ED Background Simulation - Large Faction Influence Swing Mechanics

[CMDR Endincite]

Ok the system is not 100% transaction based. Selling 2 mil cr worth in one transaction vs. 200 cr does certainly have more effect in my experience. Transaction number adds weight to any given value of transactions - which is absurd - but the two do work in tandem.

It's quite clear that value is accounted for to some extent when you consider the effect on Pilots Fed rank - entirely value.

The issue as I see it is not that 1 x 20,000 credit bounty does the same as 1 x 40,000 - afaik it doesn't. Rather that 2 x 20,000 transactions will do more than 1 x 40,000.

 

[DNA-Decay]

They're shuffling enough data about each transaction to work out the gearings. It's not just one sale equals one influence drop.
But I don't think they can pull back every transaction history.
A drop will go in each bucket and the ratchet that controls the gear gets another notch, but the transaction is forgotten immediately. There's no paper trail.

There's another video kicking around where DB OBE has a yarn about how they did the procedural generation back in '84.
Frontier are still a small company punching well above their weight. And they will still be using the style of cheats and optimisations that they did back then. What they do is not perfect. But it works well enough.

And they try to fix up the strange things and the edge cases.

The algorithm would be something like num transactions x log total value of transactions x log population x daily braking cap x state effect x log system traffic x demand x the size of Dav's underpants x forum toxicity and pH (acidity vs saltyness)

 

[CMDR Endincite]

The algorithm would be something like num transactions x log total value of transactions x log population x daily braking cap x state effect x log system traffic x demand x the size of Dav's underpants x forum toxicity and pH (acidity vs saltyness)

I haven't seen evidence of demand having an effect. Selling e.g. Imperial Slaves bought at tremendous discount to no-demand has the same effect. Also, to gauge the effect of forum toxicity, we'd need to see it actually fall...ever.

 

[Starwolfe]

I haven't seen evidence of demand having an effect. Selling e.g. Imperial Slaves bought at tremendous discount to no-demand has the same effect. Also, to gauge the effect of forum toxicity, we'd need to see it actually fall...ever.

Demand's application is more to do in the finer tune of the profits. If demands are met the price begins to drop, but in some systems the demand is so high it'll never change. It kind of depends on where from what I've observed.

 

[Jmanis]

Ok the system is not 100% transaction based. Selling 2 mil cr worth in one transaction vs. 200 cr does certainly have more effect in my experience. Transaction number adds weight to any given value of transactions - which is absurd - but the two do work in tandem.

It's quite clear that value is accounted for to some extent when you consider the effect on Pilots Fed rank - entirely value.

I've not seen that to be the case before re: profit having an effect for influence. There are point effects definitely; for example:
- Selling for 2m cr of profit, vs 200cr of profit, will massively (and proportionally) increase your reputation.
- As you mention, elite ranks are tied to value as well, but again, these are point transactions.

Point effects, like changes to supply and demand, elite ranks etc all happen *on the dot* or close to (it'll take 30 seconds or so for a faction to acknowledge you're hostile and open fire, after getting the notification of rep drop), but influence changes happen all at once on the tick, which is (probably an unresolvable) part of the problem.

The issue as I see it is not that 1 x 20,000 credit bounty does the same as 1 x 40,000 - afaik it doesn't. <snip>

In all the tests i've done, it does do the same. My faction is coming up to a conflict state soon, I'll run some numbers then. If anyone can get data sooner, go for it.

Possibly a point for it to be a mistake, but the samples I used were combat bonds in wars.
2 x 100,000 were twice as effective as 1 x 200,000
2 x 500,000 were twice as effective as 1 x 1,000,000
2 x 1,000,000 were twice as effective as 1 x 2,000,000.

And in all cases, the increases were 2% and 1% respectively, for systems =~ 50,000 pop'n

 

[anonymousanarchist123]

Possibly a point for it to be a mistake, but the samples I used were combat bonds in wars.
2 x 100,000 were twice as effective as 1 x 200,000
2 x 500,000 were twice as effective as 1 x 1,000,000
2 x 1,000,000 were twice as effective as 1 x 2,000,000.

As much as we could try and figure out what is going on exactly with transaction size / profit / system population etc the issue is very simple, all methods of commiting transactions should have the same outcome for the same overall profit/loss per population per day IMHO, spending effort trying to get to a formula to understand exactly what is going on won't fix the issues...

As as stated in the first post it's about making a 1 unit transaction done 700 times the same as a 700 unit transaction done once, and to me it's either a case of everything being calculated by aggregating the days efforts per player or using the equivalent number of 1 unit transactions and calculating with that:

Option 1 - Aggregate Daily Transactions Per Player/Transaction Type

A player’s activities would need to be aggregated for that day, so that any commodity trading, black market trading, exploration data, bounties or bonds for the 24 hour period are tallied up before being used to determine an influence adjustment, thus removing the transactional nature of the methods described above.


