I think this is meaningless. The biggest feature of an embedded system is "embedding". That is to say, your control system is embedded in your control object, so the first thing is to obey the needs and features of objects. What is the basis for separation from object null?
Each MCU has its own value, and each user's choice has its own rationale. Starting from aVR, It is a selling point of a single clock period command, compared with the classic 51 with the 12 clock at that time, and based on the characteristics of CMOS, the higher the clock power consumption, so it seems to have a significant advantage in energy consumption.
With the improvement of technology, 51 has already had four clock cycles, '2 clock cycles and even a single clock cycle chip. At this time, the speed advantage of AVR does not exist.
If you consider the maximum clock limit, you can compare it. It seems that the speed of the AVR is less than 51.
If you consider development costs, it is even harder to say.
As for arm, it has its own advantages and disadvantages. The key is what you need to attach objects? What is the final positioning of your product?
If the product has a large profit margin and the cost of MCU is limited, you may wish to upgrade the product to a higher level, which is also a promotional material and leave room for subsequent upgrades.
If the product needs to be carefully calculated and the number of products is huge, saving 1 cent can bring huge profits, then MCU is enough. Even if development is inconvenient, it is only a single expense, moreover, Chinese engineers are not expensive :(
Therefore, it doesn't matter whether you select the single-chip microcomputer. The key is whether you can grasp its nature and quickly bypass the class. Whether your product is successful depends on whether you can best choose the MCU that meets the characteristics of the embedded object.
For more information, see!
Arm vs 8051
1.8051 is 8-bit. Arm is 32.
2. Speed: the arm clock speed can reach 8051 m, and or more than 50 m is amazing.
3. Strong Arm processing capabilities. 8051 focuses on logical operations, and the arithmetic floating point operation is relatively poor.
4. Arm has rich hardware resources, and 8051 of hardware resources are relatively simple.
5. The flash and ram of arm are too large, and 8051 is too small to work.
We take Keil official mcb950 Development Board and mill technology MYD-LPC1857 Development Board to make a comparison, the two are based on 8051 Single Chip Microcomputer and arm Cortex-M3 processor.
+ ---------------- + ------------------------- +
| Function| Mcb950|MYD-LPC1857|
| Architecture | 8051 | arm Cortex-M3 |
| Clock speed | 12 MHz | 180 MHz |
| Internal RAM | 0.5kb | 136kb |
| Internal flash | 8 KB | 1 MB |
+ ---------------- + ------------------------- +
Arm vs AVR
I think the biggest feature of AVR is that it can operate on two commands at the same time. In fact, AVR is an optimization of the simplified command of the AVR,
But AVR cannot run a variety of operating systems like arm. Although it can run operating systems, the overall load is not as good as arm, but it feels that its application is relatively simple and easy to use, although arm is stronger than AVR, it feels that it must be used with the operating system to reflect its advantages. However, embedding the operating system is a relatively complex system, it takes a lot of time to understand and digest. It seems that arm has powerful computing power and portability, and is generally used in high-end instruments and equipment; this is why I want to learn about arm.
I saw hyloo's question, which I wanted to answer, but I wrote a lot about it, which is not very cost-effective. So I decided to open a post by myself, and I would like to raise my name by the way.
I have been using AVR for two months, and I have read only two books on arm. Therefore, please advise me if I have any mistakes.
The AVR and arm mentioned here are generic, that is to say, many chips, including all machine instruction sets and the two systems, are common, not just a technology or a chip.
First of all, there are similarities. Both are examples of modern CPU design. Based on the guidance of the RISC, many factors such as the CPU kernel area, speed, and advanced language support are fully taken into account at the initial stage of system design. I learned from problems exposed by commercial systems such as 51 and x86 20 years ago. from a modern perspective, they are all very advanced designs. Most importantly, GCC provides great support for both arm and AVR and is free of charge. There are many open resources available on the Internet, especially in foreign countries. Both belong to the market introduction stage in China. In terms of sales, there are two well-known domestic MCU companies acting as agents, and coincidentally, they are all profitors from Guangdong (HA, joke, do not take it seriously ).
