Table of Contents
We analyze the AMD Ryzen 5 2400G processor which is made up of four Zen cores and integrated Vega architecture graphics.
Gaming (playing video games in Spanish) has become an industry that moves millions and millions of euros worldwide in merchandising, competitions, sponsorships and especially in hardware. It is not only talking about peripherals, it is also talking about the most basic and essential elements to be able to talk about gaming and they are the processor and the graphics card, two elements without which we do not have a computer, practically. There is a very large space in entry-level gaming and AMD knows it, which is why it has developed the AMD Ryzen 5 2400G APU, an entry-level solution for those looking for a low-budget gaming device.
Basically this processor is a revision of the AMD Ryzen 5 1400 with Zen architecture but in this case with the 14nm + lithography, which allows an improvement in the working frequencies and obtain a little more performance.
Regarding its specifications, we must say that it is a quad-core processor and eight processing threads in a 4 + 0 CCX configuration, which indicates that the four cores are in the same silicon, which allows to improve performance. It has a base working frequency of 3.6GHz and reaches 3.9GHz. The encapsulation integrates Vega graphics, specifically an RX Vega 11 silicon has been arranged, which indicates that it has 11 Compute Units with a total TDP of 65W.
It should also be noted that some extra adjustments have been made in these processors, such as reducing the L3 cache, leaving it at 4MB and the main PCIe line has been reduced from x16 to x8, because it already integrates graphics and the iGPU has been left with the line x16.
To make it attractive in terms of price, other costs such as packaging and assembly have also been reduced, which means that it differs from the Ryzen that do not carry iGPU, in that it does not have the DIE and the IHS soldered, facilitating the delid. . The heatsink that is integrated in the bundle has also been adjusted, the packaging has been simplified and uses a conventional non-metallic TIM.
The two most relevant elements of this AMD Ryzen 5 2400G is that it natively supports JEDEC DDR4-2933MHz standard memory and makes use of Precision Boost 2.0 technology.
ZEN X86 MICROARCHITECTURE
The Summit Ridge models, after multiple laboratory tests allow to conclude that in some games it allows an additional benefit thanks to the additional cache that is used in a dual CCX configuration, while other titles could benefit from a reduced latency in a standalone CCX (How is the case)
Using a net zero CCX system or what is the same, a single CCX module which allows is a more compact Matrix thinking about the integration of an iGPU and also for the development of compact systems such as small computers and equipment laptops. The AMD Ryzen 5 2400G goes from 8MB of cache (compared to the Ryzen 5 1400) to a system cache of only 4MB compensated for the improved frequencies.
The main characteristics of this architecture are:
- Performance: Enhanced Instruction Level Parallelism (ILP) to improve single-threaded performance.
- Performance: Cached data speeds have been improved. Allows more efficient handling of data using Simultaneous Multi-Threading (SMT)
- Efficiency: Design thought for low consumption systems that allows efficient energy management according to the load and when there is no load.
- Scalability: CCX using Infinity Fabric enables efficient scaling, integration and communication of inter-core communication. The downside is that it is highly dependent on memory frequencies and can bottleneck multi-core processors at times of high load.
AMD SENSE MI
PrecisionBoost 2.0
It allows to scale the processor frequency beyond Boost mode in steps of 25MHz. For this to happen in the Ryzen 2000 and Ryzen APU, one of the following two conditions will be met:
- If the processor has three or more cores in two of them, the Core Precision Boost 2.0 will be applied
- If the processor has two or fewer cores, Core Precision Boost 2.0 will be applied to all of them.
This condition is limited by processor temperature and is developed to improve performance if necessary by automatically and intelligently scaling the frequency without compromising the integrity of the silicon.
Neural Net Prediction
Ryzen processors integrate an Artificial Intelligence system that, through a neural network system, performs real-time learning of the behavior of applications that allows forecasting or speculation about the next tasks or loads that will be issued to the processor.
Smart Prefetch
They are a series of state-of-the-art learning algorithms that allow reading internal patterns and the behaviors of the different applications by anticipating the data that will be necessary to carry out future executions.
Pure Power
To be as efficient as possible, a network of high precision sensors has been integrated that allow reading scales of 1mV, 1mA, 1mW and 1ºC with a refresh rate of 1000 seconds, which allows to have an instantaneous reading of the workload and an analysis of the silicon to obtain the maximum possible performance from it.
VEGA ARCHITECTURE
Next-Generation Geometry (NGG) Engine
They are a series of geometric engines optimized to deliver higher polygon performance through optimized paths within the silicon. It is a very flexible and programmable system that is accompanied by Primitive Shaders to improve performance for each of the integrated transistors.
If we talk about 3D rendering, it has an input assembly, vertex shading, Hull Sharding, Tessellation, domain shading and geometry shading for greater perfection and polish of the rendering.
Geometry Engine Lod Balancing With NGG
Primitive Shaders have a variety of uses beyond high-performance geometry, including computation of discrete vertex attributes, multi-view or multi-resolution rendering, depth preload, particle systems, and graphics processing of full scene and tour on the GPU.
The geometry engine allows the use of L2 cache memory for the storage of vertex parameters, doubling with respect to the Polaris architecture. A system is added that allows adjustment according to need and a graphic controller capable of selecting the most optimal route for any type of use.
Integrating HBM2 memories achieves bottleneck reduction while allowing or integrating improved load balancing across multiple geometric engines. The workload distribution is also automatically adjusted based on the distribution system call functions to maximize performance.
NCU with Rapid Packed Math
Increasingly greater mathematical precision is required in complex calculations, which implies that more sophisticated rendering engines are needed and this implies an increase in data value beyond FP32 and that is why NCUs have been designed to have support for Rapid Packed Math.
Within Rapid Packed Math we find solutions and elements that speed up the task such as Deep Learning, video processing, computer vision, vectorization, lighting, HDR color values and combination factorizations.
Revised Pixed Engine
It is designed to meet the demands of 4K resolutions and refresh rates of 140Hz or higher. DSBR has also been redesigned to reduce unnecessary data transfer and processing within the GPU, increasing performance and reducing power consumption.
BENCHMARK
CONCLUSION
The AMD Ryzen 5 2400G is not an ordinary processor, by integrating RX Vega graphics it directly becomes a very interesting APU for those who want a low-cost gaming equipment, especially designed for those who play titles such as CS: GO, Dota 2 or LoL, to give some examples, which are titles that fall within the eSports category. It can run heavier titles like Triple A, of course, but we already talked about stable 1080p @ 30 FPS in most cases and making some important texture adjustments.
Be that as it may, these processors seem an excellent option for those who are going to play eSports titles, as we have said or casually, they are even excellent for office automation and entertainment since we can create very compact equipment with the wide variety of micro-motherboards. ATX that we can find in the market.
