mfing project.

[WackenOpenAir]

mfing project.

some people seem to think mephi is the place to be, other think the pit, some go for travincal and some believe only in pindle.

the truth is indeed, with the relatively small number of runs we do, it is all a simple matter of luck.

now if you want real information instead of an endless pointless discussion about where to mf, couldnt we take it on a bit more serious?

we could gather information from many players by seperating the results in mf categories for comparison.

the usefullness of the items is of course just the matter of luck. the number of set / unique / rare items not however. this should be reasonably consistent.

We can get usefull information if everyone who does mf runs reports the following:

-amount of mf used
-amount of runs done
-amount of normal items found
-amount of exceptional items found
-amount of elite items found
-amount of rare items found
-amount of set items found
-amount of unique items found

while counting the number of normal exceptional and unique items in pit runs isnt very practical, it would be nice if this is done for like 10 or 20 runs, that would give consistent information of the number of exceptional and elite items. the more people participate, the more reliable the data will be. (post results only if you do 10 or more runs in the same area)

(for myself, i am quite a poor player on hardcore ladder, i am not able to do runs on hell, only nightmare. i will now start recording my results though, a comparison nightmare - hell results can also be included)

 

[wedoz]

mf'ing is 1 percent mf gear and 99 percent luck no matter which way you look at it.

you could run the pits nonstop one day and get an insane amt of items and then the next day you won't get one unique.

you could have 600 mf and not find one thing and then take your mf gear off so you don't even have 100 mf and find the "godliest" item in the game.

 

[WackenOpenAir]

yes, but that is because it is the combination of an item being the right combination of being exceptional / elite and rare / unique etc.
while that together makes small chances and thus a great range of luck, the amount of elite items and the amount of unique items should be pretty consistent. if you can compare these, you can see where you have the best chances of being lucky.

 

[Wuhan_Clan]

If you look around the forum, you'll see threads where people simply list out the 100 or 200 runs they did and what note worthy items dropped (set or unique) each run. That isn't good enough.

Frankly, most people are either too dumb or too lazy to present ALL of their mf data in an organized fashion. If you can somehow convice a lot of people to do this, I'll gladly take all the data and make a conclusion. And the key is LOTS of data for each mf lvl.

To ammend your list you should add the most important factor of all:

- they should to 20 runs MINIMUM and record the total time taken to do those 20 runs.
- all data must be submitted through excel spreadsheet form


In fact, this project has already started:
http://forums.rpgforums.net//showthread.php?t=138659

Take a look at the data. Take a look through the excel spreadsheet. That is how detailed you need to documentation to be. The guy is going to post a 2nd report soon but done with no mf. Each individual item dropped by meph was documented for type, quality, grade.

It's difficult to explain why since all there are too many kinds of statistical things to measure that are all floating through my head. Basically, anyone who wants to participate in the survey needs to read that thread, follow the link, and examine the spreadsheets.

The thing I'm afraid of is that people will be enthusiastic but will submit data that can't be used much.


Ok, nvm, crazy ideas are running through my head. People can submit whatever data that I want. The first task is collecting as much data as possible and then decide what I can use and what I can't. I don't play D2 anymore and this will be the last thing I do D2 related (other then posting on these forums).

Having said that, if people want the ulitmate conclusion of 1.10 MFing, they'll need to organize their data carefully like in the other thread. I can use general data to make some general conclusions but won't give you the GOLDEN FORMULA. That is my goal, to find the GOLDEN FORMULA. My goal is to be able to say to someone, that by doing x number of y runs will be the fastest way to find item z.

people can start submitting data whenever they want but I'll need a few days to plan out how I want to do this task - we all know that this is going to take a lot of time

 

[WackenOpenAir]

Oh very nice, those spreadsheets.
Sounds all good to me.

One thing (probably common knowledge ) i still dont know though is the treasure class.
Is treasure class only bound to the base item?
meaning a white sacred armor would be the same TC as tyraels might?
or can the treasure class be different for items that have the same base item?

 

[Wolv]

x number of y runs will be the fastest way to find item z.

...and w/ ?% of mf

Thats a tall task, but would love to see such results.

 

[ayoko]

The strategy most likely to gain results MFing is the following:

MF efficiency = #MF runs in one hour x unique drop rate

This allows for the loss of efficiency in restarting games and balances the tradeoff of footspeed, killing speed, time spent on repairs, buying pots, and the diminishing returns of MF.

I used to clock one hour's worth of Hell Meph runs with different gear to figure out the best setup. My best in 1.09....53 Hell Meph Runs in one hour with 930 MF (no map hack, bugged, hacked items).

If you are still trying to level, there are Experience Points to consider, as this will increase the potency of your character in the future.

 

[Vizier]

I think you might be right in saying that the item types should be listed. But for a Pit run, even one run would be monumental. I find a reasonable unique-drop pattern in about 30-40 runs so I would say this should be a minimum size sample. For some other posts about MF ru results in the Pit and other places:

http://www.rpgforums.net/showthread.php?t=128081
http://www.rpgforums.net/showthread.php?t=131135
http://www.rpgforums.net/showthread.php?t=136780
http://www.rpgforums.net/showthread.php?t=136096

 

[Wuhan_Clan]

Hehe, I certainly got a little overly ambitious there.

