Enchanter Charm/Cha Tests

[Splorf22]

tests:

Loraen, L60/255 cha, L53 Ilis w/tash x10: average 118 seconds
Loraen, L60/226 cha, L53 Ilis w/tash x5: average 93 seconds
Loraen, L60/226 cha, L53 Ilis shaman w/tash x 11, average 75 seconds
Loraen, L60/200 cha, L53 Ilis w/tash x10: average 139 seconds (1 outlier, without 109 seconds)
Phinger, L60/217 cha, L46 skelly w/tash x13: average 415 seconds

------


So I have been meaning to do this test for a while (255 vs 200 cha) and finally decided to stop procrastinating. I jewelcrafted up some charisma gear (platinum/electrum star ruby stuff) and swapped in my Crude Stein instead of the Insignia Protector which enabled me to get 255 cha at the cost of about 150hp/30AC. I charmed and dispelled Gruplinort's PH until it broke charm and killed me (weaponized/hasted ilis wizards are no joke). Then I charmed the commander PH until it broke charm while I was clearing NG and killed me also. The biggest problem charming these guys wasn't the charm duration per se (I was quite surprised I could charm them for 2+ minutes at a time) it was their chain resisting stuns and mezzes. So I don't recommend charming L53 stuff, despite the fantastic dps. The raw data:

255 CHA: 0-1 minutes: 3, 1-2 minutes: 3, 2-3 minutes: 2, 3-4 minutes: 3, average: 1:58 (118 seconds)

199 CHA :0-1 minutes: 4, 1-2 minutes: 2, 2-3 minutes: 1, 3-4 minutes: 3
5+ minutes: 1, Average: 2:19 (139 seconds), Average w/out the outlier: 1:49 (109 seconds)

Now I have done experiments before (basically parsed my long huge log of charms). From what I can tell, the algorithm works something like this:

int split = magic(Level, MR, CHA)
if(rand() < ~0.9)
duration = rand(0, split)
else
duration = rand(split, 15)

So with the 199 cha I was lucky enough to get 1 super charm which I did not get with the 255 cha case.

I am going to go out on a limb and guess that charm durations are linearly proportional to charisma and that charisma is softcapped at 200 like the other stats and gives a 50% return, so 255 cha is equivalent to 227.5. That should give about a +10% duration which matches nicely with experiments. But overall I would say we need more data. I'd love to see a few more people post data.

Still, on the whole I'd rather have 10% longer charm durations than 10% more HP, so I guess I'm going to have to regear a little. Stupid int/gnome builds with 75 base CHA.

full data --------------------------------------------------------------------

255

[Thu Dec 27 16:44:29 2012] You begin casting Boltran`s Agacerie. 1:11 71
[Thu Dec 27 16:45:38 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 16:45:54 2012] You begin casting Boltran`s Agacerie. 3:38 218
[Thu Dec 27 16:49:32 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 16:49:58 2012] You begin casting Boltran`s Agacerie. 1:16 76
[Thu Dec 27 16:51:14 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 16:51:37 2012] You begin casting Boltran`s Agacerie. 2:02 122
[Thu Dec 27 16:53:39 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 16:54:07 2012] You begin casting Boltran`s Agacerie. 3:50 230
[Thu Dec 27 16:57:57 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 16:58:31 2012] You begin casting Boltran`s Agacerie. 0:57 57
[Thu Dec 27 16:59:28 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 16:59:40 2012] You begin casting Boltran`s Agacerie. 0:36 36
[Thu Dec 27 17:00:16 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:01:58 2012] You begin casting Boltran`s Agacerie. 1:42 102
[Thu Dec 27 17:02:04 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:02:06 2012] You begin casting Boltran`s Agacerie. 2:59 179
[Thu Dec 27 17:05:05 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:05:30 2012] You begin casting Boltran`s Agacerie. 0:25 25
[Thu Dec 27 17:08:53 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:09:04 2012] You begin casting Boltran`s Agacerie. 3:08 188
[Thu Dec 27 17:12:12 2012] Your Boltran`s Agacerie spell has worn off.

