www.ClassicTW.com
https://mail.black-squirrel.com/

Using port reports to track player activity
https://mail.black-squirrel.com/viewtopic.php?f=52&t=26101		 Page 1 of 3
Author:  John Pritchett [ Fri Jun 04, 2010 3:04 pm ]
Post subject:  Using port reports to track player activity
I'm curious how this tactic is viewed. Is it too easy to locate players using port analysis? At the very least, I think this is a tactic that favors a helper user over a manual player because the helper can crunch so much more data. I know the game was not originally designed with this tactic in mind. Port reports are supposed to be used for planning trade activities, not hunting players. Is this generally accepted, or could it be improved?
Author:  The Mad Hatter [ Fri Jun 04, 2010 3:16 pm ]
Post subject:  Re: Using port reports to track player activity
John Pritchett wrote:
I'm curious how this tactic is viewed. Is it too easy to locate players using port analysis? At the very least, I think this is a tactic that favors a helper user over a manual player because the helper can crunch so much more data. I know the game was not originally designed with this tactic in mind. Port reports are supposed to be used for planning trade activities, not hunting players. Is this generally accepted, or could it be improved?


Well I definitly use this to find people, but more to find out what ports went missing, i.e. someone dropped a fighter in the sector, so the port no longer reads. The real problem is Ether Probes are too cheap, if the price was 100 or 1000 times what it is, it would make the tactic more difficult.

Wayne
Author:  Singularity [ Fri Jun 04, 2010 4:38 pm ]
Post subject:  Re: Using port reports to track player activity
John Pritchett wrote:
I'm curious how this tactic is viewed. Is it too easy to locate players using port analysis? At the very least, I think this is a tactic that favors a helper user over a manual player because the helper can crunch so much more data. I know the game was not originally designed with this tactic in mind. Port reports are supposed to be used for planning trade activities, not hunting players. Is this generally accepted, or could it be improved?


Well first, there's grimy trader. If people don't put a fig down, it can track their activity.
If someone grids over the fig and opens it back up, again, you can track their activity.

Next is CIM hunting. If someone is trading slowly, or by hand, then CIM hunting can be
effective. Basically you have someone that has a larger % explored (but not too much,
otherwise it's very slow), and they run a script that spits out changed or blocked ports
from CIM to subspace. Someone then follows up on that and kills.

This requires a certain percentage explored (30% or so), which is not cheap. It also
fails when the person on the other end is using a form of automation. A passive gridder
or autotrade script is simply too fast to catch this way. Considering that in an aggressive
game, the other side will also be running it's own grid defense, going after someone like
that carries it's own risks. So no, that's not too easy. There's a good level of give and
take there.

But, in your old school mode, it might not be a bad idea to put a delay on it. IE: continue
reporting ports unless they've been blocked for a certain period of time. Give someone an
hour window, problem goes away.

Hatter is right tho, sometimes it's the ether probe that's the issue. The max probe price
needs to be increased, so sysops could use 10k probes if they wanted.
Author:  John Pritchett [ Fri Jun 04, 2010 4:55 pm ]
Post subject:  Re: Using port reports to track player activity
That's good feedback, Sing. I'll make a note of that for the old-school mode and leave it as-is for everything else.

Probe price is configurable, but the max price is too low? Consider that changed.

Ok, currently Etherprobe can be set as high as 12K cr in Gold games, but cannot be adjusted in standard games. Would it be useful to increase Ether Probe standard cost to effect all games? And how high should it be allowed to go in Gold games where the cost is customizable?
Author:  Singularity [ Fri Jun 04, 2010 7:27 pm ]
Post subject:  Re: Using port reports to track player activity
John Pritchett wrote:
Probe price is configurable, but the max price is too low? Consider that changed.


5 man team, 1000 turns. 1 guy PPTs. 2nd guy busts up, goes red, makes cash. 2nd guy
busts up, builds original set of bases this way, makes a lot more cash. 1 guy builds, next
guy grids. Now you have a good grid, a good strong starting base, decent cash, and 2
strong reds that will rock the cash from day 2 on. That frees up the initial PPTer to also
grid, which will be very important early in the game.

