Calculating a monster's 'score'

What is the best formula to calculate a 'score' based on some factors (mainly, the monster's stats:EXP ratio, but we'll have to figure how to include each stat... and perhaps the monster's spawnpoints)?

I do not understand what you mean by 'score'. Is this a fancy new feature from kMS that I missed out on, or are you trying to compare monsters numerically based on their stats?

He's trying to rate monsters based on their grindability, basically.
Which mobs are the most valuable and can be killed the quickest for the most XP in the least time.

The answer however is too highly variable, historically, based on the range of attacks available.

I think he means some sort of index like HP:EXP.

Depends what you're trying to measure for and how you're applying the result, really.

It's really just a measure of the ratio of HP:XP over Availability.
Defense may finally play into it, since that actually determines how quickly you can erode the HP, but the basic concept is still pretty simple.

Too Simple, Unfortunately - Because of how dependent it is on the shape of the map, the number of spawn and the damage output, all of which are subject to their own rules of variation, the only realistic way I've ever found to do this is by comparing XP gain in 10 minute increments over various locations. The raw numbers mean almost nothing compared to geography and mob mix.

I remember doing something like this once. I started from HP:EXP, then added some other values representing some other factors. I can't remember what exactly I did though, and I lost the memo I wrote it down on.

You could approximate the map quality by measuring total ladder distance between platforms, where teleports are not available.

You definitely need to include mobs vs. platform size measures though. A big map with only a few spawns is not as good as a small map with dense spawns.

Yeah you'd have to account for both a mobility index as well as a map size in addition to the spawn rate.

Of course, it also depends on the class. Take Typhons, for example, it's a great place for Chief Bandits, but not for Hermits.

Since it's difficult to wager actual trainability in maps, a good heuristic is "area / number of mobs". The closer this value is to 0, the better training you will receive.

Pretty interesting question. To do it right you would have to score per map not monster. Also it would differ from job to job and per lvl. How to calculate: Grind for 10~15 minutes at the map with a certain job of a certain lvl (with clean class equips nearest to lvl), maybe do more than one rotation too -> Look at % increase. Tadaa thats your score =) Now repeat for every map, job and lvl. GO!

PS: Only looking at the numbers: Just exp / HP is allready good, maybe somehow calculating their def in it but I don't know how that works. Then maybe multipy that with mobs / area like fiel aid (not the inverse though since it should have higher score if better to be a score :3). Then maybe add some multipliers for certain jobs, example ice meak = *1,5 for icemages etc.

Pretty interesting question. To do it right you would have to score per map not monster. Also it would differ from job to job and per lvl. How to calculate: Grind for 10~15 minutes at the map with a certain job of a certain lvl (with clean class equips nearest to lvl), maybe do more than one rotation too -> Look at % increase. Tadaa thats your score =) Now repeat for every map, job and lvl. GO!

PS: Only looking at the numbers: Just exp / HP is allready good, maybe somehow calculating their def in it but I don't know how that works. Then maybe multipy that with mobs / area like fiel aid (not the inverse though since it should have higher score if better to be a score :3). Then maybe add some multipliers for certain jobs, example ice meak = *1,5 for icemages etc.

That would be too tedious.

I have one way. We calculate a component score from each stat, e.g EXP/HP, EXP/PAD, EXP/PDD, etc, and the lowest score is taken as the monster's score.

Obviously, for the PAD/PDD/etc we'd have to do something more complicated than a simple EXP per PAD, since that doesn't really make sense.

Perhaps use the stats to affect the HP instead? Since higher defense effectively equates to 'higher' HP.

Like AngelSL said, all of the above would be great methods, but are time consuming and tough to do. The following formula works well for me in an excel spreadsheet that I have. It uses a value called "spawn factor" that helps the mob data you are asking for be a little more accurate. "Spawn Factor" is basically a quick estimated calculation for "time attacking mobs/time walking around a map", it estimates Fiel's above suggestion quickly and without extraction data.

Here is the formula:

"Monsters Score" = (Mob XP/(Roundup((Mob HP/Averge dmg of your main atk)/hit rate%))*Spawn Factor

Where:
hit rate % = % of time you hit a mob vs. miss a mob. (There is a well known formula for this in Dusks formula compilation)

"spawn factor" is as follows:
If hits per kill > or = 3 then "spawn factor" = .7
if hits per kill = 2 then "spawn factor" = .5
if hits per kill = 1 then "spawn factor" = .3
If mob is a boss then "spawn factor" = 1

It is not perfect but if you calculate the formula for all mobs in the game and then rank the values, it will get you the information you want. Let me know what you think. I will post the excel spreadsheet sometime soon if you are interested. It already has all of the mob data in GMS loaded so all you have to do is enter your ACC and Avg Dmg for main attack and it will spit out the mob ranks.

your spawn factor is a misnomer since it has nothing to do with the number of spawn present, just the level of effort to kill a single mob.