199

[Thu Dec 27 17:12:26 2012] You begin casting Boltran`s Agacerie. 2:46 166
[Thu Dec 27 17:15:12 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:47:34 2012] You begin casting Boltran`s Agacerie. 1:22 82
[Thu Dec 27 17:48:56 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:49:30 2012] You begin casting Boltran`s Agacerie. 0:34 34
[Thu Dec 27 17:51:38 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:52:02 2012] You begin casting Boltran`s Agacerie. 1:18 78
[Thu Dec 27 17:53:20 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:53:37 2012] You begin casting Boltran`s Agacerie. 0:07 7
[Thu Dec 27 17:53:44 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:54:12 2012] You begin casting Boltran`s Agacerie. 0:33 33
[Thu Dec 27 17:54:45 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:54:58 2012] You begin casting Boltran`s Agacerie. 0:11 11
[Thu Dec 27 17:55:09 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 17:55:29 2012] You begin casting Boltran`s Agacerie. 7:11 431
[Thu Dec 27 18:02:40 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 18:03:01 2012] You begin casting Boltran`s Agacerie. 3:39 219
[Thu Dec 27 18:06:40 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 18:07:23 2012] You begin casting Boltran`s Agacerie. 3:48 228
[Thu Dec 27 18:11:11 2012] Your Boltran`s Agacerie spell has worn off.
[Thu Dec 27 18:11:23 2012] You begin casting Boltran`s Agacerie. 4:01 241
[Thu Dec 27 18:15:24 2012] Your Boltran`s Agacerie spell has worn off.

[Slave]

Thrilled someone has done any data on 200-255 cha; well done.

[Itap]

Great read, thanks Lor!

[Tecmos Deception]

Delicious, delicious charisma.

[koros]

While you very well might end up right, that's nowhere near the sample size to even begin to draw a conclusion. Especially considering with the outlier 199 cha actually lasted longer...

[Splorf22]

While you very well might end up right, that's nowhere near the sample size to even begin to draw a conclusion. Especially considering with the outlier 199 cha actually lasted longer...

That's why I used phrases like 'go out on a limb and guess' and 'we need more data'. I'm certainly not going to chuck my HP gear wholesale.

I'm quite confident in my outlier removal. I actually got a 7 minute charm on an ilis shaman earlier that I didn't record. I started out charming a frog in the tube room but it was boring so I switched to Grup's PH to PL (or NL, since I got poor Sojutsu killed. He cast bonds of force just charm broke and the ilis wiz ate him for lunch in under 10 seconds) the NG group.

[-Catherin-]

always like these posts as im always trying to improve, I consider what we have so far though as within the margin of error (and chance), thus will not currently convince me to rethink what im already doing :) keep it comming though if you feel like it

[Splorf22]

Well so far I am convinced enough to dump my granite bracer.

How about you post some logs yourself :)

[Ambrotos]

I told you this a year ago good sir!

[Snagglepuss]

I love reading threads like this. Awesome job Splorf!

[A1551]

Awesome splorf so glad to see this! There's a lot of charisma gear sitting in my bank that I was holding onto "just in case" I ever got around to doing something like this, and now I don't need to!

Based on the last conversation about this I've actually been backing off on charisma a bit and just keeping it in the 210-220 range in exchange for a bit of int stuff (hp still #1 tho!) but reading this, thinking to a fairly reasonable degree that those points over 200 get you an effective 27.5 increase in charisma with associated increases in charm duration (and assumedly lull crit resists, etc) I'm going to make some switches back.

-Propo Fol

[Splorf22]

Well TBH I think Koros is still right; the issue is far from settled. So we absolutely need more experiments to really nail things down. I'd also like to see data at various levels so we could get a nice idea about the tradeoffs there too.

If my guess is right, than each point of charisma over 200 is about a 0.25% increase in charm durations. So I'm definitely dumping the granite bracer for the golden cat eye bracer (1.75% increase in charm durations for 15hp) but I think a pair of opalline earrings vs electrum bs (2.5% increase in charm durations vs 70hp/50mana) is probably a tossup. The most interesting swap would be the Crude Stein vs the Insignia Protector (20AC vs 10cha/25hp). I need to get a Donal's BP cleric on to do some AC tests.

Of course, if you are duo with a cleric I think 255 cha absolutely. Their HP buffs will swamp your gear.

[Phinger]

I'd be happy to offer some data, but i have never done any logging or technical stuff with eq, so I'd need a lesson on how to. I charm/farm all the time now that I'm 60 and can jack up my cha to 230+ with relative ease.

[Slave]

Of course, if you are duo with a cleric I think 255 cha absolutely. Their HP buffs will swamp your gear.

At the same time, AC buffs are free AC and do not count toward your cap. So you want to gear for 255 CHA / Max AC, and of course toss HPs out the window in this case.

