Which of the following are true?

When running a regression in the presence of heteroskedasticity...

i) the confidence intervals differ from those computed assuming homoskedasticity

ii) the t-stats are the same as those computed assuming homoskedasticity

iii) our coefficient estimates differ from those computed assuming homoskedasticity

iv) the confidence intervals are the same as those computed assuming homoskedasticity

v) the t-stats differ from those computed assuming homoskedasticity

vi) our coefficient estimates are the same whether there is homoskedasticity or heteroskedasticity