There are many differences between AVR and arm:
Arm is an IP core that can be integrated into the design of major chip vendors. It is like C ++ in the software language. If you want to switch to a vendor or a product that is too expensive, there will be other vendors to compete, at least theoretically, you will not be trapped by one vendor.
In this respect, there was almost no semicolon in the Atmel family. You can select only one model in his series and cannot select the manufacturer. For example, Java in the software language. Although it is currently free (the Java SDK, not the AVR) or the price is low, the market prospects are more in the hands of vendors.
In terms of actual product cost, the AVR is better than arm. After all, the AVR is an 8-bit machine, and its configuration of any peripherals is cheap. Because the speed is lower than arm, the PCB version is also better designed, the 20 MHz digital circuit basically only needs to be connected, so you don't have to think too much about signal integrity or something. While arm's speed can easily reach 100 MIPS, 32-bit CPU is not blowing, the speed of the AVR is simply incomparable, but there are more problems, it requires four layers of PCB, moreover, peripherals are much more expensive.
In terms of functions, arm is much better than AVR. arm can be used as a PDA, mobile phone, or AVR. The worst thing is how much Linux can do on arm, although almost no one in China is running Linux on the ARM platform and is willing to publish the technology (I am trying to develop in this direction), the future is definitely bright. The functional advantages mean that arm has a wider application scope than AVR. Therefore, if "We use AVR in Middle School" is correct, we should use arm in university.
We can see that at91m55800a Based on the ARM core of ATMEL includes a lot of AVR peripherals, but still lacks Twi/I2C, variable gain ADC, EEPROM and other useful components. However, there is no doubt that arm's external expansion and peripherals are more powerful than AVR.
In terms of operating system and software source code resources, arm has advantages over AVR thanks to Linux. But the embedded operating system on the AVR is not without, uC/OS-II is good, if you can run on cheap chips such as mega8515, I think the 350 yuan arm version is hard to sell. Do you have any plans for this dual-Dragon?
In terms of debugging methods, arm should be better than AVR. I can also buy a JTAG interface Simulator for AVR, but the supported chips are limited. There are quite a few methods for debugging in the arm book, I am not familiar with and lack experience here.
Finally, let's end with a personalized conclusion. In view of GCC's excellent support for AVR and arm, I decided to study it well. Here, I will give an advertisement for GCC by the way, we can also see that the best way to use the C language is to directly use the English language. The GCC compiler is obviously the best choice.
The MSP430 will be dedicated, low-voltage, and low-power. More models should appear for measurement in different scenarios. The 430 programming method is to switch between low-power mode and task to reduce the power consumption of the system and meet the portable and energy-saving requirements.
Arm is an advanced and streamlined command Machine Based on Soft cores. In terms of high-end applications, such as the development of the motherboard of an embedded system, it requires a variety of interfaces. hardware software implementation is relatively easy to reduce costs and is widely used. Arm needs to run operating systems, such as UCOS, WinCE, and Linux, mainly because embedded system development is often multi-task and real-time.
The embedded processor does not mean that the speed will not die slowly. It has a long tradition and a glorious history, but arm has done a great job in the domestic market, the popularity of the cold fire is greatly affected. In fact, when huaheng's cold fire series of development boards were sold in full swing, arm had not yet formed a climate in China. In, arm and arm started to get popular. At that time, they were in the embedded field, the most popular keyword should be the 2410 Development Board.
The first time in the cold fire was 68 K, which was initially applied to Apple, earlier than Intel's 8088. In, sales of K chips were almost the sum of sales of all other embedded processors (including arm, MIPS, and PowerPC) on the market. Even now, we know that the customer who made an IP set-top box is a relatively large customer for Hua Heng, and the customer volume is 10 K, it's still a small customer at the Apsara stack. Well, the cold-fire chips are not outdated, but they are only well-known in China.
The cold fire series have many processor models with different characteristics.
Let me talk about the mcf52x series. Some time ago, Hua Heng and Fei Sikar went on technical training and marketing, so I also needed to study mcf52x.
