Not surprisingly, Kannada evolved gradually from Old Kannada. As such, it is tricky to specify a date when Kannada split from Old Kannada; I’ve seen dates between 1100 AD (when the first differences appeared) to 1800 AD (when, under the influence of Christian missionaries, the script was standardized).
Kannada, like most Indian scripts, is an abugida, with glyphs representing consonant+vowel “a”, with an extra diacritic decorating the consonant to change or omit the vowel. Kannada also has a large number of ligatures of consonant clusters, i.e. there are a lot of glyphs for a set of consonants with no intervening vowels. (“Stri” would be a consonant cluster in Latin script, for example. These ligatures are usually made in Kannada with what look like subscripts, as seen in the example glyph.