TGraphAsymmErrors vs TH1->Divide

ROOT
1

Hello, everyone
Is it correct that when drawing an efficiency, TGraphAsymmErrors(TH1 h1,TH1 h2,“n”) and Divide(TH1 h1,TH1 h2,“B”) should give me the same errors? I know it’s expected to get differences when h1 and h2 are not filled with weight 1, but I also get very different results for weights=1.

Thanks for any help

Ilektra

2

see discussion in new class TEfficiency
root.cern.ch/root/html/TEfficiency.html

Rene