Apr 9, 8:41 am 0 of our Mana Pools and 20 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 8:41 am Spies from Drunken Dragon Wizard VI (3:1) approached our realm, but failed to penetrate our defenses due to defense shields.
Apr 9, 8:41 am 0 of our Mana Pools and 19 of our Mana Fountains were drained by Drunken Dragon Wizard VI (3:1).
Apr 9, 8:41 am Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to defense shields.
Apr 9, 8:41 am 0 of our Mana Pools and 18 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 8:42 am Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to defense shields.
Apr 9, 8:42 am Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to defense spies.
Apr 9, 8:42 am Spies from Drunken Dragon Wizard VI (3:1) approached our realm, but failed to penetrate our defenses due to defense shields.
Apr 9, 9:06 am Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to failure to overcome defense spies.
Apr 9, 9:07 am 0 of our Mana Pools and 17 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 11:08 am Spies from utopya (3:1) approached our realm, but failed to penetrate our defenses due to failure to overcome defense spies.
Apr 9, 11:08 am Spies from utopya (3:1) approached our realm, but failed to penetrate our defenses due to failure to overcome defense spies.
Apr 9, 12:19 pm Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to failure to overcome defense spies.
Apr 9, 12:22 pm 177 of our military units were bribed by Drunken Dragon Wizard II (3:1)and have left to join their forces.
Apr 9, 12:23 pm 0 of our Mana Pools and 17 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 12:23 pm 0 of our Mana Pools and 17 of our Mana Fountains were drained by Drunken Dragon Wizard II (3:1).
Apr 9, 12:24 pm Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to defense spies.
Apr 9, 12:25 pm 0 of our Mana Pools and 16 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 12:25 pm 0 of our Mana Pools and 15 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 12:25 pm 0 of our Mana Pools and 15 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 12:25 pm Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to defense spies.
Apr 9, 12:25 pm Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to defense spies.
Apr 9, 12:28 pm Drunken Dragon Wizard IX (3:1) successfully spied on our realm.
Apr 9, 12:55 pm 0 of our Mana Pools and 14 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 12:55 pm 0 of our Mana Pools and 14 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 12:55 pm Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to defense spies.
Apr 9, 12:55 pm Spies from Drunken Dragon Wizard IV (3:1) approached our realm, but failed to penetrate our defenses due to defense spies.
Apr 9, 12:56 pm 0 of our Mana Pools and 13 of our Mana Fountains were drained, but we do not know who did it.
Apr 9, 12:56 pm Spies from Drunken Dragon Wizard IV (3:1) approached our realm, but failed to penetrate our defenses due to defense spies.
Apr 9, 12:56 pm 0 of our Mana Pools and 13 of our Mana Fountains were drained by Drunken Dragon Wizard IV (3:1).
Apr 9, 12:56 pm Spies from an unknown enemy approached our realm, but failed to penetrate our defenses due to defense spies.
Apr 9, 12:56 pm 0 of our Mana Pools and 12 of our Mana Fountains were drained, but we do not know who did it.





THIS WHOLE TIME I HAD 39% SPY SHIELDS UP.

4 missions were blocked by shields. 31 Missions could have been blocked.
4/31 = 12.9 or 13%
13% vs 39%

I dunno about you but that seems f'ing ridiculous to me. What the hell was the god damn point of keeping my shields up? Fix the freaking spy system... (This is the second war this has happened to me in already... the first war i blocked 3/19 from shields --- a grand total of 15.7 or 16%)

/signed

WELCOME TO OLW!!!

Urm ok, your complaining about the spy system yet when you start complaining your actually talking about the spy shields. the great part is that its MR and the shield system is so messed up it would give the bogeyman the willies ask bucky his magic shields state 50% but they are actually running at 1.5%

isn't it the same with spies as with spells that for every spy attempt you have a (in your case) 39% change of blocking them ...

yea there is a 39% chance to stop that 1 mission or 50% chance to stop that one spell

the shield system is very very messed up there is alot wrong with MR but nothing is being done about it so i guess we just need to live with it

we have this problem a 3 or 4 rounds already...I'm not really sure.

