AMD Radeon R5 M420 vs NVIDIA GeForce GT 635M Comparison

AMD
RADEON

AMD Radeon R5 M420

CORE STATE Jet
VRAM 4 GB
CLOCK SPEED 850 MHz
TDP —
BUS WIDTH 64 bit
ARCHITECTURE GCN 1.0
nm
PROCESS 28 nm
LAUNCH DATE 2016
VS
NVIDIA
GEFORCE

GeForce GT 635M

CORE STATE GF108
VRAM 2 GB
CLOCK SPEED —
TDP 35 W
BUS WIDTH 128 bit
ARCHITECTURE Fermi
nm
PROCESS 40 nm
LAUNCH DATE 2012

PERFORMANCE BENCHMARKS

geekbench_opencl
3,956
3,740

Analysis: AMD Radeon R5 M420 vs NVIDIA GeForce GT 635M

The AMD Radeon R5 M420 and NVIDIA GeForce GT 635M are both end-of-life mobile graphics processors, but the benchmark data clearly separates them. In the sole head-to-head test, the AMD Radeon R5 M420 scores 3956 points, which is 5.8% higher than the NVIDIA GeForce GT 635M’s 3740 points. This performance edge, combined with a significantly newer architecture and more memory, makes the AMD part the stronger option for general-purpose compute workloads, despite the NVIDIA chip’s wider memory bus and lower power draw.

FAQ

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

A: The AMD Radeon R5 M420 wins with a score of 3956, beating the NVIDIA GeForce GT 635M’s 3740 by a margin of 5.8%.

Q: How does each GPU compare to its closest rivals?

A: The AMD Radeon R5 M420 is virtually tied with the NVIDIA GeForce 830M (3957, 0% delta) and sits 0.1% ahead of the NVIDIA GeForce GT 745M (3953). The NVIDIA GeForce GT 635M is 0.6% ahead of the NVIDIA Quadro 3000M (3718) and 1.2% ahead of the NVIDIA GeForce 825M (3694), but it trails the Intel UHD Graphics 710 (3792) by 1.4%.

Q: Which GPU has a higher memory bandwidth?

A: The NVIDIA GeForce GT 635M has a substantial advantage, offering 28.80 GB/s of bandwidth over a 128-bit bus, compared to the AMD Radeon R5 M420’s 16.00 GB/s over a 64-bit bus.

Q: Which GPU supports more advanced DirectX features?

A: The AMD Radeon R5 M420 supports DirectX 12 (11_1), while the NVIDIA GeForce GT 635M supports DirectX 12 (11_0). The AMD part’s feature level is one notch higher.

Q: Which GPU has a higher transistor density?

A: The AMD Radeon R5 M420 packs 12.3 million transistors per mm², more than double the NVIDIA GeForce GT 635M’s 5.0 million per mm².

Q: Which GPU has more shading units?

A: The AMD Radeon R5 M420 has 320 shading units, compared to only 96 on the NVIDIA GeForce GT 635M, a more than three-to-one advantage.

Architecture Differences

The two GPUs represent different generations of mobile graphics design. The AMD Radeon R5 M420 is built on the GCN 1.0 architecture using the Jet chip, manufactured on a 28 nm process at TSMC. It integrates 690 million transistors on a 56 mm² die, resulting in a transistor density of 12.3 million per mm². The NVIDIA GeForce GT 635M uses the older Fermi architecture with the GF108 chip, built on a 40 nm process, also at TSMC. It contains 585 million transistors on a much larger 116 mm² die, yielding a lower transistor density of 5.0 million per mm².

The compute resources differ dramatically. The AMD part features 320 shading units, 20 texture mapping units (TMUs), and 8 raster operation units (ROPs). The NVIDIA part has only 96 shading units, 16 TMUs, and 4 ROPs. This hardware disparity translates into large theoretical throughput gaps: the AMD Radeon R5 M420 delivers 544.0 GFLOPS of FP32 compute, 17.00 GTexel/s of texture fill, and 6.800 GPixel/s of pixel fill. The NVIDIA GeForce GT 635M is limited to 182.4 GFLOPS, 7.600 GTexel/s, and 1.900 GPixel/s, respectively.

Memory subsystems also differ. The AMD Radeon R5 M420 comes with 4 GB of DDR3 memory on a 64-bit bus, running at 1000 MHz (2 Gbps effective), providing 16.00 GB/s of bandwidth. The NVIDIA GeForce GT 635M offers 2 GB of DDR3 on a wider 128-bit bus at 900 MHz (1800 Mbps effective), yielding 28.80 GB/s. The NVIDIA part’s bandwidth advantage is significant, but it is paired with half the memory capacity.

