A steam powered brain could really exist

Platinumoxicity on 21/6/2009 at 15:35

This is the explanation how Karras' robots could produce intelligence with boilers and steam. Steam-powered digital circuits!

Inline Image:

http://img301.imageshack.us/img301/9076/adder.png

massimilianogoi on 22/6/2009 at 03:53

S = TRUE in any case

C-out = FALSE on all the three inputs set to FALSE
TRUE on all the three inputs set to TRUE
TRUE on intermediate cases

The Boiler Constant is useless in both S and C-out

Really useful this circuit :p

Quote Posted by Platinumoxicity

In other news: This is the explanation how Karras' robots could produce intelligence with boilers and steam. Steam-powered digital circuits!
(http://img301.imageshack.us/img301/9076/adder.png)

Platinumoxicity on 22/6/2009 at 10:07

Quote Posted by massimilianogoi

Was that sarcasm? The scematic is an exact copy of an 8-bit full adder, I just couldn't imagine how some gates would react if there wasn't enough or was too much pressure due to the division to several pipes, that's why I made up the boiler constant. I dunno, I got lost in though about the time when I was making the last double NANDS. The boiler constant hasn't got anything to do with the logic. It's there to support the system.

Actually I got S=false, C-out=true when X=true, Y=true, C-in=false.

I think it should work perfectly just like a digital full adder.

Beleg Cúthalion on 22/6/2009 at 10:16

Could you give an example...?

Platinumoxicity on 22/6/2009 at 12:51

Quote Posted by Beleg Cúthalion

Could you give an example...?

I don't know, there might be a problem that I accidentally put triple NANDs where there should have been triple ANDs, and I also put a quadruple NAND where there should have been a quadruple OR.

Sorry, I didn't do too well on that course in school. :p
(http://www.gamezero.com/team-0/articles/math_magic/micro/fulladder.gif)

massimilianogoi on 22/6/2009 at 16:50

Case : T T T

First NAND
Inputs: F F T Results = TRUE

Second NAND
Inputs: T F F Results = TRUE

Third NAND
Inputs: T F F Results = TRUE

Fourth NAND
Inputs: T T T Results = FALSE

So S NAND
Inputs: T T T F = TRUE

Case : F F F

First NAND
Inputs: T T F Results = TRUE

Second NAND
Inputs: F T T Results = TRUE

Third NAND
Inputs: F T T Results = TRUE

Fourth NAND
Inputs: F F F Results = TRUE

So S NAND
Inputs: T T T T = FALSE

An intermediate case: T F T

First NAND
Inputs: F T T Results = TRUE

Second NAND
Inputs: F F F Results = TRUE

Third NAND
Inputs: F F F Results = TRUE

Fourth NAND
Input: F T T Results = TRUE

So S NAND
Inputs: T T T T = FALSE if the Boiler Constant is TRUE
TRUE if the Boiler Constant is FALSE

At last: if we consider the Boiler Constant in the intermediate connections, we have intermediate cases only if is set FALSE, since it has direct access to all the three first NAND results.

-------------------------------------------

Case : T T T

First NAND : FALSE

Second NAND : FALSE

Third NAND : FALSE

So C-out NAND
Inputs F F F = TRUE if the Boiler Constant is FALSE or TRUE

Case : F F F

First NAND : TRUE

Second NAND : TRUE

Third NAND : TRUE

So C-out NAND
Inputs T T T = FALSE if the boiler constant is TRUE
TRUE if the boiler constant is FALSE

Intermediate case: T F T or T F F or F T T or F T F or F F T

So C-out NAND = TRUE in every case (the boiler constant connection is useless in these cases)

It's a four bit circuit, if we consider (and we should) also the Boiler Constant as input, not an 8-bit.

Plus: a four bits sum gives ad one bit result??

For the Sum you would have an Arithmetic Logical Unit instead.

And yeah, this circuit is useless :p

Platinumoxicity on 22/6/2009 at 18:41

Quote Posted by massimilianogoi

And yeah, this circuit is useless :p

Ok, maybe I should've asked someone before wasting +2 hours in this. :p

This discussion is this |___________________________| long. (And on a very wrong thread.)

My point was this long. |_|

"It is possible to build steam-powered digital circuits that don't use electricity, by designing valves that open and close according to steam pressures applied to the different input pipes of the valves." ;)

Edit: How could I have been so foolish? Builder forgive me. These are the real half- and full adders. Now they should work flawlessly. That other circuit I made was rubbish.

Inline Image:

http://filesmelt.com/Imagehosting/pics/31de39fbe0527f88f24f044db7b0ea0e.PNG

jtr7 on 22/6/2009 at 19:34

I'm enjoying the concept.:thumb::cool:

SinisterShadow on 22/6/2009 at 19:52

Quote Posted by Platinumoxicity

The scematic is an exact copy of an 8-bit full adder,

Well, that's not quite right. It's a 1-bit full adder but you can connect several of them to become a 8-bit ripple carry adder. Despite of that, the circuit is correct – at least if this steam constant is not part of the logic and only for steam support as you declared:

Quote:

I just couldn't imagine how some gates would react if there wasn't enough or was too much pressure due to the division to several pipes, that's why I made up the boiler constant. I dunno, I got lost in though about the time when I was making the last double NANDS. The boiler constant hasn't got anything to do with the logic. It's there to support the system.

This is the same as in electronic circuits: you need a support voltage but usually, you don't draw them into those schematics. Maybe that's the reason why massimilianogoi thinks it would be useless.

Hey, how 'bout eliminating the inverters entirely by only using NANDs in your circuit? :D Great work, BTW :thumb:

Beleg Cúthalion on 22/6/2009 at 19:55

What I meant behind the example question was something like: What is it and what does it...do? :weird: I mean, it's not that it turns different 1/0 inputs into an adequate response like a Child of Karras would...?!