I dont even waste time trying to get full spy shields...I just try to get better ops ratio...this at least work a lot better for me...but also have few problems

Just get the thieves den and a 50:1 ratio like myself with 40% spy sheilds and you'll be fine... lol

yep...I had that few rounds ago...worked really well...except those players with 10k spies bribing my troops sometimes

well what really bites is the fact that when you go to rob these inactives and you know they don't have no research for anything and shouldn't have any types of spies for defense and you get the message that you failed due to defensive spies.... I mean Come on I have had the Ratio of 50% failed on inactives with no research at all.... seems alittle whacked there.. another 6 trys and only three successes..

Yes the spy system is less than perfect, unfortunatley its not a code problem but a basic design fundamentals problem, and thats my dept so the buck stops here. (I never envisaged people having 60:1+ spy ratios and the maths just dosnt work when things get that far out of whack).

Whilst I would like nothing better than to tear it down and start over Im afraid I just don't have the time for that big a project at the moment, too busy trying to pay the bills and raise kids and stuff... I have what I hope is a better system mapped out but untill we can find someone to actually implement it, or Chick or myself gets the time it isnt going to happen. I know it sucks but thats just the way of things at the moment.

One day I will do MR 2.0 and it of course will be perfect... Untill then try to remember that MR is free to play and as it has never made me a dime I can't prioritise it above paying the bills at the moment.

I believe I've made this suggestion once or twice before, but Fugs...here it comes again:

Bring in coders from the community! You *know* it can be done safe, you *know* it will be good for the game, and you *know* there are able coders out there/here!

Like us two, I just know

Btw:

Your sig can be optimized

Outland looks like a good programer for this game...maybe if u give him a chance






pssss...outlands...dont forget to suicide now!! I said what u wanted!!

Ow yeah...

KABOOM!

Yay! The issue was at least acknowledged!!!

agreed

Issue acknowledged, solution ignored. Been like that for a while.

I want to hear the EXACT REAL PROBLEM. Not just some whimpering because more than average spies got thru.

Is it just Spy Shields?

Is there any problem with other types of shields?

Is it consistently a reduced percent, or just occasionally?

Give me some hard data to work with, like the first post in this thread (which is the only one that is at all helpful).

Apparently SOMEONE hasn't EVER played MR...

No hard data, since we're in a new round, but

With 40% shields and equal ratio, the chance of the average mission coming through should be around only 10%. (50% base for equal ratio and 40% shields)

If you have more ratio you should have more chance et visa versa, also depending on what type of mission it is (Spy on realm is easy, Bribe military is hard).

So succes % formula should be something like:

succes % = (50 * (ratio self/ratio enemy)) - shields % - random % - mission %

What'd you think?

I think .... that I've looked at this code so many times that I'm tired of it, so YOU look at it and tell me if there is anything wrong with it:


$probeshieldeffect=$DefenderShieldValueFromDatabase/100;

if ... various other checks such as chaos, random chance, etc ...

elseif (rand(0,100)/100 < $probeshieldeffect)
{
$probereason="defense shields";
$proberesult=0; //failure
}
elseif ... various other checks such as personality, etc ...


If does no good to show me ONE example of where some unlucky soul exceeded the average, any more than to show me one case where someone flipped 25 heads in a row. I need to see ALL the data from ALL spy missions against one realm with shields up, and then we will see if the percent is wrong.

So if you truly still believe there is a problem, save ALL spy missions against you in an excel file for a round where your shields never drop below maximum and then let's look.

chick doesn't mention where the ratio:ratio check is performed. if it is performed before the shields, then all of a sudden possum's list is showing 4/19 "possible" blocks instead of 4/31, and so we have a percentage of about 21% instead of 13%. still not 39% as possum claims it should be, but a bit closer and certainly statistically possible.

in any case, a 4.25 hour sampling is in no way indicative of anything whatsoever with the possible exception of bad luck.

oh, btw, ever heard of Spy Invisibility?

All spy missions have an increased chance of success. There is also a greatly reduced chance of spies being identified.

gasp...

better include all-seeing eye data in your spreadsheets too.

the code is plain as day, there's nothing wrong with it.

