Discovery Gaming Community

Full Version: My opinion on why the game has constantly been losing players
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Karlotta the easiest way get your ideas rejected is by doing the same thing, getting personal and undermining your own attempts. Instead just present them and let them devs decide whether or not to use them or at least consider them fairly. Although you are a member like everyone else as pointed by a dev, truth is, you are being pro active in giving Ideas and basically pre implementing them on your own for easy understanding and testing. I'm a 100% they will take it way better and you will most likely see it in game sooner or later if you keep focus in the idea at hand. Let time do its magic, appreciate what they have done so far and defend your points without marginalizing the topic. Learn to ignore, and keep focus in the objective.
(07-09-2018, 06:17 PM)Batavia Wrote: [ -> ]
(07-09-2018, 03:26 PM)Karlotta Wrote: [ -> ]
(07-09-2018, 03:10 PM)Narcotic Wrote: [ -> ]I can't tell what the ultimate solution would be, and I don't want to sit down and figure something out*,

*we have Karlotta for that

Yes we do. Bless me.

Funny that. I replied to that thread five months ago to tell you why that doesn't work. Pretty much killed the thread. I'd love to hear your rebuttal, but I haven't.

I'm sorry, I may have intended to reply later and then forgot about it when I moved on to other projects. I'll reply to the post there after this one.

(07-09-2018, 06:17 PM)Batavia Wrote: [ -> ]Your graphs and the explanations given do not follow from each other because your underlying assumptions are incomplete and easy to circumvent. Your proposal just cuts content without giving anything in return.

My proposal doesn't cut any content , just a few jump hole connections while rerouting others. "In return" for the removed connection, a connection that allows for farther (and thereby faster) travel across Sirius along "highways" is given, while making two systems that are intended to be emptied for the benefit of the "highway" systems and connections.

(07-09-2018, 06:17 PM)Batavia Wrote: [ -> ]Network routing is essentially a pricing issue, not a connectivity issue. Data packages (players, in this case) go out of their way to avoid bottlenecks, because they're pricey, and with every node (base) being agnostic in terms of being an entry or exit point for a trade route, there's no telling exactly where players will be at any given point, meaning they'll be just as randomly distributed as they are now. Some nodes having a higher connectivity than others does not translate into higher chances of encounters, just in higher changes of repetitive most optimal paths. But without any reason to linger in a location, there is never going to be the build-up of traffic that leads to encounters, as statistically everyone remains just as likely to pass by each other as before.

It seems to me you're applying a datapackage networking analogy that is based on always choosing the fastest path, where traffic in nodes are slowed down the more the node is active, and therefore active nodes are avoided. This isn't a good analogy because 1: traffic isn't really slowed down by active "data packets". The fastest path remains the fastest path here, only the risk of piracy increases with higher activity in one "node" 2: The aim here is not only faster travel. The aim is also to give less options for travel altogether, reducing the number of possible routes and thereby make certain routes more attractive, so players choose them more.

(07-09-2018, 06:17 PM)Batavia Wrote: [ -> ]If you want to increase the chance of encounters, you actually have to concentrate node density, not edge density. Which basically translates to: put 50% of commodity points of sales in ~2% of the systems, without increasing the amount of dockable bases, and spread the other 50% among 98% of the systems. Then adjust commodities and prices to have a directional flow from low density areas ("rural production" aka mining in Freelancer) to high density areas, while labour intensive commodities (consumer goods) flow between densely populated areas only. For example: a hauler flying between New Berlin and New Tokyo should always incur an opportunity goods for selling their goods in the Sigmas, rather than being guaranteed an equal credits/second rate regardless of where they go. Ideally, doing all this, when in the most profitable areas, it should have to take a hauler several minutes of queuing behind other players and/or NPC's before they get to land, with the queuing happening in about 2% of the map where the highest profits area. That's an actual bottleneck giving pirates targets to raid, and house police targets to protect.

First of all, this is more than a trade issue. It's an issue of every kind of travel that goes from A to B.

Second, if you use a program like Freelancer companion, you'll see that profits are already balanced in the way you describe. What this alone does not do, however, is concentrate players along the same routes. You have to reduce the number of possible routes to do that. What longer profitable routes also do not do (rather the opposite), is increase the chance of encounters. Simply imagine 5 traders going back and forth a short route vs a longer one. They'll meet each other (or a pirate sitting somewhere on the route) less frequently along the long route.

Also your idea about piracy in queue locations conflicts with several realities of discovery. There is no queuing at bases before players dock, there are quick dock point. Bases are also well defended against pirates, and the best place for piracy is a lane or mining field far from dock or defense opportunities for the victims.

