A General "guide" for modding Diablo2

[Char]

Diablo2 general modding / balance "guide"

First of all, why do I make this topic? Because from what I have seen, lots of mods seem to make no sense at all. I often see things where I personally can only bump my head against the wall. This is not necessarily because the modders are idiots or something like that, but sometimes just because they have no experience and / or would need to get some tips. I want to try to give these tips here and help new modders to find a way into the world of modding Diablo2.
Furthermore lots of people calim that its insanely hard to make a balanced mod at all. I dont really think its all THAT hard, and i just wanted to give some thoughts of mine that might help.

This is seperated into several parts. I will start with more global things and after that, i will go more into detail, for example talk about items.

I do not claim to be perfect or something like this. If you see errors or parts where you don't agree with me, feel free to contact me.

I am free for new ideas and for discussions.

I just wanted to add that I do NOT dislike mods like Zy-El, although it might seem in this guide sometimes like that.... (winks to Kato). Its not MY style of mod, but I guess its good anyways. And: The last time I really played it was a year ago....


Contents

1. General
1.1 Cosmetic vs. Gameplay changes
1.2 Overpowerement Mods / Cheating
1.3 Keeping global aspects in mind
1.4 Changing things because they can be changed
1.5 Stick to the original
1.6 High vs. Low Numbers
1.7 Less is more
1.8 The two values of Diablo2
1.9 Spreadsheet Programs
1.10 Getting started / Making a plan
1.11 The size of your project
1.12 You know what you doing?
1.13 Before you ask...
1.14 Reallife and Modmaking

2. Items
2.1 Base Items
2.1.1 Weapons
2.1.2 Armory
2.2 Item Qualities
2.2.1 Magic Items
2.2.2 Rare Items
2.2.3 Unique Items
2.2.4 Set Items
2.2.5 Runewords
2.2.6 Crafted Items
2.2.7 Superior Items
2.3 Affixes
2.4 Sockets and Socketing Items
2.4.1 Gems
2.4.2 Runes
2.4.3 Jewels
2.5 Caster Equipment
2.6 Class-Specific Items
2.7 Charms
2.8 Cube Recipes
2.9 Drops

3. Characters and Skills
3.1 Active Skills
3.2 Passive Skills
3.3 Auras
3.4 Skill System
3.5 Lowlevel vs. Highlevel skills
3.6 Character Stats

4. Hirelings
4.1 Stats
4.2 Skills
4.3 Resurrection costs

5. Monster Modding
5.1 Adding Monsters / Monster Graphics
5.2 Monster Levels
5.3 Monster Stats
5.3.1 Experience Given
5.3.2 Hitpoints
5.3.3 Armor Classes
5.3.4 Attack Ratings
5.3.5 Damage
5.4 Monster Skills
5.5 Champions and Bosses
5.6 Special Bosses and Act Bosses

6. Level Modding
6.1 Act Layout / Level Distribution
6.2 Special Levels (Cow Level)

7. Making YOUR Mod

8. Miscellaneous Stuff
8.1 Huntig Bugs
8.2 Tools of the Satan

Appendix A: Formulae
Appendix B: Fileguides

[Char]

1. General

What is the sense of modding? There are two main reasons for people to come to play mods. The first is that they got bored of vanilla LOD because they simply played it for too long, the second is that they dislike vanilla LOD for some reason (too easy, to unbalanced, ...). So these are the points we have to fight with modding. We can add new things to make LOD more interesting (or leave things out to make it more interesting ), and we can enhance the difficulty, balance out things, whatever...
The basic aspect we have to keep in mind always is that the baseline is fun. If its no fun to play the mod, its pointless. A perfectly balanced but perfectly boring mod doesnt help anyone. But balance is a good point. For me, an unbalanced mod is no fun either. I dont like getting my ass kicked so hard that i have no chance at all unless i build a mainstream character with perfect items, and on the other hand it bores me to death to run around and mow enemies like a lvl 95 sorc on normal difficulty. That was, for example, the reason why i quit Zy-El shortly after reaching Nightmare difficulty. I found Zy-El's memory (20% dualleach, 100% IAS or sth like that if i recall correctly), and got my hands on boots socketed with the Zy-El runeword (freeze, 100% pierce, knockback and more i think). That was perfect for my bowazon, but i had no fun anymore. I dont want to blame Kato or Zy-El here though, and I have heard that things have changed from then.