[[well - not much anyway - i think it would be slightly more accurate if it read

elseif (rand(1,100)/100 <= $probeshieldeffect)

but i'm not certain and in any case the error is minimal in favor of success even if i am right, thus:

as coded vs. 40% shields:

0-39, or 40 values fail; 40-100, or 61 values succeed

should be vs. 40% shields:

1-40, or 40 values fail; 41-100, or 60 values succeed

maybe someone more proficient than i in coding/probability can confirm/reject this?]]

IN ANY CASE - fugs has said (more than once) that the spy system is borked due to a conceptual error he made while creating it. i know it's blasphemous to some of you children that fugs has some things to do other than make his free game run perfectly for you, but sometimes life works like that. you'll see when you graduate middle school.

ive seen ops work ridiculously well. ive seen ops get stopped ridiculously well. if i had to give my point of view through several rounds and strategies with absolutely no data to back it up, i'd say that on the whole nowhere near enough ops get through considering the percentages that should.

yawn. someone please shoot this topic and bury it so deep that even exotic cant tack some inanity onto it.

ok, ive rambled long enough, i'll crawl back under my rock for a while.

I know see the fundamental flaw in this code, and it is the same as with spells.



It's waaay to random. You should have a more *firm* chance of succes when you got a kick-ass spies ratio, spy invisibility and the opponent has no shields up.
I've modified the formula a bit after some calculating and left out Spy Invisibility, because I don't know it's exact bonus.


succes % = (100 * (ratio self/ratio enemy)) - shields % - random % - mission %

In code

for example:

Mission type is Steal Money, this is above average difficulty. So the modifier is -5%

Your ratio 26:1, his ratio 26:1, shields 20%

100 * ($self_ratio/$enemy_ratio) - $enemy_shields - $mission_type) + rand(0,50)

100 * (26/26) - 20 - 5 + 38 = 100 + 13 = 113, succes

Your ratio 46:1, his ratio 26:1, shields 30%

100 * ($self_ratio/$enemy_ratio) - $enemy_shields - $mission_type) + rand(0,50)

100 * (46/26) - 30 - 5 + 18 = 177 - 17 = 160, succes

Your ratio 26:1, his ratio 46:1, shields 10%

100 * (26/46) - 10 - 5 + 44 = 56 + 29 = 85, failure

Better?

As I've said many times, the sequence of checks is

Chaos (always successful)
Random success
Random failure
Shields (if not spy)
Personality (if applicable)
Ratio (includes invisibility and wonder effects)

The code I showed you is the Shields check, but the others are similiar with their appropriate values.

There is nothing "way too random" about this:

elseif (rand(0,100)/100 < $probeshieldeffect)

It is purely a random number computed to be between 0 and 1, compared to the shield value. If below the shields, it fails. So if you have 22% shields, the probes are blocked any time the random number generated is between 0 and 21 inclusive.

If you are lucky and you get numbers between 0 and 21 more than average, you block more than average. If unlucky, you might get 200 times between 22 and 100, and then you will post about how the shield system is broken.

and I hate my unlucky...

Meh, well I'm not really unlucky, but with the code presented it sure seems random.

How is the code random? I think the problem just lays in fugly's preconceptions about how the game would be played. He would have to rewrite the entire game to be perfect . Fixing one part wouldn't be enough as eventually everything would have to be reworked.

I like outlands' code better, seeing as it actually has something to do with your and your victim's ratio...and I think telegarn said something about WHERE the ratio code came into play about the current code??? Bah, I don't know where I was going with that...I'm soooo tired.

Anywho, Chick, I don't think people are saying that there is anything wrong with the current code per se, but that we need a NEW code. I think they're saying that the current code is fine...codewise, but it's not much to do with what you actually have vs. what your victim actually has. We need a new code to properly take into account your ratio vs. their ratio and so on...I was gonna say more, but like I already said...I'm tiiiiiiiirreddd. Bah, I'll just reread this tomorrow and hope it makes sense for now...

Haven't been in here in ages, but I can see still *nothing* (or pretty close to nothing) has changed since I left. Which is, basically, why I left in the first place...

Chick, take a second look at Outland's code. It does account more for the one single factor that should be the most important when calculating spy success/failure - ratio (which is the only real number we have for figuring out how "skilled" our spies are compared to the opponent's, since there is no spy training thingy). Ratio needs to be more important!

Another thing - are there two separate checks for random success/failure? If so, why? That's completely unnecessary...