This processor is mainly used for control. Its biggest characteristic is its cost advantage. Don't look at the price of this chip as if it is not high, but its internal SDRAM and flash, it is fully qualified for the small and compact RTOS operation requirements, making the cost of a low-end product very cost-effective.
Currently, ColdFire has four series V1/2/3/4 high-end V3/v4, which are rarely seen in domestic applications.
At present, the V2 kernel series are widely used in China. There are also many types in this series. The cf5213 mentioned above is a single-chip solution. Many other types of chips use Extended Program/data storage.
The configuration of the minimum single-chip cf5211 is as follows:
Clock speed 80 MHz Max
Built-in Mac and hardware Divisor
16 kb SRAM, kb Flash program memory
4-ch/32-bit + 4-ch/16-bit Timer
8-ch 12-bit ADC
3-ch UART
I2C; QSPI
BDM/JTAG debugging interface, k cw support for free
64pin lqfp 10 thousand pieces public reference price US $4.99
1. the ARM processor has operating modes such as system mode and user mode, and each mode has corresponding proprietary general-purpose registers. Therefore, you can quickly switch between different modes, this is very beneficial for the operating system, but DSP has no such considerations. In addition, because the ARM kernel is connected to the on-chip peripherals through VPB, the two are relatively independent. In this way, the arm's on-chip peripherals are more flexible.
I don't know if it's right. After all, I don't know much about DSPs. Thank you for your correction.
2. The two functions are different. Arm is a 32-bit CPU. It is used for mobile digital products and industrial control. It can be understood as a CPU. It can be used as a single-chip microcomputer on an operating system. DSP is used for digital signal processing. Taking TI's things as an example, 2000 is used for control, 5000 is used for static image processing, 6000 is used for dynamic image processing, and 8000 is used for multi-DSP joint control.
3. Arm is the name of a company, DSP = Digital Signal Processing
Arm has designed a variety of CPU cores, which are dominated by 32-bit CPU cores. The CPU cores designed by arm are commonly known as "ARM chips ", however, strictly speaking, an ARM chip should be a single-chip microcomputer designed with the CPU core of the arm. An SoC chip made with the CPU core of the arm is generally not called an ARM chip.
Therefore, the ARM chip mentioned by the landlord is an advanced general-purpose microcontroller.
DSP = digital signal processing, that is, digital signal processing. Like "Arm chip", "DSP chip" is commonly known as a series of single-chip microcomputer with DSP function produced by TI.
To sum up, "DSP chip" is a single-chip microcomputer with DSP function, and "Arm chip" is a single-chip microcomputer with or without DSP function;
If the difference is necessary, the "DSP chip" is much weaker than the "Arm chip" If DSP functions are not taken into account, for example, there are many differences in various modes, running OS, MMU, and various peripherals mentioned above.
4. Arm is a well-known enterprise in the microprocessor industry. It has designed a large number of high-performance, low-cost, and low-energy-consuming Proteus processors, related technologies and software. The ARM architecture is designed for the low-budget market as the first chip microprocessor, which is basically the industry standard of 32-bit single chip microcomputer. It provides a series of kernel, system expansion, microprocessor and system chip solutions, the four function modules can be used by the manufacturer to configure production according to the requirements of different users. Because all products adopt a general software system, the same software can run in all products. At present, arm has more than 90 shares in the handheld device market, which can effectively shorten the time for application development and testing and reduce R & D costs. Here is an introduction to the development of the arm structure system.
