Comments by "Samson Soturian" (@samsonsoturian6013) on "Can you calculate war? Are war games pointless? (Praxeology vs the Ossipov-Lanchester equations)" video.

  1. No AI can possibly do what a skilled human has not already done. The technology simply doesn't allow it because it takes a human programmer to reward "right" answers and punish "wrong" answers.
    11
  2. The math doesn't work when each officer and NCO starts acting independently because the insanity of war.
    4
  3. Robert Leonard in his book complained about the "battle calculus" the Pentagon was throwing around. Basically the algorithm the computer was using was the whole of both armies would start at a set distance, charge towards each other at a set rate, fire immediately upon coming into range, kill each other at a predictable rate, and the piles of bodies/wreckage not even being an issue let alone the effects of the horror of the scene upon human operators.
    3
  4.  @venislavnikolov6027  define winning. Signing a peace without German territorial loss might be considered a win even if the Nazis got purged
    3
  5. No, because no AI could predict where the battlefield would be or the condition of the opposing armies/alliances at the time of war as they change year to year and will change upon war commencing. For instance, no AI could say with certainty that the current war would be confined to Ukraine with Poland and Belarus getting heavily involved whilst no fighting on their border.
    2
  6.  @HontasFarmer80  a stellar sized computer isn't possible using current equipment. Most of its energy and processing would be spent fetching values from memory (it would take years to search for a given file). Even the speed of light is a favcor because when the function requires a billion two way messages then a light nanosecond becomes a debilitating distance.
    2
  7. Which is a polite way of saying no one in the world knows what the hell they're doing.
    2
  8. Part of the problem is that a lot of soldiers in high command actually don't know a thing about military science. They think years in uniform are infallible and every aspect of fighting is an established artisanal practice, and there's generally a lot of ego involved and little study involved. Ergo, you'll get a lot of guys who think they can reduce the outcome of any fight to literal paper-rock-scissors level of simplicity.
    2
  9. Screw those games and everything they stand for.
    1
  10.  @DS127  as a silly game, yes. As war simulation, no.
    1
  11.  @DS127  maybe it's the EU4 fanbois I'm thinking of... I never played those games
    1
  12. Screw those games and everything they stand for
    1
  13. Another effect math doesn't account for is friendship. It doesn't take a genius to figure out a team of serial killers would collapse immediately no matter how skilled they are since any one of them will sacrifice the others to save himself. However, a literal band of brothers lead by their father will just fight harder as you pick off their number. This applies on a large scale as well, as unified nations form armies that function much smoother than armies formed by large coalitions of states.
    1
  14. Socialist economists are fake economists.
    1
  15.  @nco_gets_it  ok, motard
    1
  16. The planning for any war should start with the assumption that no one in the world has any idea what they're doing.
    1
  17. The mushroom model only works if you want to dominate your coworkers, not the enemy.
    1