Ai Forums Home Welcome Guest    Friday, April 27, 2018
Ai Site > Ai Forums > The Artificial Intelligence Forum > increasing my storage capacity Last PostsLoginRegisterWhy Register
Topic: increasing my storage capacity

posted 5/29/2016  19:18Send e-mail to userReply with quote

posted 5/29/2016  20:02Send e-mail to userReply with quote
Hello R_81.

I am using the distance to matching values,
Elegant Compression in Text (The LZ 77 Method) - Computerphile

 Elegant Compression in Text (The LZ 77 Method) - Computerphile

posted 5/29/2016  23:38Send e-mail to userReply with quote
Compressing the data makes it dfficult to access the data.


posted 6/1/2016  02:02Send e-mail to userReply with quote
Thats great Keghn my man!

Working with a compressed tree->
the basic idea of compression of your "key store" which is as good as any ones mesh be it backpropagation, or just a binary tree key store. (which is my poison.) it just comes down to a distribution of so many overlapped keys.

working under compression, lets you store more, but you have to decompress it, it seems like a disadvantage, but the idea is you decompress in a tree, so then u decompress as least as possible.

I declare myself the infinity troll. :)

Ive finished the code for the infinite decompressor now, but its not debugged yet.

And.... unless you are a believer in over-unity and quantum computers, I dont think your going to believe I have a chance in hell in it working.

But it makes sense to me, and this robot has an exponential store! Which is going to be handy if you actually want to make a decent robot, your going to need something quantum, buddy... or ur robot sucks dude.

After testing is done, fail or win, I post here. Why would I bother posting if it fails? Because im still proud of it, I never even expected to get *this* far.

There is a polymorphic virus it can make, but the catch is, you have to change its decompressor as well as its virus it morphs to, so theres a tricky catch, why you actually cant do it, even with an exponential decompressor. And that makes it a different project you have to write on purpose, so ill never write it.

Last edited by rouncer @ 6/1/2016 2:16:00 AM

posted 6/1/2016  19:34Send e-mail to userReply with quote
Cool, glad it is coming along.
Having high data compression you could have your little
program rolled up in a tiny space. Using very little memory.
Like a Turing computer unravels the memory it needs.Than rolls
it back up when done. Like a Tora scroll and Turing machine combination.

Tora compuer compression program

Turing Machines:

I am going to use it to encode information for synthetic life higher than DNA!!!!

And for space craft that have to be super light:

Also for a computer simulation to model all of the universe. Because 1000 atom to make make a one bit of ram memory
cell is not going to work. I need to get like a billion byte per atom.

The compression method i was going to use is "fist in last out", FILO.

I have my designated memory zone. Squash it with my compression algorithm. Then move fresh data
in next to the compressed data and then repeat. Even 10% compression will work with this method.
But like you said compression comes at the price of taking speed away from a simulation.
So something like to speed up time in the area of where the program will be needed. May be quantum science will
come up with some in the future?

 Tora scroll Turing machine
Last edited by keghn @ 6/1/2016 7:36:00 PM

posted 6/1/2016  19:36Send e-mail to userReply with quote
Turing Machine:

I am going to use it to encode information for synthetic life higher than

 Turing Machine
Last edited by keghn @ 6/1/2016 7:37:00 PM

posted 6/1/2016  19:38Send e-mail to userReply with quote
Space craft that have to be super light:

 Space craft that have to be super light

posted 6/2/2016  07:43Send e-mail to userReply with quote
Hey Keg, the scroll idea is exactly what I have in mind. :)

heres a post about the arithmetic involved in getting the number generation to happen.

I can fit about 1500 compressor transformers in 1, so far, with my current method, and if I ever get it working, it renews its precision over by 10x!!!

Im not sure if people understand what that means, when I say it. (except keg, I think you know what im attempting to do.)

Last edited by rouncer @ 6/2/2016 7:49:00 AM

posted 6/2/2016  19:38Send e-mail to userReply with quote
Deepest Mandelbrot:

 Deepest Mandelbrot

posted 6/3/2016  16:01Send e-mail to userReply with quote
just to tell you, its going great, ive got 5 easyish lines to write to go get that fractal zoom actually frickin stored (if you could call it that?!??), but im working on another one now, and youll never believe me what it does. :)

But they will be working together of course!!!


god its been so long, head so decayed...

About molecule sims at stuff like that, it depends on the "growth ratio" things like that would require much higher growth ratios than what im getting but I wouldnt say its impossible. Im not into that anymore at all, saying things are impossible that is.

Last edited by rouncer @ 6/3/2016 4:04:00 PM
'Send Send email to user    Reply with quote Reply with quote    Edit message Edit message

Forums Home    Hal and other child machines    Alan and other chatbots    Language Mind and Consciousness  
Contact Us Terms of Use