ATMEGA at end of life

I noticed on the Mouser website that all the ATMEGA and ATTINY processors are either at end of life or not recommended for new designs. I guess this is to be expected after 20 some years.

I haven't fully explored the new line of processors, but they have a bunch of new features.

I just hate when parts I know and love go away :frowning:

Unless I have good reason not to, all my new arduino projects are either ESP32 or RP2040-based.

The only benefit of those ancient 8-bitters is that there are tons of useful libraries available.

1 Like

DIP packages. Makes prototyping on a breadboard a lot simpler and more accessible.

it's not difficult to add headers to ESP32 boards (some even come with the headers) and use them on a Breadboard.

true.
I would say also the absence of an OS makes it easier to understand what's going on. Your code is in charge.

I do not think it's completely true.

Microchip marked some chips as "not recommended for new designs". As an example probably the most famous ATmega328P has this mark. However, there is a newer version ATmega328PB which is not marked as such, so there will be AVR still available.

8-bit AVR MCU's are much simple and easier to learn in compare with e.g. ESP32 and RP2040/2350 and that's the beauty of if. Most of projects do not need such heavy weaponry anyway, speaking of blinking LEDs, driving LCD display, temperature measuring etc.

I'll be fan till the end of my days.

1 Like

True, but when a Pi Pico 2 costs £4.80 who cares?

1 Like

Very sad. What will happen to Arduino UNO boards then?

1 Like

That is true.
I spent a lot of money for these things during the years, despite of knowing that maybe I can buy "ready to use" product and without any additional effort. It's a hobby, nothing deep behind.

Just for clarification, I am not against ESP32 or Pi Pico. I actually have them too.

1 Like

It will be decades before they fully run out of stock anyway. There are batches spread out over China. The beauty of simple ATtinys is not just their cost but also some kind of ruggedness. ESP boards are a lot more fragile and tend to be a lot more power hungry.

That is not what I see…
There are in fact many relatively new Attiny chips, not to mention the new naming scheme “AVRnnnDxpp” chips.

1 Like

The datasheets say that these new chips can be powered by up to 5.5V. Does that mean their GPIO's are 5V tolerant?

I have been trying to get a Pico to work in a 5V TTL environment but the level shifting required is by no means straightforward. I have found the TXS0108E shifters to be noisy and unreliable. I am starting to look at other options, but I am thinking that boards based on these new chips might be a better way forward, although the Pico boards are still quite a bit cheaper. Then again, if you factor in the required level shifter chips, the total cost may not be that different.

The MVIO variant looks interesting for interfacing downwards with 3.3V and 1.8V environments.

Of course, if the chip is powered by 5V. The usual specification is that the voltage on an input pin must not exceed Vdd+0.5V.

Check the data sheet to be sure of the specification, of course.

the level shifting required is by no means straightforward.

It is for me, using these level shifters. But by far the most straightforward approach is to not mix 3.3V and 5V parts.

Not when running at a lower voltage.
Absolute maximum ratings of AVR DB series:

What new chips?
You need to be a little more specific.

A good question.

I think the strength of the Uno and similar is they are good tools for learning, they are aimed at beginners and have lots of support for that purpose. I would not recommend a Pico for a beginner.

I worked in telecoms all my life. A telephone exchange used to be a big building with lots of expensive equipment filling the whole building. Telephone exchanges will soon be part of history, with telephone call routing now done by software running on a server in a server farm. You can build a very capable PABX using a Raspberry Pi and open source software. Similar capability would once have cost you thousands of pounds, now it cost next to nothing. I no longer work in telecoms because it no longer exists. Ever cheaper hardware puts companies like Arduino at risk of failure, I hope their mission to educate is what keeps them going.

1 Like

I know this is getting a bit far from the original topic but I remember being given a tour of the telephone exchange of my employer back in the '80s. It was a local government somewhere down in the south west of England. The most outstanding feature I noted was rows and rows of large glass containers on shelves. These were sulphuric acid filled accumulators for the emergency power supply.

1 Like

My apologies. The new Atmel ones mentioned by westfw in the post just above:

Regarding level shifting:

Yes, I have used those in projects before and they work OK but I need to shift 16 pins. Having 16 transistors and 32 resistors is going to take up quite a bit of space so I was looking for a more compact solution, which is why I tried the TXS0108E. I do have some of those 4-way shifter boards somewhere and since my first test was unsatisfactory, I do intend to dig them out and give them a try. I have a couple of other chip options to test on order but the problem is the cost. A handful of transistors and resistors is quite cheap by comparison.

I agree that its better if level shifting can be avoided, which is why the new Atmel chips piqued my interest.

Thank you for finding that. I did search the datasheet for all four chip families and couldn't see that bit of information, although I was sure it must be in there somewhere. So yes, basically VDD + 0.03, so if the chip is run at 5V, then 5.3V max.