Experience Irregularity?

[Ephirith]

So I'm a 53 wood elf ranger duoing with my brother's 55 dark elf enchanter in teh Burned Woods, fear-kiting things around with a charm pet and just generally massacring the zone.

Everything was going fine today until my brother started getting upset about his slow experience, at which point we began tracking it closely. We assumed it was just because I was a ranger and my penalty was brutalizing his xp gain, but everything I can find tells me our experience pools are roughly the same size, and we are both supposed to gain roughly the same amount of experience per kill (as a proportion of our pools):

-

http://i.imgur.com/JCb8gYw.jpg

-

We tracked our experience at various points. For every blue I (53 ranger) gained, my brother (55 enchanter) gained only half a blue. This happened consistently throughout our entire session today. My experience was pretty good, his was terrible.

I was under the impression that experience in a group was shared such that every member advanced at the same rate. Am I fundamentally misunderstanding the way the exp system works in this game? Are the equations in these various EXP-share calculators inaccurate?

[Jimjam]

This seems strange.

What level were the mobs you were fighting? Perhaps they were within five levels of you so his xp was being halved but yours wasn't? I thought the 'within five levels non penalty' was applied to the whole group though...

[quido]

I thought you were about to pitch a fiber supplement.

[Swish]

54+ XP does slow down a lot, I can only assume its that... or that P99 end game XP isn't truly classic.

[khanable]

I thought you were about to pitch a fiber supplement.

lol

[biggbowler]

Isn't level 55 considered a "Hell Level"? Where it takes almost twice as much exp as someone on a non-hell level.

[Kender]

Isn't level 55 considered a "Hell Level"? Where it takes almost twice as much exp as someone on a non-hell level.

it's 54 AND 59

51 also but it's hardly noticable

[Tasslehofp99]

To put things into perspective for you:

Halfway from level 1 to 60 in terms of experience = 54.5

Your brother is just experiencing the slowdown that occurs for every class 54+

[Swish]

54 onwards is just slow and gets worse til 60 :/

[blondeattk]

folk who exp in a low zem zone deserve slow exp rates <mr spock>

[fastboy21]

I thought you were about to pitch a fiber supplement.

kekeke

[Swish]

folk who exp in a low zem zone deserve slow exp rates <mr spock>

naw, folks wanting to split the MM pond camp 3 ways deserve no XP :p

[Kristian]

54+ XP does slow down a lot, I can only assume its that... or that P99 end game XP isn't truly classic.

To go from 55 to 56 for an enchanter (10% penalty) requires 36,925,240 xp, ending with 366,025,000 total xp.

To go from 53 to 54 as a ranger (40% penalty) requires 36,394,820 xp, ending with 354,327,260 total xp.

This is with the 50-60 level multipliers taken into account. Given the numbers and the idea that xp distribution is based off the total xp of the players involved, it seems they should both be sharing xp and leveling at approximately the same rate (~3% difference). Can anyone figure out why the ranger would be getting 2x the xp the enchanter is getting?

[Ephirith]

So today we tried testing it again in skyfire.

Our experience is most definitely not moving at the same rate:

http://i.imgur.com/BBJeNMd.jpg

Why is this? I thought as far as the equations for experience distribution are concerned, we are both the same level.

[Victorio]

If the ratio: (xp required to level)/(total current xp) of each character is the same, you should be advancing at the same rate on your experience bar. I don't know why your two characters aren't. That being said, I've seen similar irregularities before.

[Daywolf]

Why is this? I thought as far as the equations for experience distribution are concerned, we are both the same level.
:rolleyes: they are probably using the same mathematical formula on that as used for drui's today with dot and root resistances. Oh and forget trying to keep cammo up for more than two minutes today lol. Fuzzy maths, changes by the day me thinks :p

[Schwing]

So hilarious.

[Dirtnap]

Ignoring the tranny porn...

Exp has worked like this on this server since launch. The calculations for how experience is split up are either wrong, or there is something wrong in the code.

[Tecmos Deception]

I thought as far as the equations for experience distribution are concerned, we are both the same level.

Did you consider mobs going light blue and green for the 55 before they do so for the 53? IIRC, once a mob is a green that is worth no XP to the highest member of a party, no one in the party will get XP for that mob. But when a mob is a green that gives xp (or a light blue, which of course gives xp), how much XP you get depends a LOT on your level... it's something like, the first level a mob is light blue it still gives half the XP it gave as a dark blue, but from there the XP for that same mob is drastically less and less down to nothing.