1.1 Cosmetic vs. Gameplay changes

You should think about if your mod is mostly about cosmetic changes (coming up with new graphics, new monsters, new uniques, perhaps even new skills but leaving the gameplay more or less untouched, just spicing Diablo2 up a bit) or if it should be about gameplay changes. Of course, a major mod might want to cover both.
Cosmetic changes are the ones that seem to tie players to a mod from first sight. Everything looks so new, so fresh, so not-vanilla-D2-like. I always really appreciate it when something is different, like the starting screen (nearly a must for a larger modification), the graphics for items or if i find new, unknown uniques based on new, unknown base items.
But if you have stopped playing Diablo2 because of the gameplay, this will probably not help to satisfy you. The same old gameplay and character imbalances are still there, and the blessed aim aura is still useless. In that case, you need to start changing the gameplay, and this is what this guide is mostly about, since that is the more difficult part of modding. Okay, you realized that the vanilla gameplay is crappy, but you still don't know how to make it better. I will try to give some tips here.
Both modding types have their right to exist and you cannot say that the one is better than the other. They just serve different purposes, and often they even come together.

1.2 Overpowerement Mods / Cheating

Some words to the "god" mods. I dont see any point in making a mod that gives you powers so extremely insane that you will never die and monsters cannot even attack you because they are dead before they can attack. Modding is not just a better form of cheating. Modding is a way to enhance the game and the gameplay and to enable new, unknown things in Diablo2. This is not done by giving players "uber-leet" items. This is done by giving them the chance to defeat monsters while giving monsters the chance to defeat players.

1.3 Keeping global aspects in mind

In everything we do, we have to keep in mind its effects. For example, someone wants to make the game harder by making enemies harder to kill. There are several attempts we could take to do that. The first, and maybe most obvious, is to increase the monsters hitpoints. But this has some negative sides: A barbarian with 6000 defense, high damage and high leech will laugh a bit, take the double time to kill enemies, never loose more than 25% of his hitpoints because of enough leech and get bored because enemies just take more time to be killed now. Not exactly what we wanted to reach.
Another attempt might be to decrease the weapon damage. This is not the most taken attempt as i have seen. People seem to dislike low numbers for some reason. However, this has some effects that have to be seen, too. Leech becomes all of a sudden only half as effective (halved damage -> halved leech). Furthermore, this doesnt affect sorceresses in any way, or any other things that add damage in points rather than percentages (min damage / max damage jewels, charms, the "of evisceration" affix, lots of uniques, the +damage bash gives besides it +% damage, the paladin auras, ....). Lots of things the modmaker has to take care of.
But perhaps someone takes a totally different route and goes to give enemies 50% higher resistances. After having altered monstats.txt, adding +50% to each resistance, lots of monsters turn immune all of a sudden. After having corrected this (only 25% for a monster that has already 50% resistance, please), other strange things appear. The sorceress skill "static field" has halved its effectiveness, while skills like "conviction" and "lower resist" skyrocket in effectiveness. Leech is halved, too.

You see what is the basic problem. Everything is linked to everything else. There are no easy solutions, and if you want to do a larger mod, you have to dig very deep into ingame mechanics. If you change one thing, you have to keep hundreds of other things in mind.
This is one of the most essential things you have to have. The view for the "big total". If you change one thing, its very likely that other things get touched without you wanting it. Enhance the weapon damage and defense becomes even more useless cause you just leech everything. Increase the enemies' attack ratings and defense becomes useless because the difference between 1000 and 5000 defense is the difference between 75% of getting hit and 50% of getting hit. Increase the value of armor and you decrease the value of blocking. Increase the maximum character level and you increase the value of "per level" affixes. Decrease the block values on shields and you especially hit paladins. Increase the maximum character level while decreasing the stats per level and you hit the block chance in particular because of the way the block formula works.
If you increase all values of all skills, like for example multiplying all damages by two, you get several effects: The first one is that you reduce the value of strength as damage adder, the second one is that you reduce the power of skills that dont add damage (doubleswing for example). Furthermore, you help paladins in particular because they get damage added twice: Through their aura and through their skill (well this is at least true for the mainstream zealot build). On the other hand, reducing the damage on skills has the opposite effects. This can be nice if you want to give strength a higher value. But some skills get very very powerful if doing this, for example doubleswing. Doubleswing effectively doubles your dealt damage, while other skills might only add a small percentage.

Multiple shot gets very powerful in this system, too. This list can be enhanced as much as you like.
But it goes even further. Improving all skills (like for example setting the maximum skill level to 50) increases the differences between characters even more than they are already. It doesnt help the paladin's hitpoints, they stay the same. But the druids and barbarians get significantly more hitpoints when doing this. If you balance your monster damages to druids and barbarians now, the paladins are... screwed. If you balance the damage towards paladins, barbarians and druids will very rarely die.

This is just some reminders I can give you. I can never tell all the things that can be changed and all the side effects that arise because of this.
You can do whatever you want, but always keep in mind: Changing a number doesnt just change one number. Lots of things are linked together very much, and because of that, changing one number might change the whole gameplay.

