Cohen-d issues with pingouin ttest function, paired samples

Question

I have two lists of time values, 'times0' and 'times1'.
The data are paired (first value in times0 is paired with first value in times1, second value idem etc etc). I want the cohen-d of the t-test between these data and I'm using pingouin library in python.
If I use pingouin.ttest(times0, times1, paired = True), it gives a cohen-d = 0.395.
If I create deltatimes = times1 - times0, and do the t test for one sample compared to 0 : pingouin.ttest(deltatimes, 0), it is supposed to give the same result but it give cohen-d = 0.96.

Any idea why ?

Thanks !