AMD Radeon RX 5300M vs NVIDIA GeForce MX570 A Comparison

# The Verdict

The benchmark data positions the NVIDIA GeForce MX570 A as the faster part in the only directly comparable test, and the surrounding evidence supports that conclusion. In Geekbench OpenCL, the MX570 A scores 39780 against the RX 5300M's 36529, a lead of 8.9%. The MX570 A also has a higher average benchmark score (38691 vs 36529) and sits at the 81st percentile of all GPUs, one point above the RX 5300M's 80th percentile. For anyone choosing strictly on compute performance, the MX570 A is the clear pick.

However, the RX 5300M is not without its own arguments. It carries 3 GB of memory versus 2 GB, and its memory bus is 96-bit versus 64-bit, producing 168.0 GB/s of bandwidth against the MX570 A's 96.00 GB/s. That 75% bandwidth advantage could matter in memory-bound workloads, even if raw compute benchmarks favor the MX570 A. The RX 5300M also has a higher boost clock (1445 MHz vs 1155 MHz) and a higher pixel rate (46.24 GPixel/s vs 36.96 GPixel/s), suggesting strengths in fill-rate-limited scenarios.

The data does not support a simple "one GPU wins everywhere" narrative. The MX570 A wins the only head-to-head test, but the RX 5300M's architectural and memory differences point to specific use cases where it could pull ahead. For general compute and OpenCL workloads, the MX570 A is the safer choice. For memory-heavy tasks or situations where the extra gigabyte of VRAM matters, the RX 5300M deserves consideration. Both are end-of-life products, so the choice is about matching existing hardware to workload profiles rather than future-proofing.

# Architecture Differences

The two GPUs come from fundamentally different design philosophies. The MX570 A is built on NVIDIA's Ampere architecture using the GA107SB chip, fabricated on Samsung's 8 nm process. The RX 5300M uses AMD's RDNA 1.0 architecture with the Navi 14 chip, built on TSMC's 7 nm node. The process difference is notable: the RX 5300M's 7 nm node is denser and more modern, though the MX570 A's die is actually larger at 200 mm² versus 158 mm². The MX570 A packs more transistors overall (8,700 million vs 6,400 million) and a higher transistor density (43.5M / mm² vs 40.5M / mm²).

The shader configurations diverge sharply. The MX570 A has 2048 shading units, 64 TMUs, and 32 ROPs. The RX 5300M has fewer shading units (1408) but more TMUs (88) and the same 32 ROPs. This mix explains some benchmark behavior: the MX570 A has 45% more shading units, while the RX 5300M has 37.5% more texture units. The MX570 A also includes 16 ray tracing cores and 64 tensor cores—features entirely absent from the RX 5300M, which lists no RT or tensor core counts.

FP32 compute figures tell a similar story. The MX570 A delivers 4.731 TFLOPS of FP32 performance, about 16% ahead of the RX 5300M's 4.069 TFLOPS. However, the RX 5300M has a significant FP16 advantage: it produces 8.138 TFLOPS at a 2:1 ratio, while the MX570 A only manages 4.731 TFLOPS at 1:1. This means the RX 5300M is nearly 72% faster in FP16 throughput, which could matter in mixed-precision workloads.

API support also differs. The MX570 A supports DirectX 12 Ultimate (12_2), while the RX 5300M tops out at DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The MX570 A's newer DX12 feature level reflects its later release (December 2021 versus November 2019 for the RX 5300M).

Power draw is a major differentiator. The MX570 A is rated at 25 W TDP and described as an IGP (integrated graphics processor) with no power connectors. The RX 5300M consumes 85 W TDP—more than triple the power—and also has no power connectors listed. This makes the MX570 A dramatically more power-efficient in theory, though the RX 5300M's higher clocks (boost 1445 MHz vs 1155 MHz) and wider memory bus explain some of the power gap.

# Where Each One Wins

The MX570 A wins in raw compute throughput. Its FP32 figure of 4.731 TFLOPS exceeds the RX 5300M's 4.069 TFLOPS, and its texture rate of 73.92 GTexel/s is lower than the RX 5300M's 127.2 GTexel/s—but the shading unit advantage (2048 vs 1408) gives it an edge in general-purpose shader workloads. The OpenCL benchmark confirms this: 39780 vs 36529, an 8.9% win. The MX570 A also has the higher average benchmark score (38691 vs 36529) and sits one percentile higher (81st vs 80th).

The RX 5300M wins in memory bandwidth and capacity. Its 168.0 GB/s bandwidth is 75% higher than the MX570 A's 96.00 GB/s, and its 3 GB memory capacity is 50% larger. The wider 96-bit bus (versus 64-bit) and faster effective memory speed (14 Gbps vs 12 Gbps) both contribute. This makes the RX 5300M better suited for workloads that stream large datasets or require more VRAM than 2 GB can hold. The RX 5300M also has a higher pixel rate (46.24 GPixel/s vs 36.96 GPixel/s), suggesting faster rasterization in fill-rate-bound scenarios.

The FP16 advantage belongs to the RX 5300M. Its 8.138 TFLOPS FP16 throughput is nearly double the MX570 A's 4.731 TFLOPS. Applications that use FP16 arithmetic—certain machine learning and media workloads—would see a substantial speedup on the AMD part. The MX570 A counters with its tensor cores, which are designed for AI acceleration, though the data does not include a direct tensor-core benchmark.

The MX570 A wins on power efficiency. Its 25 W TDP is less than a third of the RX 5300M's 85 W, and it achieves a higher average benchmark score per watt by a wide margin. For thin-and-light laptops or fanless designs, the MX570 A is clearly the better fit. The RX 5300M's 85 W draw requires more substantial cooling and battery capacity.