1.4 Changing things because they can be changed

With the patch 1.10 and the new possibilities we have through code editing, there are lots of things to discover and to do. We can have lots of fun and introduce totally new gameplay features.
But what I have seen more than once is this ... lets call it actionism. People do things because they can be done, but do not think of the consequences. I have seen that in 1.09, and in 1.10 it seems to happen even more... perhaps because even more is possible now. Doing things because they can be done is odd. Doing things because you want to have the effects that has (on gameplay for example) is good.
For example, whats the point in doubling monster hitpoints as well as weapon damages? The only real effect that has is to enhance leech, and i dont really think that is what people wanted to produce. Often, this is not even thought about.
I also dont see the point in just changing all skills, if there is no real reason. Of course, I see the point in changing all skills if you make a mod like Shadow Empires: Realms of the Unseen or Dark Alliance. It is just necessary. I also see the point if you change skills as part of a major modification. But just changing them because it is possible? I personally dont see what is wrong with the current skills. Of course, some are crappy / make no sense at all. Most of them are not balanced. But, why not take the skills we have as a "good basis" and change some of them?
This is true for hundreds of other topics, too. People change things because it is possible. This might help players who got to modding because vanilla LOD was just boring. But it doesnt help gameplay, and sometimes it even hurts it.
Diablo2 modding is in a large part not about doing things, but about thinking. Thinking about what might be cool and what effects it has. On special classes, on special builds, on players / monsters / whatever in general). If you stop and hold a while to think before you start wildly editing things, your success will be greater and you will have less things to redo later.

1.5 Stick to the original

Not because of this, my advice is, especially if you are new to modding, to take vanilla LOD as a basis for modifications.
Everything in the actual system (meaning the technical system) of Diablo2 was made for the Diablo2 world. Because of this, you will get less technical problems when staying close to the original, and technical problems can be really annoying. For example, if you change the maximum level of characters to 200, you got problems with the reduced experience highlevel (80+) characters receive in 1.09 times. You had to code edit.
It's not like that anymore in 1.10, but a lot of other problems persist. For example, for changing the maximum level to 200, you need to rewrite the entire experience table because the maximum experience that can be stored is 2^32 - 1 = 4294967295. If your damage goes too high, besides the fact that your leech effectiveness goes up very much, you get problems with damage rollovers.

There are many other things that get problematic if you leave the "standard" path of Diablo2.
Of course, if you are experienced and / or you really need it for your mod, just do what you think is right regardless of possible technical problems. Technical problems are there to be solved. One thing though: You mod Diablo2 for a reason. The reason is that you played Diablo2 and enjoyed it. Other players think so as well. So Diablo2 has to have something positive. Try not to kill that with your modding.

1.6 High vs. Low Numbers

This is more than just a question of taste. Both variants, using high numbers and using low numbers, have their positive and negative sides. The positive sides about high numbers are that they seem to be more popular and that you do not run into problems like the one that your Frozen Orb dealing 4-12 damage is overpowered on hell (like I did), and displaying the same damage on different skill levels. On the other hand, low numbers are better readable, and it is nice when you can beat Hell difficulty even if you are not equipped with kickass items, but wearing kickass items doesnt make you a god either. But the most important point is not readability nor displaying different damages on different skill levels. The most important fact is the scaling between lowlevel items and highlevel items.
If your character deals 3-6 damage on level 1 and 100000-200000 damage on level 90 (in act5 hell), that means that the end damage is 33333 (approximately 2^15) times the damage that he did on start in act1 normal. To achieve this, he needs to double his damage output about every act. On normal, that might be achieveable, that would mean 12-24 damage on start of act3 and 48-96 damage on start of act5. But on hell, that would mean this:

act1 start: 3k - 6k
act2 start: 6k - 12.5 k
act3 start: 12.5k - 25k
act4 start: 25k -50k
act5 start: 50k -100k
act5 end: 100k - 200k

