real fct > standalone fct ?

[Wsky51]

Hi , I rerun all test cases you provided. That's really a big help. Thanks a lot for the repo. But when I rerun the test case according to your guidance.I found something strange to me. in the output fct files. some data are like this

0b013f01 0b00a501 10000 100 20030 2000002133 13938 13923
0b005d01 0b005801 10000 100 8043 2000012313 4807 4838
0b00b601 0b00b901 10000 100 52835 2000017268 8706 8590
0b012001 0b011501 10000 100 60369 2000013263 13417 13263
0b00f701 0b009701 10000 100 4542 2000015038 12587 12622

according the source code , the last one is standalone fct, and the second to last is the real fct（calculated by time now minus the flow begin time）. So the real fct must be less than the standalone fct. but the output data is not, which is really frustrated me as shown above . Could you explain why the real fct of some flows are less than the standalone fct ? Thanks for you answer!