When I heard this familiar voice, Dejie is about to meet you again. In the past two years, Dejie Shishi has contributed 45 exciting conversations to everyone, with themes covering all aspects of the semiconductor industry.
Slide to see more
022 Dejie Moment Schedule
022 Dejie Moment Online Super Party is coming soon, please take a look at the schedule:
There are really a lot of things to learn for embedded Internet of Things. Don’t learn the wrong route and content, which will lead to the salary being unavailable!
Free sharing a package for everyone, almost 150G. The learning content, articles and projects are relatively new and comprehensive! It is estimated that it will cost at least dozens of fish to buy.
Click here to find the assistant to get 0 yuan:
Hot Topics
Here, let me reveal a few topics that Sister Ni is most interested in:
✔ All MCUs can run TinyML? After the device is connected to the network, I hope it is smarter and more convenient. But reality also tells us that not all conditions can be met. For smart devices, bandwidth, storage, and computing are currently a very large consumption, and they will also cause delays due to factors such as distance, reducing real-time. At this time, if some devices can be more intelligent and autonomous, the demand for networking will be reduced a lot.
TinyML is such a technology. With the support of neural network , it can help smart devices complete some low-level edge computing without the need for the host or remote to obtain data before processing.
TinyML further compresses the models generated by the traditional machine learning framework, allowing MCU to perform deep learning inference. It can do something that seemed impossible before, such as the use of gesture control turning on the light, or a magic wand to control the sweeping robot at home.
The following demonstrates that TinyML is deployed on ESP32, identifying, lighting or turning off the LED through air gestures.
Since TinyML can be deployed on ESP32, does it mean that all MCUs can use TinyML to complete edge computing? The answer is yes. comes from Xia Qing, senior engineer at DFRobot, , in the Dejie Time interview, revealed to us why Arduino can play TinyML.
✔ Everything can be the universe?
Scan the mobile phone camera, and a three-dimensional three-dimensional scene appears on the phone. Various scenes and characters can be set up. How can your girlfriend not like such a unique birthday card?
This is one of many application scenarios in the meta universe - adding a real-life AI greeting card. Metauniverse can also do AI navigation, AI medical teaching, three-dimensional description of products, and can also make obscure and difficult-to-understand technologies use AI technology to model and explain them.
Technological tycoons all over the world have joined the meta-universe position. How much technology is behind the popularity of the meta-universe concept? Dejie invites our old friends from and Tongji Zihao, the owner of artificial intelligence UP on B station. For the novices who learn artificial intelligence computer vision , the "three mountains" that must be crossed on the road of popular science learning, and the "secret" of the popularity of the meta-universe.
✔ Are there any disadvantages of silicon carbide?
The raw materials of the chip devices we usually use are mostly silicon, and about 90% of semiconductor devices are made of silicon as the raw material. However, due to the limitations of silicon itself, traditional silicon devices are a bit "stretched" in some applications where device power requirements are high.
So the third generation semiconductor material , silicon carbide, one of the wide bandgap semiconductor materials, began to emerge in more and more applications. What are the advantages of silicon carbide with wide bandgap in
compared with traditional silicon substrates?
) Conductivity loss is small. Wide bandgap material can hinder more voltage per unit thickness, one tenth of that of traditional silicon-based devices. When on, device losses are greatly reduced.
) Fast speed and low switching loss. The internal carrier of wide bandgap devices is much faster than that of traditional silicon-based devices. At the moment of switching, the speed is faster and the switching loss will also drop greatly.
When discussing silicon carbide devices, we often only see their advantages. Little do we know that silicon carbide also has disadvantages. ON Semi's chief wide bandgap device expert Dr. Wu Tong summarized that before wide bandgap devices are used on a large scale, there are still three pain points that need to be solved urgently:
Pain point 1: It is necessary to produce wide bandgap material chips on a large scale and reduce costs. If costs are reduced, the market for silicon carbide will become wider and wider.
Pain point 2: A brand new packaging technology to cope with high temperature and high voltage working environments, requiring more advanced design and processing technology.
Pain point 3: System integration and upgrading, you cannot simply use a system for silicon-based devices. More complex designs for wide bandgap devices are required, with more functional drive systems integrated for optimal performance.
Article link:
https://mp.weixin.qq.com/s/zJlPgKe_POuhLpPIIzUVSA
Reprinted from: Daerwen Says, Author Nimo
Article link: 100 Popular Technologies in the Electronics Industry, All you are interested in are here
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-