51

(49 replies, posted in Bugs)

Celebro wrote:

@Ozy I don't really understand what's the issue with the formula tbh , Hungarian math or not the bonus is applied directly to the target.

Now that's out of the way, if I have 90% locking time bonus on my SS then a any bot with 7.35sec x 1.9 = 13.965sec that is exactly 90% suppressed, if you add T4 tuner then 13.965sec x 1.175 = 16.4sec exactly 17.5% increase in locking time. My raw data proves this is so.

My basic issue was the effective formula that was only used for SS tunings is

y = base * (1+%)^# + #*% while every other tuner is
y = base * (1+%)^#

I didn't like this inconsistency and how it overamplified the tuner effect. Problem is, I only looked at it from the input side of things without considering the output. Seeing how the end results are consistent with expectations, I should probably rest the case now as it's pretty apparent that this is intended behaviour, even though it's accomplished by rather inconsistent application of hungiarian math, lol.

52

(1,455 replies, posted in General discussion)

Maybe it's just me, but spark teleports provides a very basic form of convenience, which is something there is very little of in the game. Stuff like 8-1 hour cooldowns would basically kick us back into the stone-age in regards to mobility, which i don't think is a good idea, at least when we're talking about mobility between alpha islands.

To make life a little easier for new players, removing the extension and just giving everyone 10 points doesn't sound like a bad idea, to be honest.

53

(49 replies, posted in Bugs)

Rage Rex wrote:

Having read hundreds of posts from Celebro I have opined that he is Objective and Sincere. You Ozy, I've never noticed in the forums (and being in STC does not help your Objective Credibility on this topic).

Since Burial started this thread in Bugs and Burial is as Biased as they come for ewar I will defer to Celebro's opinion on the math.

This smells like another "nerf ewar" thread now with phony maths.

Rather than watch you guys all poke around in the dark with sticks perhaps a Dev can just post the formula, end the 'discussion' about a factual formula.

I certainly have my opinions regarding ewar, but I don't think they have any place in the bug section of the forum. As for why you've never seen me ... uhm ... i've been absent? Also, this could be funny: http://forums.perpetuum-online.com/topic/5186/rip-ozy/

Anyway, there's no phony math here. All relevant formulas have now been figured out and it's clear that there still is a potential issue (less so with the actual suppressors). Check the formulas a post above for yourself and it should be pretty clear what's going on and why it could be a bug. I clarified the stuff a bit more for easier reading too.

Tl;dr: No BS, no agenda, just weird hungarian math that may or may not be working as intended.

P.S. My main's actually running zeniths and intakts a lot, so there's no personal vendetta against ewar.
P.P.S: By the way, i'm not sure celebro and i are disagreeing on anything here, necessarily. I would appreciate his opinion re the latest and full formulas though.

54

(49 replies, posted in Bugs)

The T3 cuts one off too .. that's 16.5%. If nothing else, this thread did increase the amount of available information by uncovering that.

This means that the formula(s) for tuners above are indeed correct.

--

All in all, the suppression calculation itself could be accepted. What stands as really problematic is how the math implementation affects the tuners--while pretty much every other tuner in the game applies the bonus as you'd expect, SS tuners effectively convey twice their stated bonus per tuner. I'd like a statement of whether this is intended behaviour, ideally.

--
Problem description:

The general tuner formula is as follows
1) Y = Base * (1 + %change)^number of tuners.

Strictly speaking, the same formula is used for sensor tunings, but the caveat is that instead of base, (base + 1) is used. Like so:

2) Y = (Base + 1) * (1 + %change)^number of tuners.

An alternative way to write this is like so:

2) Y = Base * (1 + %change)^number of tuners + number of tuners * %change.

The second formula demonstrates the issue more clearly as it's apparent what tangible effect the (base + 1) has (+ number of tuners * %change). This means that, effectively, tuners do apply their bonus double on suppressors while pretty much every (most?) other item in the game simply applies the bonus from the tuners directly (= just one time). The result is that the suppressor tuner effect is significantly more potent than the bonuses other tuners convey

--
Solution:

1. Either modify the base value for tuners internally so that it's .4 and not 1.4 and then change the formula that's used to calculate the bonus to celebro's formula (see below*). This would allow the tuner formula to remain the same is it is now.

2. Specifically modify the tuner formula that's used to calculate the sensor suppressor bonus to
y = (base - 1) * (1 + %change)^number of tuners.

*
Celebro's Formula: Y = Base * (1 + Robot Bonus *.03 + SS Skill * 0.03) + Robot Bonus *.03 + SS Skill * 0.03

55

(49 replies, posted in Bugs)

Rage Rex wrote:

Despite all the maths what's the purpose of this thread? Um let me guess...

An item is not behaving as expected, mathematically speaking, and we're trying to determine what's intended here. If you're looking for agendas and stuff, corp dialogues is the right forum tongue

As for the formula .... your base formula worked, celebro.