Digital singnal processor (DSP) is a unique microprocessor with its own complete command system. It uses digital signals to process a large amount of information. A digital signal processor contains control units, operation units, various registers, and a certain number of storage units in a small chip. It can also connect several memories on its peripheral, it can also communicate with a certain number of external devices. It has all-round software and hardware functions and is itself a microcomputer. The DSP adopts the Harvard design, that is, the data bus and the address bus are separated, so that the program and data are stored in two separate spaces, allowing the overlapping of commands and execution commands. That is to say, when executing the previous instruction, the next instruction can be taken out and decoded simultaneously, which greatly improves the speed of the microprocessor. It also allows transmission between the program space and the data space, because it increases the flexibility of the device. The working principle is to receive analog signals, convert them to 0 or 1 digital signals, and then modify, delete, and strengthen the digital signals, in addition, the digital data is interpreted back to the analog data or the actual environment format in other system chips. It is not only programmable, but also runs at a rate of tens of millions of Complex Instruction programs per second, far exceeding the general-purpose microprocessor. It is an increasingly important computer chip in the digital electronics world. Its powerful data processing capability and high running speed are two of the most commendable features. Because of its powerful computing capability, fast speed, small size, and high flexibility in using software programming, it provides an effective way to engage in a variety of complex applications. According to the requirements of digital signal processing, DSP chip generally has the following main features:
(1) One multiplication and one addition can be completed within one instruction cycle;
(2) programs and data spaces are separated and commands and data can be accessed simultaneously;
(3) Fast RAM in the chip, which can be accessed at the same time through an independent data bus;
(4) hardware support with low overhead or no overhead loop and redirection;
(5) Fast interrupt handling and hardware I/O support;
(6) There are multiple hardware address generators operating in a single cycle;
(7) multiple operations can be executed in parallel;
(8) supports pipeline operations so that operations such as finger fetch, decoding, and execution can be performed in overlapping ways.
Of course, compared with general-purpose microprocessor, other general functions of DSP chip are relatively weak.
What is the difference? : Arm has strong transaction management functions and can be used to run interfaces and applications. Its advantages are mainly reflected in the control aspect, while DSP is mainly used for computing, for example, encryption and decryption, modulation and demodulation, and other advantages are powerful data processing capabilities and high operating speed.
If it's so cool, just try again.
The main advantage of DSP is speed. It can perform a multiplication and an addition at the same time in a command cycle, which is very suitable for the needs of Fast Fourier transformation. DSPs have specialized instruction sets, mainly for communication and multimedia processing, while arm uses the RISC Instruction Set (of course, the arm's E series also supports DSP instruction sets) for general processing. A very vivid analogy is that arm is a high-end vehicle, while DSP is a high-end sports car. However, there is also a dual-core chip that uses DSP to take the finger and perform computation. When the floating point computing volume is large, arm assists in processing usually, and arm is only responsible for peripheral control.
The driver of different application requirements is also one reason.
Any technology must be well developed without the drive of the market. arm can be called a high-performance single-chip microcomputer. After the traditional single-chip microcomputer fails to meet the requirements, people began to seek a higher-performance single-chip microcomputer. Therefore, this batch is driven by the needs of single-chip microcomputer users, arm began to develop gradually. DSP is designed for users who are pursuing high-speed computing. these are two independent development directions. Each technology maintains its own personality and has evolved to the present. let's talk about it.
5.
The memory architecture and Instruction Set have different characteristics.
For the convenience of Memory Management (easy to support the operating system), the single-chip microcomputer generally adopts the Von noriman structure with unified instruction and data space encoding. To speed up data throughput, DSP basically adopts an independent Harvard structure with commands and data spaces.
The single-chip microcomputer has fewer instructions on digital computing. In order to perform fast digital computing and improve the efficiency of common signal processing algorithms, the DSP has added many instructions, for example, commands for single-cycle multiplication and addition and subtraction in reverse order (which is particularly useful in FFT, rather than reverse order of ARM) and block repetition commands (reducing jump latency), etc, you can even design a sequence of many commonly used operations to complete a single instruction in one cycle (for example, an instruction is used as a multiplication to accumulate the result, at the same time, the operand address is added to 1 in reverse order, which greatly improves the signal processing speed. Because the reading and write-back volumes of digital processing are very large, in order to increase the speed, commands and data spaces are separated to access two spaces using two buses. At the same time, generally, there is high-speed RAM in the DSP. Data and programs must be loaded to the high-speed slice Ram before they can run. To improve the efficiency of digital computing, DSP sacrifices the convenience of memory management and has many poor support for multiple tasks. Therefore, DSP is not suitable for multi-task control.