You don't seem to have anything to account for availability of the mob.

your spawn factor is a misnomer since it has nothing to do with the number of spawn present, just the level of effort to kill a single mob.


You don't seem to have anything to account for availability of the mob.

You are right. "spawn factor" is a terrible name for it. As for factoring in availability of the mob, that is the major weakness of my spreadsheet. For example, even though the mob "Fairy" has a very low spawn rate, they rank very high for a lower lvl character because they have a decent XP/HP ratio. If I were able to multiply the formula by the "area / number of mobs" ratio that Fiel talked abot, I believe it would increase the ranking's data integrity greatly. If I could get my hands on the data it would be easy to incorporate.

I believe with my current formula, as long as you have some general knowledge about the game you can ignore the anomalies in the ranking. I know this is not a satisfactory answer for the Southperry crowd because of all the loose ends but it’s the best I can contribute.
Also I forgot to mention before that the “Averge dmg of your main atk” portion of the formula takes into account mob def.

Or just calculate how much % you get in X place for 1min, then:

%/min x 60 = %/hour.

It would be impossible to calculate it without visiting the map, cause you need at least these factors: time spent between spawn points (ladders etc), monster hp, monster exp, your dmg, your accuracy, mob skills(stuns and stuff), skill delay, skill range etc etc...

Okay. So again it depends how specific you want to be.

If all you want to do is incorporate defense, you can just use EXP/HP * (100 - DRate). Ideally you would compute this separately for physical and magical damage, but to combine them you can do this: EXP/HP * (200 - PDRate - MDRate)

 Spoiler

I used every 20th mob to make a quick sample, higher is better:

Monster	EXP/HP * (200 - PDRate - MDRate)
Snail	36.0
Potted Sprout	28.8
Flyeye	24.6
Toy Clown	24.4
Dark Axe Stump	19.9
Mateon	18.7
Curse Eye	16.8
Cold Eye	16.1
Wild Boar (PC)	15.0
Patrol Robot S	13.2
Master Chronos	12.0
Mummydog	11.1
Black Ratz from Another Dimension	6.9
Crimson Balrog	4.1
Cronos	9.6
Fire Sentinel	9.8
Transformed Snack Bar	13.7
Play Seal	8.8
Poison Golem Level 3	5.6
Trixter	8.5
Jr. Lucida	7.8
Pink Teddy	7.5
Bloctopus	7.1
Lazy Buffy	0.0
Dark Drake	5.8
Gold Captain	7.0
Krip	6.0
Barnard Gray	6.1
Emergency Guard Robot L	5.7
Guard Robot	5.5
Tick-Tock	5.3
Giant Snowman	6.7
Octobunny	43.3
Racaroni	4.7
Athena Pierce's Clone	3.0
Griffey	12.5
Dark Sand Dwarf	4.1
Security System	4.0
Firebrand	3.9
Sage Cat	3.8
Gigantic Viking	3.7
Master Guardian	0.2
Chaos Blue Wyvern	1.0
Gatekeeper Nex	3.1
Cold Shark	3.5
Skelegon 1 in Cave	3.4
Yeti	3.3
Freezer	1.5
Maverick Type D	3.4
Coolie Zombie	3.3
Krexel	1.2
Lilynouch	2.8
Chaos Horntail's Legs	0.6
Witch Cat	15.0

Then you can multiply this by mobs/area to give a crude overall measure.

If you're allowed to use data from your own damage range, you can calculate expected hits per kill, and from there approximate time per kill (factoring in mobs/area; either more hits or fewer mobs means more time) and then calculate exp/time.

Actually, calculating such score doesn't really prove that a monster is good training wise.
Some monsters have very good ratios, but there are many factors that can turn them into complete pieces of crap.
A good HP:EXP ratio with a bad spawn / map layout turns a monster into a bad deal.
.
But if you want to consider HP, EXP, Map Layout, Spawn and such factors, it'd be the same as considering common knowledge, it'd be easy as that, you'd just use numbers on unnecessary calculations.
Some things can't really be numerically expressed or calculated. It'd be better to analyze the things on how they are in the game itself.

Here is my take on a "Monster Score" tool.

http://rapidshare.com/files/434189078/Mo...g_V0.9.xls

All you have to do is enter your ACC, LVL, and Average damage of your main attack and the sheet will tell you the top 20 regular and top 10 boss mobs that you should try for training. Is it perfect? no.... but it brings you 95% of the way there.

I am working on a post BB version of the tool which will obviously be more relevant. The post BB tool should also help people know where to train in the future. Seeing as how the new ACC penalties for higher lvl mobs will make us jump from mob to mob as we level.

All comments are welcome..... even death threats!

You'd have to quantify their spawn rate and relate it to the HP:EXP ratio.

I'm happy to say that after BB it'll become a lot more obvious what you should be on. Based on what I've seen/heard, monsters will steadily get harder as you move away from Victoria.