I'd expect that for two reasons. The first is that all array keys are interred strings (names) so integers need to first be converted to strings and then hashed into a name and then hashed to the slot in the arrays hash table.
The second is that byte based version is essentially using a byte sequence as a bit set - each setting of a byte in the set is about as close as you get to a truly atomic (in the sense of using '1' instruction) as you can get in LC. It is what you would do in C if you couldn't be bothered using bit operations to manipulate individual bits in machine words (which would be substantially faster again - as you would be touching 1/8th of the memory). Warmest Regards, Mark. Sent from my iPhone > On 6 Aug 2018, at 20:24, Mark Wieder via use-livecode > <use-livecode@lists.runrev.com> wrote: > > Interesting. I tried a variation using arrays, and it's three times slower > than yours. > > function get_primes pN > local tMroot, tPrimes, tIsItPrime > > if pN < 2 then return empty > if pN = 2 then return 2 > put trunc(sqrt(pN)) - 1 into tMroot > > repeat with i=1 to pN step 2 > put 1 into tIsItPrime[i] > end repeat > > repeat with i = 3 to tMroot step 2 > if tIsItPrime[i] is empty then next repeat > repeat with j = i^2 to pN step i > delete variable tIsItPrime[j] > end repeat > end repeat > > return the keys of tIsItPrime > end get_primes > > -- > Mark Wieder > ahsoftw...@gmail.com > > _______________________________________________ > use-livecode mailing list > use-livecode@lists.runrev.com > Please visit this url to subscribe, unsubscribe and manage your subscription > preferences: > http://lists.runrev.com/mailman/listinfo/use-livecode _______________________________________________ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