Arm vs FPGA
First of all, I can only give comments without forcing your right to choose.
Second, I can only tell you what it will look like after learning both FPGA and arm.
Let's take a look at the above two points.
(Ha, I am the opposite of you. I am working on FPGA. I am also working on FPGA at the company, but I have to learn about arm due to hardware architecture issues)
I don't think FPGA and arm have to be compared. However, the same way is achieved. Arm is not enough to deal with problems that cannot be handled using a classic FPGA structure. The final purpose of the device is to help us solve the problem.
1. If you decide to focus on arm, the advantage of FPGA learning is that you understand arm, so you will have an advantage after learning EDA:
The architecture of the combination of CPU and FPGA, which is a solution and widely used in embedded systems (You have to give me a cell phone for example, I can't help, laugh ~~) Because the system is always layered and the CPU always works at the application layer, as FPGA, although it works at the link layer (helping the CPU store data, rate matching, interface implementation, etc) however, in this architecture, FPGA runs more stably than the CPU. This architecture is flexible. External CPU interfaces can be expanded at will, as long as you write your own FPGA Program Implementation interface.
Advantage: a better understanding of the system architecture while learning about the Core algorithms of the system. This is advantageous, because the system must be structured before using the system. In the above CPU and FPGA architecture. FPGA acts as memory from the CPU perspective. From the perspective of the underlying layer, FPGA acts as a driver. If FPGA cannot be used under this interface, arm startup is only half-left. All arm functions cannot be used.
(I am talking to software programmers about how FPGA and PXA255 work collaboratively. They are surprised that FPGA can reuse addresses or use transparent dual-block memory, I am surprised to write different results for reading the same address because they do not understand how FPGA works. They cannot understand the flexibility of FPGA. But you can .)
Option 2: Focus on FPGA.
I think there are a certain number of misunderstandings about FPGA and arm in the school, as if students or a few teachers will think: "Linux and Windows are similar things", "VHDL is changed according to VC", etc, these may make sense to the extent above (and this meaning is almost a counterexample that can be overturned ). FPGA is very powerful, and both DSP and CPU can be embedded. It is also a soft core, which is very good. The main reason why FPGA is not available is the cost, and its technology is mature.
So your advantage is: After you understand FPGA, You can embed the soft kernel into FPGA. At this time, the company does not need to configure your arm or CPU.
Refer to the website to see what FPGA has developed:
Xilinx market solutions: http://china.xilinx.com/technology/
Altera's market solution: http://www.altera.com.cn/technology/tc-index.html
To sum up one sentence: A hero does not care about the source, so it doesn't matter what he learns, how he chooses, and how he or she interests himself.
Arm is a company name, also known as a processor, or a technical name. Authorization for Chip Design Technologies. At present, we use the arm intellectual property (IP) core microprocessor, which we usually call an ARM microprocessor, has been used in various product markets, including industrial control, consumer electronic products, communication systems, network systems, and wireless systems, the Application of ARM-based microprocessor occupies a market share of more than 75% of 32-bit Proteus microprocessor, and ARM technology is gradually penetrating into all aspects of our lives. Arm is a company specializing in chip design and development based on the RISC technology. As an intellectual property supplier, arm itself is not directly engaged in wafer production. by transferring the design license, the partner company will produce distinctive chips, major Semiconductor manufacturers in the world (rfid rf express Note: Philips, Ti, Intel, Broadcom, Atmel, etc.) have purchased the ARM microprocessor cores designed by arm companies, according to their different application fields, add appropriate peripheral circuits to form their own ARM microprocessor chips to enter the market. At present, dozens of large semiconductor companies around the world use arm's authorization, so arm technology is supported by more third-party tools, manufacturing, and software, in addition, the cost of the entire system is reduced, so that products can easily enter the market and be accepted by consumers and more competitive. The ARM architecture is designed for the low-budget market as the first chip microprocessor, which is basically the industry standard of 32-bit single chip microcomputer. It provides a series of kernel, system expansion, microprocessor and system chip solutions, the four function modules can be used by the manufacturer to configure production according to the requirements of different users. Because all products adopt a general software system, the same software can run in all products, which can effectively shorten the time for application development and testing and reduce R & D costs.
