Search For Powder Intelligence returns! by perogiepro

perogiepro

26 / 7

22nd Dec 2015

10th Jan 2016

We are back, refurbished and with new ideas. Read all of our old discussion on the forum under a topic called Evolution in a video game! Anyways the main plan is to create small bots that eat to survive, reproduce, and then change (genes). Check list.

sfpi intelligence evolution electronics test group science life workinprogress

Comments

• 

  msasterisk 30th Dec 2015
  Also, DLAY encoding seems like a good idea. I'll work on it.
• 

  msasterisk 30th Dec 2015
  Kevino, I already tried a wifi-less hivemind. The signals don't travel fast enough, the bot would run into itself and act more like a wave than a single creature. That is not good.
• 

  msasterisk 30th Dec 2015
  Hm, a lot has happened while I was off.
• 

  Windspren 30th Dec 2015
  Also, we should test each bot before letting it "reproduce" by subjecting it to an obstacle course where it needs to reach the PLNT. That way, the best genes will be passed on.
• 

  Windspren 30th Dec 2015
  Now, I'm going to find a way to make it work inside of a bot. Also, I had an idea that we could make a "hivemind" for multicellular bots without WIFI by making each cell as small as possible, linking each one with a "bridge" of MERC randomizers, and having one large cell with the gas randomizer inside, that powers the MERC randomizers. That way, the cells can move with the bot, but each still has random motion.
• 

  Windspren 30th Dec 2015
  I did it!!!!!! ID:1913931
• 

  Windspren 30th Dec 2015
  Also, it would be easy to make my gas randomizer work if it was housed outside of the bot, with WIFI connecting it to the bot. However, we would need to manually set the channels to have more than one bot on screen at a time, which kinda defeats the purpose. I'll make an example later today.
• 

  Windspren 30th Dec 2015
  I was thinking that if we found a way to encode a randomizer in DLAY, we could put a MERC "neuron" on the output wire, thus making it more likely that a signal would pass through that neuron each time. This would also help with some randomizers changing state too fast.
• 

  perogiepro 30th Dec 2015
  Good find. Yes SFPI is a reference to SETI and pseudo-randomizer would be great, that way the bot will have certain directions favoured. Then if we can make it replicate the bot will carry the good genes on (the ones that make the bot run into plnt, for example) so bots pass knowledge down to offspring. Anyways the heat mediation would be great if it was triggered by say, finding plnt or reproducing (that would cause the bots to stick together as well.)
• 

  Karakanlud 30th Dec 2015
  (Also it has a heat mediation, so if a randomiser recieves more signals, it will more likely pass it to the next one)