Ok, I think I have it figured out now. Ignore everything else I've said above.

The list of things to record (from the original poster at the top) is good enough. If that information was recorded, we can use known rules from the game code to figure out how many items, % of each grade, % of each quality, drops per run.

Forget the time issue. I'll say this because everyone is going to be doing different runs at different speeds. This is going to vary from character to character and from player to player, offering too many possiblities to account for.

Now it doesn't matter that 1 Pit run will net you way more items then 1 Meph run. And 1 meph run should net more items then 1 pindle run. What is important is that we can build towards a model that says: you will find X number of elite uniques using Y mf after Z runs.

When we can calculate that, it is then up to each individual MFer to determine which area is best for him or her. This is where the time comes in to play.

Lets say that the model predicts after 100 meph runs you find 3 elite uniques and after 20 Pit runs you also find 3 elite uniques. If you can do 5 times as many meph runs in the time in takes to for you to do 1 pit run, you should do meph runs. If you can't do 5 meph runs in the time it takes for you to do 1 pit run, then you should do pit runs.

This model would be much simpler to construct: the figure items per run is constant (given enough data) but runs per unit time is not.

 

[Cujo]

One thing (probably common knowledge ) i still dont know though is the treasure class.
Is treasure class only bound to the base item?
meaning a white sacred armor would be the same TC as tyraels might?
or can the treasure class be different for items that have the same base item?

TC is bound to the base item, but the difference betweeen the Normal/Unique/Set version is Qlvl, which means that monsters has to be high enough Mlvl to match the Qlvl..

-a common example is Pindle, who can drop (TC87) sacred armors, he can also drop Templars Might, but not Tyraels Might due to it being Qlvl87, and Pindle is only Mlvl86..

 

[Vizier]

The list of things to record (from the original poster at the top) is good enough. If that information was recorded, we can use known rules from the game code to figure out how many items, % of each grade, % of each quality, drops per run.


What is important is that we can build towards a model that says: you will find X number of elite uniques using Y mf after Z runs.

The game's formulas can tell you exactly what you're after without needing any empirical evidence.

Namely, you figure out what classes of monsters are in the target areas and the varietal statistics that usually show up. Then you figure out which TCs each one can drop from, then you figure out the odds that an elite TC will be chosen when the monster drops, then the odds of a particular item in one of those TCS (if you want to go that far) and so on. I think I'm correct in believeing that this is how the old drop calculator worked, without even considering MF. You can factor MF's effects in on top of the odds you've already calculated for any given item. There are some exceptions I think when it comes to boss monsters since they give an inherent MF boost to the odds. Hope this helps. I think Alainpp66, Thrugg, and Shade could help you understand this stuff better because they seem to have near-game-code knowledge of how this all works.

 

[Nevre]

Nail on the head

Vizier has the right idea guys.

The information exists within the game code to determine what will drop when and how and with how much MF, it has to exist for the game to know what actually drops. I believe this information has been extracted from the game code. Using this you can make statistical statements about when, where and how things will drop, however, as many people have pointed out, these are just statistical statements. Luck does play a large part. (Well, the not quite absolutely random drop calculator plays that part).

Working backwards from actual drops is a waste of time because
1. The info already exists. (see above)
2. You would need far more data than you are ever reasonably going to collect to determine the formulas from the drop data.

This is assuming there are no server-side modifications of item drop calculations going on, even then number 2 would still apply.

Neat idea, but you are probably better working from the game code to make you chart of where to look for things with what and who and how.

Good luck!!

 

[Wuhan_Clan]

So why hasn't someone done this already? Or has it been done and where can I find the info? Seems to me, as long as you know how to read the mpqs and you know basic math, you can easily calculate the best place to MF per/run wise. People can then use that as a basis to see where they as individual players will be most efficient since each person MFs at different speeds depending on area.

 

[Vizier]

So why hasn't someone done this already? Or has it been done and where can I find the info?

It's scattered all around the forums in bits and pieces but usually posted in a theme of "where's the best place to find this item?" for which the drop calculator is best to answer that.

Conversely, the oringal intent here is to compare one MF place against another. Really, you should have some sort of context in mind as far as what sort of all the possible drops you would find "profitable" or could call the "best" place to MF. Many people are usually after a small subset of elite/exceptional uniques, or runes, or unique rings or some such. Within this context, it's far easier to find out what areas would be best for what you want, based on the drop calculator. From there, you coudl figure out if you need to build a specific type of character to run that area for efficiency and so forth.

 

[Dracoy]

with 600 mf and soloing baal.

Every item is either rare or unique. So far ive done like 30-40 runs only once have i seen a magical item.

Since almost everything is unique drops are very good ^_^

can baal drop everything?

In like 30 runs ive found 13 elite usable good items O___________________o

i get a good one every 1-2 runs. Its usually a bad unique. It all depends on the quality of the item. ^_^