Digital singnal processor (DSP) is a unique microprocessor with its own complete command system. It uses digital signals to process a large amount of information. A digital signal processor contains control units, operation units, various registers, and a certain number of storage units in a small chip. It can also connect several memories on its peripheral, it can also communicate with a certain number of external devices. It has all-round software and hardware functions and is itself a microcomputer. The DSP adopts the Harvard design, that is, the data bus and the address bus are separated, so that the program and data are stored in two separate spaces, allowing the overlapping of commands and execution commands. That is to say, when executing the previous instruction, the next instruction can be taken out and decoded simultaneously, which greatly improves the speed of the microprocessor. It also allows transmission between the program space and the data space, because it increases the flexibility of the device. The working principle is to receive analog signals, convert them to 0 or 1 digital signals, and then modify, delete, and strengthen the digital signals, in addition, the digital data is interpreted back to the analog data or the actual environment format in other system chips. It is not only programmable, but also runs at a rate of tens of millions of Complex Instruction programs per second, far exceeding the general-purpose microprocessor. It is an increasingly important computer chip in the digital electronics world. Its powerful data processing capability and high running speed are two of the most commendable features. Because of its powerful computing capability, fast speed, small size, and high flexibility in using software programming, it provides an effective way to engage in a variety of complex applications. According to the requirements of digital signal processing, DSP chip generally has the following main features:
(1) One multiplication and one addition can be completed within one instruction cycle;
(2) programs and data spaces are separated and commands and data can be accessed simultaneously;
(3) Fast RAM in the chip, which can be accessed at the same time through an independent data bus;
(4) hardware support with low overhead or no overhead loop and redirection;
(5) Fast interrupt handling and hardware I/O support;
(6) There are multiple hardware address generators operating in a single cycle;
(7) multiple operations can be executed in parallel;
(8) supports pipeline operations so that operations such as finger fetch, decoding, and execution can be performed in overlapping ways.
Of course, compared with general-purpose microprocessor, other general functions of DSP chip are relatively weak.
FPGA is the abbreviation of field programmable gate array (Field Programmable Gate Array). It is a product of further development on the basis of PAL, gal, PLD and other programmable devices. It is a dedicated Integrated Circuit (ASIC) the most integrated. FPGA adopts a new concept such as logic cell array (LCA), which includes configurable logic modules CLB (Programmable Logic block) and output input module IOB (input output block) and interconnect. You can reconfigure the logic modules and I/O modules in FPGA to implement your logic. It also features static repeatable programming and dynamic restructuring of the system, so that hardware functions can be modified through programming like software. As a semi-customized circuit in the specialized IC (ASIC) field, FPGA not only solves the shortcomings of the custom circuit, but also overcomes the shortcomings of the limited number of original programmable device door circuits. It is no exaggeration to say that FPGA can be used to complete the functions of any digital device, from high-performance CPU to simple 74 circuit. FPGA is like a piece of white paper or a pile of wood. Engineers can design a digital system using a traditional schematic input method or a hardware description language. Through software simulation, we can verify the correctness of the design in advance. After the PCB is complete, you can use the online modification capability of FPGA to modify the design at any time without changing the hardware circuit. Using FPGA to develop digital circuits can greatly shorten the design time, reduce the PCB Area, and improve the system reliability. FPGA is set by the program stored in the On-chip RAM. Therefore, you need to program the on-chip RAM when working. Users can adopt different programming methods based on different configuration modes. When power-on is enabled, the FPGA chip reads the EPROM data into the on-chip programming Ram. After the configuration is complete, the FPGA enters the working state. After power loss, FPGA becomes white, and the internal logic relationship disappears. Therefore, FPGA can be used repeatedly. FPGA programming does not require a dedicated FPGA programmer. You only need to use a universal EPROM or prom programmer. When you need to modify the FPGA function, you only need to change an EPROM. In this way, different programming data of the same FPGA can generate different circuit functions. Therefore, FPGA is very flexible to use. It can be said that FPGA chip is one of the best choices for small batch systems to improve system integration and reliability. Currently, Xilinx and Altera are the leading FPGA companies.
