Is the anti-interference ability of PIC microcontroller stronger or that of AVR microcontroller?

The anti-interference performance of single-chip microcomputers has always been valued by everyone. There are about ten single-chip microcomputers on the market that I have come into contact with, including Samsung and Hyundai from South Korea; Mitsubishi, Hitachi, Toshiba, Fujitsu, NEC from Japan; EMC, Songhan, McIntire, and Hetai from Taiwan; Motorola from the United States, National Semiconductor's cop8 series, microchip series, TI's msp430 series, AVR series, 51 series, and ST series from European STMicroelectronics. . . . . 

Most of the anti-interference performance of these single-chip microcomputers has been tested by me personally. The machines used are two high-frequency pulse jammers produced by Shanghai Sanji, one is the standard adopted by Europe, and the other is the standard adopted by Japan; the Japanese standard is adjustable from 0ns to 250ns, and the European standard is pulse intermittent (interval time and emission time are adjustable), and the pulse width is also adjustable from 50ns to 250ns; our country adopts the European standard. Under 

normal circumstances, it is considered good if the pulse interference can withstand more than 2000V (it seems that the national power standard in China is 1200V), and some can reach 3000V, so many people are very proud of this. 

Whether the program of the single-chip microcomputer runs normally under high-frequency pulse interference, or whether the anti-interference is passed, some people take the program not flying away, or "freezing" as the standard, and some people take the program running normally without reset as the standard. In many cases, the chip reset program can continue to run, which is not very clear on the surface. I usually check whether the MCU resets under interference. If it resets, I think it is not working. It is better if it does not reset and the program runs normally than if it resets. 

Many people are very happy to see that the anti-interference of their circuits reaches 2000V or 3000V. In fact, the anti-interference of the chip is not necessarily very good. Here I have to talk about the Japanese standard, the form of high-frequency pulses sent continuously. Don't underestimate the difference between a continuous and an intermittent. In fact, if you have the opportunity, test your chips and circuits with Japanese standards, you will find that they are almost very different from European standards. You will be very sad if you use Japanese standards, because most MCUs cannot pass! 

The Japanese standard is 1600V. The dozen or so MCUs I mentioned above: 

ST ≥1800 
Mitsubishi ≥1800 
Fujitsu and Hitachi ≥1600V 
NEC ≥1500 
Toshiba ≥1300V 
Motorola ≥1300 
Samsung ≥1300 Hyundai 
≥800 
Microchip ≥700 
National Semiconductor cop8  ≥500
AVR and 51 series ≥500 

I haven't tested the ones that don't have data here, but I know that EMC's 28pin has design defects (EMC's own people said); Hetai's is said to be able to pass 3000V under European standards. 

Everyone can compare the MCU you use to see what level it is. However, if anyone is hit, don't be too sad, because I have compared it, and there are also circuits that are 5-600V in Japanese standards but can reach 1500V in European standards, on the same board; I have encountered such situations several times. In most cases, if it fails the European standard, it also fails the Japanese standard; if it passes the Japanese standard, it may occasionally fail the European standard. 

It should be noted that many MCUs often reset under high-frequency interference pulses, but the program operation does not seem to be affected. If it crashes, you can increase it by 1-200V, Japanese standard. The European standard will be even higher. 

When you choose a MCU, take a closer look at the information. If the reset time is at the us level, it will "die ugly" if it adopts the Japanese standard. This is why the cop8 series and avr series can't reach the Japanese standard of 1000V no matter how hard they try, and the avr can only reach 700V at most. 
As for the price of MCUs, I don't need to say more here. ST is the most expensive, followed by Japan and the United States, followed by South Korea, and then Taiwan. 

ST MCUs are expensive in anti-interference performance, which is excellent! And it has built-in eeprom. Avr's flash + built-in eeprom is a good selling point. I don't know if you know that National Semiconductor also has very good flash, and the price is slightly lower than avr. But I don't know if it has built-in eeprom. The characteristic of Msp430 series is a/d. Actually, if you want to use a cheap 16-bit a/d microcontroller, I recommend one from Taiwan Songhan and one from Samsung, which are cheaper than most msp430s. Especially Samsung's 16-bit a/d+flash, super cheap! But there is also a prerequisite, quantity! Development systems from ST, Mitsubishi, and Samsung are very expensive, all above 5-6000 RMB. Most of the others are between 1000 and 3000. 

In terms of microcontroller structure, I think the cisc structure is better than the RISC structure. The risc structure has fewer instructions, and I don't think there are many benefits other than that. The inconvenience brought by this is quite a lot. For a simple function, several instructions are repeated, and a program with dozens of instructions is produced; and some rams need to be paged, and some roms also have paged. It's annoying to jump around when programming! The instructions of the CISC structure are many, but not difficult to remember. You will remember them naturally if you use them more often. The structure is neater, so you don't have to worry about paging (even if there is paging). You can program wherever you want. 

Now I basically only represent and use Samsung and Hetai microcontrollers. They are cost-effective. If you need to use LCD drivers, I suggest you use these two companies. Hetai doesn't have much flash, but with special methods, you can burn a chip several times. Samsung Hetai's microcontrollers have been selling very well in the past year or so! In addition to having LCD drivers, the anti-interference performance is also good. 

In comparison, Samsung's microcontrollers are slightly more expensive than Hetai's, and the development system is very expensive (but you don't have to buy them, the agency will usually lend them to you - you have to pay a deposit, hehe!)! But Samsung's microcontrollers are really good. Many people use 9454, right? What about 9228? It's not bad either, right? But I tell you, there are better ones than these two, flash+4*18 LCD driver+10bit a/d. In addition, Samsung also has a chip specifically used for remote control, 64pin, 4*32 LCD driver, which is also very cheap. Even if it is used as an LCD driver chip, it is cost-effective. The price is about the same as two LCD driver chips ht1621. 

But Samsung microcontrollers are also not easy to use. Most of them are streamlined instructions. A small function requires you to use several instructions several times. 
Hetai's microcontrollers have good performance and the development system is also cheap. Especially the ones with LCD drivers may be worth considering. 

Many of the technical issues mentioned here are just personal experience. You can learn from them. After working in technology for a long time, everyone has their own characteristics and strengths. Please criticize and correct if there are any inaccuracies!