If your ascent is other than a direct ascent to the surface at 60 fpm, the additional time used in the ascent must be added to you bottom time.
I don't disagree. But the instructions for the RDP do not exactly say that, and instructors do not always make it clear, either. Moreover, most teachers I know nowadays are encouraging a 30 ft/min ascent, especially above 60 ft.
The calculations of a square profile can be challenging enough to new divers, much less getting into multilevel dives, compartments, half-times, the pros and cons of deep stops, and all that.
Which gets back to the question that the OP asked. If I understood his question, he was wondering why they don't teach new divers how to calculate NDLs from actual depths and times rather than using tables which end up taking away bottom time by all the rounding and stuff. He then went on to say that he had just done a dive which he could not make work using the RDP.
Now, I assume that he used a computer to make the dive. (I sure hope he didn't make a dive that he already thought wouldn't work!)
So I tried to give him two reasons why he was able to make a dive that didn't work on the RDP.
(1) Computers and tables figure Bottom Time differently. Tables assume a direct ascent at a fixed rate whereas computers are making real-time estimates.
(2) The RDP and PADI Nitrox tables are for square dive profiles. He did a multi-level dive. Had he had a Wheel or an eRDPml, he could have come closer to planning the dive that he had made -- or, more correctly, closer to retro-calculating the dive -- but even then you have to know how Bottom Time is defined by the tables compared to a computer.
I really just wanted to make the point that definition of terms is important, and they can change depending upon the type of dive, tables and equipment you use.
Maybe I'm overly sensitive to this because early in my career I had the same problem as the OP. I dove with a computer, but then when I logged the dives and tried to retrofit the dive to the RDP (just for fun and for practice), I found that it didn't always work. When I investigated, I learned that part of the problem was in how each defined Bottom Time.