Why is the value of res.lis different from the value of tab.lis?

Continuing the discussion from What is the difference between a usrsuwev.f file and a usrsuw.f file?:
Dear FLUKA experts,
Indeed, as you mentioned, I am aware of the reason for the fewer nuclides in the res.lis file. However, there is still an issue that I hope you can address. I conducted a simple calculation of the radionuclides in the res.lis file (under identical conditions), dividing the values in the column (Bq/cmc) by the (Beam intensity). The resulting values matched those in tab.lis file, but there are a few radionuclides whose values do not align. For example, when calculating 8Li, 3.312E+16/6.74E+11 yields approximately 4.9146E+04, which matches tab.lis; however, for 8Be, 7.332E+16 /6.74E+11 results in approximately 1.089E+5 quite different from 5.4393E+04 in tab.lis. 6.74E+11 is the Beam intensity
The contents of res.lis and tab.lis are provided below and a visual representation of the values for 8Li and 8Be is also included,res.lis on the left and tab.lis on the right.
I look forward to your response and thank you.

Guanzhou Ding

test15_23_tab.lis (45.7 KB)
testwater2_res.lis (17.7 KB)

Dear @dinggz,

thanks for your reporting. Could you please send me the input file and the irradiation profile that you used so that I can try myself to reproduce the problem? You can even send them via direct message if you do not wish to upload them publicly.

Many thanks,

Dear expert,

Thank you for your reply, I will upload the source file here as flair.source-test.f indicates a user-defined SOURCE file.The irradiation time is one year, and the cooling time is only 0 moments.

Guanzhou Ding
test1.flair (13.9 KB)
source-test.f (7.2 KB)

Dear Guanzhou Ding,

Thanks for your patience. I ran your simulations with the routine provided to get enough statistics on the nuclide activities. Unfortunately, I cannot reproduce your problem.
My guess is that you are misusing the usrsuwev program: in particular, if you want to obtain with usrsuwev the same numbers that you would obtain with a one-step simulation (FLUKA RESNUCLE + IRRPROFI + DCYSCRE) you should provide as input to the program a file that contains radionuclide yields (FLUKA RESNUCLE without DCYSCORE). If you are using the test15_23.rnc the result would probably be meaningless.
To better frame the issue and better help you, it would be useful to know exactly what you type as input to usrsuwev.