You see the problem? If someone doesnt find any good items in acts 2 and 3 on hell, he / she will be in act4 still dealing 6-12.5k damage, and the enemies just don't die. All non-up-to-date items become basically useless. This is of course not so relevant for casters, but another problem arises here: If you use the standard skillsystem, and your skills deal 25-50k damage at lvl 20, you can get this damage at very low levels and be totally overpowered. if you let the damage at lvl 20 be lower, you have to scale the damage extremely up to lvl 25 or 30, depending on what you can get with +skill items. This only makes +skill items more powerful, and especially makes staffmods very powerful.
What I want to tell you is: The higher the difference between low- and highlevel damage on your characters is, the more important are up-to-date items. This means that you have to constantly provide good items for your characters, good drops are very important in this case. Furthermore it means that the difference between a lucky character or a character that gets its items from higher level chars and an unlucky character is quite extreme. Depending on your difficulty settings, either only mainstream / really good builds will be able to survive or other builds will be okay, but those builds will just rock the house. It is very much harder to balance such a highnumber system.
I do not tell you not to use high numbers, actually there seem to be really good mods that use high numbers (Zy-El is the most popular probably). But if you do so, remember what I said and develop techniques to balance that. I cannot tell you what you can do there exactly except for better drops that are quite essential in that case, cause I personally use low numbers . But low numbers have their balancing problems, too. For example, all highlevel items lie quite close to each other, and so if you make one unique a bit too powerful, it will immediately be one of the best items in the game, even if it was only an exceptional item.

1.7 Less is more

Sometimes, reducing numbers or amounts (of items or skills for example) serves the goal of your modding activities more than increasing further and further. For example, what is the purpose of having 1000 different weapons? Aren't they so close to each other that they anyways make no difference at all? In Diablo1, there were about 20 skills (5 of them were actually useable), 50 monster types, and 50 different items, and was it fun? Definitely it was.
Before adding more and more, be sure that it has some purpose and does not just hang around making no difference at all.

1.8 The two values of Diablo2

Everything you do moddingwise, and everything that exists in Diablo2, balance-wise, comes down to exactly two values: The kill speed and the ability to survive. Thats it.
And its even worse than that, because increasing kill speed increases the ability to survive as well. The monsters hit you less, you leech more, ....
Furthermore, the killspeed is more or less only interesting in one area. If a character cannot deal non-physical damage (Multiple Shot Amazon for example) or has problems with lightning immune monsters (Lightning Fury Amazon), who cares? Just go to some area where no physical or lightning immunes exists (The Cow Level).

1.9 Spreadsheet Programs

Spreadsheet programs are your best friend. You can use them to quickly clone lines in the .txt files, but more importantly, they can deal with formulas.

Formulas are your best friend.

It is way easier to balance things with formulas than to balance things by hand. You can correct the small mistakes your formulas made later by hand.
If you do not own Excel, you can use things like StarOffice or OpenOffice. You will quickly learn to love formulas .

1.10 Getting started / Making a plan

Okay, now you know what your problem is (what you want to cure with modding) and you have received some general advice. So... where to start now? After heaving read the Beginner's Guide to Modmaking (MS Word / PDF), you mod some uniques yourself and then decide to make a real mod. Now, everything lies before you and you have no idea what to do.
First of all, make a plan. Yes, that is the thing that can be done without computers , a pencil and a piece of paper is probably enough. Because if you just start modding, you will have to redo certain parts over and over again. But also be sure that your plan will be just an outline, and will not stay. My initial plan was to make a nearly affix-only mod because I didnt like the fact that uniques were so strong and rares so weak, and what do I have now? It's grown beyond my wildest dreams from that days.
Now the question is where to start. That depends on your modding skills and on your personal preferences. If you are new, items are probably a good place to start because they are (compared to skill editing for example) relatively easy. If you already did some modding, character layout might be the place you should start at.
For each modding subpart, there are places where you should start if you modify them. I stated them out at least in the skills section (well, actually Brother Laz did) and in the items section.

1.11 The size of your project

If you do a large project, be sure that it will take a LOT of time. I started modding with the opinion that my mod would be ready within two or three weeks, and now I have been modding for nearly a year. Modding takes a lot of time and a lot of patience. Modding is nothing you do between brushing your teeth and going to bed. It is something beautiful and it is very rewarding, but if you are not motivated, it will lead you to nothing. There will be some really hard times during your modding. I myself was several times really close to quitting modding entirely.
For really large projects, modding teams should be considered, like there are already for several mods (like Dark Alliance, Shadow Empires and so on). But be sure that you are all modding for the same goal before you start .

1.12 You know what you doing?

No, this is not just my poor english. Look here for more information .
I often see people post things like
"well, I changed this, and I dont know what really happened, but somehow it works, now I just got one problem, ..."

STOP.
Do not go any further. If you do not know what exactly happened, you should stop, look back, see what you changed and understand. You learn the most about modding this way. If you don't know how something works, it doesn't help you. Even if it works, it will quite for sure lead to unexpected behavior and / or crashes somewhere else.

1.13 Before you ask...

Try to figure out how to solve a problem yourself before running around and asking questions. This way, you get much more knowledge about modding. For those who pioneered things in Diablo2 modding, there was noone answering their questions either, and they made it anyways. The more you do yourself, the more you will understand how Diablo2 works internally. And the more you understand that, the better you can mod, that's one thing I can guarantee you.
Even if you do not figure out a problem yourself, you quite for sure learnt something while trying to, which you can later on use in other parts of your modding. Knowing what is possible and what is not is very important for modding. And before you ask, you can still search the forums. You learn more that way, too.

