General > General Technical Chat

End of the analog era

(1/5) > >>

As MCUs and FPGAs gradually replace complex analog ICs, we may look back and wonder when the analog era ended, much like we look back and see the end of the vacuum tube era in the 60s.  Of course, the odd example always lingers on, but I think you could fairly uncontroversially say that the transition happened in that decade.

For analog, I'd say we could pin the date down more precisely to a decade ago, June of 2011.  Jim Williams and Bob Pease both passed away that month--Pease exactly 10 years ago on June 18, 2011 and Williams the week before.  Of course we'll still be using analog products for a long time, but their passing seems to me to be the end of an era.

This mantra has been hummed for the last four decades. It wasn't true then, and it isn't true now.
The fact is, that the "analog border" has moved, meaning that digitization has moved closer and closer to the signal source (and the actuator output).
This has resulted in higher and higher demands being placed on the analog part. Analog today is very high performance.

Try digitizing sunlight directly (as in brightness measurement). Won't work, no?

"End of the analog era?" Pah!

Sometimes a 0.1$ analog circuit can make the difference between a 1$ MCU and a 0.1$ MCU. These solutions are not going away anytime soon.

Many of DMMs still have an analogue true RMS converter IC.

I am not sure about the future of discrete analog components in commercial products (although DC/DC converters will continue using analog feedback loops for a while longer), however analog design is far from dead.

With moore's law slowing down you are probably doing to see some interesting analog computing approaches, especially for machine learning where the accuracy of the individual math operations is relatively low.


[0] Message Index

[#] Next page

There was an error while thanking
Go to full version