Manufacturing Nodes: Why They Appear First in PostPC

If we look at the market for PostPC devices we will see how there are processors using the 5 nm node, while in PC we see how we seem backward at 7, 8 and even 10 nm. What is it that makes new manufacturing nodes such as 7nm and 5nm debut first on PostPC devices, is it a trend that we will see in the future?

We understand as a manufacturing node all the technology that is used to manufacture processors with a certain size in their transistors. New nodes appear regularly that allow chips with higher capacities.

Manufacturing Nodes

But for some time now we have seen how each new node is not released on PC but on PostPC devices such as smartphones, tablets and the like. But all this has a very simple reason to understand.

New nodes debut on PostPC devices

Apple M1

In recent years we have been able to observe how the new manufacturing nodes of processors of all kinds have not been released with high-performance products, whether they are CPUs, GPUs and even SoCs.

Every time a new node appears, we get our hopes up to see which high-end processor we are going to see, but this is not true from the start and we have to wait several years to finish seeing the use of these manufacturing nodes in high-end processors .

This has the consequence that for example the 7nm node from TSMC was released for the first time in a smartphone SoC, as well as the recent 5nm node from the same foundry.

Manufacturing maturity and cost of nodes

Evolución Coste Nodos

Processor manufacturing nodes mature over time, this means that they improve their performance. But what do we understand as performance in this case? Well, due to the fact that the amount of errors per mm of manufactured chip is getting lower and lower. When the error rate is high, then it reaches the point where manufacturing a chip from a certain size becomes counterproductive due to the fact that the number of chips that would come out per wafer becomes so little that it is not profitable to manufacture.

Smartphone SoCs are usually below 120mm2, they are small enough to launch a new manufacturing node. In addition, the high margins of this market allow them to finance the high cost of deployment of the new manufacturing node. Another reason has to do with clock speed and consumption, this is directly related to voltage and PostPC devices use speeds clock and thus lower voltages than desktop.

Manufacturing nodes start with clock speeds much lower than their ideal, which is why data is often leaked from future PC processors that are manufacturing samples that operate at much slower speeds than the final processor, especially if they are they use to “release” a new node.

In PC, the first cores to release a node are those with low performance

Intel Lakefield Board

As manufacturing nodes have passed, the deployment of each new node has become more and more expensive. If for example we had seen AMD, Intel or NVIDIA deploy one of their new architectures directly under one of the new nodes then not only the stock would have been negatively affected but also the cost.

A curious case is the decision of AMD with Zen 2, whose first version was the Ryzen 3000, built by several chips in which one part uses a less advanced node while the Zen 2 cores used the 7 nm node, but measuring less. 100mm. Over time AMD released a monolithic version of Zen 2 in the form of the Ryzen 4000 but it took more than a year to do so.

Another example is the Intel Lakefield, whose half is built at 10 nm and it is a chip with very low power consumption and size, being the first chip built under Intel’s 10 nm node.

New nodes will continue to debut at PostPC

CPU comunicación

Since each new manufacturing node is increasingly expensive and part of the initial deployment cost includes the development cost of the new node, by logic we will not only see how the trend will continue, but also the high-end PostPC devices they will be more and more expensive and the maturity period of the nodes will be much longer.

Although the marketing and the predictions speak of a certain node for a certain date in the future, they are really very optimistic predictions and we may find that architectures designed for a specific manufacturing node are delayed or backported to a less advanced node.

Although future manufacturing nodes in many cases can already manufacture this does not mean that they are ready to be used in the latest high-end processor, whatever its nature. that is why we will continue to see them for the first time on PostPC devices.