1.14 Reallife and Modmaking

If your cat looks like it is starving, if your girlfriend / boyfriend starts getting on your nerves (pleeease, stop this modding crap....), if you cannot play with your Skat friends because you got to mod, if the pile of pizza boxes is higher than your monitor....

STOP.
Go out. Get some fresh air. Visit friends. Have social contacts. Get some haircut (in the last month of modding, your hair grew quite a lot...). Care about your girlfriend or boyfriend. There is a life beyond modding.

Reallife owns modmaking. It is just more important. Never forget that.

[Char]

2. Items

Items are an interesting thing to do, and furthermore, they are one of the easiest things to do. You can do entire mods with only changing items, several older 1.09 mods (or even earlier) were done this way.
Items are one of the topics too that can ruin your mod as well as make your mod really great. If everything is perfect but the items are crappy, your mod is gonna suck, so think about this topic. Remember 1.09 Buriza Do-Kyanon? This one item ruined an entire class, the bowazon. Noone had a windforce, and the Buriza was THE standard item for all bowazons, although it was no bow after all. The Cruel affix in 1.09 ruined the rare weapons. What i want to tell you: BE CAREFUL about what you do.
First, do the base items. Then, do the affixes. Then, do runes and jewels. Set and unique items as well as runewords are the last things you should do (unless you really like redoing things cause they are unbalanced now because the possible affixes on rare and magic items changed).

2.1 Base Items

Most mods out there will either leave the base items unchanged, or only add to the existing ones. If you decide to rethink items, rethink every single slot and the values the item can have. For example, you can have shields with either high defense, or high block rating, or both (well, or nothing at all). High block rating shields serve a different purpose than high defense shields, and require more dexterity (which a character using blunt weapons will usually not have).

2.1.1 Weapons

Weapons are essential because they deal most of the damage ingame and therefore kill the enemies. But i guess this is not new to you.
There are several characteristics of a weapon that influence its ingame "value":

1. The average damage dealt
2. The difference between minimum and maximum damage
3. The speed of the weapon
4. The range of the weapon (for melee weapons)
5. Twohanded / onehandedness
6. The stradd / dexadd
7. The item type (important for barbarian's masteries and skills that rely on special weapons like weapon block or poison dagger)
8. The standard bonus like blunt weapons dealing 1.5 times damage versus undead / automagics
9. The strength and dexterity requirements

The best thing is to develop a formula that takes all these things into account, weights them and calculates a weapon "value" which i will call weaponvalue from now on. Then, you set everything but the weapon damage and pre-choose a weaponvalue. Play around with the damage until the weaponvalue has the chosen value.
Keep in mind that there are two more interesting things about weapons: The rarity (how often does it drop) and the level / level requirement (which should be in a given ratio perhaps, like level requirement = 0.8 times item level). Weapons with higher level requirement should have higher weaponvalues, and more rare items should also have a (slightly, since the level the item can drop at is more important) higher weaponvalue.
You should also pre-choose a ration between onehand and twohand weapons, like onehanddamage is always half (or two third) of the damage a similar twohand weapon would deal.
Be creative here. Its quite nice to have weapons have no stradd and dexadd (the additional damage gained through strength and dexterity) at all, in my mod i use crossbows for that. On the other hand, it might be funny to have items that profit especially from high strength and dexterity values. I used axes for that. You could also have some items having special automagics (IAS, crushing blow, RIP on scepters, attacker takes damage, + to strength, -% to dexterity, - movement speed, weapons that increase damage with high vitality values, whatever).
Just so you know what is possible, i made an entire group of weapons that deal elemental damage only and profit from high energy values (they deal no physical damage, but the elemental damage increases with energy, just like physical damage increases with strength / dexterity).

Divide your items into caster and fighter weapons. Make this now so you can later make no caster attributes appear on fighter weapons and vice versa. Nothing is more annoying than attack rating and damage enhancing attributes on rare staves for a sorceress. If required, clone the stave class and make one with only caster attributes and one with fighter attributes.

2.1.2 Armory

Depending on weather you use a shield or not, there are five or six different armor pieces. The primary thing they should do, logically, is give you armor class. The problem is that in vanilla LOD the armor class is mainly done by the body armor (up to 600 base AC here, only 60 on belts). This means that nice attributes like godly are quite rediculous on items like gloves and belts. think about every armor part and what its purpose should be. If it is for giving you armor class, than let it have high enough armor class to actually make a difference. If it is for blocking (shields), than probably leave the armor class low. If it is for sure not for giving you armor class, than perhaps don't let armor enhancing attributes spawn on it (make a notice in your head or on a ToDo list for the affixes that come later).
You can be creative here, too, with automagics / autoattributes that influence the quality of your items. Make special items that always have three sockets, even if they are rare. Make items that let your character walk slower or faster (heavy armors are heavy to wear). Make items that drain your stamina. Be creative.