(07-09-2018, 06:17 PM)Batavia Wrote: [ -> ]Your proposal of just increasing edge connectivity density does nothing when players can bypass each other just as well as before, since edge capacity is basically unlimited in Freelancer.

What do you mean with "increasing edge connectivity"? I'm actually reducing the number of connections between systems.

(07-09-2018, 06:17 PM)Batavia Wrote: [ -> ]It is node capacity that is limited and which should be focused on. Systems are not nodes, just a collection of edges. Traders just going past each other and going "Hai, bye!" is not an engaging encounter.
What do you mean with "node capacity"?

I'm not only focusing traders, I'm focusing everyone who travels. The aim is not to get more traders to say hi to each other, the aim is to make everyone's routes more predictable and limited, so players can find each other more easily, be it friend or foe.

I really have the impression you're applying a network theory which isnt really applicable here, because it's based on wrong or over-simplified assumptions that don't match Freelancer gameplay.

(07-09-2018, 06:17 PM)Batavia Wrote: [ -> ]To illustrate, have a picture of where economic activity is located in the US:

Your map has absolutely nothing to do with any of this. I'm not trying to build a realistic economy on a continent with unlimited travel connections between two points.
(07-09-2018, 07:01 PM)Manu Wrote: [ -> ]Karlotta the easiest way get your ideas rejected is by doing the same thing, getting personal and undermining your own attempts. Instead just present them and let them devs decide whether or not to use them or at least consider them fairly. Although you are a member like everyone else as pointed by a dev, truth is, you are being pro active in giving Ideas and basically pre implementing them on your own for easy understanding and testing. I'm a 100% they will take it way better and you will most likely see it in game sooner or later if you keep focus in the idea at hand. Let time do its magic, appreciate what they have done so far and defend your points without marginalizing the topic. Learn to ignore, and keep focus in the objective.

It's easy to say "just ignore it", but people have believed and repeated what other people said about me in the past because I didn't object to it.

Even when someone calls me a dick and then white-knights admins and devs claiming I'm insulting them personally by criticizing some of their actions (or inaction), and even if it's actually pretty plain to see what he did and why, if enough people do the same (and at least 3 of them posted here), there will be more people who will take them seriously and believe them.

Replying or not replying is the choice between cholera and plague, and I choose to reply because the times when I didnt (or when the replies got deleted by admins/mods) have proven to be even more damaging than the times that I did.
well all I can see is that every ones are really good at telling people what needs to be done . WALLSSSS of text that are so boring to read that after the first 2 sentences I want to fall asleep. This stable is not big enough for all the high horses. Why people don't want to stick around. *ponders* I really hate to say it is the game community . That is a fact. I there been countless times I am setting some where and I get someone new PMing me that need questions answer and a good 80% of them tell me I was the only one that answered back to the response. BAM !!!! There is part of it right there. Most of you are to busy to even help a new player asking you your valued opinion. When people don't feel welcomed they will just leave and there is no way you can cast that off and blame someone else for that. YOU FAILED.


*puts soap box away*
(07-10-2018, 12:11 AM)Karlotta Wrote: [ -> ]
(07-09-2018, 07:01 PM)Manu Wrote: [ -> ]Karlotta the easiest way get your ideas rejected is by doing the same thing, getting personal and undermining your own attempts. Instead just present them and let them devs decide whether or not to use them or at least consider them fairly. Although you are a member like everyone else as pointed by a dev, truth is, you are being pro active in giving Ideas and basically pre implementing them on your own for easy understanding and testing. I'm a 100% they will take it way better and you will most likely see it in game sooner or later if you keep focus in the idea at hand. Let time do its magic, appreciate what they have done so far and defend your points without marginalizing the topic. Learn to ignore, and keep focus in the objective.

It's easy to say "just ignore it",.



Normal human behavior: "This person might have a point but I don't want to accept it so I'm gonna sort of derail the topic, nit pick about a minor detail that is often not even related to the problem at hand, get personal or incite the person in order to marginalize your idea or concept" CAUSE THEY DON'T HAVE A SOLID ARGUMENT!! THEY DON'T!!!. Look a the thread I made about my personal first impression about the game and look at how many people tried to derail and marginalize even using examples non related to the issue at hand. You can clearly see how some desperately tried to brush off the issue. For the Grace of God don't fall for such silly and rudimentary tricks. If they get personal with you is because they know your triggers! Learn this now and put at end to it!
You know, you could just start ignoring Karlotta. Most people do, or at least try to. Works out best.
(07-10-2018, 01:08 AM)Chubicega Wrote: [ -> ]You know, you could just start ignoring Karlotta. Most people do, or at least try to. Works out best.



Best for what?
Best for not getting threads derailed into pointless discussions.
(07-10-2018, 01:14 AM)Chubicega Wrote: [ -> ]Best for not getting threads derailed into pointless discussions.


