Design, cost and development details

I have gone through many things over the past two days. To sum up, I have a new idea:

I went to a software bidding meeting the day before yesterday. The software itself was worth yuan, and the company that included the test mobile phone sold 5 million yuan. Then I won it in Zhongguancun 1 K. At that time, I was surprised that the software was made in China. The 17W was just a retail version, not the source code... he sold a copy of it, which is the profit of the 17W!

These two days, I was very concerned about Apple's macbook air. I was amazed at its light and thin nature. I carefully checked the disassembly diagram. Originally, most of the size was occupied by huge thin lithium batteries. In order to minimize the volume, SSD-based nand flash on-board hard disks, on-board memory, graphics card and CPU are all ultra-small packages. Simply put, that is, all components are placed on the PCB that accounts for 1/5 of the notebook area. So can we be so light and thin.

In the design of my X-MOVE 4.0, in order to ensure that the wireless sensor node as light as possible, I from the selection to design, it is not easy to compress the PCB Board in the nail size, 3mm thickness, integrated accelerometer, gyroscope, magnetic field meter, processor and RF. Unfortunately, the battery has become a short board of system size and design. The battery volume occupies 60% of the total system size, but it cannot be smaller. Otherwise, the power consumption is lower and it cannot guarantee a long endurance. In order to make the handheld node as beautiful as possible and reduce the thickness by half, I had to re-design the circuit and encapsulation, spend more time and rethink all the problems. However, even if there is a 3mm error in size, the Board may be discarded.

Beijing mail has so many innovative projects every year that people are very curious about it, because there is no such thing in the market. But is it true? This kind of things has been around for a long time, or has been thought of for a long time, but it is unreliable and does not meet the requirements of details, so it cannot be promoted (of course, I am also included ).

Someone is doing 3G. when I do LTE, web pages, and software development, I look down on people who are "Single Chip" and think that the small broken chip can only run to 8 MHz. I use Linux, the hardware has a CPU and GPU of G! This is like standing on the first floor. Standing on the 20th floor, you are taller than me? As we do not know, the more advanced things are, the more encapsulated and easier to use. Think about making a Phase-lock loop and GPS. Who is complicated? Phase-Locked Loops are much more troublesome!

Why is the success of "Fruit Ninja", a mobile game that has been used all over the world? Its technology? Its algorithm? Its Interface? None of them seem to be. Its success comes from people's simplest game desires-relaxation, venting, and scoring desires. Maybe a complicated maze game introduces complicated algorithms, but it is rare.

So I realized that it was a long and complex process for Apple to complete technical verification-prototype design-prototype-product-exquisite product-perfect product, they can only be exquisite products. It may take us one hour to do one thing, but it will take us 50 hours to do it well. If we want to do it perfectly, it will not be enough for 200 or even 1000 hours. Too many details need to be considered, and each step should be carefully planned. A complete failure will be caused by a slight error.

I do X-MOVE handheld terminal, color screen interface, sensor and system is I carefully designed, I think there is almost no bug. But there will always be problems. In the cold environment, the sensors become very rigid, the screen becomes blurred, and the rubber of buttons start to get faulty!

It turns out hard to understand why Apple is so expensive to sell and why millions of ads are produced. Why is my work always faulty and often crashes, military-purpose chips are more than 10 times more expensive than civilian chips. Why is it difficult for somatosensory technology, image recognition, and muscle sensors to integrate into the lives of the masses for so many years. Because of the details. One failed in the first test, that is, it is immature. The button for launching a nuclear bomb still adopts the most reliable key design, and there will be no "brainwave recognition detection" function in the hundreds of years. It takes too much time and effort to reach perfect details. The so-called "Body Feeling" and "image recognition" are at most what people play when they watch the Internet after a meal. Games are not even willing to use it because they are too tired, what operations are not as simple, fast, convenient, and accurate as buttons, seriously affecting scores ...... sadly, isn't the work I have done for a year without value ?.

What is value? Value cannot be measured by a ruler. I remember there was a story in the past, saying that the gas pipeline in the factory was broken and I couldn't find a problem. So I invited a German expert. The expert went to a certain place and knocked it out, the fault will be lifted immediately. The experts will charge 20 million yuan. People are puzzled. Isn't it just a knock? Experts say that they know where to knock 19 W, and the value is 1 W.

Is what I do worth it? How can this be measured? A song can be sold for millions, and it is not uncommon to draw one million songs. The value of a huge chemical formula patent generates billions of dollars each year, which may be just a small parameter requirement. Many scientific theories cannot be used for practice, but they cannot be regarded as "useless". For example, one of the core concepts of navigation theory and computer graphics is "Quaternary ", people thought it was only a digital dance before 100, but now it is a very important mathematical tool. What's more, the linear algebra of pattern recognition and machine learning is just a boring mathematical game when I was a freshman!

The article wants to express a lot of meanings, to sum up:

1. hardware is very valuable, software is more valuable, design is more valuable, and creativity is the most valuable. The difficulty of system design increases exponentially as required, so both development time and cost prices will increase. So, let alone the price of some expensive items.

2. Maturity is the main factor for entering the market. Even if it is a little unreliable, it may fail.

3. The Details determine everything. Different details of the same thing are put on the Himalayas for a test, and the victory and defeat will soon be seen. In order to satisfy the details, the entire solution should be retried.

4. value cannot be measured by one person, a group of people, or even an era.

5. Don't be too easy-it's better to write code carefully if you are full of "architecture. What is the developer of Android Development who is more powerful than resistor and capacitor for model power? Haha.

Well, all of the above are personal opinions. Those who disagree are welcome to discuss them.