ability to sort by kills would be nice, not sure how the detection by compairing bases v time in game will work, a lot of players transfer money to new chars, keep up the good work Cryogenius

I was thinking that if a character has a very high rank (say above 20), but has hardly any kills, and hasn't been outside of New York, then it's pretty suspect! I don't propose that the character should be banned straight away, but if admin could be alerted, they could investigate further.

Any further suggestions on how to detect cheats this way?

Cryogenius - Author of FLstat - Freelancer Rankings for the Web.

---

I was thinking that if a character has a very high rank (say above 20), but has hardly any kills, and hasn't been outside of New York, then it's pretty suspect!

---

Nope its possible with a bit of luck, raiding the depots and some of the better wrecks can get you to 15 easily in NY.

Suggestion: As a admin maybe having a sort all player characters to together. See who is hiding in different characters.

Cryo,

FLStat already does an awesome job! They only real suggestion I have to add (other than those you've already heard from me) is instead of displaying the last time a player was online, display the amount of time SINCE the player was online. I'm constantly looking at this column and doing the math in my head and it's a pain in the ass For example, you'd display 2 days, 24 hours and 43 minutes, or 2d, 24h, 43m. Whatever floats your boat

JoeBoomz
IONCROSS

Pinger wrote:

---

its possible with a bit of luck, raiding the depots and some of the better wrecks can get you to 15 easily in NY.

---

This still takes time though, doesn't it? How long do you think it'll take to do this?

Cryogenius - Author of FLstat - Freelancer Rankings for the Web.

raid flint, 20 tonnes of cardamine so 30k straight of. Buy a patriot or rhino and raid the police depots, if lucky they have a decent cargo, another 10k or so. Hit a few Mules for cardamine/diamonds 70 tonnes a kill. So about a hour for a good pilot.

JoeBoomz wrote:

---

...instead of displaying the last time a player was online, display the amount of time SINCE the player was online... For example, you'd display 2 days, 24 hours and 43 minutes, or 2d, 24h, 43m.

---

The problem is, these times are relative to when the page was updated, and someone may view this page some time later. On the other hand, this is independant of timezone.

Alternatively, how about:
1) Use of words "Today" and "Yesterday" for logins in that last 24 and 48 hours, instead of the date.
2) Highlighting players who have logged in to the server in the last N hours.

Anybody else have a preference or comment?

PS: Item number 2) above has already been partially implemented (in FLstat v1.1). If you add the following code into your file "table.css", players who have logged in in the last hour will be highlighted in bold:
<pre><font size=1 face=Courier>
TR.active {
font-weight: bold;
} </font></pre>

You could also change the font colour, or (if you really know your stuff) the background colour.

Please try this and let me know if you think it's any good. If I get a positive response, I'll work on incorporating it properly. It works best on servers that run FLstat every hour or so.

Cryo.

Cryo,

Yeah, I see your point about delays and such.. Currently my sever updates every 30 minutes, so this wouldn't be much of a factor, however servers that do it hourly or daily might not work so well....

--JoeBoomz
IONCROSS

Okay, here's a question for you: I believe that I have developed a simple but effective cheater detection algorithm (already in place for Global Stats). So, if I were to intergrate the detection of cheaters into FLstat, should any potential cheaters be:

a) Highlighted in shame for all to see, and for admin to deal with,
b) Reported in a seperate log file, only viewable by admins,
c) Simply omitted from the ranking table, as if they never existed.

If you have any thoughts on this, please let me know.

Cryo.

a + b

I believe in name and shame, thats why it would be helpful if players could be sorted to the account number to show which characters are theres.

As using FLserver it takes to long to search for a character name, so difficult to see if someone has been trying a trainer in more than one character to hide the fact they are trying to cheat.

Edited by - Pinger on 09-08-2003 18:27:21

Just so it doesn't go without notice:

There are lot of ways to boost character level very high and very quickly, some without ever even leaving the manhattan dock once.

1: A high level character can use the command /invite CharacterName to invite the new (still docked) character, and begin doing lvl 43 missions; half of the proceeds from which will go to the new character. 3 missions, and the new character will have about 300k under his belt, all without having ever left the dock.

2: Wrecks

****3: This one is COMMON and deserves some stars, because it is used by almost every experienced player in the game imo: A high level character flies to manhattan and uses a friend to transfer large sums of cash to one of their new startup characters. I have done this personally with sums of ~20 million going to my new startup character. I also know that *alot* of other players do this quite often. Alternatively, the new character may just fly to Theta or what-have-you, and have the transfer done there.

The trick is that no one wants to work every new character up the hard way when they have 60 mil sitting on their mission character. The community has figured out many ways to avoid this problem.

I think that cheat detection routines would be of benefit, but it might be worth a disclaimer or two at the very least in the readme file. I know that 90% of the experienced player-base on old Browclops would have been eliminated in this way if Brow had simply banned new characters that obtained cash too fast. On the other hand, there might be more merit to banning new ACCOUNTS that show too much profit too fast, although that still doesn't eliminate the possibility that they actually have friends who play on the same server.

Perhaps if there was a method to simply track the overall rise and fall of an account's 'total character assets' by including all the characters in an account. Unusual gains might be matched with unusual drops manually to show where the money went, if necessary.

Gung_Ho,

The cheat detection system is based on how much money you acquire over a period of time, so getting money from a mission (as you describe in point 1) really is cheating. Finding wrecks (point 2) takes more time for less return and therefore wouldn't be flagged as a cheat.

Your last point (3) is definately considered cheating on some servers, especially between accounts. I could look at the total money/time for each account, and detect cheats that way, but I'll have to think about that some more...

Cryo.

Thank you for the clarification. It is appreciated.

Please note that although definitions of cheating may vary from server to server, none of those three are empirically considered cheats. Personally I see no reason why they should be. I believe the game allows for items and money to be traded in one direction (recieving nothing in return) by intention. That being the case, trading from gratuity is built into the very fiber of the game. And this is to say nothing about the fact that swapping items and money between characters is the only way to change your name within the game, and hence the only way to attain a clan tag. Also, I think the incidental cost of creating a new character and swapping items to him is more than high enough to justify the move of assets. Besides which, out of the hundreds of players I have talked to within the game, I have only met one who didn't use method #3, and this was to satisfy role-playing considerations.

I do however think the detection still has value. The important thing is that each server admin balance the importance of a cheat flag based on their own understanding of what causes the flag, and vs their own server rules they have posted or otherwise made the players aware of.

Edited to better clarify what i meant. One sentence could have been interpited as a minor flame, which was not my intention.

Edited by - Gung_Ho on 27-08-2003 19:45:38