john2218

There everywhere

Did you review what it was as advertised, could be or what it is??

You're going into a very strange realm by asking this question. When you say "electricity" are you referring to AC or DC? Do you mean current, voltage or electrical energy? These are all important if you want to talk about electrical "speed". Current, in simple terms, can be thought of as a flow of electrons (charges, more accurately) within a conductor. In a DC circuit, these charges move VERY slowly in one direction. About as slowly as warm putty, or a few inches per hour. In an AC circuit the charges just wiggle back and fourth and don't really go anywhere. Voltage can be thought of as the magnetic field surrounding the conductor. Think of an electromagnet. This magnetic field can build up and collapse at the speed of light, but only at right angles to the wire, so it doesn't really go anywhere at all. In a DC circuit, the magnetic field builds up and then just sits there. In an AC circuit, the magnetic field builds up and then collapses and builds up again 50 or 60 times per second, but again, doesn't really go anywhere. Electrical energy travels throughout a circuit at the speed of light and is propagated by the voltage, believe it or not. It's important to note that in an electrical circuit, there is no starting point and there is no ending point when it comes to the energy transmitted. In other words, the energy doesn't come from one spot only to flow to another along a single path. Think of it like a belt connected to pulleys. If you move any part of the belt, the whole thing moves all at the same time. Electrical circuits are the same.

There is no reason for this what so ever, in real life power moves at the speed of light and is the same strength across the whole network, so there is no reason to use agents to simulate it, an equation represents reality better in all cases. (Power created-Power used= Surplus or deficit). Water is not quite as simple but still as long as there is a surplus all areas should be fine, so why not save the computing power for an area where individual agents actually might simulate the world better (like traffic) rather than using them in areas where they can literally never simulate the world better than a simple equation??

Agreed 100%, should not use too much more computer power and would make everything work much, much better even without improved pathfinding.... Jealous I didnt think of it myself.

Yeah I had a city with 90K people and only 8000 were workers... Dont think they reduced the number of shoppers though, still had bad traffic in the commercial areas and there seemed to be millions of shoppers all over the place.

Power, water and sewage all use agents, which why even if you have plenty you will still have areas without them sometimes. Seems like the patch made it much better in that regard. (Although I still lost power a couple times even though I had capacity which in real life is not possible if I am on the grid) As far as police fire and busses I was talking about how they decide to go to a place, not how they actually get there. Right now individually each one looks for the nearest thing to do (Put out a fire, nab a criminal ect) rather than going to the next place it is needed, this is why all the fire trucks and police respond to the same event even if you have 20 others occuring elsewhere. Another solution would be to have them ignore a fire if it is already being fought.

So the Sims live and work at different places each day? That's extremely realistic..... Its efficient, the only problem is they never taught the agents to communicate with each other. So you get this massive herding instinct. As each agent leaving a certain property (be it home, factory, school etc) is all trying to get to the nearest goal. You saw it in my Jude thread. Due to that lack of agent communication, the ideal of keeping zones separate from each other is actually a very bad thing. As the longer the herd is in the streets, the worse the traffic (foot traffic included) gets as herds run into other herds and it just builds and builds. (you can see this effect very easy by watching an elementary school let out, as the walking school kids all head to the closest house, then ping pong house to house till the mob of kids is devoured by houses). I guess agent communication just took too much processing power..i dunno. I think it was done for memory reasons, to have each sim remember their house and jobs location would take alot of memory(I think, I am not a programmer so maybe it would not, in which case it was just a stupid decision) The thing is they did it in simcity4 which makes me think they could have done it now, especially considering the cities are so much smaller. As far as the communication goes yes, they could have done it in many cases pretty easily. For fire fighters and police just do this 1.Is there a crime Y or no 2. Is there an available responder? Y or N 3. If there is send closest responder Thats not hard, and is super obvious, they didnt do it that way do to stubborness on someone in the design team I am guessing. For busses it would be more complex but first bus could tell the others where it was going and so therefore no demand would be left so now figure out what your new closest spot with demand would be and on down the line. The thing is they used "Individual actors" in situations where individual actors dont exist, busses have routes, police and fire have dispatchers, power is either connected or not, no matter the distance from the source... Why use a system that was designed for traffic (Poorly but thats not my point here) for things that do not behave like traffic??

Its actually MUCH worse than them just taking the shortest route, they also have no memory of where the live or where they work so they all are trying to take the most direct route to the closest available house (or job ect) Once that fills they all take the most direct route to the next available un occupied house. I confirmed this by building 2 residential areas, one a bit close to my industry and one a bit further. Every day all the cars would go to the first and when it all filled the would U turn and go toward the second. This is why all the busses follow each other, they are all trying to get to the same place.

