Post reply

Warning: this topic has not been posted in for at least 120 days.
Unless you're sure you want to reply, please consider starting a new topic.

Note: this post will not display until it's been approved by a moderator.

Name:
Email:
Subject:
Message icon:

shortcuts: hit alt+s to submit/post or alt+p to preview

Please read the rules before you post!


Topic Summary

Posted by: sloanjh
« on: August 09, 2015, 01:50:57 PM »

Not sure yet if this is a good idea :) but I have completely rewritten the civilian code and everything else that relies on path-finding, such as refuel at nearest colony. Instead of a four jump limit, there is no longer any limit at all.

I am setting up a very complex campaign at the moment with fourteen (!!) player races so if the new code can handle that, it should be fine. If not, I will have to place some type of restriction but I will make it configurable by the player.

Look on the bright side - at the very least you got rid of the existing bugs :)

John
Posted by: Bryan Swartz
« on: August 09, 2015, 01:08:03 PM »

Sounds like a big step forward regardless of how it turns out Steve.

*Moment of silence for the pathfinding code observed*
Posted by: Steve Walmsley
« on: August 09, 2015, 12:47:41 PM »

Not sure yet if this is a good idea :) but I have completely rewritten the civilian code and everything else that relies on path-finding, such as refuel at nearest colony. Instead of a four jump limit, there is no longer any limit at all.

I am setting up a very complex campaign at the moment with fourteen (!!) player races so if the new code can handle that, it should be fine. If not, I will have to place some type of restriction but I will make it configurable by the player.
Posted by: Arwyn
« on: August 03, 2015, 09:37:36 PM »

I've haven't played for a while due to finding out about X-COM Long War :)

Long War is the Devil!! It is MUCH more like the original Xcom, punishing! :D

(like 2300AD if anyone remembers that)

Love that game. One of the best Sci Fi RPGs from GDW, played the hell out of that and Renegade Legion.

The cool thing about 2300 is that it hit quite a bit of the hard science, and the aliens WERE alien. The Kafers were a cool concept for bad guys.
Posted by: Steve Walmsley
« on: August 02, 2015, 10:04:47 AM »

Yes, the above is along the lines I was thinking. There are now fewer systems within a given radius, so increasing the radius of civs should be OK. It's years since I looked at that code so I'll see if I can do anything to improve it as well.
Posted by: Prince of Space
« on: August 02, 2015, 08:44:49 AM »

Now we are discussing so much about a thing that may likely not even be in the game...

True, but this is the Suggestions subforum. It seems like the place for it.  :P

The simplest change might be to readjust the limit from 4 to something greater. Take the current average number of jump points, raise that number to the fourth power, and that gives you the average number of systems that the civvies search through looking for contracts. Take the average number of systems in the upcoming version, raise it to the power of x, and set that equal to the previous result. Solve for x and the result is the civvy search radius.

So for example, if the average system now has 6 JPs, and the new system has an average of 3:

6^4 = 3^x

x = 6.5 jumps (give or take; round in whichever direction you feel is most appropriate)

Would that be simple enough and satisfactory?
Posted by: Vandermeer
« on: August 02, 2015, 06:59:35 AM »

I'm no economist, but I thought supply and demand determined prices, not the cost to bring the product to market. [...]
Of course. This is already in those consideration though. You could interpret it in a way like this: "High distance between colonies makes ship contact more rare, so the supply and demand levels experience greater disparity, making the trips more worthwhile."

Now, if fuel only costs 10 cent per AE, then that would be all we need to account for, but in reality (and likely much more dominant as factor in lengthy age-of-the-sail like space trips), the increased gains from high demand need to override the increased shipping/fuel costs that come with greater distance to make such a trip really worth more. (+risk and some other factors, but those are details..)
This is what that equation above was about too. d("demand")+d("risk")+d("fuel/shipping cost")<0 means you get a more worthwhile shipping once the advantage of the first two outpace the greater investment of the third.
..Counts for a mere choice consideration between differently distant locations, just as much as for how pricing develops when technology becomes better.(..or you strap on wasteful engines)


@alex_brunius
Because if in the future fuel costs are added distance works perfectly but time dont. You dont pay 10 times more for a shipping that takes 10 times longer to do the same work. It's common sense.
That is only true in the realm of the same technology where faster is indeed more expensive. If you are slower and have more traveling expenses (mostly fuel) because of your technology being inferior though, you indeed end up paying more for less speed. So at a tech jump you see the opposite effect, and distance measurement can not emulate that.
Quote
It's also desirable for higher tech designs to earn more even in the current model without fuelcosts because the high tech designs is a much higher investment that will be harder to pay off with constant earnings (locked to time).
That the initial technology (/build-cost) investment works to offset this behavior is a valid point, but it is a one time investment at first (making it fade in relevance over time), and then secondly likely not the dominant expense. (shipping is after all cheaper today than it used to be before, despite of expensive container ships)

The reason why I thought that time would work out as a measurement in Aurora though, is that in Aurora the engines for civils are fixed as always being the 0.5 power factor 25 HS design with a fixed percentage of engine allocation in a freighter too. So the whole "design fast wasteful taxi" thing is factored out in standard Aurora, meaning we never would actually have to face this effect were a fuel wasting 'fighter-engine' freighter would make cheaper trips despite blasting the fuel. What stays then is instead that only overall tech progression can make a freighter faster, and then we see this cheapening effect where it actually also makes sense.(partly because of the reaction time and reduced risk, but mostly because of the also saved fuel)
Fuel would probably be the best measurement of pricing, but it may be difficult or impossible to track for civils in Aurora. So then you have to decide between the other two options of either measuring time or distance.
- Measure distance: Freighting returns (per freighter) increase as technology goes up.
- Measure time: Freighting returns go down as technology progresses.

