Just in case you weren't aware, the best way to build the stack is from the top down, as the pumps are evaluated in the reverse order in which they were built. e.g., build pumps A, B, C, D and E, in that order, and the game pumps with E, then D, then C, B, and A, in that order. This is how you can pump a full 7/7 of water or magma the entire height of the pump each tick (assuming the bottom-most pump is always pulling from a 7/7 tile, of course).
Also, I remember a thread about how to build a pump stack that kept all the tiles surrounding it at the correct temperature so you could leave temperature on while pumping and not have to worry about the lag from the game recalculating temps in each tile as the pump-stack did its work. I don't remember who started that thread, however, as it was close to a year ago.
Actually, order of pump construction doesn't matter as regards throughput. Yes, it does affect how soon you'll get fluid of the top of the stack when you first turn it on, but after it's been running a while? Not at all. You'll still get 7/7 per tick, every tick regardless of if you've built the stack from the top down, or the bottom up, or any other order. The ONLY effect that build order has is how soon the first delivery of fluid happen when you start the stack. For example, let's take your 'A B C D E' pump order as an example. Pump A built first, Pump E acts first.
Example 1. Stack built from top to bottom.
Pump E transports a tile of fluid for pump D to transport for C to transport, etc, until pump A transports it. Yup. Bottom to top in a single tick just like you said.
Example 2. Stack built from bottom to top.
Tick 1. Pumps B through E don't have anything. Only pump A. Fluid is available at the output of Pump A only.
Tick 2. Pumps C through E don't have anything. Pump B transports fluid. Pump A then transports fluid replacing what pump B moved. Fluid at outputs of A & B.
Ticks 3, 4, 5.... Each tick gets the fluid to a level higher. Once there, the higher level pump will always see fluid every tick.
In fact, when I build a tall pumps stack, I'll contruct a "support gear" every 3rd level of the stack under construction. Then I'll build a pump attached to every one of those gears simultaneously. After those pumps are finished, I'll then designate and build the remaining two thirds of the pumps. And finally, after the pumps are all constructed, I'll deconstruct the support gears. After all, why should I spend the time for those pumps to be built one at a time when I can instead have 'em built en-mass and let the horde of dwarves I have a available all do the work in parallel?
As for the thermal issues with magma stacks, yes, it's quite important.... And easily handled. Just make the output from each level of the stack a 1x3 reservoir. In fact, if you look at the 3 images I posted for a stack I have under construction, you'll see on the "power distribution" level a space for 1 pump that will be supplying magma to higher levels (it's toward the top of the image. A length 1 axle is leading from the top and bottom of it. The 1x3 reservoir is positioned to the left against the magma pipe.
I will agree with you on magma replenishment in pipes. If you're going to need a LOT of magma, pumping straight from the sea near the map border is best. But there's a limit to how close to the edge you're allowed to build and that increases the distance that is subject to flow mechanics.
Frankly, tapping from the bottom of a magma pipe gives the best compromise between available flow rate vs available magma over time. Given that the original poster is simply wanting to perform some large scale obsidian casting, flow rate and total magma really isn't his or her bottleneck. Instead the issue is simply getting each layer ready for casting. And that's not a pretty or fast issue. For instance, the "ideal" method would be to
1. Fill mold area at level Z with magma.
2. Flow at level Z+1 water until filled, converting level Z to obsidian.
3. Flow at level Z+2 with magma until filled, converting level Z+1 to obsidian.
Keep alternating between flooding a level with magma or water until the desired casting is completed.
Unfortunately, step 3 above doesn't work. So instead step 3 becomes something like
3. Drain off excess water and let dry. After dry, fill level Z+1 with magma and flood Z+2 with water converting Z+1 to obsidian.
And this is SLOW. REALLY REALLY SLOW
Or instead construct two pits that are 20x44 spaced apart from each other by 10 tiles. And at the top of each of those pits have some retracting bridges that can be flooded with magma or water as designed and dropped into the pit. Then proceed to start filling and emptying them until the original poster has two cast columns of obsidian. At least the casting in this case will go as fast as fluid can be obtained. Once the two columns are completed, then build a few more bridges between them. Clean out the space. And cast the 10x44 region between the columns in a similar manner with obsidian.
Author
Topic: Looking for framerate and workspeed improvements (Read 2970 times)