Arm has strong transaction management functions and can be used to run interfaces and applications. Its advantages are mainly reflected in the control aspect,
DSP is mainly used for computing, such as encryption and decryption, modulation and demodulation. Its advantage is its powerful data processing capability and high operating speed.
FPGA can be programmed with VHDL or VerilogHDL, which is flexible. Due to programming, debugging, re-programming, and repetitive operations, it can be fully designed for development and verification. When the circuit has a small number of changes, it can better show the advantages of FPGA, its field programming capability can extend the product's life in the market, and this capability can be used for system upgrades or debugging. The current trend of FPGA is the possibility of replacing the first two. The high-speed DSP processing capability is achieved by adding multiplier and DSP Blocks in FPGA. Built-in hard core CPU or soft core CPU in FPGA (Xilinx has a PowerPC hard core product with a microblze soft core. Can be a comprehensive device capable of implementing digital logic and adapting to embedded development.
DSP-data processing; CPU-control; FPGA-Interface Conversion
FPGA-circuit implementation through programming
DSP-data processing and control through programming guidance Processing Units
CPU, complicated stuff
I think FPGA is a programmable device. It can also be used to implement DSP or CPU, but the possibility cannot meet the requirements.
DSP is used for digital signal processing,
CPU is a type of controller. There are many types of controllers, such as Pentium and single-chip microcomputer.
CPU is a circuit, while FPGA is a circuit implementation method relative to ASIC.
To a large extent, the DSP and the bed are similar in that they all perform operations to process data. The internal structure is complicated. FPGA is a controllable logical device and can be programmed as needed.
In terms of operating frequency, the CPU on the PC is measured in GHz. However, if the frequency of an FPGA chip exceeds 1 GHz, basically the current MHz is capped.
Single Chip Microcomputer Low Speed
DSP High Speed
FPGA ultra-high speed
FPGA programmable devices allow users to implement interface protocols, protocol parsing, data processing, and control based on their own needs. The advantage is that users can program as they like.
DSP is actually an ASIC embedded with CPU and some dedicated digital processing modules (micro-engines, multiplication circuits), dedicated to high-speed data signal processing. However, the main execution process is implemented by software.
CPU is undoubtedly a central processor. It performs operations based on user instructions. In fact, the most basic operation is multiplication.
Because DSP and CPU are dedicated integrated circuits, the internal clock frequency can be very high and the density can be very high, especially the CPU, often representing the latest and highest semiconductor technology. Although FPGA does not have as high execution frequency as DSP and CPU, and the process may not be so advanced, FPGA can achieve higher data processing capabilities than CPU and DSP through parallel processing and pipeline. In addition, FPGA is now embedded with DSP modules and CPUs, which can be designed into on-chip systems. For boards that do not have high CPU requirements, you only need to use the CPU in FPGA. You do not need to attach another CPU chip. Because the three devices have their own advantages, no one can replace them. The main competition is between FPGA, DSP, and low-end CPU.
DSP is a software implementation algorithm
FPGA is a hardware implementation algorithm, so FPGA processing speed will be higher
One important reason why FPGA is faster than DSP is that FPGA can implement parallel operations. Due to hardware structure constraints, DSP mainly relies on software to extract commands for execution.
This article is from mill technology. The original Article address is http://www.myir-tech.com/resource/502.asp.
Comparison between ARM, 8051, AVR, MSP430, ColdFire, DSP, and FPGA systems
Start building with 50+ products and up to 12 months usage for Elastic Compute Service
1 on 1 presale consultation
24/7 Technical Support 6 Free Tickets per Quarter Faster Response
Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.
A comprehensive suite of global cloud computing services to power your business
Payment Methods We Support