[Splorf22]

I'd be happy to offer some data, but i have never done any logging or technical stuff with eq, so I'd need a lesson on how to. I charm/farm all the time now that I'm 60 and can jack up my cha to 230+ with relative ease.

Type /log on to turn on logging. Open the resulting text file in EverQuest/Logs and search for "You begin casting Allure" and "Your Allure spell has worn off" (or the equivalent with Boltrans).

The one complicating factor is that we need to know the level of the mob (every 4 on its hit from 140 is 1 level, 144 = 51, 136 = 49, etc). Also it's probably a good idea to dispell them.

[Splorf22]

bump for more data

[Raemius]

Don't need experiments, have a read over the eqemu source code and you'll see how CHA is included in the break chance when it triggers.

[Llodd]

bump for more data

You'll be lucky.

The best person to get this data is you, which clearly you need as the sample size is tiny.

[Dagner]

good stuff loraen... surprised some of the names in this thread had enough time to afk from their farming to post in this thread!

[Splorf22]

I did read the EQEMU source code; it's why I went full int on Loraen. However the charm code here is totally different.

And yes, I would prefer 100 samples for each but that's just a bit much work for one person!

[gotrocks]

any chance of doing another ~100 vs 200 test splorf?

i read the first one and for the most part agree with you - the data points to charisma having a noteworthy affect on charms up to 200, but is still inconclusive due to lack of data. would love to see someone repeat this test, parse, an do it about 10-20 times to see what the result is. I would do it but im currently too poor to afford charisma gear to put my dark elf ench at 200 :(

plus im saving all my money for my shaman :D gotta afford torpor at 60 ;)

[Raemius]

I did read the EQEMU source code; it's why I went full int on Loraen. However the charm code here is totally different.

And yes, I would prefer 100 samples for each but that's just a bit much work for one person!

How is it different? Is there something factual you are basing this on? Based on my experience with charm vs. the src code, it feels pretty close to me. That's why I went MAX charisma except when raiding.

It could be placebo effect, but there seems to be a small but definite difference of 220 cha vs 255.

[Raemius]

Also, Rogean has been known in the past to post code snippets of certain mechanics. Perhaps he or Nilbog would be willing to just paste the mechanics portion of charisma on charm?

[Splorf22]

When I read the EQEMU source code the only effect of charisma on charm was +1 tick for every 25 or something. If you get the charm mechanics out of Kanras, please let me know :D

[A1551]

I told you this a year ago good sir!

Just sayin!

[Raemius]

When I read the EQEMU source code the only effect of charisma on charm was +1 tick for every 25 or something. If you get the charm mechanics out of Kanras, please let me know :D

Actually that's incorrect according to the eqemu code. Each tick, there is a fixed 25/100 (25%) chance of triggering a charm break check. The code then basically says each 10 points of charisma count for 1/200 reduction in charm break chance (after taking into account level and MR). The effects of Level, MR, and Charisma all add up to a total out of 200. Charisma is 25.5 (assuming 255 max CHA) points out of the total 200. Please fact check me though since its been a few months since I looked at the code.

[Splorf22]

hmm either the code has changed in two years or I read it incorrectly the first time. In either case it is obviously not being used here. A fixed chance to break per tick (for any given mob/cha/MR combo) would generate an exponential distribution, which is not the case here.

[webrunner5]

Well this thread is anything but conclusive lol. But interesting none the less.

[Raemius]

hmm either the code has changed in two years or I read it incorrectly the first time. In either case it is obviously not being used here. A fixed chance to break per tick (for any given mob/cha/MR combo) would generate an exponential distribution, which is not the case here.

As a fun exercise (you seem to like these sort of things), could you please have a look over the code again and give me your specific thoughts vs my observations? It is true that p99 may be running modified code for the charm portions, but its also quite possible a lot of it is leveraged.

The code can lead to a lot of interesting test cases. Also glad to help you do some testing in the game if you like.

[Extunarian]

... but its also quite possible a lot of it is leveraged.

I believe Haynar, Uthgaard, Kanras, and maybe even rogean, have all worked on the charm mechanics in the past, and it's my understanding that at this point they've completely replaced whats in the stock eqemu code for charm duration.

Anyone who has used charm regularly throughout the server's lifetime can tell you that there have clearly been some major overhauls to those mechanics.