Also, it looks like the numbers in that screenshot of yours are of the xp to leave levels 53 and 55. If the 53 rang is 53.01 and the 55 ench is 55.99... the split will be quite a bit less even, also. Edit - just noticed your newer screenshots, and this is pretty close to the case.

[Ciroco]

Did you consider mobs going light blue and green for the 55 before they do so for the 53?

That makes sense, although I always thought the exp penalty for killing mobs that were green to someone in the group would be applied to the entire group. If that's what's going on here, it must only apply to the person who the mobs are green to.

[Ephirith]

Did you consider mobs going light blue and green for the 55 before they do so for the 53?

Also, it looks like the numbers in that screenshot of yours are of the xp to leave levels 53 and 55. If the 53 rang is 53.01 and the 55 ench is 55.99... the split will be quite a bit less even, also.

That makes sense, although I always thought the exp penalty for killing mobs that were green to someone in the group would be applied to the entire group. If that's what's going on here, it must only apply to the person who the mobs are green to.

We suspected that as well, but we aren't sure specifically how our XP is penalized by something being LB to him. I was trying my hardest to primarily pull things that were still DB to him, and it didn't seem to help much. That's why we tried Skyfire, where everything was DB. It was still bad for him and good for me.

[Ephirith]

Also, it looks like the numbers in that screenshot of yours are of the xp to leave levels 53 and 55. If the 53 rang is 53.01 and the 55 ench is 55.99... the split will be quite a bit less even, also. Edit - just noticed your newer screenshots, and this is pretty close to the case.

Wouldn't that just split it even more in his favor? He's the one getting bad exp. I find it hard to believe that even if he was 56, his exp would be progressing less than half as fast as mine, but that's the case. I always thought it tried to make all group members XP at the same rate.

[Tecmos Deception]

I missed the skyfire comment, but yeah, the only thing I could think of was the interaction of the 55 needing significantly more XP points to get through his level than the 53, coupled with many kills giving reduced XP to the former but not the later because of light blue cons.

Otherwise, no idea.


Is the chanter running off to pull another mob and outranging XP before some mobs die?

[Messianic]

I thought you were about to pitch a fiber supplement.

[GnashingOfTeeth]

Aww man, saw the title and was scared to click, I thought it was either the tranny dick guy posting more pooping pics, or a thread about constipation, either way not a win at first glance LOL.

[Scrubosaur]

Maybe it has to do with the pet doing most of the damage during your kills. It could be bugged where the pet owner is only receiving 50% of the exp while the rest of the group (in this scenario only 1 person) receives the full unmodified amount.

[Shannacore]

Shout out to the <Rants & Flames> XP calc~

www.rantsandflames.com/xpcalc

[Whirled]

Experience just sucks after lvl 50
/thread

[Daywolf]

Experience just sucks after lvl 50
/thread
They're all hell levels. Then of course some double-hell levels to make it interesting :D

[August]

Experience is only shared truly equally (in percentage towards level) when operating on the same level modifier. Every level post 51 has a different modifier.

This is why it takes longer during your own hell level and why people level 'twice as fast as you' while you're in your own hell level.

You people really should learn how to do math


Level Multipler
1-29 1.0
30-34 1.1
35-39 1.2
40-44 1.3
45-50 1.4
51 1.5
52 1.6
53 1.7
54 1.9*
55 2.0
56 2.1
57 2.2
58 2.3
59 2.5*
60 2.6

Just because experience is equally split doesn't mean you level at the same rate.

[Ephirith]

Experience is only shared truly equally (in percentage towards level) when operating on the same level modifier. Every level post 51 has a different modifier.

This is why it takes longer during your own hell level and why people level 'twice as fast as you' while you're in your own hell level.

You people really should learn how to do math


Level Multipler
1-29 1.0
30-34 1.1
35-39 1.2
40-44 1.3
45-50 1.4
51 1.5
52 1.6
53 1.7
54 1.9*
55 2.0
56 2.1
57 2.2
58 2.3
59 2.5*
60 2.6


Ok, so you've explained the same thing that's already been pasted a few times in this thread, in other threads, and on the wiki. Maybe you can explain, maybe mathematically, why the modifiers at 55 and 53 are 2.0 and 1.7, yet the 53's experience was moving more than twice as fast. The calculator (if it is even correct) lists the 53 ranger as actually requiring more experience to hit 54, than the enchanter requires to hit 56.