The second effect makes realistically more sense for the various reasons of the former answer, so I chose to mention that one instead.
Only way to tilt the effect is the technology investment cost you mentioned, but those would have to be quite big in comparison, and it would also be impossible to track them as they depends on how long designs stay in flight (/a tech era lasts), which is different from game to game. Too much consideration in the end, as the effect would be unnoticeable to players anyway. Even the time measurement would just be a make-shift approximation to a realistic answer, as you will probably research fuel efficiency (the realistically most saving factor) at different points to a total engine upgrade (which increase speed and thus would cause the actual saving to appear ingame). First priority should always be to hold the calculation simple before realistic, and this could come close enough.


Now we are discussing so much about a thing that may likely not even be in the game...
Posted by: TheDeadlyShoe
« on: August 02, 2015, 03:49:30 AM »

Yes, price would be the natural method of bringing local vs distance runs under control.  There could also be price for colonists, determined by employment levels.  Colonies with overemployment would be willing to pay a premium for more passengers.  *shrug*. A price system might get pretty complicated though, it has to be reasonably comprehensive to get real benefit from it comparing to just fudging it.

I also think loading times for goods should go up.  Right now there really isn't a lot of point for commercial spaceports except for extremely short runs.  The loading times right now make it pretty important to fit in cargo handling on troopships or dropships, even with their sometimes-tight tonnage requirements... but for cargo ships they are total gimmes and there's not even much point to teching them up.  Longer basic loading times would add a higher minimum time overhead for commercial runs, disadvantaging short runs.

Posted by: Prince of Space
« on: August 01, 2015, 06:13:05 PM »

I'm no economist, but I thought supply and demand determined prices, not the cost to bring the product to market. If so, then the way to get civvies to deliver commodities to colonies beyond the Earth-Luna milk run is to allow prices to fluctuate locally.

If a colony in Alpha Centauri has pent up demand for consumer electronics, the price of consumer electronics there will go up. Somebody is going to see that as an opportunity to break away from the pack and ship some iPads to the outer colonies.

If I understand the reason for the four jump limit correctly, it's to keep the shipping lines from calculating an excessive number of buy-here/sell-there permutations, and thus bogging down the game. But if there was an up to date list of trade runs, sorted by profitability, the shipping lines just need to pick from the top of the list.

Now, I'm not sure if this is computationally simpler than the current model, or simple enough to justify dropping the 4-jump limit, but some approximation of market economics would provide a self-regulating mechanism for ignored outsystem colonies, regardless of the jump point density.
Posted by: alex_brunius
« on: August 01, 2015, 02:59:38 PM »

//Double Post, sorry
I wrote so much, but here is a far easier to understand disproof of distance as the governing pricing factor:

If time or fuel don't matter and you get a fixed reward whenever you finish a freighting trip, then the fastest ship will be the best, because 'more finished trips per [time]'.
-> Hence, a hauler with fighter engines would be the most optimal choice, because it is definitely the fastest, even though it burns through millions of liters of fuel on the way.
Of course this is not how the real market could work.

Sounds like a good argument for distance and against time.

Because if in the future fuel costs are added distance works perfectly but time dont. You dont pay 10 times more for a shipping that takes 10 times longer to do the same work. It's common sense.

It's also desirable for higher tech designs to earn more even in the current model without fuelcosts because the high tech designs is a much higher investment that will be harder to pay off with constant earnings (locked to time).

Even a same tech design with higher speed will cost more investment.

But the best argument is probably that it is more similar to how the game currently works. Right now faster ships will make more traderuns and earn more in the same time if they stick to the same route, so lets not remove this nice and realistic feature, ok?
Posted by: Vandermeer
« on: August 01, 2015, 02:11:30 PM »

//Double Post, sorry
I wrote so much, but here is a far easier to understand disproof of distance as the governing pricing factor:

If time or fuel don't matter and you get a fixed reward whenever you finish a freighting trip, then the fastest ship will be the best, because 'more finished trips per [time]'.
-> Hence, a hauler with fighter engines would be the most optimal choice, because it is definitely the fastest, even though it burns through millions of liters of fuel on the way.
Of course this is not how the real market could work.
Posted by: Vandermeer
« on: August 01, 2015, 02:03:45 PM »

The root cause is a constant reward regardless of distance. IMO the only way to fix it properly is to scale traderoute income with distance * cargo.