You'll end up, eh, maybe 3m spare cash first day. 2nd day, 1900 turns (1000 minus, say 50
for moving around) running SDT at 17k per turn (which is quite feasible, actually you can
do more on better ports). 1900*17000=32m. With really good ports, 19k per, that's 36m.

Now lets break this down in terms of probing. Probe by depth, 20k, 3500 DEs will cover
approx 70% of the universe. Lets say it takes 5000 probes to do that. 5000 * 3000
(standard probe cost is 3k per) and that's 15m. It's possible then, on day 2, to probe
70% of the universe. If you then probe deep tunnels too, you can get 90% of the uni
for maybe 7000 probes. Still, do-able.

Now is it financially feasible? Yes. You can now isolate the enemy grid and intelligently
counter grid. Taking out their most likely base sectors, you stand a good chance of taking
out their starter base set. Now they will rebuild, of course, but if you can do that on day
2 or day 3, you can delay their development. This means their grid defense suffers, their
cashing suffers, and if it's an MBBS game, they're stuck doing SDT while you're doing
megas. Given the turn efficiency difference, probing like that pays for itself in no time.

Now some edits don't work like this. If planets are mobile quickly, (3 days or less), then
it really doesn't matter. Likewise, if the enemy is clever they'll grid around dock several
layers deep and force ppl to remote probe. Add in cheap 1tpw ships, and grid grows too
fast to isolate like this.

But in slow games, games where planets take a week or two, this is THE way to win
the game. Isolate the enemy, take out their stuff, marginalize their cashing, their grid,
and then hammer them till they surrender. This is why people run those aggressive
grid defense scripts... it forces the other team to slow down.

Different games then need different probe costs. Right now, in an MBBS game, you can
only go up to like 5k. If you made this an option, say up to 65k, it would be much easier
for sysops to configure this option to fit the edit and style of play they want.
Author:  The Mad Hatter [ Sun Jun 06, 2010 4:24 pm ]
Post subject:  Re: Using port reports to track player activity
Singularity wrote:
John Pritchett wrote:
Probe price is configurable, but the max price is too low? Consider that changed.


5 man team, 1000 turns. 1 guy PPTs. 2nd guy busts up, goes red, makes cash. 2nd guy
busts up, builds original set of bases this way, makes a lot more cash. 1 guy builds, next
guy grids. Now you have a good grid, a good strong starting base, decent cash, and 2
strong reds that will rock the cash from day 2 on. That frees up the initial PPTer to also
grid, which will be very important early in the game.

You'll end up, eh, maybe 3m spare cash first day. 2nd day, 1900 turns (1000 minus, say 50
for moving around) running SDT at 17k per turn (which is quite feasible, actually you can
do more on better ports). 1900*17000=32m. With really good ports, 19k per, that's 36m.

Now lets break this down in terms of probing. Probe by depth, 20k, 3500 DEs will cover
approx 70% of the universe. Lets say it takes 5000 probes to do that. 5000 * 3000
(standard probe cost is 3k per) and that's 15m. It's possible then, on day 2, to probe
70% of the universe. If you then probe deep tunnels too, you can get 90% of the uni
for maybe 7000 probes. Still, do-able.


Note that I made a huge cut from Singularity's post.

I had a game where Ether Probes were priced at $30,000 a piece (I used a hex editor, which isn't something I'd recommend to anyone other than a hacker like myself). This totally changed game play. Where it was reasonable to use all of the techniques that Singularity mentioned, the cost of the probes made it uneconomical to do so (I'd also raised ship prices dramatically, to the point where it was impossible to buy anything with lots of holds easily). The result was a game where it was actually cheaper to explore. Now this was 5 years ago, and I'm going by memory, but I think that the E-Probe price is stored as a signed integer, and that was as high as I could go (2 byte storage). If it had have been an unsigned long integer, then the price could have been jumped to $100,000 dollars or higher.

Let's look at the numbers:
a) 5 person corp - earning $50,000,000 per day could buy 16,666 eprobes in one day at $3,000.
b) 5 person corp - earning $50,000,000 per day could buy 1,666 eprobes in one day at $30,000.
c) 5 person corp - earning $50,000,000 per day could buy 500 eprobes in one day at $100,000.

