There are just too many areas for the government to focus on. It is as simple as that. Also China initially had neither the capital nor demand for such products so it did not make large inroads into it. I still remember the talk about Grace Semiconductor and other efforts from way back in Jiang Zemin's time. What a huge flop. But SMIC basically cloned the TSMC business model and has had a lot of success with it.
The military has little demand for semiconductors compared with the civilian market so it is easier for them to acquire what they need on the gray market or even the black market. Also, they do not need to respect patents and other commercial shenanigans. Nor do they need the latest process technology and the smallest transistors. Typically processors like that are not radiation hardened. Remember when the US blocked Intel Xeon CPU sales for Chinese supercomputers? China quickly developed its own CPU designs in response based on the MIPS architecture. Then the US started harassing Chinese commercial companies which made MIPS architecture derived processors like Longsoon. Later Chinese investors purchased Imagination Technologies which held the MIPS patents. Now, I do think it will take way more than 2 years for China to dominate most of the cycle, but quite simply they might not need to do it all.
Already we have heard that Samsung allegedly has built a production line devoid of US equipment just because they expected this after the ZTE and Huawei bans. I don't know if this is true or not but it would go a long way to help existing Chinese designs get to manufacture until China can produce the chips with its own equipment or have its own production lines devoid of US equipment. As for why China spent the money on US equipment in the first place? Well, they had to recycle those dollars somehow, so might as well get something useful out of it. That was the thinking behind it I think. Plus the expectation was this made them less liable to be sanctioned in the first place. But it backfired.
It is important that China invests in equipment from foreign companies which challenge the US sanctions regime as this will make the global semi market spin out of the US's orbit even faster. China needs to manufacture the process inputs by itself as quickly as possible. I think this is even more important than having the machine tools ready first.
China needs to convince the rest of the Asian semiconductor powers that it isn't in their best interest to follow the US into their path to mercantilism. Taiwan basically threw down the towel and sold its memory industry to Micron (US) for peanuts a couple of years back. Probably sour from that. In Japan, Toshiba could have been acquired for peanuts by Western Digital (US), had it not been for Japanese government intervention.
But, like I said way back, the US sanctions and Trump came way too late. China already has most of the technology it needs to develop its own cycle and even if they need to reverse engineer the rest they have the resources to do it. In a decade I doubt we will be talking about this. China has the manpower, the capital, and the will to continue on this path and the US has long since stopped being the leading semiconductor manufacturer. While they still dominate some key technologies I think it is a matter of time until they don't. The fact they dominate EDA software tools is simply driven by the fact most of the advanced chip design is still made in the US. When it isn't the main EDA software tool vendors will change as well. Of course, it isn't enough that China manufactures immersion lithography tools, or even the EUV tools as an example, China needs to have research efforts not only to reach the current state of the art but to get to the step beyond that.
But no, I don't think the step beyond that will be quantum computing, that is a bridge too far still. I expect some other computing architecture to be dominant before that.
