Their semiconductor manufacturing was 10-15 years behind the Western technology. They just didn’t have the capability. Despite that they had good brains and delivered efficiently with what they had.

But they simply weren't able to sustain it.

In the West, while the military industry initially pushed computer development, private companies quickly adapted those technologies for the consumer market. Over time, the Western consumer market became vastly larger than the military one.

In the USSR, this cross-pollination wasn't possible because anything that even touched the military was immediately classified as a state secret. This obsession with secrecy even affected civilian infrastructure like nuclear power plants. Plant operators weren't fully trained on how the systems worked under extreme conditions, and they were kept completely in the dark about inherent design flaws—because in the Soviet system, everything was by definition perfect and superior to the West.

Furthermore, because the consumer market was strictly controlled by the government and the party, the Soviet economy lacked any organic market signals regarding what people actually wanted or needed. Apparatchiks had to look elsewhere for data, so they resorted to copying Western solutions—sometimes just copying the basic concept (like a radio where users could choose their own stations), and sometimes cloning the entire machine.

While Soviet scientists had some highly innovative and interesting ideas in the beginning, central planners eventually decided it was faster and easier to copy a Western solution that was already 5, 10, or 15 years ahead in mass production.

There'd be other interesting implications as well, socialist systems were more open to the idea of cybernetics and with a proper computer industry the Soviets might have had more room to explore it.

Mind you I still think it would have likely been impossible for political reasons, there were many structural incentives to falsify economic data in the USSR due to the high degree of corruption and patronage among the nomenklatura. The whole point of cybernetics is to treat economic problems as systems problems and expose data transparently, and given the USSR was structurally dependent on falsifying this data suddenly having an accurate picture might have actually been destabilising kind of like how Glasnost turned out to be.

Another interesting 'Soviets had decent computers' counterfactual is that the Chernobyl disaster might have been prevented, since the Kurchartov Institute would have been better able to characterise the processes in the bottom of the fatally flawed RBMK in low power regimes before it was put into mass production. Again this might not have actually helped, the overconfidence the Soviet system had in its scientific and technical institutions was high and genuinely really interesting.

That was my feeling when I first heard about Lisp Machines. It's unfortunate that I never got to see or use one in person.