The most massive misconception around Moore's law

^{Figure from the original 1965 article.}

What will happen if you ask almost anyone, from readers of this blog to Wikipedia, what was the observation made by Gordon Moore about the development of electronics? The answer will be doubling every few years of everything and everything, starting from the number of transistors and ending with performance.
But all this is only a consequence, but let's understand how it really is.
* I just want to say that I will not write about what has already been said more than once (a dozen?) Times. I will try to be brief and to the point.

The original article in 1965 was written at a time when no Intel company existed yet, and its future founders were looking for ideas for their business. In particular, the driving force capable of making electronics truly mass. And the main obstacle for this was, first of all, the price.

Accordingly, the initial observation consisted in how the cost of a single transistor behaves, depending on their quantity and production technology.
')


For small (by the standards of a specific technology) devices, the cost of a transistor is almost inversely proportional to their number. But, with an increase in the number of transistors, the crystal area grows, the share of the yield of usable crystals during production decreases and, accordingly, the price increases.
It turns out that for each production technology there is a certain level of device complexity (expressed by the number of transistors in their composition), for which the use of this technology is most beneficial.

The transition to the new process technology is achievable when the cost of the transistor at the new design rate is significantly lower than for the current technology. Here a forecast was made about a tenfold reduction in the cost of the transistor when changing the process technology.

As for the number of transistors in the device for which production is most cost-effective, then, according to forecasts, it actually doubles when switching to a new technology.
But, firstly, this is already a consequence, which can be obtained by constructing minimum cost points in the coordinates of the “year” - “number of transistors”.
Secondly, this forecast has not been fulfilled for a long time in its original form - it should be adjusted to the area of ​​the crystal itself. The last few technical processes instead of doubling the number of transistors bring only a marginal increase in their number and a significant reduction in the area of ​​the crystal itself.

In fact, now that a new microprocessor is being developed, the main constraint is not what capabilities it should have or what performance should be. The main limiter is how much it should cost, from where everything else already flows.

Moore's law was and remains the law of value. And many things would be clearer if people remembered that.

Did you ever know about this side of Moore's law?