# FAQ

Q: Which GPU has the higher benchmark score?

A: The NVIDIA GeForce MX570 A scores 39780 in Geekbench OpenCL, which is 8.9% higher than the AMD Radeon RX 5300M's 36529. The MX570 A also has a higher average benchmark score of 38691 versus 36529.

Q: Does the RX 5300M have any advantages in memory?

A: Yes. The RX 5300M has 3 GB of GDDR6 memory versus 2 GB, a 96-bit bus versus 64-bit, and 168.0 GB/s bandwidth versus 96.00 GB/s. That is 75% more bandwidth and 50% more capacity.

Q: How do their compute capabilities compare?

A: The MX570 A has 2048 shading units and delivers 4.731 TFLOPS FP32, while the RX 5300M has 1408 shading units and 4.069 TFLOPS FP32. However, the RX 5300M produces 8.138 TFLOPS FP16, versus the MX570 A's 4.731 TFLOPS FP16.

Q: Which GPU supports newer DirectX features?

A: The MX570 A supports DirectX 12 Ultimate (12_2), while the RX 5300M only supports DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4.

Q: What about power consumption?

A: The MX570 A is rated at 25 W TDP and is classified as an IGP. The RX 5300M has an 85 W TDP, which is more than three times higher. Neither uses external power connectors.

Q: Are these GPUs still in production?

A: No. Both are listed as end-of-life products. The MX570 A was released in December 2021, while the RX 5300M was released in November 2019.

# Head-to-Head Benchmarks

The only direct benchmark comparison available is Geekbench OpenCL, and it favors the MX570 A. NVIDIA's part scores 39780 against AMD's 36529, a delta of 8.9%. This is the sole head-to-head result in the data, so it carries significant weight. The MX570 A's win count stands at 1, with 0 wins for the RX 5300M.

The margin is meaningful but not overwhelming. An 8.9% lead in OpenCL suggests the MX570 A's extra shading units (2048 vs 1408) and higher FP32 throughput (4.731 vs 4.069 TFLOPS) translate directly into compute performance. The MX570 A also has the advantage of tensor cores (64) and ray tracing cores (16), which could accelerate certain workloads beyond what the raw FP32 numbers suggest.

However, the RX 5300M's closest rivals in the benchmark database provide context. Its average score of 36529 puts it in the same league as the NVIDIA GeForce GTX TITAN X (36530, 0% delta) and the NVIDIA T1000 (36289, 0.7% delta). The AMD Radeon PRO W6400 scores higher at 37157, a 1.7% advantage over the RX 5300M. The MX570 A, meanwhile, sits near the AMD Radeon Pro 580X (38706, 0% delta) and the NVIDIA GeForce RTX 5080 Mobile (38349, 0.9% delta). These rival clusters suggest the two GPUs occupy adjacent performance tiers, with the MX570 A roughly 5.9% above the RX 5300M's average score.

In FP16 performance, the RX 5300M would likely win if a comparable benchmark existed, given its 8.138 TFLOPS versus 4.731 TFLOPS. But the data only provides OpenCL, where the MX570 A takes the crown. The texture rate differential (127.2 GTexel/s for AMD vs 73.92 GTexel/s for NVIDIA) also points to potential wins for the RX 5300M in texture-heavy scenes, though no direct test confirms this.

# Specification Differences

The core specification differences between the two GPUs are substantial. The MX570 A uses the GA107SB chip on Samsung's 8 nm process, while the RX 5300M uses Navi 14 on TSMC's 7 nm node. The MX570 A has more transistors (8,700 million vs 6,400 million) and a larger die (200 mm² vs 158 mm²), but the RX 5300M has a slightly higher transistor density per area (40.5M / mm² vs 43.5M / mm²—the MX570 A actually leads here).

Clock speeds differ notably. The RX 5300M has a higher base clock (1000 MHz vs 832 MHz) and boost clock (1445 MHz vs 1155 MHz), plus a defined game clock of 1181 MHz that the MX570 A lacks. Memory clocks also favor AMD: 1750 MHz (14 Gbps effective) versus 1500 MHz (12 Gbps effective).

Memory configuration is a major split. The MX570 A offers 2 GB GDDR6 on a 64-bit bus with 96.00 GB/s bandwidth. The RX 5300M offers 3 GB GDDR6 on a 96-bit bus with 168.0 GB/s bandwidth—a 75% bandwidth advantage and 50% more capacity.

Compute resources differ in configuration. The MX570 A has 2048 shading units, 64 TMUs, and 32 ROPs, plus 16 RT cores and 64 tensor cores. The RX 5300M has 1408 shading units, 88 TMUs, and 32 ROPs, with no RT or tensor cores listed. Pixel rates favor AMD (46.24 vs 36.96 GPixel/s), as do texture rates (127.2 vs 73.92 GTexel/s). FP32 favors NVIDIA (4.731 vs 4.069 TFLOPS), but FP16 heavily favors AMD (8.138 vs 4.731 TFLOPS).

Power and physical specs differ sharply. The MX570 A is rated at 25 W TDP and classified as an IGP, while the RX 5300M draws 85 W. Neither uses external power connectors. Both use PCIe 4.0 x8 interfaces and have portable-device-dependent display outputs. API support: the MX570 A supports DirectX 12 Ultimate (12_2), the RX 5300M only DirectX 12 (12_1); both support OpenGL 4.6 and Vulkan 1.4. The MX570 A belongs to the GeForce MX (5xx) generation and was released in December 2021; the RX 5300M is from the Navi Mobile (RX 5000M) generation, released in November 2019, with the Polaris Mobile as its predecessor.