How exactly does that modifier factor into experience distribution, and why are you thinking it ignores the total experience pools of both players? Are these modifiers applied to each person's experience after everything else? I'm legitimately curious, because this seems to explain what we've been experiencing.

Just because experience is equally split doesn't mean you level at the same rate.

In our case, our total experience pools are about the same size, and we need about the same amount of experience to reach the next level. What specific mechanic causes us to level at different rates?

Is the chanter running off to pull another mob and outranging XP before some mobs die?

No, he stayed at camp and I pulled things to him and they never died very far away from where he was sitting. I was the only one who missed EXP occasionally, but not during our skyfire test session that produced the screenshot.

[August]

Ok, so you've explained the same thing that's already been pasted a few times in this thread, in other threads, and on the wiki. Maybe you can explain, maybe mathematically, why the modifiers at 55 and 53 are 2.0 and 1.7, yet the 53's experience was moving more than twice as fast. The calculator (if it is even correct) lists the 53 ranger as actually requiring more experience to hit 54, than the enchanter requires to hit 56.

How exactly does that modifier factor into experience distribution, and why are you thinking it ignores the total experience pools of both players? Are these modifiers applied to each person's experience after everything else? I'm legitimately curious, because this seems to explain what we've been experiencing.

Because you asked so nicely, I will explain to the best of my ability.

The experience needed to level is said to be equal to the formula of:

LEVEL^3 (that's your level, to the third power) * C * EXP_MOD

C = your race/class combo penalty/bonus. This is a constant in the equation.

EXP_MOD is equal to 1.0 for the first 30 levels of your life. In this case, you will see an inverse relationship, while in a group, to your penalties/benefits. Meaning that if you have a 40% penalty you will see an inverse to that (say 2.5X) in experience while in a group with someone with 0% penalty. Since you're both operating on the same level mod you should expect your "Percent of bar" to be the same for both parties.

Now here's where it gets interesting. At 30 the experience mod ticks up by one tenth. The important fact you have to realize is that this is applied RETROACTIVELY to the 'amount of exp i need to hit the next level' of the first 30 levels. Let's math it out:

29^3 = 24389 - you need 24,389 exp to get to level 30
30^3 = 27,000 - you need 27,000 exp to get to level 31

That means that, without the level mod in the equation it would take approximately 2,411 exp to level from 30->31

Now let's add the increased exp mod:

29^3*(1.0) = 24,389 - you need 24, 389 exp to get to level 30
30^3*(1.1) = 29,700 - you need 29,700 exp to get to level 31

This means that WITH the level mod in the equation it would take approximately 5,311 exp to get to level 31. This is what is known as a hell level.

Now, the exp mod stays the same for 31-32, so, some more math here:

31^3*(1.1) = 32,770 exp, or only 3070 exp is needed to get to 32 - less than the amount it took to get from 30-31.

The reason for this is because your experience mod did not change from 31-32.

Everytime the experience mod changes, the change is being multiplied towards your sum of experience

The above statement is why 35 is worse than 30. It's why 40 is worse than 35 and why 45 is the worst of them all pre-50. Each uptick is acting on a higher (Level^3) sum and thus causing more and more experience to be required to 'bridge the gap' as it were.

Now, on to your specific problem.

EVERY LEVEL past 50 represents a different experience modifier. In this way, every level past 50 is a hell level.

What's more, there are 'double hells' that increase that mod by .2 - meaning you're getting an even larger 'required exp' to level.

This means that almost never will you get the 'same experience as a percentage' when leveling with others in the group. You may be getting the 'same proportion' of experience - but you may have a much longer path to get your experience.

Now, I will grant you that using the expcalc it seems that you should be getting the same amount as your experience split is even and your 'exp to level' is even. It is not well understood what our exact exp mods are. There is an alternate set out there that scales all the way up to 3.0 - and it very well may be that P1999 uses those (and would explain your discrepancy way better).

However in general, you shouldn't expect to 'level at the same rate' as people in the 50s - the exp mods make each level it's own 'special hell' - remember the level 30 example where the person in level 30 takes 2x as much as the person at 29 and 31 - and it changes the rules.

[August]

Just thought I'd post the 'alternate' set of experience mods that exist out there:

51 1.5
52 1.6
53 1.7
54 1.9
55 2.1
56 2.3
57 2.5
58 2.7
59 3.0
60 3.1

[lecompte]

Fastest way to 60 for you both: Enchanter chardok AoEs the 60, then ranger gets chanter exp spot to 60 ;p

[August]

Also, the expcalc has a bug in it - which is probably why there is some confusion. It doesn't appear to be doing it's calculations quite correctly.