Clock behavior and power are also divergent. The AMD Radeon R5 M420 has a base clock of 780 MHz and a boost clock of 850 MHz, while the NVIDIA GeForce GT 635M has no listed base or boost clocks. The NVIDIA chip is rated at a 35 W TDP, whereas the AMD part has no TDP specified. Both are integrated into laptops (slot width: IGP), and both feature portable-device-dependent display outputs.

API support shows the AMD part is more modern. The Radeon R5 M420 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The GeForce GT 635M supports DirectX 12 (11_0) and OpenGL 4.6 but has no Vulkan support listed. The AMD part also uses a PCIe 3.0 x8 interface, while the NVIDIA part uses PCIe 2.0 x16.

The Verdict

The AMD Radeon R5 M420 is the clear winner for raw compute performance. Its Geekbench OpenCL score of 3956 is 5.8% higher than the NVIDIA GeForce GT 635M’s 3740, and its shading unit count (320 vs. 96) suggests a large advantage in parallel workloads. The AMD part also doubles the memory capacity (4 GB vs. 2 GB) and offers a more modern feature set, including Vulkan support and a higher DirectX feature level.

The NVIDIA GeForce GT 635M does have one clear strength: memory bandwidth. Its 28.80 GB/s is 80% higher than the AMD part’s 16.00 GB/s, which could benefit memory-bound tasks. Its 35 W TDP also provides a defined power envelope, while the AMD part’s power draw is unspecified. The NVIDIA chip’s PCIe 2.0 x16 interface is also wider than the AMD part’s PCIe 3.0 x8, though the newer PCIe standard on the AMD side may offset this.

For users choosing between these two, the data favors the AMD Radeon R5 M420 for general-purpose GPU compute, as reflected in the benchmark. The NVIDIA GeForce GT 635M may be preferable in scenarios where raw memory throughput is the bottleneck, but its lower compute throughput and older architecture make it the weaker overall choice. Given that both are end-of-life products, the decision rests on whether the workload is compute-bound (choose AMD) or bandwidth-bound (choose NVIDIA).

Specification Differences

The two GPUs differ across nearly every major specification category. The AMD Radeon R5 M420 uses a 28 nm process node, while the NVIDIA GeForce GT 635M uses a 40 nm node. The AMD chip contains 690 million transistors on a 56 mm² die, whereas the NVIDIA chip has 585 million on a 116 mm² die. Transistor density is 12.3M / mm² for AMD versus 5.0M / mm² for NVIDIA.

Clock speeds are only listed for the AMD part: 780 MHz base and 850 MHz boost. The NVIDIA part has no base or boost clock data. Memory clocks differ as well: the AMD runs at 1000 MHz (2 Gbps effective), while the NVIDIA runs at 900 MHz (1800 Mbps effective). Memory capacity is 4 GB on the AMD side versus 2 GB on NVIDIA. Both use DDR3, but the bus widths differ: 64-bit for AMD versus 128-bit for NVIDIA. This yields bandwidths of 16.00 GB/s and 28.80 GB/s, respectively.

Compute resources are heavily skewed toward AMD: 320 shading units, 20 TMUs, and 8 ROPs versus NVIDIA’s 96 shading units, 16 TMUs, and 4 ROPs. Theoretical rates follow: AMD’s pixel rate is 6.800 GPixel/s and texture rate is 17.00 GTexel/s, while NVIDIA’s are 1.900 GPixel/s and 7.600 GTexel/s. FP32 performance is 544.0 GFLOPS for AMD versus 182.4 GFLOPS for NVIDIA.

Power and interface details also diverge. The NVIDIA part has a TDP of 35 W and no power connectors, while the AMD part has no TDP listed and no power connector data. The bus interface is PCIe 3.0 x8 for AMD versus PCIe 2.0 x16 for NVIDIA. API support shows AMD with DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170; NVIDIA has DirectX 12 (11_0) and OpenGL 4.6, but no Vulkan. Release dates also differ: the AMD launched on 2016-05-14, while the NVIDIA launched on 2012-03-21.

Head-to-Head Benchmarks

The only benchmark available is Geekbench OpenCL, and it produces a decisive result. The AMD Radeon R5 M420 scores 3956, while the NVIDIA GeForce GT 635M scores 3740. The delta is 5.8% in favor of AMD, and the AMD part wins the single head-to-head test, with NVIDIA winning zero.