Option 2 - Breakdown All Transactions To Units

A player’s activities would all need be broken down to individual unit transactions regardless of how they are submitted, effectively making all influence changes happen as they do using the mechanisms described above. Once all methods of committing transactions are working the same way the effect of these could be balanced to work as intended with influence swing.​

Trying to factor most of this change in the process into the service call when a player submits the transaction(s) rather than at the point of the tick would be wise too...and the client wouldn't need to wait, e.g. client call -> server acknowledgement -> client completed -> server processing ----> daily tick completes all as per today

 

[McIranta]

Also there is this one thing: It slows down the servers to sell 700t of goods in 1t patches. Ever wondered why the servers are so slow? Or are they using different servers for missions and markets?

 

[anonymousanarchist123]

Also there is this one thing: It slows down the servers to sell 700t of goods in 1t patches. Ever wondered why the servers are so slow? Or are they using different servers for missions and markets?

That's a very good point, if the transactions based approach is addressed and doesn't bring any benefits no one will bother and server load is reduced, same goes for missions stacking, mode swapping etc etc etc, get all these areas addressed and server load is much lower...

So.....probably best then to go for an a process like option 1 above where as a player commits transactions (single units or batches, doesn't matter) the existing transactions they've already committed in the system bucket for that 24hour period are incremented against rather than more added in separately, that way the tick based process remains as is and the only overhead is on the server at the point of the client committing these.

 

[Jmanis]

<snip>

So, you won't get any disagreement from me?

My post was implying that those combat bond results were expected behaviour, because the value of those bonds doesn't matter, just the number of times you hand in bonds.

 

[Schlack]

So, you won't get any disagreement from me?

My post was implying that those combat bond results were expected behaviour, because the value of those bonds doesn't matter, just the number of times you hand in bonds.

Our experience in high traffic/activity systems strongly suggests that the tick equation/formula is not solely transaction based. Other mechanisms and interactions come into play.

 

[anonymousanarchist123]

So, you won't get any disagreement from me?

My post was implying that those combat bond results were expected behaviour, because the value of those bonds doesn't matter, just the number of times you hand in bonds.

Gotcha I guess my post was more for all concerned as based on your post and others before it, I could see where this all might go with lots of conversation on how it all exactly works right now, but that is a distraction away from what really needs looking at...and would clutter this thread with a back and forth. We know there's a variation for combat bonds/bounties/trading/exploration data based on how you hand in due to transaction size / number of times per day / profit/loss / system pop etc etc and that is what is wrong IMHO. The KISS principle should be applied rather than further complexities with caps etc, e.g. all variations on commits have the same affect over the course of a day.

 

[anonymousanarchist123]

Feedback

If anyone has any feedback of BGS mechanics and would like to share information to update this content please feel free, I'll try when I can to update the details above to be as accurate as possible.

So.....any more BGS mechanics not listed in the first post that I'm unaware of...don't be shy, let's get all the info out there for the betterment of the game! I don't mean more detail on the understood scenarios but any other scenarios where influence can be swung up and down?

 

[Jmanis]

So.....any more BGS mechanics not listed in the first post that I'm unaware of...don't be shy, let's get all the info out there for the betterment of the game! I don't mean more detail on the understood scenarios but any other scenarios where influence can be swung up and down?

There's already a thread for that https://forums.frontier.co.uk/showt...ide-to-Minor-Factions-and-the-Background-Sim/

(In actuality, I'm surprised these threads haven't been merged yet...)

 

[anonymousanarchist123]

There's already a thread for that https://forums.frontier.co.uk/showt...ide-to-Minor-Factions-and-the-Background-Sim/

(In actuality, I'm surprised these threads haven't been merged yet...)

Well unless you live and breath in that thread any particular details are not obvious at all, there is so much waffle, smart aleck replies and mis-understanding in there finding specific details is far from easy, something I want to discourage in here!....in all reality we need a set of sub threads related to each aspect of the BGS, the topic is just too big! Having a first post in each that is updated to detail all the relevant info would also be a good idea....

 

[Schlack]

Well unless you live and breath in that thread any particular details are not obvious at all, there is so much waffle, smart replies and mis-understanding in there finding specific details is far from easy, something I want to discourage in here!....in all reality we need a set of sub threads related to each aspect of the BGS, the topic is just too big! Having a first post in each that is updated to detail all the relevant info would also be a good idea....

An excellent suggestion. One that FD have recently decided against, instead advising us to use the Guides subforum where none of us have permission to post threads!

waffle, smart replies and mis-understanding

This is in the nature of the game. The BGS is a black box, has been from the start. Until the "bible" posted by Michael Brookes in Jan this year every little bit of information had to be theorized, tested and figured out, with the odd bone thrown to us by FD. This is one of the interesting challenges of the BGS. Experience to date suggests strongly that there will never be a definitive guide to the BGS.

