Exercise 3: What's Named Out of Scope, and Why Naming It Is More Honest — Possible Solution ==================================================================== AT LEAST TWO SPECIFIC OUT-OF-SCOPE ITEMS ------------------------------ Per this chapter's own "Honestly Scoping This Topic" section, at least these are explicitly named as left out: 1. The many packed integer/float data-type variants (beyond the single-precision float example this chapter actually showed) 2. Each SIMD instruction set's own feature-detection requirements (checking, via CPUID, which extensions a given CPU actually supports before using them) 3. Real memory-alignment rules — some SIMD instructions fault on unaligned memory operands 4. AVX-512's own further masking and broadcasting capabilities Any two of these satisfy the exercise. WHY NAMING THEM EXPLICITLY IS MORE HONEST THAN SILENCE ------------------------------ If this chapter had simply shown the ADDPS example and stopped, without naming what it left out, a reader would have no way to know whether that one worked example represented "basically all of SIMD" or just a small, deliberately chosen slice of a much larger topic. Explicitly naming specific missing pieces (data-type variants, feature detection, alignment rules, AVX-512 masking) tells the reader exactly how much MORE there genuinely is, and roughly what shape it takes, without this course pretending to have delivered complete SIMD coverage it never actually attempted. This matches the same honesty pattern the wider arc has used repeatedly — cpu8bit1-12's own explicit "still out of scope" section, assembly2-7's own refusal to state a false-precision instruction count — where naming a real limitation directly is treated as more trustworthy than either overstating completeness or quietly saying nothing about what was skipped. WHY THIS WORKS AS AN ANSWER ------------------------------ It lists concrete, specific items drawn directly from the chapter's own stated scope note rather than invented ones, and explains the honesty argument in terms of what a reader could otherwise wrongly assume (that the one example was representative of the whole topic) if those omissions had gone unstated.