As we can see, as the price goes up, the ability to eprobe for exploration becomes more limited. This gives less experienced players a chance to hide for a bit. Oh, they are still going to get their buts kicked, but it won't happen on day 2...

Wayne
Author:  Singularity [ Sun Jun 06, 2010 5:12 pm ]
Post subject:  Re: Using port reports to track player activity
The Mad Hatter wrote:
The result was a game where it was actually cheaper to explore. Now this was 5 years ago, and I'm going by memory, but I think that the E-Probe price is stored as a signed integer, and that was as high as I could go (2 byte storage). If it had have been an unsigned long integer, then the price could have been jumped to $100,000 dollars or higher.


Nod, which is exactly why this would be a useful change.

I figured it was an unsigned word (16bit), since other items can go up to 65k in value. But
yeh, make it a 32 bit value, signed or not, and you have all the control you'll ever need.

The Mad Hatter wrote:
As we can see, as the price goes up, the ability to eprobe for exploration becomes more limited. This gives less experienced players a chance to hide for a bit. Oh, they are still going to get their buts kicked, but it won't happen on day 2...


It affects experienced players too. Probe day 1, take each avoid and run a nearest unfigged
DE and unfigged tunnel search on each avoid. Take that list, grid it early before the other
team gets their map distributed. Odds are good you'll find a base, if they have one.

It's hard for any team, no matter how skilled, to get enough grid down to stop this. People
start building in low traffic 3-ways and the like, but those are easily found day 2 or 3. There
are work-arounds, but they get into the realm of tactic/counter-tactic.

Eprobes are crazy powerful for the price.
Author:  The Mad Hatter [ Sun Jun 06, 2010 5:15 pm ]
Post subject:  Re: Using port reports to track player activity
Singularity wrote:
The Mad Hatter wrote:
The result was a game where it was actually cheaper to explore. Now this was 5 years ago, and I'm going by memory, but I think that the E-Probe price is stored as a signed integer, and that was as high as I could go (2 byte storage). If it had have been an unsigned long integer, then the price could have been jumped to $100,000 dollars or higher.


Nod, which is exactly why this would be a useful change.

I figured it was an unsigned word (16bit), since other items can go up to 65k in value. But
yeh, make it a 32 bit value, signed or not, and you have all the control you'll ever need.

The Mad Hatter wrote:
As we can see, as the price goes up, the ability to eprobe for exploration becomes more limited. This gives less experienced players a chance to hide for a bit. Oh, they are still going to get their buts kicked, but it won't happen on day 2...


It affects experienced players too. Probe day 1, take each avoid and run a nearest unfigged
DE and unfigged tunnel search on each avoid. Take that list, grid it early before the other
team gets their map distributed. Odds are good you'll find a base, if they have one.

It's hard for any team, no matter how skilled, to get enough grid down to stop this. People
start building in low traffic 3-ways and the like, but those are easily found day 2 or 3. There
are work-arounds, but they get into the realm of tactic/counter-tactic.

Eprobes are crazy powerful for the price.


Amen.

Wayne
Author:  John Pritchett [ Sun Jun 06, 2010 5:37 pm ]
Post subject:  Re: Using port reports to track player activity
I'm definitely willing to provide more flexibility on setting eProbe pricing, but if at all possible I'd like to fix this problem in standard gameplay rather than provide the means for experienced gameops to fix it through price balancing.

One idea I had was making eProbes follow a simulated supply/demand curve. The more eProbes are purchased, the more expensive they become, and with less use, price comes down, with a configurable delay on the cost recovery. The idea being, rather than setting a specific price based on potential cashing, let the game set the cost based on actual cashing. If someone's making a bunch of money and wants to do some blanket eProbing, the cost of eProbes will rise quickly as they buy them up. This would give an advantage to the first person to start buying eProbes, but I don't think it would be too much of an advantage, and very quickly the cost of eProbes would rise to a level that makes it prohibitively expensive. Done correctly, this would create an automatic restriction on eProbing so that you can't just dump a ton of eProbes in the first few days.