[Clark]

good thread

[Trader1]

Yes charm on p99 has most definitely changed. When the server first opened through probably hate release or so charm used to last almost full duration 15 min or so. This was one of the main reason I rolled a chanter. I mean there was a very small chance of it breaking earlier then the full duration but I am talk really small chance. Like, I would not be hesitant to charm anything them afk for 10 minutes.
So there has been modifications to the charm code here most definitely.

[Phinger]

Here's some charm data from last night in HS...didn't have enough time to get a ton of data so i kept my same gear setup, 217cha and level 46 mob. Pardon the formatting, doing it from my phone. Looks pretty random to me with a constant cha.

Start Break Duration total sec

11:20:25 PM 11:28:22 PM 07:57 477

11:28:38 PM 11:47:49 PM 19:11 1151

11:48:06 PM 11:57:08 PM 09:02 542

11:56:24 PM 11:59:57 PM 03:33 213

12:01:03 AM 12:03:15 AM 02:12 132

12:03:35 AM 12:04:58 AM 01:23 83

12:05:04 AM 12:11:11 AM 06:07 367

12:11:34 AM 12:14:41 AM 03:07 187

12:16:07 AM 12:18:58 AM 02:51 171

12:25:00 AM 12:29:38 AM 04:38 278

12:39:27 AM 12:44:18 AM 04:51 291

12:45:00 AM 12:50:52 AM 05:52 352

12:51:22 AM 1:10:31 AM 19:09 1149

total 5393 seconds

total 89.88 minutes

average 415 seconds

average 6.914103 minutes

[Splorf22]

L53 Ilis Wizard, 226 cha: average 93 seconds

[Mon Jan 14 21:19:06 2013] You begin casting Boltran`s Agacerie. 111s
[Mon Jan 14 21:20:57 2013] Your Boltran`s Agacerie spell has worn off.
[Mon Jan 14 21:21:15 2013] You begin casting Boltran`s Agacerie. 193s
[Mon Jan 14 21:24:28 2013] Your Boltran`s Agacerie spell has worn off.
[Mon Jan 14 21:24:53 2013] You begin casting Boltran`s Agacerie. 48s
[Mon Jan 14 21:25:41 2013] Your Boltran`s Agacerie spell has worn off.
[Mon Jan 14 21:25:54 2013] You begin casting Boltran`s Agacerie. 65s
[Mon Jan 14 21:26:59 2013] Your Boltran`s Agacerie spell has worn off.
[Mon Jan 14 21:27:23 2013] You begin casting Boltran`s Agacerie. 42s
[Mon Jan 14 21:28:05 2013] Your Boltran`s Agacerie spell has worn off.

[Splorf22]

Also thanks Phinger, I'm collecting results at the top of the thread. Hopefully more people will chip in.

[Splorf22]

Froglok Ilis Shaman, 226cha, tash, 11 reps, 75 seconds (I wonder if I had some kind of MR?)

[Tue Jan 15 21:30:47 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:31:59 2013] Your Boltran`s Agacerie spell has worn off. 72
[Tue Jan 15 21:32:29 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:36:36 2013] Your Boltran`s Agacerie spell has worn off. 253
[Tue Jan 15 21:37:03 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:37:18 2013] Your Boltran`s Agacerie spell has worn off. 15
[Tue Jan 15 21:37:29 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:38:12 2013] Your Boltran`s Agacerie spell has worn off. 43
[Tue Jan 15 21:38:33 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:39:06 2013] Your Boltran`s Agacerie spell has worn off. 33
[Tue Jan 15 21:39:20 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:39:48 2013] Your Boltran`s Agacerie spell has worn off. 28
[Tue Jan 15 21:40:00 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:42:55 2013] Your Boltran`s Agacerie spell has worn off. 175
[Tue Jan 15 21:43:12 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:44:25 2013] Your Boltran`s Agacerie spell has worn off. 73
[Tue Jan 15 21:45:20 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:46:37 2013] Your Boltran`s Agacerie spell has worn off. 77
[Tue Jan 15 21:46:56 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:47:25 2013] Your Boltran`s Agacerie spell has worn off. 29
[Tue Jan 15 21:48:03 2013] You begin casting Boltran`s Agacerie.
[Tue Jan 15 21:48:31 2013] Your Boltran`s Agacerie spell has worn off. 28

[maker]

Nice info! Have an up vote!

[Calabee]

agacerie means teasage in french, that's all what your enchanters are... teasing me to make one so i can have a reliable charm -.-