2.2 Item Qualities

There are 8 item qualities: normal, magic, rare, set, unique, runeworded, crafted, (tempererd). This is, modding-wise, not exactly correct since there are low and high quality as well, and runewords are normal items just socketed. But its correct for ingame use.
I guess the perfect idea would be to have all these item qualities make sense. The perfect thing would be that all item qualities are used throughout the game, even on hell difficulty. Well, maybe normal items dont have to be used on hell anymore, and if noone uses magic rings on hell difficulty, i can live with that.
But this means that all the item qualities have to be in a way equally strong at least to a certain extent. In vanilla 1.10, unique items are extremely strong, and because of this, noone seems to use rares. Well, I am not that firm with 1.10, but i am very sure that it was like that in 1.09.
There have been several discussions about item qualities and item strengths on the Phrozen Keep, and they all seem to tend to the conclusion that the more rare an item is, the better it should be.
Remember that set and unique items don't necessarily have to be better. They just have to have a purpose. For example, make you get tired less fast (stamina recovery) or protect you from fire (15% fire absorb, -15% cold absorb).

2.2.1 Magic Items

Magic items are somewhat the backbone of items. They are better than normal (white) items, but somewhat common, and, most important, they can be bought. It should probably be possible to complete the game with magic items only. They should be strong enough for that, especially the weapons, since it could always be that someone is unlucky and finds no good weapons. In that case, he / she should at least be able to buy a weapon that deals enough damage to get some further.
Magic items have higher chances of getting good mods than rares, because there are only two mods that have to be good, and its more likely that two mods are good than that six mods are good at the same time. This should be remembered when designing the items. Furthermore magic items can have better automods than rare items.

2.2.2 Rare Items

Toprares are, as already said, the most unlikely drops in Diablo2. Because of that, they should also be the best possible items. If a weapon that deals 400% enhanced damage, has 30% increased attack speed, 2 sockets, dualleach and some other nice mod is possible, that might be okay, be sure it will never drop anyways. Finding the best possible rare is an illusion.
Also realize that rare items are the real uniques. Any rare item is unique, meaning that it only exists once.

2.2.3 Unique Items

There are several ways to deal with unique items. The first, "usual" way is to make them be special items with funny names, having properties that cannot spawn on magic or rare items. If you are using this approach, you should take some time for doing the uniques. If you have half an hour modding time, and you can either do 15 bad uniques or one good unique, make one good unique.

Think about a nice name, make stats that fit that name, and make it really a unique item, meaning that there is no other item that is like it. Unique items can be something for very specialized characters (I made a ring that gives +2 to magic skills, +1 to number of necro summons, 50 mana and 10% faster cast rate). Unique items can be something that is just funny. Unique items can be something that is just very extraordinary. Do not hesitate to make special stats and properties for the unique items alone. Make uniques that have properties that look or sound really nice, but dont effectively really do anything. Make uniques that do other things than what you might think they would (Wizardspike, Dreamflange in Diablo1). Make Thundermauls that are the perfect item for an elemental druid. Make Mage Plates that are perfect for melee fighters, but suck for Mages since they got +% requirements. Make a lvl 90 short sword unique that has +2500% enhanced damage and is quite nice for lvl 80 characters. You got the idea.
Also don't forget to perhaps add negative modifiers to your items (has to be enabled in itemstatcost.txt problably). Good items should perhaps have a drawback. Make a good caster item that has -100 mana or something like that . You have definitely won a personal victory if people in your mod start using unique items not because they are the best available items, but because they are just cool in name and stats.

On the other hand (less taken approach, actually i havent seen it yet), they can be "standardized" items that are there for general purposes and serve the standard character builds, and you use them until you find a better rare item. You could for example make a unique weapon that has leech, increased attack speed, increased damage, but all of these items in moderate form, with rares being

possibly much better. Realize that the term "unique item" is a joke. Unique items are the least unique items in the game (well, maybe only runewords and nonmagic items are less "unique"). How many Windforces are there on the realms...?

If you are using this approach, the chances for getting unique items have to be really high.
To make clear what i mean with this approach, you could have unique rings in this mod, the lowest level giving +2 to all attributes, the second +5 to all, the third +8 to all and the fourth +12 to all. Same with resistances.

You can also combine the two approaches with making two unique versions of each item, and making the second ("standardized") unique type drop 10 times more likely.