If we want to include tuners, this seems to work somewhat:

PTR * (1+Tuner%^#tuners)+#tuners*tuner%

It's accurate for the first tuner, but then produces results that are slightly off. In absolute and rounded terms, the results are off by 1% per added tuner.

Edit: Also, annihilator .. there seems to be a lack of understanding on your part right now (even though you helped me understand what was going on with the sup initially). The problem ONLY occurs when you're transforming a base percentage into a multiplier and use that for your stacking calculation. This is not the case with either amps or demobs. Those simply apply percentages as you'd expect--i.e., base * (1-%)^#ofamps for amps. Same goes for demobs really. This is expected behavior. What isn't expected is base +- x * (1 + bonus)

56

(49 replies, posted in Bugs)

Annihilator wrote:

ozy:
weapon damage (any damage extension)
mining module amount, (any mining/harvesting amount extension)
supression (supressor, thats the topic about)
nexus (any nexus buff is % based)
sensor amps. (they have their own extension, % based)
geoscanner accuracy (not sure, but i think that one too)
etc.

I did some very brief testing and weapons and their tuners definitely don't follow this formula (i.e., the raw percentage is used). Mining modules do, however. Amps don't and geoscanners might not as well (will check later). Some of the nexuses could be using it, but the majority isn't.

Just to clarify again, this whole thread is basically about the usage of the raw percentage (40% = .4) versus the multiplier of the percentage (1.4) as base value in formulas, and not multiplication with percentage changes.

57

(49 replies, posted in Bugs)

@celebro there's no real technical difference in how the bonuses are applied. the difference comes from the base value. with AP values (and almost everything else in the game) you're operating on normal numbers, while you work with a percentage on suppressors. Now, instead of using the raw percentage and then modifying that, the percentage is transformed into a multiplier (.4 -> 1.4) before applying the buffs. This is kind of a special case, but i'm not convinced that the current values are intended.

58

(49 replies, posted in Bugs)

Annihilator wrote:

in that universal formula, there is a variable exponent that is either +1 or -1, that is -1 whenever the extension is reducing the input number, so it can never reach zero (eg. cycletimes)

Mathematically, this has pretty extreme effects when the base values are percentages. Just out of curiosity, are there any other cases where percentages are being modified like with suppressors?

59

(49 replies, posted in Bugs)

Annihilator wrote:

(1+ 0.4) * (1+0.27+0.15) = 1.988 aka lockitme of (rounded) 199% aka a locktime increase of 99%

Celebro's formula for the base locking time is correct as well though, but I'm not sure the tuner math is correct (Edit: based on quick testing, it comes pretty close but is not completely correct).

Sheet with formula: https://docs.google.com/spreadsheets/d/ … edit#gid=0

Either way, your formula does help explains that there's at least some logic behind how the value is calculated. The question now becomes whether this intended behaviour or whether it's an oversight. The somewhat exponential effect tuners have should/could be considered here as it's over-amplified due to how the math works out.

that's very non-committal answer tongue

personally, im thinking that the ewar biased fleet has the advantage in most cases due to three points:

1. ewar mechs can take a single opposing bot of any strength out of the fight in most scenarios. When vagas and zeniths are combined and communication works (hardest part about this, lol), this number can go up.

2. Ewar has both a speed, range and masking advantage over similar bots, and especially masked ewar not being retaliated against by a majority of the opposing fleet is pretty damn strong.

3. Augmenting the ewars with ictuses means that hostile bots can be rendered permanently useless/shields can be overcome.

In general, the move away from follows made ewars stronger, i think, as follows used to mitigate some of these issues. Just out of curiosity, how would you say the general equation changes when we're moving from heavies to either mechs + ewar mechs or mechs + light ewar?

--
Just for the record, i haven't really formed an opinion regarding the balance of ewar yet.

Lemon wrote:

It all comes down to being big enough to win or being fast enough to go home.

In that regard, ewar has an inherent advantage over heavy mk2s due to their higher speed--i.e., the ewar fleet should be able to flee in most scenarios that don't involve login traps.

Also, feel free to assume that both fleets only utilise mk2s or mk1s. this is primarily about a theoretical match, after all, so we can control conditions however we like. Either way, what's your verdict re advantage etc?

Lemon wrote:

I can already think of quite a few scenarios where either gang would come out on top.

Bro, is your theory crafting even over 9000?

If you're up for it, then let's go ahead and play some theorycrafting.

Location: Let's say somewhere on nova. An encounter where both meet or one is defending dana seem like reasonably common scenarios.

Size: Heavies and ewar mechs and their mk2 version.

Fits: I'm a bit out of the loop, to be honest, so feel free to determine the fits yourself. My only requests would be that you choose reasonably (meaning no dps fleet with a max range of 200m lol) and under the assumption that
neither gang knows what the other is going to bring.

Goal: I'd like to add one here--a roam encounter.