You can view this easily by making a

52 WOOD ELF RANGER 314,961,920 36,394,820
53 WOOD ELF RANGER 354,327,260 39,365,340
54 WOOD ELF RANGER 418,854,240 64,526,980
55 WOOD ELF RANGER 465,850,000 46,995,760
As you can see 314.9 + 36.4 = 351, not 354
354.3 + 39.3 = 393.6, not 418,854
418.8 + 64.5 = 483.3 not 465.

My guess is that they're using the wrong exp mod to calculate previous experience. This is a bug that mainly only happens in the 50s.

[Ephirith]

Appreciate the input August. We're going to do one more test without a charmed, hasted, dual-wielding pet to rule out a glitch in pet xp, but I think we'll ultimately need to conclude that our duo just isn't generating enough XP to get him through 55/56 at a bearable rate.

Fastest way to 60 for you both: Enchanter chardok AoEs the 60, then ranger gets chanter exp spot to 60 ;p

I wish! Apparently Chardok mostly only takes 60 enchanters now, though. And we like duoing... the XP was pretty good for the ench until recently.

[August]

Let's see how it would work out using the 'alternate' set of exp mods, shall we

53 Wood Elf Ranger - 40% penalty.
55 Dark Elf Enchanter - 10% Penalty

RANGER:

Total experience needed to COMPLETE 53:
53^3 * (1.4) * 1.7 = 354,327

Total experience at the START of 53:
52^3* (1.4) * 1.6 = 314,961

Total experience NEEDED TO LEVEL:
354,327 - 314,961 = 39,366

ENCHANTER:

Total experience NEEDED to Complete 55:
55^3 * (1.1) * 2.1 = 384326

Total experience at START of 55:
54*3 * (1.1) * 1.9 = 329,100

Total experience NEEDED TO LEVEL: 55,226

So in this hypothetical, Let's assume that Enchanter is at the beginning of his level, and we'll give him 335k experience. Let's also assume ranger is about mid-way through his level, so we'll give him 335k experience as well. That gives us a 50/50 split.

At 2k experience per mob, you'd each get 1k.

It would take the enchanter 55 kills to level
It would take the ranger 39 kills to level.

Let's skew it even further. Let's put the ranger at 90% in, or at 350k experience, and let's put the enchanter at 'freshly dinged' - 330k.

Each kill of 2000 would net the ranger (350 / 680) 1030 experience
Each kill of 2000 would net the enchanter (330 / 680) 970 experience

It would take the ranger 39366 / 1030 = 38.2 kills to level
It would take the enchanter 55226 / 970 = 57 kills to level

57 / 38.2 = ~1.5 - so you're leveling about 50% slower than the other party. I'd be complaining, too! (And these are fake numbers, if you're killing level 45 mobs in a ZEM of 75 you're only getting 45^2 * .75 = 1518 exp per kill split between you)

[Ephirith]

Let's see how it would work out using the 'alternate' set of exp mods, shall we

53 Wood Elf Ranger - 40% penalty.
55 Dark Elf Enchanter - 10% Penalty

RANGER:

Total experience needed to COMPLETE 53:
53^3 * (1.4) * 1.7 = 354,327

Total experience at the START of 53:
52^3* (1.4) * 1.6 = 314,961

Total experience NEEDED TO LEVEL:
354,327 - 314,961 = 39,366

ENCHANTER:

Total experience NEEDED to Complete 55:
55^3 * (1.1) * 2.1 = 384326

Total experience at START of 55:
54*3 * (1.1) * 1.9 = 329,100

Total experience NEEDED TO LEVEL: 55,226

So in this hypothetical, Let's assume that Enchanter is at the beginning of his level, and we'll give him 335k experience. Let's also assume ranger is about mid-way through his level, so we'll give him 335k experience as well. That gives us a 50/50 split.

At 2k experience per mob, you'd each get 1k.

It would take the enchanter 55 kills to level
It would take the ranger 39 kills to level.

Let's skew it even further. Let's put the ranger at 90% in, or at 350k experience, and let's put the enchanter at 'freshly dinged' - 330k.

Each kill of 2000 would net the ranger (350 / 680) 1030 experience
Each kill of 2000 would net the enchanter (330 / 680) 970 experience

It would take the ranger 39366 / 1030 = 38.2 kills to level
It would take the enchanter 55226 / 970 = 57 kills to level

Very interesting. The reality of our situation was the opposite of the skewed figure: http://i.imgur.com/BBJeNMd.jpg.

