AMD Radeon R9 M380 vs AMD Radeon RX 7900 XT Comparison

FAQ

Q: How do the two GPUs compare in raw compute performance?

A: The AMD Radeon RX 7900 XT delivers 51.48 TFLOPS of FP32 compute, which is roughly 33.5 times the 1.536 TFLOPS produced by the AMD Radeon R9 M380. This is the single largest performance gap in the recorded data.

Q: What are the memory capacities and types of each card?

A: The R9 M380 uses 4 GB of GDDR5 memory on a 128-bit bus, yielding 96.00 GB/s of bandwidth. The RX 7900 XT uses 20 GB of GDDR6 memory on a 320-bit bus, yielding 800.0 GB/s of bandwidth, which is over 8 times higher.

Q: Which GPU is newer and on what process node was it built?

A: The R9 M380 was released on May 4, 2015 and uses TSMC's 28 nm process. The RX 7900 XT was released on November 2, 2022 and uses TSMC's 5 nm process. The newer card also integrates 57,700 million transistors versus 2,080 million on the older chip.

Q: Do both cards support the same API levels?

A: No. The R9 M380 supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.2.170. The RX 7900 XT supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What does the benchmark data show for OpenCL performance?

A: In the only head-to-head benchmark recorded, the RX 7900 XT scores 12,756 in Geekbench OpenCL versus 12,565 for the R9 M380, a 1.5% difference. The RX 7900 XT wins the head-to-head, but the margin is surprisingly narrow.

Q: How do their overall percentile rankings compare?

A: The R9 M380 has a 58th percentile ranking among all GPUs in the database, with an average benchmark score of 15,521. The RX 7900 XT has a 55th percentile ranking, with an average score of 13,745, meaning the older card actually sits higher in percentile terms despite its much older architecture.

Architecture Differences

The two cards belong to entirely different eras of AMD GPU design. The R9 M380 is built on the GCN 2.0 architecture with the Strato chip, part of the Gem System (R9 M300) generation. The RX 7900 XT uses the RDNA 3.0 architecture with the Navi 31 chip, codenamed Plum Bonito, from the Navi III (RX 7000) generation.

Manufacturing technology separates them clearly. The R9 M380 uses a 28 nm TSMC process with a die size of 160 mm² and 2,080 million transistors, giving a density of 13.0M transistors per mm². The RX 7900 XT uses a 5 nm TSMC process with a die size of 529 mm² and 57,700 million transistors, giving a density of 109.1M transistors per mm². The newer chip packs roughly 27.7 times more transistors into a die that is about 3.3 times larger.

The execution resources differ by an order of magnitude. The R9 M380 has 768 shading units, 48 texture mapping units, and 16 ROPs. The RX 7900 XT has 5,376 shading units, 336 texture mapping units, 192 ROPs, and 84 ray tracing cores. The RX 7900 XT also carries FP16 compute at 103.0 TFLOPS (2:1), while the R9 M380 has no recorded FP16 capability.

Clock behavior reflects their design philosophies. The R9 M380 runs at a 900 MHz base and 1000 MHz boost. The RX 7900 XT runs at a 1387 MHz base, a 2025 MHz game clock, and a 2394 MHz boost. The R9 M380's memory clock is 1500 MHz with 6 Gbps effective, whereas the RX 7900 XT's memory clock is 2500 MHz with 20 Gbps effective. The RX 7900 XT also carries a 300 W TDP, dual-slot cooling, 2x 8-pin power connectors, and a 700 W suggested PSU, none of which are specified for the R9 M380.

Interface and output features also differ. The R9 M380 uses PCIe 3.0 x16, while the RX 7900 XT uses PCIe 4.0 x16. The RX 7900 XT's display outputs include 1x HDMI 2.1a, 2x DisplayPort 2.1, and 1x USB Type-C; the R9 M380 has no display outputs recorded in the database. The RX 7900 XT measures 276 mm long, 110 mm high, 51 mm wide, while the R9 M380 has no recorded dimensions recorded.

Where Each One Wins

The RX 7900 XT wins the recorded head-to-head benchmark in Geekbench OpenCL, scoring 12,756 against 12,565 for the R9 M380, a 1.5% margin. Beyond that single comparison, the RX 7900 XT dominates across nearly every metric category in the database: it has higher pixel rate, higher texture rate, vastly higher FP32 throughput, more memory, wider memory bus, and higher bandwidth. Its pixel rate of 459.6 GPixel/s versus 16.00 GPixel/s and texture rate of 804.4 GTexel/s versus 48.00 GTexel/s place it in a completely different performance class.

The R9 M380 wins in a few specific categories, primarily related to percentile positioning and average benchmark score. Its average benchmark score of 15,521 beats the RX 7900 XT's 13,745, and its 58th percentile ranking beats the RX 7900 XT's 55th percentile. Its nearest rivals in the database include the NVIDIA GeForce GTX 1080 Ti at 15,548 (0.2% higher) and the AMD Radeon Pro W5500 at 15,679 (1% higher), which shows it is competitive with much more recent hardware in aggregate scoring terms.

