When we talk about a chip, we often think about it having a really complicated circuits, like the Apple M2 where it has 20 billions of transistors and etc. Back in the days, chip weren’t that complicated so I could digest that they could design it. But nowadays since they’re more complicated, do they really design it when they’re very small like nanometer level?
Ps: I’m not a computer engineering student but feel free to use some technical terms if needed since I take related course to this.
In: 13
There are varrying levels of ‘abstraction’ that VLSI engineers use to build these computers.
You start out at the device level, which is literally 1 transistor. You design it to have a specific width, area, etc to match some form of equivalent reisstence and current capability.
Then you move up a level and put that device into a circuit. The easiest of which is the CMOS inverter. Its 1 PMOS and 1 NMOS transistor tied together. Plenty of resources online to read up on, and I highly recomend you get started there. Its what most college courses use to introduce the CMOS concept.
Now you’ve made a small circuit that could represent a single AND, NOR, XOR, etc gate. You build a component using the abstact symbols for gates to build a component. Examples of these are adders, multipliers, flip flops, etc.
You then use these larger absteact symbols to build modules, such as the ALU, Cache, Control & Decoder units, etc.
Then you use these to build your larger system. Place cores, memory, etc and thats the schematic done. So schematic designers aren’t going in and placing individual transistors on their schematic, thats done for them using design tools.
The actual layout of the chip has many more factors to consider. How many layers are there, how many metals, how big of a chip, the package of the chip, what metals, what size node is there, etc etc. I’m also not a layout engineer, my experience is much more on the analysis side of things, so I can’t really give too much insight on that side of the fence, hopefully someone else chips in.
Early on people could actually design chips manually, drawing the circuits and locating components by hand. Now though with chips having billions of components there is no way for people to do it, so instead custom computer programs form chip layouts based on sets of rules. The specific workings of these programs are closely guarded trade secrets.
To start off with they only need to design one part of the circuit and then copy past that all over the place. You design a one bit ALU and then copy past it to get ten 64 bit ALU for a processor core.
But already in the 80s they came up with gate arrays. You might recognize the name as part of FPGA which was one path the technology went in. A gate array is a matrix of AND and OR gates that given any input can produce any output that have been programmed into it. This did replace read only memory for some tasks but was fast and cheap enough to replace a lot of logic circuits with this as well. A gate array is designed by a computer using a simple algorithm but you can run optimization passes on it to remove transistors and pathways that are not needed. So instead of designing a one bit ALU you just put all the possible inputs and the corresponding output into the gate array algorithm and it will spit out the design of an ALU for you.
This is when things start to get fun, because you can easily take machine code like assembly language and make each operation a gate array which allows you to write code and render it in transistors. Again you have optimizes on multiple levels that helps produce efficient hardware using fairly simple algorithms. But why stop with machine code, you can build higher order languages that compiles down to this code. And this is what people do today. They are not actually placing down each transistor but rather are coding in object oriented languages and just have compilers generate the silicon.
20 billion is only about 142,000 x 142,000.
The transistors aren’t laid individually or even by a robot. Someone draws a grid which has 142,000 cells in each direction, that gets printed as a large template, and then a light is shone through that template onto a much, much, much smaller area of silicon.
By doing so it makes tiny “transistors” (not a physical object like you might expect, but just a layer of two types of different semiconductor, which you can make by exposing a template as above in different ways on the right substance).
So long as the layout of the grid, the interconnects between the grid items, etc. are correct, you get a working processor. Basically it’s like printing a pattern onto a T-shirt but using a magnifying glass in between so that the large “pattern” is shrunk down onto a tiny area of silicon.
Simply put no one designs things from scratch(probably different in the cutting edge tech) but everyone uses programs that helps via the use of macro blocks and logic testers (simulate the processor)
Chips with billions of transistors are designed and manufactured using very advanced machines and technologies. The first step in designing a chip is to create a blueprint or a plan for the chip, which shows how the transistors and other components will be arranged on the chip.
Once the blueprint is ready, the chip designers use special computer programs and tools to create the detailed instructions for making the chip. These instructions are called “computer-aided design” (CAD) files.
Next, the CAD files are used to make a small, physical model of the chip using a process called photolithography. This involves using light and chemicals to etch the design of the chip onto a tiny piece of silicon, which is the material that most chips are made of.
The silicon wafer with the chip design on it is then placed into a machine called a “stepper,” which uses light and other technologies to create the tiny transistors and other components on the chip. This process is repeated many times to create many identical chips on the silicon wafer.
Finally, the silicon wafer is cut into individual chips and tested to make sure they are working properly. The chips are then packaged and sold to companies that use them in their products, such as computers and smartphones.
So, as you can see, designing and manufacturing chips with billions of transistors is a very complex and high-tech process. It involves many different machines and technologies, and it is done at a very small scale, on the nanometer level.
Latest Answers