The previous posts in this topic are hinting that there really is no particular relation between the algorithm chosen and the resulting instrument timbre.
This post: Algorithms and their effects - #3 by RobertSyrett provides the best explanation of the differences between the algorithms, because it describes the various modulation paths for each type of algorithm. Each algorithm has two audio output signals (X and Y) which may or may not be controlled by the operator envelope.
So the two outputs can be used to provide two separate elements for a sound. They could provide the tonal and atonal partials of a pad sound, or the initial attack and the sustained portion of a percussion sound, for example. Bass and noisy sounds should be easy to produce using any algorithm.
The actual timbre of any element of a sound is determined by the FM parameters (and the subsequent overdrive and filtering). Time variance is controlled by envelopes and LFOs. This is described in Appendix B of the DN’s manual.
If you want to learn how to make a particular sound, there are many, many tutorials on FM synthesis, most of which are generally applicable to the DN. A little searching on the forum will help you get started. For example, here’s a post that links to an online resource that shows how to recreate the sound of acoustic instruments based on knowledge of their timbre: Book recommendations for learning FM Synthesis - #7 by mickeyziggyk
If I understand correctly from your other posts, you have a DN so it would be a useful exercise to select a preset sound and study how the interactions between the four operators affect the timbre.