It still doesn't seem to match those numbers

[Pringles]

Edit: NM didnt see there were 4 pages, only read page 1 :)

[August]

Very interesting. The reality of our situation was the opposite of the skewed figure: http://i.imgur.com/BBJeNMd.jpg.

It still doesn't seem to match those numbers

Yes, it doesn't match exactly, and the reality of the situation is I don't believe we know exactly how the EXP formula is computed here.

What we do know is that experience required is on the scale of a 3rd degree polynomial, and experience earned is on the scale of a 2nd degree polynomial.

We know there are well-defined classes/penalties.

We know that there are modifiers that apply linearly to experience requird to leveled - these are in theory numbers like '1.6' and '2.1'. I don't know if we have confirmation with what they are specifically for this server, but if we did know that we could pinpoint the problem.

The major point I'm trying to make is that experience is not guaranteed to move you in exact percentages when you level. People think this because it is true for 1-30 and to a large extent 1-50 assuming your group is within the same experience mod group (1.0, 1.1, 1.2, 1.3, and 1.4).

If you two characters are equal in 'total experience earned' they absolutely will earn the exact same AMOUNT of experience

However, the amount of experience required to level may vary wildly based on what experience mod they reside in, and so it may appear that one is leveling much faster than the other.

[Clark]

I thought you were about to pitch a fiber supplement.

Lol

[Victorio]

Yes, it doesn't match exactly, and the reality of the situation is I don't believe we know exactly how the EXP formula is computed here.

What we do know is that experience required is on the scale of a 3rd degree polynomial, and experience earned is on the scale of a 2nd degree polynomial.

We know there are well-defined classes/penalties.

We know that there are modifiers that apply linearly to experience requird to leveled - these are in theory numbers like '1.6' and '2.1'. I don't know if we have confirmation with what they are specifically for this server, but if we did know that we could pinpoint the problem.

The major point I'm trying to make is that experience is not guaranteed to move you in exact percentages when you level. People think this because it is true for 1-30 and to a large extent 1-50 assuming your group is within the same experience mod group (1.0, 1.1, 1.2, 1.3, and 1.4).

If you two characters are equal in 'total experience earned' they absolutely will earn the exact same AMOUNT of experience

However, the amount of experience required to level may vary wildly based on what experience mod they reside in, and so it may appear that one is leveling much faster than the other.

There's obviously something going wrong with the calculation. Yeah, your "alternate" set of mods matches their leveling situation much better, but your jumps from 54 to 55 to 56 etc are all the same at 0.2 . I don't think this matches the reality.

[iruinedyourday]

this usually fixes my problems and gets everything, regulated.

http://dailystache.net/wp-content/uploads/2013/12/42271_franchise_icon0009.jpg.png

[August]

There's obviously something going wrong with the calculation. Yeah, your "alternate" set of mods matches their leveling situation much better, but your jumps from 54 to 55 to 56 etc are all the same at 0.2 . I don't think this matches the reality.

Yeah my point is that we know that exp required scales on a 3rd degree polynomial x=level. What are the actual coefficients? I don't know. It's conjecture. All that we know is the points at which they change. It could be the scale on the wiki, it could be the alternate set I posted (which I grabbed from another classic EQ site) or it could be something completely unique to P99.

Point still stands that just because your total exp earned is equal, you shouldn't expect to level at the same percentage of level per kill in the 50s. You will get the same amount of exp per kill, yes, but the amount of exp required to level will be different depending on those coefficients.

[Clark]

Edit

[Victorio]

Point still stands that just because your total exp earned is equal, you shouldn't expect to level at the same percentage of level per kill in the 50s. You will get the same amount of exp per kill, yes, but the amount of exp required to level will be different depending on those coefficients.
I don't think I've seen anyone in the thread claiming otherwise, so I'm not sure who you've been trying to convince lol. The original poster even explicitly stated that because they should be getting around the same xp per kill AND should be taking about the same xp to level that they should then be leveling at the same rate.

[August]

I don't think I've seen anyone in the thread claiming otherwise, so I'm not sure who you've been trying to convince lol. The original poster even explicitly stated that because they should be getting around the same xp per kill AND should be taking about the same xp to level that they should then be leveling at the same rate.

Yes, using data that we have no idea if it's correct. If you alter the coefficients (in my earlier examples) you see that exp required to level is vastly different and easily explains the discrepancy.

If you start with a false assumption, anything can be true.