Aardvark Daily

New Zealand's longest-running online daily news and commentary publication, now in its 25th year. The opinion pieces presented here are not purported to be fact but reasonable effort is made to ensure accuracy.

Content copyright © 1995 - 2019 to Bruce Simpson (aka Aardvark), the logo was kindly created for Aardvark Daily by the folks at

Please visit the sponsor!
Please visit the sponsor!

Sometimes old is still good

27 September 2021

As we all know, a chip shortage is crippling the car industry.

Manufacturers are pulling their hair out as they watch assembly lines grind to a halt and contemplate the costs of being unable to build $50,000 cars for the lack of a $1 microcontroller.

Meanwhile, some of the world's leading chip fabrication companies are saying "hey, we can make you chips, just no the ones you're looking for".

What's going on here? Why can't those plants with spare capacity (and yes, apparently there are some), make the chips that the car industry so badly needs to start its production lines running again?

Well it seems that it's nothing to do with copyright or anything similar, it's all about the transistor-size and other qualities of the required semiconductors.

Most of the newer chip plants are being built to produce semiconductors that have increasingly smaller transistor sizes. Indeed, we've seen a race to reduce transistor sizes with the likes of AMD now pushing the boundaries to technologies that are right down to 7nm.

Unfortunately for the car makers, the chips they're seeking are much older in design and are built on sizes as large as 90nm -- more than an order of magnitude bigger.

Apparently, those fab plants designed to work with the much smaller technologies can't simply be pivoted to make the older chips that are in such demand right now.

So why don't car-makers simply shift to chips that use the newer fabrication technologies?

Well that's a complicated story involving many factors.

Firstly there's the cost of re-engineering the electronic systems of a modern car to use an entirely new microcontroller. That is far from trivial.

Even once the redesign is done, there would have to be a huge amount of testing and "proving" undertaken to ensure that there were no bugs that might later result in safety issues or mass recalls being necessary. A single recall can cost a maunfacturer billions of dollars so these are things that are to be avoided at all cost.

Then there's the issue of robustness and reliability.

The smaller the junction size on a chip, the less "robust" it becomes in terms of its ability to withstand disruptive forces such voltage spikes, transient currents and even bit-flip due to the effect of cosmic rays.

And no, I'm not kidding about that last one. Apparently bit-flip caused by cosmic rays is really a thing, watch this video for some very interesting data:

It's easy to see why the electronics in most of our interplanetary space craft are built around "old-school" processors and devices; components that have proven to be robust, reliable and trustworthy over decades of use.

So there are many reasons why a manufacturer might want to stick with old but proven technology for the electronics of its vehicles and many other reasons why jumping to new technology in haste could be a very bad idea.

I suspect that quite a few car makers will use the switch to EVs as an opportunity to review the type of tech being used in the electronic systems used onboard so we may see some change in the years ahead. However, as is the case today, once they've saddled a new horse, they'll be reluctant to change again for quite some time.

This situation actually reminds me of the aviation industry, where "antiquated" technology still abounds for exactly the same reasons as cited above, except even moreso.

For example, modern light (GA) aircraft such as those from Cessna and Piper are still using engines that are basically designs from the 1940s that have seen very little change in over 70 years of manufacture. By comparison to modern car engines, these aircraft engines are incredibly low in power per litre of displacement and have only a tiny fraction of the development applied to the engines in our daily drives.


Because the number one requirement for an aircraft engine is that it is reliable and safe. Over seven decades of experience has shown these engines to be exactly that and nobody wants to compromise that reliability and safety by tweaking them to produce gobbs more power. In short... they work extremely well so nobody's trying to change that.

And so it is with the electronic subsystems in our modern cars.

Please visit the sponsor!
Please visit the sponsor!

Have your say in the Aardvark Forums.

PERMALINK to this column

Rank This Aardvark Page


Change Font

Sci-Tech headlines



The EZ Battery Reconditioning scam

Beware The Alternative Energy Scammers

The Great "Run Your Car On Water" Scam


Recent Columns

Do your job, get an award?
As someone who has been self-employed for pretty much their whole life...

They did what?
Oh no, I've been at it again! ...

Fight fire with fire
It looks as if things are getting worse in respect to the CV19 situation in Europe right now...

Electric aviation?
I just watched a video published by Rolls Royce...

Buy a piece of history
Regular readers will be well aware that I predicted the failure of the Martin Jetpack right from the get-go...

Pragmatism or principles?
I have to admit that recent developments in this whole pandemic thing and the government's response have left me torn...

Sometimes old is good
Today we live in a throw-away culture...

The danger overhead
The sky is falling...

The great chip glut of 2028?
Right now the world is suffering from an acute shortage of key semiconductor devices...

Do we have an EV bubble?
Telsa has become the world's sixth most valuable company, with a market captitalisation of over a trillion dollars...

Youtube infuriates users
When I write a column about the idiocy that sometimes drives decision-making at YouTube it does bad things to the daily stats for Aardvark...