ESP32 Forum

The modem clock is here
i guess, the independent clock enables efficient time-sharing between WiFi and BLE

some food for thought: if a new product would not have radios, wouldn't it make more sense to call it the ESP32-P41?

If a new product would be RISC-V with radios, wouldn't it make sense to call it ESP32-C7?

Code: Select all

#define SOC_MODEM_CLOCK_IS_INDEPENDENT            (1)

i guess, the independent clock enables efficient time-sharing between WiFi and BLE

Or... CPU can sleep/stop while modem stays up?

some food for thought: if a new product would not have radios, wouldn't it make more sense to call it the ESP32-P41?

If a new product would be RISC-V with radios, wouldn't it make sense to call it ESP32-C7?

I would assume the idea is to replace the current S3 so it can be set to "Not recommended for new designs" as soon as the S31 is ready. Having similar names makes it easier to communicate such a step.

If a new product would be RISC-V with radios, wouldn't it make sense to call it ESP32-C7?

We've communicated before that the CPU architecture actually is one of the less important things in our chips, as ESP-IDF and the compiler plaster over that pretty efficiently allowing you to do more or less the same thing on both architectures. The fact that the ESP32-Sx line (as in, the currently-releases -S2 and -S3) only consists of Xtensa CPUs is more of an happenstance than a defining characteristic.

I would assume the idea is to replace the current S3 so it can be set to "Not recommended for new designs" as soon as the S31 is ready. Having similar names makes it easier to communicate such a step.

That sounds very plausible.

This choice of name is going to be causing confusion down the road though. The S2 and the S3 have a lot in common, and the S3 can be reasonably seen as an extended version of the S2; the S31 is a completely different chip, i.e. the step between S2 and S3 is much smaller than between S3 and S31...

However, I have now learned that I don't understand Espressif's naming scheme at all. I guess the names are chosen based upon some characteristic (target application/product?) I'm not aware of.

Let's keep looking forward to when the S31 comes and hope that it'll become reasonably available for tinkerers

This choice of name is going to be causing confusion down the road though. The S2 and the S3 have a lot in common, and the S3 can be reasonably seen as an extended version of the S2; the S31 is a completely different chip, i.e. the step between S2 and S3 is much smaller than between S3 and S31...

I would not be surprised if we see a S21 next or the S31 might replace the S2, too.

Let's keep looking forward to when the S31 comes and hope that it'll become reasonably available for tinkerers

It's like the wait for ESP32-P4 all over again. It was worth waiting but not easy.

...no ethernet MAC

...no ethernet MAC

#0
#1
#2