The problem is every sim wants to go to the closest available non occupied structure that needs them... They have no memory of where they lived before they went to work or school and no memory of where they worked when they wake up each morning. Once the structure they were going toward fills they will all decide to go to the next closest... Once thats full they all start moving to the next ect ect. Its shockingly horribly designed.

Simulating every electron, piece of crap and liter of water was a huge waste of developer resources that should have been obvious from the very beginning. Lets assume that it works perfectly and moves everything in the most efficient (or at least an efficient enough) manner. Why bother? In the end the equation does not change. Power created-power used=Surplus and or deficit. Piece of crap created-piece of crap processed= Surplus or deficit of pieces of crap Liters of water created-Liters of water used= Surplus or deficit of water Now assume a couple of variations where simulating the actual movements would make a difference. 1.There is plenty of supply but the agents due to either poor programming or lack of processing power cannot get from start to finish. HUGE problems for the city creator, requiring solutions such as moving power plants or water processing plants, for no other reason than the simulation is broken. 2.The city is not all connected to all other parts of the city. AH! Now actually simulating is neccesary to represent the fragmented nature of the city!! Oh wait, no its not just have the more than one "grid" and do above calculation and then determine if each grid has a surplus or deficit. 3.What if there is a deficit? AH! Now once again there is a need to simulate using actual actors to see what areas would not have power ect... Well actually no, if there is a deficit there are a few easier (and just as realistic) ways of dealing with it. You could have random brownouts (which is what occurs in real life in the case of power deficit) OR you could have the furthest from the source be without power (which is what would happen anyway if the "agents" were working perfectly) So in conclusion why spend processing power and more importantly coding time working on a system that tries to replicate something that is actually much better done with a simple equation. NOTE: Agents would be relevent if there were chokepoints and different sized piping throughout the city (IE traffic simulation) and although glassbox has issues in that regard it is not what I am talking about here.

Your approach has buildings lose power not because there isn't enough power... but because the power agent doesn't get there even though it is connected to the grid. There is nothing natural about that. THAT is an unnatural consequence to glassbox's poor hard coding.

It's not that simple. Here's what would happen: On day 1, all sims take route A. Route A was congested so on day 2 all sims take alternate route B. Everyone took route B so it was congested. They need a new route. On day 3, all sims take route A... It's called oscillation. If they only had a percentage of the sims recalculate their routes each day then pretty quickly virtually all the sims would be taking the fastest routes possible. The roads that are underutilized would be used properly. Gridlock traffic would naturally be aleviated. Best of all, it would take much less processing power so the cities could be larger. Right now it recalculates all the sims every single day. There is no reason for that.

Does the glassbox approach ever work better than mine as far as utilities go? I can't think of a scenario where it would so why make things more complicated than they need to be? I do understand your point, which is that glassbox was being developed anyway and why not extend it to the utility system. My point is that there is a much more elegant and simple way to handle utilities which they ignored because they had a new way of doing things. Now, since they went with their novel approach they are left with much more work to do to get it working than they would have had to had they used a system that actually made sense in the beginning.

@Mayorgeek I really do not understand what exactly your point is. If I get rid of capacity on a system that WAS overwhelmed and all of a sudden it works perfectly, which has happened in my game play and many others game play something is broken. YOUR point seems to be that making it more complicated somehow replicates reality better and mine is that a simple equation would actually represent reality better. With my system having more utilities actually increases the the capacity of the system, whereas your system makes it so that in many cases you need to remove uitlities for places to get coverage. Which simulates reality better in your opinion? Which system needs more debugging before release in your opinion? (Note: Clearly there wasn't enough debugging done on the glassbox system to fix the glitches inherent within it.)

My basic point is if you want to count how many llamas you have do you count the llamas or do you count how many trails the llamas make, and which one will result in more errors in counting?

Right, not to mention it works pretty well. If you played enough, you would know there's nothing wrong with it. Incorrect, I had a sewage plant that was over capacity so I built another in a different part of the city. The original was still over capacity with sewage backing up and the one I built was barely recieving any sewage. To fix this I built a larger one that still did not recieve any of the sewage. I figured out that the agents are to stupid, they always go to the closer place, so I deleted both the original sewage plants and all of a sudden problem solved!! By reducing my capacity I made the system work!! So no, the way its set up actually does not work.

If you had read the entire post you would know that thats exactly what would happen if you just simulated it rather than using "agents".