I'm sure there are some exploits, so let me know how this would be abused, or what's wrong with the idea in general. If there's any interest in trying this out, I'll code it up and allow people to try it out.
Author:  Singularity [ Sun Jun 06, 2010 6:14 pm ]
Post subject:  Re: Using port reports to track player activity
John Pritchett wrote:
I'm sure there are some exploits, so let me know how this would be abused, or what's wrong with the idea in general. If there's any interest in trying this out, I'll code it up and allow people to try it out.


People will pre-stock ships w/ probes, but otherwise it might work. Does create a tactic
whereby people will buy and probe as early as possible to shut out the enemy corp's
probing. That will definitely disadvantage manual players, since automated players
will have a time advantage. Another option would be to make eprobes cost less as the
game ages.

Just make a way to turn it off, if you would, since there are some games where this can
be fun. Experienced, aggressive corps set traps around dock when probing starts. Lots of
mayhem ensues as people get torped, killed, etc. Eventually people start gridding out
from dock in layers to make it more turnly, and a whole run of tactics goes from there.
It can be a lot of fun. To that, it'd be nice to be able to price eprobes higher.
Author:  John Pritchett [ Mon Jun 07, 2010 11:55 am ]
Post subject:  Re: Using port reports to track player activity
It's hard to predict how a change like a floating eProbe price will effect gameplay. The best way to find out is probably to just stick it in there and give it a try. If it plays well, we can keep it, and if it doesn't, pull it back out. But yeah, make it optional either way.

If I set the standard base price of eProbes to a higher value than it currently is, what would be a good amount? 10x the current amount? That would be a base price of 30K cr and a range of price from 7,500 cr to 120,000 using the standard 1/4th to 4x range.
Author:  Maniac [ Mon Jun 07, 2010 12:29 pm ]
Post subject:  Re: Using port reports to track player activity
EProbes possible suggestions

EProbes ... Lets make them a limited supply (Sysop Configurable) just like planets and ships and/or a price increase.
Make this a Corp/Personal limit... You can only buy X amount of EProbes.
<Edit>
EProbes cost a turn each launch. (Still cheaper than moving)

We already have this... make the ship only carry 1 eprobe.
</Edit>


Course all of this should be Sysop Configurable.
Author:  the reverend [ Mon Jun 07, 2010 12:32 pm ]
Post subject:  Re: Using port reports to track player activity
along the lines of experienced sysop edits, one thing i've done in the past for eprobes is make ships only carry 1 eprobe. this forces players to spend turns to probe the universe, which gets expensive. needless to say, players don't like spending turns to eprobe, so they don't play edits like that. there is also another lever that players don't like and that is probe delay. i like setting probe delay because it gives people time to respond to probes hitting their grid. so the way i see it, there are three levers:

1. probe delay (time cost)
2. probe price (credits cost)
3. probe capacity (turns cost)

#1 could be more configurable, i.e. specify what the probe delay is in milliseconds
#2 could be increased to 65k or more

which leaves #3 - one thing you could do is have a setting for probe turn cost which would default to zero (normal) and could be any integer value. maybe as an option you could set probe turn cost to be whatever the ship movement cost is, so you would want to probe from 1 tpw ships.

all that being said, i think it should be some sort of iron triangle - i.e. you can't set them all low - if you want low delay, you have to have higher price or higher turn cost.

just an idea.
Author:  Vid Kid [ Mon Jun 07, 2010 2:42 pm ]
Post subject:  Re: Using port reports to track player activity
Rev has some kool thoughts on that ...

But as far as price increasing , I agree .. the low limit should stay at its current level
just because the way it is .. but the max price limit raise to as high as it can coded.
That way if a SysOp wanted to stop this tech. he could put it to 1 mil each
If it could be that high.

just my 2ยข
Author:  John Pritchett [ Mon Jun 07, 2010 2:45 pm ]
Post subject:  Re: Using port reports to track player activity
If ops just wanted to remove them from the game entirely, I could have a toggle to turn them off, just like with Photons.
Page 1 of 3 All times are UTC - 5 hours
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/