The reason you want to use distance rather then time is because this gives an advantage to fast new designs. In reality a shorter traveltime would rather cost an extra premium, not the other way around.
I thought about distance instead of time or fuel as well, but aside from being surely more difficult to determine in a system where civils already are the calculation hog nr.1 already, I dismissed the idea because distance is an illusionary since relative price factor.
I see it like this: 200 years ago, shipping some nice clothes or sweets that you wanted from US to Europe would have cost a lot, and they would still arrive slowly. Shipping technology advanced beyond the sail eventually, and transportation cost went down.(of course as secondary factor also due to tighter logistic networks) However, we can of course still get a fast delivery for more money (like airmail), but is that really because of technology advance? No, not really, still more or less same fossil ovens, yet it costs more because of the less efficient transportation (fuel/energy expense per weight unit) or having to act outside of established supply lines.(personal taxi at 10 is more expensive than 9 O'clock train)
If technology were to advance to give us say personal around-the-world jet/vtol drone delivery with a cheap as hell He3 fusion motor, transportation cost would not go up, but instead shrink even further. In an eventual extreme, a star trek transporter system would probably cost you a stamp mark worth of electricity in the end.(..and this is true even if we had it between planets and stars)

So the point is that, yes, within the realm of your current technology level faster transportation will cost extra, as a faster "ship" or whatever will use less efficient engines, or you need to charter(or build in aurora) a special clipper for an outside of routine job.
Space and distance are relative though, and if your means to get around become more efficient (and at the same time sometimes faster), the expenses of shipping fall down as well, so as long as there is competition on the market, that will push providers to attract with lower prices eventually, if not immediately.( as a "sell-point" of the new technology for example. the flyPhone! faster than your old hauler and microwave oven combined, and can be located via galactic-positioning-system at all times!)
Faster transportation technology virtually 'shrinks' the world around you, and with that all kind of things happen:
1 - highest possible delay of ware supply decreasing (mitigating the "readiness" price component of rare wares)
2 - better reaction time ensures that business can make more secure deals (reducing the "risk" price component. Risk is not really pirates, but actually more inherent long term investment danger since long delays between request and delivery might mean you will be beaten to it. What you do is usually you factor in some sort of self insurance each time, - otherwise you are just gambling on your luck.)
3 - if it is also more fuel efficient, you might actually save on energy, which is probably the base factor to the whole transport cost equation
? - ...possibly more things I can't think of right now

So in some cases the 3rd point's price increase (less efficient engine) might override the advantage of the two speed advantages, which creates these fast "taxis" that might be situationally useful at some point, but wont evolutionary survive on the mainstream market.
An advance in technology is therefore made if that doesn't happen, so if you get at least a price of d1+d2+d3<0, you have found a cheaper way to do it despite being faster.
...Which means the economic returns of transportation tax go down as technology advances. :P


...Consequently also meaning that time and fuel are the actual influencing factors, as distance is only a influence, but what really counts is how fast+cheap you can cross it.
Posted by: sloanjh
« on: August 01, 2015, 08:08:09 AM »

The root cause is a constant reward regardless of distance. IMO the only way to fix it properly is to scale traderoute income with distance * cargo.

The reason you want to use distance rather then time is because this gives an advantage to fast new designs. In reality a shorter traveltime would rather cost an extra premium, not the other way around.

And the revenue (or score) should probably prefer longer trips, i.e be faster than "proportional to".  In "reality" this would show up in load/unload time, i.e. revenue = rate*distance - fixed_cost_per_trip_due_load_unload.  In practice (game mechanics) it would probably be easier to scale with a soft exponent, e.g. revenue - rate*distance^(1.1).  The former would severely impact the current revenue from the earth/mars run; the latter would simply prefer longer routes.

John
Posted by: alex_brunius
« on: July 31, 2015, 12:02:21 PM »

Yes, I agree. I might reduce the benefit of in-system runs. Need to look at it in more detail.

The root cause is a constant reward regardless of distance. IMO the only way to fix it properly is to scale traderoute income with distance * cargo.

The reason you want to use distance rather then time is because this gives an advantage to fast new designs. In reality a shorter traveltime would rather cost an extra premium, not the other way around.
Posted by: Vandermeer
« on: July 28, 2015, 07:31:54 PM »

I thought about that there could be some sort of timer to a civil ship (or just a note that specifies when it was loaded). With this it could be introduced that a ship's contents are worth more the longer it stayed underway, thus naturally creating fairness between long and short distance trade. It also makes sense from the realistic perspective, as goods from distant worlds should be more rare exactly because of the more costly shipment, thus increasing in value. Faster ships would reduce the prices later on, just as it goes when logistic becomes easier, more competitive and wares more readily available.(bananas were exotic once too)
An alternative (or upgrade) could be to just start counting the fuel that a ship would have lost by transporting the goods, if that is even in the realm of possibilities for civilians. The result of that would however be even more logical to be a base of pricing, and would only drop prices with fuel efficiency tech or larger engines(/ships) of course. ...It would also be more calculation I presume. The first method would only make a time stamp on loading, and then do the 1 calculation on unload to determine final value of goods by comparing to the time passed 'now vs. time stamp'.

However, a system like this would just make the returns more fair, but solving how civilians even decide on trade routes is another and surely more complicated issue.