AMD Radeon R5 M420 vs NVIDIA GeForce GT 745M Comparison

The AMD Radeon R5 M420 and NVIDIA GeForce GT 745M are two end-of-life mobile graphics processors that land in the same performance tier, but the data shows a clear, if narrow, winner. In the only direct benchmark comparison available, the AMD Radeon R5 M420 takes the lead, but the NVIDIA part counters with a significantly wider memory interface and a more advanced feature set. The benchmark results place both GPUs at the 23rd percentile of all GPUs, indicating they are entry-level solutions, yet their architectural approaches to achieving that performance are fundamentally different.

Head-to-Head Benchmarks

The sole head-to-head benchmark in the data is the Geekbench OpenCL test, and it delivers a decisive result for the AMD Radeon R5 M420. The AMD part scores 3956, while the NVIDIA GeForce GT 745M scores 3580. This translates to a 10.5% advantage for the AMD chip, a substantial margin that puts it clearly ahead of its rival in this compute-oriented workload. This is not a marginal victory; it is a significant gap that indicates the AMD architecture is better suited to the type of parallel processing that OpenCL measures.

This result is corroborated by the average benchmark scores and the nearest rival data. The AMD Radeon R5 M420 has an average benchmark score of 3956, which is nearly identical to its OpenCL score, suggesting that this single test is a strong indicator of its overall compute performance. The NVIDIA GeForce GT 745M, while having a higher average score of 3953 than its OpenCL score, is still behind. Its average is pulled up by its performance in other APIs, specifically Vulkan and Metal, where it scores 5502 and 2777 respectively. However, when both are tested on the same OpenCL workload, the AMD chip is unequivocally faster.

Looking at the nearest rivals for each GPU further contextualizes this performance. The AMD Radeon R5 M420 is a near-perfect match for the NVIDIA GeForce 830M, with a delta of 0% (3956 vs 3957). It is also 0.1% ahead of the GT 745M itself. Conversely, the GT 745M's nearest rival list shows it is 0.1% behind the R5 M420. This tight clustering around the 3950-3960 score range confirms that these are all very closely matched parts, but the head-to-head data provides the only direct evidence of which is faster, and it clearly favors the AMD product.

Architecture Differences

The two GPUs are built on the same 28 nm process at TSMC, but their internal designs diverge sharply. The AMD Radeon R5 M420 uses the "Jet" chip, based on the GCN 1.0 architecture, while the NVIDIA GeForce GT 745M uses the "GK107" chip, based on the Kepler architecture. This fundamental difference in design philosophy explains their performance characteristics.

The AMD chip packs 690 million transistors into a 56 mm² die, resulting in a transistor density of 12.3M / mm². In contrast, the NVIDIA chip is significantly larger and more complex, with 1,270 million transistors on a 118 mm² die, but a lower density of 10.8M / mm². Despite having nearly twice the transistor count, the NVIDIA chip's architecture does not translate to a compute performance advantage in the OpenCL test. The AMD chip compensates for its lower transistor count with a higher clock speed, running at a base of 780 MHz and a boost of 850 MHz, whereas the GT 745M's core clocks are not specified in the data.

Memory configuration is another major point of divergence. The AMD Radeon R5 M420 is equipped with 4 GB of DDR3 memory on a 64-bit bus, yielding a bandwidth of 16.00 GB/s. The NVIDIA GeForce GT 745M, on the other hand, has 2 GB of GDDR5 memory on a much wider 128-bit bus, delivering a bandwidth of 64.00 GB/s. This gives the NVIDIA part a fourfold advantage in memory bandwidth, which is typically crucial for gaming at higher resolutions and texture detail. However, this advantage does not manifest in the compute-focused OpenCL benchmark.

Feature support also differs. The AMD Radeon R5 M420 supports DirectX 12 (11_1), while the NVIDIA GeForce GT 745M supports DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan, but the NVIDIA chip has a slightly newer Vulkan version (1.2.175 vs 1.2.170). In terms of raw compute units, the AMD chip has 320 shading units, 20 TMUs, and 8 ROPs, while the NVIDIA chip has 384 shading units, 32 TMUs, and 16 ROPs. The NVIDIA part has more of everything, yet it still loses the compute benchmark.

FAQ

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

A: The AMD Radeon R5 M420 is faster, scoring 3956 compared to the NVIDIA GeForce GT 745M's 3580, a 10.5% advantage.

Q: What is the memory bandwidth of each GPU?

A: The NVIDIA GeForce GT 745M has a memory bandwidth of 64.00 GB/s, while the AMD Radeon R5 M420 has a bandwidth of 16.00 GB/s.

Q: How do the average benchmark scores compare?

A: The AMD Radeon R5 M420 has an average benchmark score of 3956, while the NVIDIA GeForce GT 745M has an average of 3953, making them nearly equivalent overall.

Q: What is the difference in their DirectX support?

A: The AMD Radeon R5 M420 supports DirectX 12 (11_1), while the NVIDIA GeForce GT 745M supports DirectX 12 (11_0).

Q: What is the transistor count for each GPU?

A: The NVIDIA GeForce GT 745M has 1,270 million transistors, whereas the AMD Radeon R5 M420 has 690 million transistors.

Q: Does the NVIDIA GPU have a higher pixel rate?

A: No, the AMD Radeon R5 M420 has a higher pixel rate of 6.800 GPixel/s, while the NVIDIA GeForce GT 745M has a pixel rate of 4.392 GPixel/s.

