Search For Powder Intelligence by Windspren

Windspren

30 / 4

2nd Jul 2016

29th Sep 2016

The reborn SFPI! Our goal is to develop and identify forms of pseudointelligence and life within TPT in the forms of PSTN bots, computing systems, genome emulators, brain architecture simulations, and anything else we (or TPT AIs we find :D) can think of

sfpi tpte3

Comments

• 

  Schmolendevice 7th Jul 2016
  Hmm. Wave division multiplexing in TPT may be something... do note that BRAY and PHOT ctypes work the same way. Technically it's 30 wavelength (colour) "channels," but indeed 2^30 _combinations_. Once again, if you look at mine and LBPHacker's saves, you'll realize that we've very much solved the "BRAY decay problem" with solid spark + DRAY and are very much updating FILT ctypes at 60 Hz.
• 

  QuanTech 7th Jul 2016
  the wavelengths of photons can also be easily divided using glass prisms and such devices.
• 

  QuanTech 7th Jul 2016
  yes, i have heard of FILT technology before, but photons have many more wavelengths. In TPT, there are 2^32 photon wavelengths! Plus, you can add multiple photons at once in a cable, allowing a higher data transfer rate (in FILT, you have to wait for the BRAY to dissapear, but in optical computing, you can have 100 photons in a fiber cable at once!
• 

  Schmolendevice 7th Jul 2016
  Sir, I do hope it is not the case that you have not seen this save before, id:2004222, mind any of my own saves where I very much deal with high speed FILT electronics independent of PHOT. Solid spark and subframe BRAY already permits 60 additions to be performed in a second, an operation per frame.
• 

  QuanTech 7th Jul 2016
  Optical computing could allow more parallelism and speed, which can also be advantageous to AI for optimal performance!
• 

  QuanTech 7th Jul 2016
  sadly, light is slower than electricity in TPT, so i might need to add ACEL to accelerate the light to faster-than-electricity speeds.
• 

  QuanTech 7th Jul 2016
  i'm starting a new project (not exactly an SFPI project) that i will do in my free time. This project is to build an optoelectronic processor (not quantum though, as you can't do that in TPT!) It is a classical optical computer that will hopefully help perform computations faster than regular tpt electronics!
• 

  QuanTech 7th Jul 2016
  ai that can learn could also allow complex things like CleverBot (though not likely in TPT!)
• 

  QuanTech 7th Jul 2016
  I'm thinking of a "military bot" or "game bot" in my head. Like a tic-tac-toe bot that learns your strategies and can think of ways to counter your moves. That sounds like a fun but difficult project.
• 

  QuanTech 6th Jul 2016
  +NorthMustang We currently have 3 projects: 1. Finding ways to encode cell genomes 2. Adaptations for evolutionary cell 3. AI and neural networks. I'm mainly working on project 3 (most of my work now will be experiments that won't really be published until completion!)