I totally agree! By not making back handed personal comments to diminish or degrade a person. Right Cubicegaaaaa?
(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]My proposal doesn't cut any content , just a few jump hole connections while rerouting others. "In return" for the removed connection, a connection that allows for farther (and thereby faster) travel across Sirius along "highways" is given, while making two systems that are intended to be emptied for the benefit of the "highway" systems and connections.

You've not demonstrated that at all. If you'd actually do the math, you'd fine that the increased chance of encountering another player along these highways is a negligible percentage increase because all haulers are randomly distributed along all equally profitable paths. Chance increases in the 10's percentage range, not the multiple 100's of percentage increase that you need to notice any effect. You're dramatically overestimating the efficacy of cutting paths.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]It seems to me you're applying a datapackage networking analogy that is based on always choosing the fastest path, where traffic in nodes are slowed down the more the node is active, and therefore active nodes are avoided. This isn't a good analogy because 1: traffic isn't really slowed down by active "data packets". The fastest path remains the fastest path here, only the risk of piracy increases with higher activity in one "node" 2: The aim here is not only faster travel. The aim is also to give less options for travel altogether, reducing the number of possible routes and thereby make certain routes more attractive, so players choose them more.

Speed is irrelevant, as are degrees of freedom. Giving less options doesn't equate to players hanging around when they meet each other. It's all about price and cost; opportunity cost of choosing higher or lower density space specifically, and whether to move or stay put in a particular situation.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]First of all, this is more than a trade issue. It's an issue of every kind of travel that goes from A to B.

No, it is a trade issue at its core. Haulers and miners follow profit. Pirates follow haulers and miners. Police follows pirates. If you want to fix player density, you start with fixing trade.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]Second, if you use a program like Freelancer companion, you'll see that profits are already balanced in the way you describe. What this alone does not do, however, is concentrate players along the same routes. You have to reduce the number of possible routes to do that. What longer profitable routes also do not do (rather the opposite), is increase the chance of encounters. Simply imagine 5 traders going back and forth a short route vs a longer one. They'll meet each other (or a pirate sitting somewhere on the route) less frequently along the long route.

None of this is relevant as long as everybody keeps moving. Encounters dissipate quickly when you're travelling faster than light.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]Also your idea about piracy in queue locations conflicts with several realities of discovery. There is no queuing at bases before players dock, there are quick dock point. Bases are also well defended against pirates, and the best place for piracy is a lane or mining field far from dock or defense opportunities for the victims.


As far as I know you can't jump queues at docking rings and jump gates.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]What do you mean with "increasing edge connectivity"? I'm actually reducing the number of connections between systems.

Amount of incoming/outgoing connection points. And no, you're not reducing the number of connections. Systems are not real locations. They're transient. Bases are the real locations, and you're proposing nothing for them. As I said before: less connections would be a meaningless change as long as nobody has a reason to stop and just keeps racing past each other because their delivery point is somewhere else.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]What do you mean with "node capacity"?

Amount of players visiting a particular base as a percentage of all bases visited and the throughput capacity of said base in terms of ships per unit of time.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]I'm not only focusing traders, I'm focusing everyone who travels. The aim is not to get more traders to say hi to each other, the aim is to make everyone's routes more predictable and limited, so players can find each other more easily, be it friend or foe.

And I'm telling you your math is way off. You won't get the increase in encounters that you're looking for simply by cutting a few routes.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]I really have the impression you're applying a network theory which isnt really applicable here, because it's based on wrong or over-simplified assumptions that don't match Freelancer gameplay.

Network theory is fundamental to the study of trade networks. If anything, it's your assumptions that are too simple because you're not doing any cost accounting and are ignorant of the fact that optimal paths are not impacted by the amount of connected nodes and edges but by the opportunity cost of picking one connection over another.

(07-09-2018, 11:49 PM)Karlotta Wrote: [ -> ]Your map has absolutely nothing to do with any of this. I'm not trying to build a realistic economy on a continent with unlimited travel connections between two points.

Unlimited travel connections? Do 18-wheelers drive through fields in your neck of the woods? Tongue If you hadn't noticed .. travel in Freelancer happens along the same repetitive lines as roads do, from point to point. You can pretty much construct a road map for Freelancer with travel distances and everything. The roads just happen to be in space, but most travellers don't really deviate from them. 99,9% of space in a Freelancer system is almost never used, with less than 1% of space in a given system facilitating almost all travel in that particular system. The economic principles of both real world shipping and Freelancer shipping are mechanically the same, with the same opportunity cost principles applying.

Sorry, Karlotta, but you're basically ignorant of network topology as it applies to trade and transport, and your projections don't work out as you think they would because of that.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19