Having a first post in each that is updated to detail all the relevant info would also be a good idea

Nice of you to volunteer

Seriously though Walt Kerman did an excellent job updating the primary post, but Real life has intruded. it would be something of a mammoth task to do so particularly as there is not always agreement on what the actual mechanics are.

 

[Jmanis]

Seriously though Walt Kerman did an excellent job updating the primary post, but Real life has intruded. it would be something of a mammoth task to do so particularly as there is not always agreement on what the actual mechanics are.

A while back I thought about knocking up an "idiots guide to the BGS" with just the basics. Maybe I actually finish that.

 

[goemon]

Gotcha I guess my post was more for all concerned as based on your post and others before it, I could see where this all might go with lots of conversation on how it all exactly works right now, but that is a distraction away from what really needs looking at...and would clutter this thread with a back and forth. We know there's a variation for combat bonds/bounties/trading/exploration data based on how you hand in due to transaction size / number of times per day / profit/loss / system pop etc etc and that is what is wrong IMHO. The KISS principle should be applied rather than further complexities with caps etc, e.g. all variations on commits have the same affect over the course of a day.

as said before, and brought up various times:

- counting transactions is in favor of small ships and giving beginners a chance to influence a system. your system would pretty much reward huge ships and render BGS work to a rich mans hobby, isntead of rewarding effort.
- the problem with 1T trading is imho special: it rewards not leaving the station.

a (maybe simple) fix for that would be to aggregate all transactions of a type as long as you did not undock (or even better: not leave the station isntance). still that could be gaimed .... i already see people exploiting this by jumping to sc between each 1T transaction ;-)

 

[anonymousanarchist123]

as said before, and brought up various times:

- counting transactions is in favor of small ships and giving beginners a chance to influence a system. your system would pretty much reward huge ships and render BGS work to a rich mans hobby, isntead of rewarding effort.
- the problem with 1T trading is imho special: it rewards not leaving the station.

a (maybe simple) fix for that would be to aggregate all transactions of a type as long as you did not undock (or even better: not leave the station isntance). still that could be gaimed .... i already see people exploiting this by jumping to sc between each 1T transaction ;-)

We'll have to agree to disagree then I think Add a couple of decimal places to the influence figures if you want to see a eagle pilot making a difference....the point is that anyone starting out in the game wont really be all that interested in the BGS, they'll be busy with earning credits, upgrading ships, ranking up etc, only when they get to a position of having a good trading ship might they actually care about the BGS, like most of us here I assume....

 

[Schlack]

as said before, and brought up various times:

- counting transactions is in favor of small ships and giving beginners a chance to influence a system. your system would pretty much reward huge ships and render BGS work to a rich mans hobby, isntead of rewarding effort.
- the problem with 1T trading is imho special: it rewards not leaving the station.

a (maybe simple) fix for that would be to aggregate all transactions of a type as long as you did not undock (or even better: not leave the station isntance). still that could be gaimed .... i already see people exploiting this by jumping to sc between each 1T transaction ;-)

No-one is going to play the BGS like that for too long. Even a sustained assault against another faction would drive the players doing it mad, unless they were particularly motivated. That's my main problem with the exploit. Not that it can be used to crush other player factions (such conflicts appear thankfully very rare ), rather that it breaks the game and takes away the challenge. If 1T exploiting is all there is, there is no game.

Jmanis, One of our cmdrs has prepared such a guide, it rapidly dated. We were waiting until 2.2 settled down before redrafting. I would suggest waiting and seeing how 2.2.03 changes things before developing anything. This is the problem with all guides! by the time the BGS has settled and we have figured out the mechanics new changes come in.

It looks like the next major patch may be giving us asteroid based (new I presume instead of just re-skinning outposts). If they open more planet types in season 3 - will mean more settlements and possibly different mechanics and missions. FD also intend developing the BGS into something far deeper. Its a never ending task to keep track of all this.

 

[anonymousanarchist123]

An excellent suggestion. One that FD have recently decided against, instead advising us to use the Guides subforum where none of us have permission to post threads!

Nice of you to volunteer

Seriously though Walt Kerman did an excellent job updating the primary post, but Real life has intruded. it would be something of a mammoth task to do so particularly as there is not always agreement on what the actual mechanics are.

Here's an idea....have a main discussion thread with the first post listing other threads dealing with specific topics and first posts within containing summary info, you don't need a sub-forum to make it work....Canonn sort of do this with their threads right?

e.g.

• BGS - Discussion
  • BGS - State and Influence Mechanics - this could start out being based on the first post in the main BGS thread and expanded on later
  • BGS - State Issues
  • BGS - Influence Issues - this could start out with this thread
  • .....

I can help with the "BGS - Influence Issues" thread cause it'll basically be this with any new input through discussion...just need someone willing to get the ball rolling for each of the other threads....I'm sure the mods would be happy to lock the old threads and add links in the first and last posts of them once new ones are established