Control Table - Disadvantages

Disadvantages

• training requirement - application programmers are not usually trained to produce generic solutions

The following mainly apply to their use in multi-dimensional tables, not the one dimensional tables discussed earlier.

• overhead - some increase because of extra level of indirection caused by virtual instructions having to be 'interpreted' (this however can usually be more than offset by a well designed generic interpreter taking full advantage of efficient direct translate, search and conditional testing techniques that may not otherwise have been utilized)
• Complex expressions cannot always be used directly in data table entries for comparison purposes

(these 'intermediate values' can however be calculated beforehand instead within a subroutine and their values referred to in the conditional table entries. Alternatively, a subroutine can perform the complete complex conditional test (as an unconditional 'action') and, by setting a truth flag as its result, it can then be tested in the next table entry. See Structured program theorem)