Homer's findPeaks: normalized Tag counts, Peak Score, and RPKMs

Question

[fr]

I was reading this and this posts where it is mentioned that

normalized tag count [is] (equal to RPKM reflecting peak density).

But I wonder if this is true: from homer's docs it mentions

Column 6: Normalized Tag Counts - number of tags found at the peak, normalized to 10 million total mapped tags (or defined by the user).

Column 8: Peak score (position adjusted reads from initial peak region - reads per position may be limited)

From looking at the relationship with peak size, I see a clear association between Normalized Tag Counts vs Peak Size, but not between Peak score vs Peak Size.

I have the following questions:

• From the docs, it seems Column 6: Normalized Tag Counts only accounts for library size. Is this correct?
• Does Column 8: Peak score then account peak size? It is unclear to me what it means "reads per position may be limited"
• Does this mean that Column 8: Peak score does not account for library size? or should I normalize it in some way to 10m mapped tags?
• Given that I want to look into both wide and narrower peaks equally, which (if any) of the above metrics would you use?
• Which of these metrics could I use as "RPKM"?

Thanks a lot in advance