I'm curious where this is gonna go now!

Jita wrote:

A better comparison would be six kains vs three Kain's and three cameleons. Even then without ECCM your gonna have real trouble and that's how it should be. I well organised gang should be beating moar guns!

What happens when we escalate numbers a bit and poise two unbalanced gangs against each other? So, how about 25 v 25 in the following mix:

Gang 1
20 DPS
5 ewar/tackle

Gang 2
5 DPS
5 Ictuses
15 Ewar

Arga wrote:
Oriamen wrote:

The "problem" with ECM is that it's good in large engagements but even better in small engagements.

The 'Problem' is when agressors only bring DPS and they run into a smart group of players that have a balanced squad with dedicated ECM bots.

I've been out of the game for a while, so correct me if i'm wrong here, but it wouldn't it be more correct to state that a DPS bias can indeed be disadvantageous, while an EWAR bias will most likely not be (unless the opponent is either using loads of ECCMs or the DPS in question are ranged gropho mk2s).  So, it's not necessarily about balanced groups but having a sufficient amount of ewar.

65

(102 replies, posted in Recruitment forum)

Fred  Phelps wrote:

My personal experience with STC:  As I was looking for a corporation to join, I decided to join the RPS channel to see what all the rift raft was about.  Upon entering the channel I immediately saw 2 yellow names and 4 people who were in STC in white.  I then was immediately kicked from the channel and banned without even a word.  If your interested in joining STC be weary, they are narrow minded individuals who think everyone is a spy.  After that day, I decided to form the Westboro Baptist Church and continue the crusade.  Be warned.

There's no tolerance for those who glorify intolerance.

Why are NPC vs Player kills on the killboard then? They have no business being there, and the fix is to remove them, and not the kill coordinates.

Aye Pod wrote:

Its there. Just have to sift through the STC propaganda. Heat maps are cool but only without the info on specific kills.

I give up. I've read your summary of the counter arguments and the pattern analysis argument. The only conclusion i can make based on this is that I like the specific info and think that it should stay in. I still think it'd be awesome if someone could provide an actual list of arguments in favour of the removal though.

Aye Pod wrote:
Ozy wrote:

Would you mind summarising the case for the removal of specific coordinates? I didn't actually see any compelling arguments for that while skimming through the last page.

Points have been made throughout this thread.

Im mostly seeing lots of hyperbole and sarcasm, mostly by people wanting to get rid of this feature, it seems. Are you guys honestly worried about pattern analysis of the kills and heat-maps? Because i'm not seeing many arguments that are making other points ...

Aye Pod wrote:

So let me get this straight. So far I've heard:
- Its on a delay so its not that big of a deal.
- More data is better.
- EVE api has way more stuff.
- Both sides get to use it.

So far STC has not offered up one reason why the feature should be kept for the betterment of the game. All politically motivated BS.

Would you mind summarising the case for the removal of specific coordinates? I didn't actually see any compelling arguments for that while skimming through the last page.

70

(18 replies, posted in General discussion)

First loss
http://www.perp-kill.net/?m=view&id=237601

First Kill (lol)
http://www.perp-kill.net/?m=view&id=238081

71

(15 replies, posted in General discussion)

pretty sure that explosions are 2d, actually. either way, unlike chaos we didnt use any exploits to attack the base.

72

(10 replies, posted in Balancing)

missiles also ignore cover and dont use any cap.

there won’t be an official tournament this time. Last year it was really great and it provided us with a lot of experience, and we’d really love to do it again in the future. But the truth is that it takes away so much development resources and manpower that at this time we’d rather spend that on actual development.

development resources, really? to be frank, this is very disappointing. i believe the tournament is important to a lot of players, and if there's any way to make it happen ... do it. a 2 day tournament cant really take that much away from the development time, can it?

Goffer wrote:
Merkle wrote:

Kains and Mesmers are not sub par, at least as a fully maxxed out blue pilot I do not feel that way.

I do think what your seeing tho is that when you go against red you see a major bias toward them winning, and rightly so.

Kain vs Artemis = bias towards Artemis
Kain vs Tyrannos = strong bias towards Tyrannos if shielded

Kain is even required to drop a weaponslot if like to equip a drainer or neuter to overcome at least shielded castel.

For Messmer situation is other, but you don't see that much Messmer in Pvp atm.

an artemis mk2 cant kill a kain mk2 rep tank ... which is alright, but any kind of balance change should be considered carefully.

not being able to overcome shields is an entirely different issue though (one that's relevant for both reds and greens).

wat ... how's that even supposed to work? just dont fly with stuff you cant replace. and the psychological barrier ... man, if you can only pvp if you are in t4 gear, then that's a problem. pvp is supposed to be fun. so fly what you can afford.

Also, the potential for abuse is absurdly high, and the benefit is .... marginal. I.e., insured items would drop as loot and thus 'generate' nic for all involved parties.