Unfortunately, while her multitasking ability lets her design equations in parallel, she can only cast/test one iteration at a time, two with Hive's aid. This limits her ability to acquire test data, providing the single largest limiter on her ability to develop useful spells.
No, this is the limit on her ability to
verify the simulation. Assuming that the TSAB has a simulation, it's already well-tested unless they're working at the bleeding edge, so testing in the real world would be much less dangerous for the vast majority of spell researchers...Taylor is starting near zero there.
Initial development of e.g. simulators or emulation tools are typically undertaken by small teams of people, so until her failure rates for spells cast successfully in simulation are less than 1%, she's probably not any slower (still limited by her available time to cast spells, but not as significant). The real challenge for Taylor is that she can't really focus on one area of research...which does have an impact on how far she can go, but on the other hand she'll understand the magic she uses in ways that even TSAB research mages don't. Multitasking may offset this a bit if different instances focus on different areas, and she may well see more synergies between areas than are easily found by more deeply-focused researchers.
In a few decades, she might be able to match a century's worth of normal development for one person, but to develop a library sufficient to create a useful high level language would take one person millennia, not centuries.
I also get the impression that Taylor is working on multiple things in different threads (or, in some cases different approaches to the same thing), so those things tested outside of simulation have already been tested *in* simulation (and e.g. results in a 10->1 reduction for her, which ratio is improving as she and Hive tune the simulation; for TSAB research mages perhaps this is 1000->1 or better), so...if we assume 24x7 x 20 real-time-equivalent threads that's 3360 Taylor-hours/week, vs. e.g. 40 TSAB-hours/week per researcher, so 1 Taylor = 84 TSAB researchers. That means that, if Taylor focuses, she could easily get ~2.5 millenia of research in 3 decades at that level of multi-tasking. The real-world testing time obviously is still limited, but the development time and test-in-simulation time is much less so, perhaps to the point that her real-world testing is nearly equivalent of that used by a team of that size.
It's not clear to me whether an instance of Taylor-in-a-drone can cast spells without guidance or not (and how useful they are for research/simulation); if she can, then this significantly increases the frequency of tests she can run (and how much research she can do). If not, as she can communicate with her drones across dimensions, she could still cast spells remotely, controlled by herself or one of her multitasking instances. This also could reduce the limits on real-world testing.
[T]he TASB has manged to do things as fast as they have because they have large portions of populations working on low level development and compiling scripts into high level languages. [<snip>] In other words, she simply will not have time to experiment enough to match the tens of thousands of researchers the TSAB employs both directly and indirectly.
Do you have a Nanoha-canon reference to these numbers, and the relative proportion of the population that is doing this low-level research? Note that I compared this to a typical population, as e.g. >10% doing low-level research is extremely unlikely on a long-term basis.
The Mid-Childan style has also inherited more or less incomplete versions of several ancient magical styles to use as foundations and touchstones, where Taylor and Hive only have fragmentary accounts of what various spells do, not even how they did it. In that respect they have already accomplished miraculous feats of spell design just by getting as far as they are now.
Yes, but do they understand the low-level parts of those magical styles? Based on my (limited) understanding of Nanoha canon, they do not, and a number of things those ancient styles can accomplish aren't really possible in the Mid-Childan style without significantly more effort (if at all). Taylor may have duplicated some of these effects, and certainly would be in a good position to reverse-engineer or re-create them.
