我是否需要在 conda 基础环境中更新/升级 Python 版本？

Question

Python version at my base conda environment is 3.7.12. Do I need to update it to a newer version (3.10) in order to properly create new environments with Python 3.10?

Is there any problem in maintaining base environment Python version at 3.7 if I'd like to use newer version of packages and Python itself in other environments?

I've just reinstalled anaconda ( conda install anaconda ) after trying to update conda and python but being unsuccessful because of conflicts. After reinstalling anaconda I managed to update conda from 4.10.3 to 4.13. Now I'm trying to conda install python=3.10 and it's taking HOURS ! Literally: I started it 24h ago and it keeps Examining conflict for ... a list of packages that updates every 10 or 15 minutes, with just slight changes.
Is this list of packages being examined so long because anaconda installs all of them at the base environment when it's installed for the first time?

Answer 1

Do I need to update it to a newer version (3.10) in order to properly create new environments with Python 3.10?

no. You can have python version X in your base, create an env with python version Y and another one with python version Z. That is the advantage of having conda managed environments that you can switch to different installed packages using a conda activate command.

Is there any problem in maintaining base environment Python version at 3.7 if I'd like to use newer version of packages and Python itself in other environments?

No, the general recommendation is also to not mess with your base env at all. Create an environment if you have specific needs (eg one for each of your projects)

Is this list of packages being examined so long because anaconda installs all of them at the base environment when it's installed for the first time?

Yes, anaconda installs a very long list of packages when first installed with a complicated mix of dependencies. When you now try to update your python version, probably all packages need to be changed to be compatible, which creates a very complex net of dependencies, hence the recommendation in the two answers above: don't try to mess with your base env and create a new env with your desired python version and only those packages that you need for a specific project (ie don't just install all anaconda packages into new environments)