Are more bits better in an MCU? That’s similar to asking: are more cylinders better in an internal combustion engine (ICE)? Peripherals and other features matter, not just bits (or cylinders). For example, in the 1980s, Volvo offered a station wagon with a turbo-V4 engine that was faster off the line than Chevrolet Corvette with a很多larger V8. Of course, the station wagon was toast once the V8 was wound up after a hundred yards. But the first hundred yards made a great visual for the Volvo commercials. In the case of MCUs, total power consumption, availability of peripherals, latency demands, and other factors are important, not just the number of bits.

MCUs are everywhere. They are found in cell phones, toys, vehicles, white goods, renewable energy systems, Internet of things (IoT) devices, and on and on. MCUs are essential for controlling hardware functions from motors to flashing LEDs. The basic architecture of MCUs includes a central processing unit, some memory, and peripheral functions(Figure 1)。The MCU gets inputs from sensors, switches, and other devices and controls the function of peripherals such as drivers for displays, actuators, and motors, as prescribed with preprogrammed instructions. Depending on the complexity of the functions being controlled and the amount of processing needed by the various inputs, MCUs are available with 8-bits, 16-bits, and 32-bits. In general, the more bits, the more processing potential.

“位”是“二进制数字”。在8位MCU中，数据存储在8位，称为字节（或在某些情况下是八位字节）。8位内存寄存器可以存储可能的2个⁸值。根据所使用的整数表示，实际值的范围有所不同。例如，使用未签名的二进制数，范围为0到255（2⁸负1），使用两个的补体表示，范围为-128（-1 x 2）⁷）至127（2⁷——1)。在一个8位单片机8位数据总线wide, the address bus widths can vary. Address bus widths of 12- to 16-bits are common. A 16-bit wide address bus, for example, results in a directly addressable memory space of 65,536 (2¹⁶）字节（称为64KB）。但是这些分类只会开始暗示特定的8位MCUS的实际功能(Figure 2)。

8位MCUS具有通用性；它们提供简单的编程，能源效率和小包装尺寸（有些只有六个引脚）。但是这些MCU通常并非旨在实施网络和通信功能。最常见的网络协议和通信软件堆栈是16位或32位。通信外围设备可用于某些8位设备，但是16位和32位MCU通常可以是更有效的选择。尽管如此，在各种控制，传感和界面应用程序中通常使用8位MCU。

More power with 32 bits

While 8-bit processors are versatile, they can be equally limited in performance. When 8-bits is insufficient, designers can turn to 32-bit MCUs that can pack quite a performance punch. 32-bit MCUs are used across a wide range of applications, and they can support high-end algebraic operations and floating-point mathematics. And while 8-bit MCUs are typically limited to several MHz of processing speed, 32-bit units are available with clock rates into the GHz range, although several hundred MHz units are more common.

这种提高的性能伴随着32位MCU的编程复杂性的成本，包括多个状态登记册，复杂的中断管理，多层固件执行特权等。这就是使32位单元如此强大的原因。他们可以快速处理数千种复杂的计算，以用于需要大量数据操纵的应用，例如图像处理或对高速系统的实时控制。此外，通常需要32位处理器的较大内存地址空间来缓冲高带宽数据流。

The higher clock speeds generally associated with 32-bit MCUs can also lead to increased energy consumption. In general, 32-bit MCUs can support more intense computation by trading off higher energy consumption. But the choice between 8-bits and 32-bits is not quite that simple. For example, a 32-bit MCU can complete a series of computations faster than an 8-bit MCU and then enter sleep mode for a longer time. So, is the 32-bit MCU more energy efficient? That’s possible, but not necessarily true. An 8-bit MCU can have lower overall energy consumption and contribute to longer battery life in portable devices. The choice is not simple, and the optimal energy consumption solution must be determined on a case-by-case basis.

有一个中间立场

回到冰类比，在4个圆柱体（8位）和8个圆柱体（32位）之间有一个中间地面。通常，6缸（16位）设计可以提供效率和性能的正确组合。8位MCU的性能本质上是有限的，只能通过添加外围设备在限制内提高。使用32位MCU的系统提供更高的性能，但比替代解决方案可能是过度杀伤和消耗更多的功率。

That’s where 16-bit MCUs can come in; they can combine the simplicity of 8-bit units while providing enhanced performance without an excessive power penalty. Especially in applications that don’t need massive multi-threading and moderate memory needs, 16-bit MCUs can provide an optimal middle ground solution(Figure 3)。

外围设备的关键。有16位单片机math co-processors that support intensive mathematical calculations, similar to some 32-bit units at the same clock speed. And, there are numerous communications stacks that support 16-bit MCUs. For example, with the correct implementation, Ethernet, controller area network (CAN), universal serial bus (USB), and Zigbee are among the communications protocols that can run efficiently on 16-bit MCUs. And the use of 16-bit MCUs can provide for simpler circuit board implementations compared with many 32-bit solutions.

Adding peripherals

The availability of peripherals and where they reside can be important factors driving the choice of 8-bits, 16-bits, or 32-bits. Adding an external communications peripheral to an 8-bit MCU can be an option. But, the addition of a peripheral chip can often eliminate the cost-benefit of using an 8-bit MCU. It can be less expensive to use a 32-bit unit with the communications functions already integrated. Moving up the scale finds 32-bit MCUs often include more features and can handle multiple peripheral functions more efficiently than 8-bit and 16-bit devices.

涉及更大数量和快速计算的应用程序受益于16位和32位MCUS的应用。可以从16位或32位MCU中受益的示例包括FFT计算，高质量的音频或视频，高分辨率图像处理以及各种边缘计算应用程序。此外，通常使用特定于应用程序的加速器，机器学习和人工智能实施通常需要32位。

在复杂性量表的另一端，传感器或其他来源的处理模拟信号并不一定受益于使用32位或16位MCUS。8位MCU可以包括支持8位，10位，12位甚至16位的内置模数转换器（ADC）。

8-bit MCUs are available that can support some cryptographic functions for increased security. But those functions are often implemented in hardware that increases costs or software that places an increased burden on a relatively low-power processor. If higher-level security functions are important, it may be wise to use a 16-bit or 32-bit MCU.

总结一切 - 哪个MCU最好？

So, are more bits better? No, not necessarily. It’s a complex decision process(Figure 4)。Successful designs identify the MCU architecture that delivers the needed performance in the most cost-effective manner. And for battery-powered devices, in the most energy-efficient manner.

确定最佳MCU不一定是一个简单的问题。有时，这不仅仅是当今的绩效要求；随着时间的流逝，可能需要提供具有增长潜力的解决方案，以满足更具挑战性的应用程序需求。例如，安全需求可以不断发展。适应未来对新安全功能的需求的能力可能是一个重要的应用程序。

如果特定应用需要大量的随机存储器（RAM），则与8位甚至16位设备相比，使用32位MCU通常会提供更多的RAM。在比较各种MCU替代方案时，延迟是要考虑的另一个因素。可能可以使用8-，16或32位MCU执行某些功能。但是，如果应用程序需要较低的延迟，则可能需要以更快的时钟速率的16或32位处理器的速度。

参考

8-bit microcontroller summary，，，，Microchip
8-bit vs. 32-bit MCU: Choosing the Right Microcontroller for Your PCB Design，，，，Altium
IoT: choosing 8-bit vs. 32-bit MCUs, Lynnette Reese
MCU Basic Structure/Operation, Renesas

---

Filed Under:Featured，，，，Tech Articles
Tagged With:16-bit，，，，32-bit，，，，8-bit，，，，微控制器，，，，microprocessors

---