Lowlevel uniques are rarely used just because they drop too late. You should think about making lowlevel uniques drop more frequently than highlevel ones (can be done in treasureclassex.txt, if you dont know how, just ask).

Do not forget to add perhaps-existing automagics on the item manually (unique items do not get the automagics automatically).

2.2.4 Set Items

Set items are quite interesting because they only reveal their true strengths if put together. And exactly here, a problem arises.
For lowlevel sets, this might be okay in singleplayer, assuming that enough sets items drop (you can increase the lowlevel set drops especially if you like just as you can for lowlevel uniques). In multiplayer though, this means that you get lowlevel sets that can be worn at level 15 and are to be completed at level 35 (since it takes time until all items of the set are dropped to you) at the level you can wear them via trading and your character is all of a sudden overpowered. But this isnt the worst problem I guess, but needs to be kept in mind.
For highlevel sets, noone wears any items of the sets if you only have one or two items probably, and highlevel sets are (and should be) hard to complete. So for most of the world except for some hardcore (multiplayer) players, highlevel sets are useless. But there is a solution: Make highlevel set items be nearly as good as good unique items. Let their built-in (blue) bonuses be really good, and make the bonuses when combined with the other items (green and gold) be quite low. This way, highlevel set items can still be used as nice single items. Highlevel sets probably dont even need green bonuses at all (you can have more blue bonuses then).
In general, you should not add too much gold and green bonuses. This way, the set can either be worn completely or not at all, which is not all that nice (at least i think). Furthermore, realize that if you do a four-item set and you add three green bonuses on all items, your full set bonus will be the last green bonus for all items as well as the full set bonus and the four-item completion bonus if that exists (quite a lot if you think about it).
Play around with the different adding functions. You can have set items that get their green bonuses depending on which other set items are equipped rather than how many. You can make set items that have more blue bonuses but no green ones.

Set items is the field you can be most creative with. I once made a half-necromancer half-paladin set. You could do a set where most items have no attributes at all, but the ring that is in the set combines all bonuses and has, if the set is fully equipped, +200% defense, +250% enhanced damage (if the set has a weapon), +10% to max life, +25 to all attributes, 10% dualleach and 60% resist all. As with uniques, you should take the time to make good sets and to actually test them (ALL of them). It's not always just the items and affixes that make a set fun to wear, but also the name, the colors, the special affixes that do nothing useful but are just fun, ...

Just as unique items, set items do not get the automagics usually, so you have to add them by hand.

2.2.5 Runewords

Runewords are in a way a kind of handmade unique items. The player has a set of runes and can then choose from the different available runewords. Almost everything I said about unique items can be said about runewords, too, so be sure to read the part about unique items, too.
Depending on your rune drop frequencies, runewords are more or less hard to aquire. Runewords have a nice bonus over unique items: They can be socketed into different item types, and therefore the player can choose between more defense and more strength requirement or less defense and less strength requirement, or between different weapon types for example. Because of this, you should be careful not to overpower them, even if they require highlevel runes.
Remember that runewords have another bonus: If the base item has automods, runewords can have the best possible automods that can usually only spawn on normal and magic, but not on rare items. This can be especially strong in class-specific items. Runewords can also be socketed in high quality items which have some built-in bonuses. Hint: You can set those built-in bonuses in qualityitems.txt (found in d2exp.mpq iirc).
Also remember that it is sometimes more difficult to get the required socketed base item than to get the runes required for the runeword, and that some items have limited amounts of sockets (Mythical swords can only have three sockets....)

Important for runewords is how many sockets they require. There are a lot of items that simply cannot have more than three sockets (Mythical Sword for example). If you make runewords, remember that you also should add some for such items.
Be careful to think about all item types that a runeword works in. If you make a runeword for all armor parts for example, the runeword might be balanced nicely in torso armors, but overpowered in helms or, if you made them socketable, gloves.

Runewords are not necessarily just something for highlevel players. If you add some runewords for lowlevel runes, too, they might be used in act2 normal already. Don't add five-socket runewords for lowlevel runes though, because the first five-socket items drop at about act3 nightmare or so.

2.2.6 Crafted Items

Crafted Items are basically rare items which you can choose some attributes on. Crafted items (like used in vanilla LOD) get some fixed attributes and 1-5 random attributes from the rare attributes pool. They cannot get six random attributes like rares can, though.
Be careful: The random and fixed attributes can stack and make things like 11% lifesteal rings in vanilla LOD. If you use this on weapons and use dmg% as fixed attribute, you can surpass the max dmg% this way, which makes these weapons extremely valueable. So just add some small (and maybe unusual) attributes. I personally think this is quite well done in vanilla LOD.
Be careful to make the ingredients for the crafting recipe not too rare. If you would need a ZOD rune for the craft, you have two possibilites: Either noone uses the recipe, or you have to make the output so powerful that it imbalances the game.