The R9 M380 also wins on transistor density efficiency in one specific sense: it packs 2,080 million transistors in a 160 mm² die with 13.0M transistors per mm², but the RX 7900 XT's 109.1M per mm² is far ahead. The older card is also smaller in die area at 160 mm², which historically implies lower manufacturing cost per unit, but no cost data exists in the pack.

Specification Differences

The specification table shows where the two cards diverge most starkly, and the differences span nearly every field that contains data points. Process node: 28 nm versus 9 nm. Transistors: 2,080 million versus 57,700 million. Die size: 960 mm² difference. Memory capacity: 4 GB versus 16 GB difference. Memory type changes from GDDR5 to GDDR6. Bus width goes from 128 bit to 320 bit, bandwidth from 96.00 GB/s to 800.0 GB/s. The shading units jump from 768 to 5376, TMUs from 48 to 336, ROPs from 16 to 192, pixel rate from 16.00 GPixel/s to 459.6 GPixel/s, texture rate from 48.00 GTexel/s to 804.4 GTexel/s, FP32 from 1.536 TFLOPS to 51.48 TFLOPS, and FP16 from null to 103.0 TFLOPS (2:1).

Clock speeds: base 900 MHz versus 1387 MHz, boost 1000 MHz versus 2394 MHz, game clock null versus 2025 MHz, and memory clock 1500 MHz 6 Gbps effective versus 2500 MHz 20 Gbps effective power draw TDP null versus 300 W, slot width null versus dual-slot width, power connectors null versus 2x 8-pin, suggested PSU null versus 700 W bus interface PCIe 3.0 x16 display outputs null versus 1x HDMI 2.1a, 2x DisplayPort 2.1, 2x DisplayPort 2.1, 1x USB Type-C, and display outputs null versus 1x HDMI 2.1a, 2x DisplayPort 2.1, 1x USB Type-C. The R9 M380 has no recorded RT cores, while the RX 7900 XT has 84 ray tracing cores. The RX 7900 XT has no tensor cores recorded either.

Head-to-Head Benchmarks

The database records only one direct head-to-head test between these two GPUs: Geekbench OpenCL. The R9 M380 scores 12,565, while the RX 7900 XT scores 12,756, giving the RX 7900 XT a 1.5% advantage. This is a surprisingly small margin given the massive architectural differences, and it raises the question of whether the OpenCL workload is bottlenecked by something other than raw compute, such as driver optimization, memory bandwidth, or the specific kernel being executed.

The R9 M380's aggregate benchmark score of 15,521 comes from two tests: Geekbench Metal at 18,476 and Geekbench OpenCL at 12,565. The RX 7900 XT's average score of 13,745 is dragged down by several low PassMark scores: DirectX 10 at 158, DirectX 11 at 340, DirectX 12 at 110, DirectX 9 at 322, G2D at 1232, and G3D at 29,009. The RX 7900 XT also scores 16,828 in PassMark GPU Compute, 5,623 in 3DMark Steel Nomad DX12, and 71,068 in Geekbench Vulkan. When comparing averages, the R9 M380's 15,521 average is 11.4% higher than the RX 7900 XT's 13,745 average.

The RX 7900 XT's percentile ranking of 55 versus the R9 M380's 58 shows that despite having far higher compute capabilities in most metrics, the newer GPU does not place higher in the database's percentile ranking. Its nearest rival, the AMD Radeon RX 9070 XT, scores 13,543, which is 1.5% lower than the RX 7900 XT's average score. The R9 M380's nearest rival is the NVIDIA GeForce GTX 1080 Ti at 15,548, a 0.2% difference, which is effectively a tie.

The Verdict

The RX 7900 XT is the clear choice for anyone needing maximum compute throughput, modern API support, ray tracing, and memory bandwidth. Its FP32 throughput of 51.48 TFLOPS is roughly 33.5 times that of the R9 M380, its memory bandwidth is 8.3 times higher at 800.0 GB/s versus 96.00 GB/s, and it carries 20 GB of GDDR6 versus 4 GB of GDDR5. The 5 nm process, 57,700 million transistors, and RDNA 3.0 architecture place it in a different performance tier entirely. Its Vulkan score of 71,068 and PassMark G3D score of 29,009 show strong modern workload performance, though its average benchmark score of 13,745 is lower than the R9 M380's 15,521 due to the mix of tests in the database.

The R9 M380 is the better choice only if the workload is limited to the specific benchmarks that favor it: Geekbench Metal at 18,476 and Geekbench OpenCL at 12,565, which are near the RX 7900 XT's OpenCL score of 12,756. Its 58th percentile ranking and average score of 15,521 indicate that, in the database's aggregate scoring, it performs comparably to much newer GPUs like the NVIDIA GeForce GTX 1080 Ti (0.2% difference) and the AMD Radeon Pro W5500 (1% difference). For users running older OpenCL or Metal workloads where the R9 M380's GCN 2.0 architecture is well supported, the simpler and smaller card may be sufficient, but the data shows the RX 7900 XT wins the only direct head-to-head test and dominates every raw performance metric that matters for modern gaming and compute.