Why do the Nunchaku text encoder node and the Dual Clip loader node produce different results even when using the same settings and seed ?

[Greatz08]

Both node work well and that is to be expected as nunchaku text encoder node is based on dual clip loader only, it has one extra option of t5 token length which i can observe visibly without code inspection, but i don't think it should generate different image with same setting+same model+same seed, so i thought to ask here first.
( In the meantime, I will do my best to understand this behavior through code inspection with the help of an LLM. However, if anyone can explain the behavior of both nodes, I would greatly appreciate it. :-) )