Question on the friction setting for randomization

[Pengzhi Yang]

Hi, thank you for your great work!

However, I have a question about a setting for randomization. In the paper, you mentioned the training range for the lateral friction is [0.05, 1.25] which is a relatively big range. However, in the code it seems to be [0.5, 1.25].

Which one did you choose in the paper's experiments during training and what's the reason? Is that because friction is a very crucial impact factor?

Thank you!

[erwincoumans]

I think we used [0.5,1.25] for the paper, it maybe a typo.