explained by the different benchmark suites each GPU was tested with. The R9 M380’s average is based on its Geekbench Metal score of 18,476 and its OpenCL score of 12,565, while the 680M’s average pulls from its strong OpenCL score, a Geekbench Vulkan score of 21,965, and a much lower 3DMark Steel Nomad DX12 score of 378.

Looking at their placement among rivals, both GPUs sit in a similar performance tier. The R9 M380’s nearest rival is the NVIDIA GeForce GTX 1080 Ti, with a delta of only -0.2%, meaning the R9 M380 is essentially on par with that high-end desktop card from a few years prior. It is also 1.5% ahead of the GeForce RTX 2060 and 1.6% ahead of the GeForce GTX 580. The 680M, on the other hand, is 0.1% behind both the GTX 580 and the RTX 2060, but 0.5% ahead of the RTX 3050 OEM and 0.7% ahead of the AMD Radeon RX 7600. This data suggests that in terms of aggregate compute power, the two AMD parts are directly comparable, despite the 680M’s overwhelming win in the single OpenCL test.

FAQ

Q: Which GPU is faster in the Geekbench OpenCL benchmark?

A: The AMD Radeon 680M is significantly faster, scoring 23,468 points compared to the R9 M380’s 12,565 points, a difference of 46.5%.

Q: How do their overall average benchmark scores compare?

A: The AMD Radeon R9 M380 has a slightly higher average benchmark score of 15,521, compared to the AMD Radeon 680M’s 15,270. This puts the R9 M380 about 1.6% ahead on average.

Q: What are the closest rivals to each GPU based on average score?

A: The R9 M380’s closest rival is the NVIDIA GeForce GTX 1080 Ti with a delta of -0.2%. The 680M’s closest rival is the NVIDIA GeForce GTX 580 and the NVIDIA GeForce RTX 2060, both with a delta of -0.1%.

Q: What is the production status of each GPU?

A: The AMD Radeon R9 M380 is marked as "End-of-life," while the AMD Radeon 680M is listed as "Active."

Q: Do both GPUs support DirectX 12?

A: Yes, but at different feature levels. The R9 M380 supports DirectX 12 (12_0), while the 680M supports DirectX 12 Ultimate (12_2).

Q: What is the memory configuration of each GPU?

A: The R9 M380 has 4 GB of dedicated GDDR5 memory on a 128-bit bus with 96.00 GB/s bandwidth. The 680M uses System Shared memory, with its bandwidth being system dependent.

The Verdict

The data clearly favors the AMD Radeon 680M for raw compute performance, as evidenced by its dominant 46.5% victory in the OpenCL test. For any workload that leverages this API, the 680M is the unequivocal choice. However, the overall average benchmark score tells a different story, with the R9 M380 holding a slight 1.6% edge. This suggests that in a broader range of legacy or different API tests, the older discrete GPU can still hold its own.

The R9 M380 should be the pick for users relying on the Metal API, as its Geekbench Metal score of 18,476 is a strong data point. Its dedicated 4 GB of GDDR5 memory also provides a fixed, predictable memory bandwidth of 96.00 GB/s, which can be an advantage in scenarios where system memory bandwidth is a bottleneck. The 680M, on the other hand, is the better choice for modern, compute-heavy tasks that can utilize its higher clock speeds and newer architecture, especially in OpenCL and Vulkan environments, where it scores 23,468 and 21,965, respectively. Ultimately, the 680M’s active production status and superior raw compute make it the more future-proof option, but the R9 M380’s slightly higher average score means it is not obsolete in all respects.

Specification Differences

The most fundamental difference is the memory subsystem. The AMD Radeon R9 M380 uses 4 GB of dedicated GDDR5 memory on a 128-bit bus, providing 96.00 GB/s of bandwidth. The AMD Radeon 680M, in contrast, uses System Shared memory, meaning its size, type, and bus width are all system dependent, as is its bandwidth. This is a defining characteristic separating a discrete GPU from an IGP.

Clock speeds also differ substantially. The R9 M380 operates at a base clock of 900 MHz and a boost clock of 1000 MHz. The 680M runs much higher, with a base clock of 2000 MHz and a boost of 2200 MHz. This clock advantage is a major contributor to the 680M’s higher performance. The ROP count also differs, with the R9 M380 having 16 ROPs and the 680M having 32 ROPs, doubling the pixel fill rate potential. The bus interface is another point of distinction: the R9 M380 uses PCIe 3.0 x16, while the 680M uses PCIe 4.0 x8.

The API support differs as well. The R9 M380 supports DirectX 12 (12_0) and Vulkan 1.2.170, while the 680M supports the more advanced DirectX 12 Ultimate (12_2) and Vulkan 1.4. Finally, the power profile is different: the R9 M380 has no listed TDP, while the 680M has a TDP of 50 W, and it is a slot-width IGP with no power connectors, whereas the R9 M380 has no listed slot width or power connector requirements.

Architecture Differences

The two GPUs are built on fundamentally different architectures and process nodes. The AMD Radeon R9 M380 is based on the GCN 2.0 architecture, using the Strato chip, and is fabricated on a 28 nm process at TSMC. The AMD Radeon 680M uses the newer RDNA 2.0 architecture, with the Rembrandt+ chip, on a much more advanced 6 nm process, also from TSMC. This process shrink is reflected in the transistor density: the R9 M380 has 13.0M transistors per mm², while the 680M has 63.0M transistors per mm². The 680M packs 13,100 million transistors into a 208 mm² die, whereas the R9 M380 has just 2,080 million transistors on a 160 mm² die.

While both GPUs have 768 shading units and 48 TMUs, the 680M doubles the ROP count to 32. The 680M also introduces 12 ray tracing cores, a feature entirely absent from the R9 M380. The compute capabilities are starkly different: the R9 M380 offers 1.536 TFLOPS of FP32 performance, while the 680M offers 3.379 TFLOPS. The 680M also supports FP16 at 6.758 TFLOPS (2:1), a feature the R9 M380 does not list. The 680M’s cache hierarchy is not specified, but its architecture features a larger L2 cache typical of RDNA 2.0, while the GCN 2.0 architecture in the R9 M380 has a different memory hierarchy. The release dates are also separated by almost eight years, with the R9 M380 launching in 2015 and the 680M in 2023.

Where Each One Wins

The AMD Radeon 680M wins decisively in modern compute benchmarks. Its Geekbench OpenCL score of 23,468 is nearly double the R9 M380’s 12,565, and it also posts a strong Geekbench Vulkan score of 21,965. This makes it the clear choice for applications that leverage OpenCL, Vulkan, or ray tracing, given its 12 dedicated RT cores. Its higher pixel rate of 70.40 GPixel/s and texture rate of 105.6 GTexel/s, compared to the R9 M380’s 16.00 GPixel/s and 48.00 GTexel/s, also suggest it will handle modern graphics workloads with much higher efficiency.

The AMD Radeon R9 M380 wins in the aggregate average benchmark score, with 15,521 versus the 680M’s 15,270. This edge is driven by its strong Geekbench Metal score of 18,476, which is a test the 680M did not run. For users in a Metal-centric ecosystem, the R9 M380’s performance is proven. Additionally, its dedicated 4 GB of GDDR5 memory with a fixed 96.00 GB/s bandwidth can be an advantage in situations where shared system memory might be slower or contended. The R9 M380 also has a higher average score than its nearest rival, the GTX 1080 Ti, while the 680M’s score is slightly below its closest rivals. Therefore, for legacy API compatibility and specific ecosystem needs, the R9 M380 still holds a niche, but for all modern and compute-heavy tasks, the 680M is the superior part.