I have a very active hardware discussion group with hundreds of tech experts. Among them, there are dozens of entrepreneurs whom I know are the bosses of startup companies. In addition, there are also many top technical talents working in various companies. The most intuitive feeling I get from this group is that these bosses seem to work even harder than the tech experts. They often post screenshots of themselves laying out circuit boards and sometimes ask technical questions in the middle of the night when everyone else is chatting and having fun.
I also have a friend who is a boss in Shenzhen. He often stays up late to design circuit boards and debug code. He would even send me messages in the middle of the night to show off that he had finished designing a board or fixed a certain function. Of course, I usually reply to him the next day at noon because I am sleeping at night, and he is sleeping in the morning. He makes fast – moving consumer electronics and he does make a lot of money. He often changes houses and cars, but he is also really exhausted. He has to be hospitalized several times a year mainly due to overwork. I am curious why bosses in hardware startups all seem to live such a tiring life.
I also know several managers of medium – sized companies in Xi’an. When they encounter some hardware problems that are difficult to solve for a long time, they will come to ask me, or even consult other experts through me. Their hardware is not stable. Once it goes into mass production, problems arise. As a result, even though they have sales channels, they dare not produce and sell in large quantities. Today, another friend contacted me. He encountered some obstacles in product development. Some problems have not been solved for a long time. He doesn’t know whether they are hardware problems or software problems. His colleagues are all very hard – working and responsible, but they still can’t solve the problems. The product cannot be finalized and mass – produced.
The above problems seem to be common issues faced by hardware – oriented startup companies. My suggestion to them is either to outsource the hardware – related development work of the company and let professionals handle professional matters, or to hire top – notch hardware experts to lead the team in hardware development. Rome was not built in a day, but with more people involved, it can certainly be built faster. However, hardware development is not like that. Sometimes, what ten junior hardware engineers can’t solve can be easily handled by a senior hardware engineer in a short time. It’s not that the ten junior engineers are not hard – working enough, but that they lack experience and need more training. This is why it is even more necessary to have a top – notch expert to lead and form a tiered hardware development team. Of course, this is a problem at the team organization and management level, not the main topic of today.
Now, let’s get back to the main topic: Why don’t many companies that make electronic products have hardware teams? First of all, it is difficult to find top – notch experts who can lead the hardware team. To be more accurate, it is even more difficult to afford them. This will create a vicious cycle that is hard to break. Without top – notch experts, it is difficult to quickly mass – produce products, which leads to difficulties in making profits and, in turn, makes it even harder to afford top – notch experts.
Secondly, it is not cheap to support a hardware team. Hardware is not like software, where you can get by with just a computer worth a few thousand dollars. For hardware, everyone needs a multimeter, right? You need an oscilloscope, right? You need to set up a separate area for soldering and debugging, right? You need test personnel, right? If you are making a product with wireless functions, you need a spectrum analyzer, a vector network analyzer, and a comprehensive tester, right? You need RF engineers, right? If you are making power – related products, you need an oscilloscope that can measure the loop, right? If you are making high – speed signal – related products, you need an oscilloscope that can measure the eye pattern, right? In short, a hardware team requires continuous investment and is often the high – ground of a company’s R & D costs. So, if you are not willing to take my advice and outsource these tasks, what other options do you have? Of course, there are. Use more mature products.
For example, if you are making a non – standard product, first of all, there is no need to design the entire system by yourself. There are all kinds of development boards on the market, and you will most likely find one that meets your needs. For example, if you are making a non – standard product with a motor drive, you may even consider using a PLC or a frequency converter, etc., instead of designing your own motor drive board. If you are making a communication – related product, there is also no need to do everything by yourself. Find a mature module, get the product up and running first, sell it, and occupy the market. That is the key. By following this path, you will find that things become much easier. Why? Because the module manufacturers are also competing fiercely. For example, I saw today that a development board of the T113 chip from Allwinner has already been sold for $99. If you design a board yourself, you need to pair the T113 with DDR, eMMC, PMIC, and other components to form the core. You also need to add all the necessary peripherals to make the board work. After several board runs and various tests, several months have passed, and you find that your competitors have already been selling their products in large quantities. So, just buy the $99 T113 development board, verify the functions, use their core board, design a two – layer base board, connect the components, run the board, finish the software, and start selling in large quantities. By this time, your competitors may still be busy testing their boards! Yes, today I want to recommend to the bosses of hardware product startups four development boards based on Allwinner chips from Feiling Embedded.
OK, OK, OK, OK! Too OK! I would be willing to buy these boards at this price even if it meant I could live forever! Originally, it would take our own hardware team several days to design the hardware architecture, then draw the board, run the board, do SMT, get the board back, and debug it. After more than half a month, you would get a board that may or may not be OK. It is really not as good as directly using the above – mentioned OKOKOKOK boards. Ambitious bosses may choose to design their own boards, but smart bosses will definitely choose the OK boards! So, let me reiterate the rules here. Bosses, pay attention and place orders with your company as soon as possible. What? You say you are not a boss yet? Here comes the chicken soup for the soul: You won’t have a company no matter how many times you think about it, but if you really go and register one, you will immediately become a boss!
No comments