This 216-point gap places the two GPUs in different performance tiers relative to their peers. The AMD Radeon R5 M420 sits at the 23rd percentile of all GPUs, while the NVIDIA GeForce GT 635M sits at the 22nd percentile. The AMD part’s nearest rival is the NVIDIA GeForce 830M (3957, 0% delta), indicating a statistical tie. The NVIDIA GeForce GT 635M’s nearest rival is the NVIDIA Quadro 3000M (3718, 0.6% delta), which it slightly beats.

The benchmark results align with the architectural differences. The AMD part’s 320 shading units and 544.0 GFLOPS of FP32 compute give it a theoretical advantage that shows up in real-world OpenCL performance. The NVIDIA part’s higher bandwidth (28.80 GB/s) and wider bus (128-bit) were not enough to overcome its lower compute throughput (182.4 GFLOPS) and older Fermi architecture.

Where Each One Wins

The AMD Radeon R5 M420 wins in compute-heavy workloads, as evidenced by its 5.8% Geekbench OpenCL lead. Its 320 shading units and 544.0 GFLOPS of FP32 performance make it better suited for parallel processing tasks such as general-purpose GPU compute, video encoding, and any workload that leverages OpenCL. The AMD part also wins on memory capacity (4 GB vs. 2 GB), which can be beneficial for holding larger datasets in local memory. Its support for Vulkan 1.2.170 and DirectX 12 (11_1) gives it a more complete modern API feature set.

The NVIDIA GeForce GT 635M wins on memory bandwidth, offering 28.80 GB/s versus the AMD’s 16.00 GB/s. This could make it faster in memory-bound scenarios where data transfer speed, not compute throughput, is the limiting factor. Its 128-bit bus is twice as wide as the AMD’s 64-bit bus, and its 35 W TDP provides a defined power budget for system designers. The NVIDIA part’s PCIe 2.0 x16 interface may also offer higher bandwidth to the host system than the AMD’s PCIe 3.0 x8, depending on the platform.

In practical terms, the AMD Radeon R5 M420 is the better choice for tasks that stress the GPU’s compute units, such as OpenCL acceleration or shader-heavy workloads. The NVIDIA GeForce GT 635M is preferable when the workload is dominated by memory access patterns, such as texture streaming or large buffer operations, where its higher bandwidth can be exploited. Overall, the AMD part’s single benchmark win and superior compute specifications make it the more capable of the two, but the NVIDIA part’s bandwidth advantage gives it a specific niche.

DETAILED SPECIFICATIONS

SPECIFICATION
R5 M420
GT 635M
Core Specs
Shading Units
320
96 -70.0%
Shaders
320
96 -70.0%
TMUs
20
16 -20.0%
ROPs
8
4 -50.0%
Compute Units
5
—
SM Count
—
2
Clocks
Base Clock
780 MHz
—
Boost Clock
850 MHz
—
GPU Clock
—
475 MHz
Shader Clock
—
950 MHz
Memory Clock
1000 MHz 2 Gbps effective
900 MHz 1800 Mbps effective
Memory
Memory Size
4 GB
2 GB
VRAM (MB)
4,096
2,048 -50.0%
Memory Type
DDR3
DDR3
Memory Bus
64 bit
128 bit
Bandwidth
16.00 GB/s
28.80 GB/s
Cache
L1 Cache
16 KB (per CU)
64 KB (per SM)
L2 Cache
128 KB
256 KB
Performance
Pixel Rate
6.800 GPixel/s
1.900 GPixel/s
Texture Rate
17.00 GTexel/s
7.600 GTexel/s
FP32 (TFLOPS)
544.0 GFLOPS
182.4 GFLOPS
FP64 (TFLOPS)
34.00 GFLOPS (1:16)
15.20 GFLOPS (1:12)
Power
TDP
—
35 W
TDP (W)
—
35
Power Connectors
—
None
Architecture
Architecture
GCN 1.0
Fermi
GPU Name
Jet
GF108
Generation
Gem System (R5 M400)
GeForce 600M
Process Size
28 nm
40 nm
Transistors
690 million
585 million
Die Size
56 mm²
116 mm²
Foundry
TSMC
TSMC
Density
12.3M / mm²
5.0M / mm²
API Support
DirectX
12 (11_1)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
1.2.170
—
OpenCL
2.1 (1.2)
1.1
CUDA
—
2.1
Shader Model
6.5 (5.1)
5.1
Physical
Slot Width
IGP
IGP
Outputs
Portable Device Dependent
Portable Device Dependent
Bus Interface
PCIe 3.0 x8
PCIe 2.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
Solar System
GeForce 500M
Successor
Polaris Mobile
GeForce 700M
View Radeon R5 M420 Details View GeForce GT 635M Details