Bah, last post was only a year and a half old... not quite a skeleton yet.
Besides, if it wasn't him it woulda been me; realized what the issue is with being able to shove around multiple blocks of water... it *isn't* "due to the way water descends z-levels" as stated in the wiki.
Rather, it is because of the nature of computers; which leads to a possible solution.
The computer is really activating one pump at a time, even if it looks like they are all active at once.
Assuming you are using a simple 4 pump system, with 4/7 water blocks places on opposite corners, the first 4/7 water block would be pumped through pump 1, and deposited in the empty square there... but then pump 2 kicks in, and picks it back up, and drops it right onto the *other* 4/7 water block, making 8/7, which leaves 1/7 leakage; and continues pumping around a single 1/7 block of water.
If this is how this works, then it wont matter how long you make the chain... the single block will be transported around the entire loop, colliding with whatever other water blocks it hits.
The trick to fixing this will take some experimentation; which, unfortunately,I don't have the time to perform myself.
I am assuming that Toady, in code, sticks all the pumps in a list (or array), and steps through the list one at a time, checks if it is active, then tries to pump the fluid, then goes to the next one... that would be the easiest and most direct way to do this.
If that is, in fact, how this works... then there is some means that the system has for deciding which pump goes first. The easiest and most direct way of doing this is build order.... although, knowing Toady, it could be something really screwy and obscure (like the algorithm he uses to get miners to decide what square to dig).
Most people, I am assuming, build their pumps in the same direction that they feed... so if pump A feeds pump B, then most people will build them in that order. As can be seen above, that is guaranteed to make a collision happen as that first chunk of water is fed around the system.
However, if you build the system 'backwards'... by placing one pump, then the one that feeds into it, then so on... this might not occur. Instead, the first block would move forward in the system, then the pump behind it would actuate, and water would be pumped into the square that the first block just left. If this idea is correct, you *should* be able to push around as many blocks of water as you have pumps minus one... the empty space that the first pump fills.
=====================================
There are other unanswered questions I have about this device... Most of my forts are on paper right now (due to work and kids), but I was thinking about some implications for this device in the infamous dwarven super soldier program.
If you had two of these systems, timed properly, and fed a block of magma into one, and a block of water into the other... you *should* be able to repeatedly douse someone with magma to catch them on fire, then immediately douse them with water to put them out.
Again, assuming you could time it right, and assuming that the ideas above on pump timing are correct (big leaps, I know)... then you would never have to worry about the two systems mixing, even though they dump into and pull from the same square; the water would already be gone, and the pump that pulled the water would no longer be the pump that the computer was looking at (so you wouldn't end up with magma in your water chain, or visa versa).
First problem is timing it properly.
If the two circuits are each 4 pump circuits (for simplicity), is there any way to guarantee that they are exactly a half phase apart? Hopefully *without* having to trigger each pump [the repeater] due to it slowing the system down, and thus causing the lava to sit too long...
This might be another build order issue. (i.e. using build order to determine which pump pumps first, etc... assuming that even works as I expect)
The second issue is with accidentally coating someone in obsidian... after being doused with water, the water block is immediately lifted back up and off of them... but they and their clothes are still wet. Would this cause them to be entoumbed if immediately hit with magma?
============================
Well, thats my 2 cp... thank you to anyone who read through my rambling, and a beer to anyone that can confirm or deny my thoughts.
Author
Topic: Yet another mist generator! (now with 4x the mist!) (Read 108233 times)