Specification Differences

• Chip: AMD Radeon R5 M420 uses "Jet"; NVIDIA GeForce GT 745M uses "GK107".
• Architecture: AMD Radeon R5 M420 is based on GCN 1.0; NVIDIA GeForce GT 745M is based on Kepler.
• Transistors: AMD Radeon R5 M420 has 690 million; NVIDIA GeForce GT 745M has 1,270 million.
• Die Size: AMD Radeon R5 M420 is 56 mm²; NVIDIA GeForce GT 745M is 118 mm².
• Transistor Density: AMD Radeon R5 M420 has 12.3M / mm²; NVIDIA GeForce GT 745M has 10.8M / mm².
• Base Clock: AMD Radeon R5 M420 has a base clock of 780 MHz; NVIDIA GeForce GT 745M has no base clock specified.
• Boost Clock: AMD Radeon R5 M420 has a boost clock of 850 MHz; NVIDIA GeForce GT 745M has no boost clock specified.
• Memory Size: AMD Radeon R5 M420 has 4 GB; NVIDIA GeForce GT 745M has 2 GB.
• Memory Type: AMD Radeon R5 M420 uses DDR3; NVIDIA GeForce GT 745M uses GDDR5.
• Memory Bus Width: AMD Radeon R5 M420 has a 64-bit bus; NVIDIA GeForce GT 745M has a 128-bit bus.
• Memory Bandwidth: AMD Radeon R5 M420 has 16.00 GB/s; NVIDIA GeForce GT 745M has 64.00 GB/s.
• Shading Units: AMD Radeon R5 M420 has 320; NVIDIA GeForce GT 745M has 384.
• TMUs: AMD Radeon R5 M420 has 20; NVIDIA GeForce GT 745M has 32.
• ROPs: AMD Radeon R5 M420 has 8; NVIDIA GeForce GT 745M has 16.
• Pixel Rate: AMD Radeon R5 M420 is 6.800 GPixel/s; NVIDIA GeForce GT 745M is 4.392 GPixel/s.
• Texture Rate: AMD Radeon R5 M420 is 17.00 GTexel/s; NVIDIA GeForce GT 745M is 17.57 GTexel/s.
• FP32 Performance: AMD Radeon R5 M420 is 544.0 GFLOPS; NVIDIA GeForce GT 745M is 421.6 GFLOPS.
• TDP: NVIDIA GeForce GT 745M has a TDP of 45 W; AMD Radeon R5 M420 has no TDP specified.
• Bus Interface: AMD Radeon R5 M420 uses PCIe 3.0 x8; NVIDIA GeForce GT 745M uses PCIe 3.0 x16.
• DirectX Support: AMD Radeon R5 M420 supports 12 (11_1); NVIDIA GeForce GT 745M supports 12 (11_0).
• Vulkan Support: AMD Radeon R5 M420 supports 1.2.170; NVIDIA GeForce GT 745M supports 1.2.175.
• Release Date: AMD Radeon R5 M420 was released in 2016; NVIDIA GeForce GT 745M was released in 2013.

Where Each One Wins

The AMD Radeon R5 M420 is the clear winner in compute-heavy workloads. Its 10.5% lead in the Geekbench OpenCL test, combined with its higher FP32 performance (544.0 GFLOPS vs 421.6 GFLOPS) and higher pixel rate (6.800 GPixel/s vs 4.392 GPixel/s), demonstrates its strength in tasks that rely on raw shader throughput and parallel math. It also offers double the memory capacity (4 GB vs 2 GB), which can be beneficial for holding larger datasets in memory, even if the slower DDR3 interface limits transfer speeds. The AMD part also holds a slight edge in DirectX feature level support with 12 (11_1) over the NVIDIA's 12 (11_0).

The NVIDIA GeForce GT 745M wins decisively in memory bandwidth, offering 64.00 GB/s compared to the AMD's 16.00 GB/s. This is a massive advantage for memory-intensive tasks like texture streaming in games. It also has a wider 128-bit memory bus and a newer Vulkan API version (1.2.175 vs 1.2.170). While it loses the compute benchmark, its higher texture rate (17.57 GTexel/s vs 17.00 GTexel/s) suggests it could be competitive in fill-rate-bound scenarios. The NVIDIA chip also has a higher transistor count and more shading units, TMUs, and ROPs, indicating a more robust geometry and pixel processing pipeline.

The Verdict

The data presents a clear split: the AMD Radeon R5 M420 is the better choice for general compute performance, while the NVIDIA GeForce GT 745M is the better choice for memory-bandwidth-sensitive applications. If the primary use case is running OpenCL-based applications, video encoding, or other tasks that leverage FP32 compute, the AMD Radeon R5 M420 is the superior option. Its 10.5% lead in the head-to-head benchmark is a definitive data point that cannot be ignored.

However, for gaming, the NVIDIA GeForce GT 745M's 4x memory bandwidth advantage is a critical factor. The GT 745M's 64.00 GB/s bandwidth allows it to feed its 384 shading units far more effectively in a gaming environment, where textures and geometry data are constantly being fetched from memory. The AMD part's 16.00 GB/s bandwidth is a severe bottleneck that would likely negate its compute advantage in real-world gaming scenarios. Given that both GPUs sit at the 23rd percentile, neither is a high-performance part, but the NVIDIA chip's memory subsystem makes it the more practical choice for its primary mobile use case: gaming. Ultimately, the AMD Radeon R5 M420 wins on raw compute metrics, but the NVIDIA GeForce GT 745M wins on the bandwidth that matters most for game rendering.