Tapering does not restore tunes

[bogomyag]

Tapering does not restore tunes in PTC_TWISS and in TWISS.

I understand from manual (chapter 38) that TWISS calculates optic functions in 4d only, therefore it is expected that with RF cavities and RADIATION=ON only the X,Y,PT will be approximately correct, but the tunes will not.
However, PTC_TWISS should correctly calculate the orbit and optical functions, which is not the case as demonstrated in run-test-1.madx

twiss (radiation off)
@ Q1 %le 220.544999999977
@ Q2 %le 219.60999999979

twiss (radiation on)
@ Q1 %le 220.546533628453
@ Q2 %le 219.612476383926

ptc-twiss (radiation on)
@ Q1 %le 0.546583283413037
@ Q2 %le 0.612737897188153

twiss tunes (radiation on, tapering on)
@ Q1 %le 220.56335366833
@ Q2 %le 219.621529278762

ptc-twiss (radiation on, tapering on)
@ Q1 %le 0.676054382201816
@ Q2 %le 0.5

I use
+++++++++++++++++++++++++++++++++++++++
test-1.zip
+++++

• MAD-X 5.09.01  (64 bit, Windows)                                              +
  

• Support: mad@cern.ch, http://cern.ch/mad +
• Release date: 2023.12.04 +
• Execution date: 2024.05.06 10:51:32 +
  ++++++++++++++++++++++++++++++++++++++++++++

[rdemaria]

I have never used PTC with tapering. I need to check whether data is passed correctly. Presently, twiss calculations are correct only to first order in PT, and therefore the tune is not correct with orbit with large pt.

[bogomyag]

What is the purpose of tapering then? I thought it was not only to calculate PT and closed orbit, but to adjust magnet strength for tracking and twiss with energy loss due to synchrotron radiation.
How to do tracking in PTC with radiation when tapering is not possible to apply. The results will be incorrect.