2.3 Affixes

This belongs to rare and magic items a bit, but i decided to make it an extra point. It is hard to structure all this....
The most important point here is: The most needed affixes are the ones that spawn most often. That means: Drastically increase the chances for the enhanced damage% and enhanced defense% attributes on weapons and armors respectively. No more finding five magic weapons and they all have some crap, but none of them has enhanced damage. That is simply odd. This also makes rares stronger, since it affects them twice. For a good rare weapon, you will need one attribute from the "King's" tree and one from the "Merciless" tree (if you didnt change that in your mod).
Remember what an item is meant to do. Weapons have primarily the use of dealing damage. Increased attack speed, leech, attack rating, .... this are all nice attributes, but they can also be accomplished from other items. The damage can only be dealt by the weapon. So this is and will always be the primary attribute for weapons. Unless you change things really heavily.

Another thing that is important: Do not make affixes that render your skills useless. I saw things like 100% pierce on items (remember Buriza?), which makes the entire pierce skill of the Amazon useless. It´s not a good idea to do such things.

One special attribute I want to warn you about: Leeching. Making especially lifeleech too high renders defense useless cause who cares if you get hit? You leech the damage back anyways within half a second.

2.3.1 Automagics / Staffmods

Automagics are nice. Automagics are powerful. Automagics are evil.
If you make the automagics be too high, it makes the item that spawns with them too strong. Remember: They are added on top of all the affixes the item has anyways. They are added to runewords. Furthermore, making an item have automagics makes especially this magic enchantment be very common. For example, adding fastcast on orbs as automagic makes fastcast a really common attribute that practically every sorceress gets automatically.
Staffmods are really powerful, too. You can get skills you usually would not have that way, and you can get items that have +3 to three different skills plus eventually even +3 to a certain skilltree which comes out as +6 to single skills. Very evil. Remember that when dealing with them.
It is very rare though to get an item with perfect affixes and perfect automagic or staffmods.
By the way, automagic.txt is your best friend cause it makes a lot of softcoded workarounds possible.

2.4 Sockets and Socketing Items

In vanilla LOD, every item can have one socket via the act5 quest. Magic items can get up to two sockets. This is another fact that makes magic items stronger than rares. Even if your maximum enhanced damage on rare and magic items is the same, magic items will still be able to deal higher damage because of their two sockets which you can socket damage enhancing runes into.
The problem is (or was in 1.09) that you can only socket an item once. Noone would socket a good item with the next jewel he found, but everyone would wait for this kickass 40% ed / 15% IAS jewel to socket that one. Because of that, lots of people would wait for ages and never socket their items. Perhaps it is a good idea to add an unsocketing recipe (be careful not to make one for normal items, i have heard that leads to problems since runewords are technically normal items). You can make it really simple (item + scroll of TP + scroll of identify -> unsocket) if you like, since the rune / jewel / gem / whatever will be gone anyways. You could also determine the costs based on the item quality (magic, rare, set, unique, crafted, tempered) and the item difficulty level (normal item, exceptional item, elite item).
Socketing an item makes it even stronger, so if only really good characters in really high stages can socket items, that means that you make especially those characters stronger, but not the others. Depending on what you would like your mod to be like, it might be better to make socketing possible for all characters and already quite from the beginning. That might require that you add recipes for adding sockets to items (at least for magic and rare items, perhaps not for set and unique ones, that would render the act5 quest useless).
The problem of highlevel chars profiting especially from sockets becomes stronger if you can add more than one socket to items via cube recipes, so my tip would be to stick to the maximum one socket in items. By the way: Recipes that add more sockets furthermore render the act5 quest useless.
Of course you could also make more than one socket available in all item qualities (and eventually even allow runewords in some of them), but in that case my tip would be to make the socketing recipes cost practically nothing and add cheap unsocketing recipes too. Every item has two affix parts then: The random one that it usually has, and the handmade part (the socketed), which should probably be about equally strong. Perhaps you should also make socketing item (runes, jewels, gems) drop more frequently.

You should not make socketing items too strong even if they are rare. I have seen mods with ZOD runes giving +100% mana / life in armors. Very funny. IF someone finds a ZOD rune in that mod, he / she will be immediately way stronger. Being too strong is not funny either.
As there are gems, jewels and runes in vanilla LOD, you should think of something to make them be different. If you are not careful, the difference between gems and runes will be just the frequency they drop with.

2.4.1 Gems

Gems are usually easier to find than runes. They seem in my eyes to be the perfect items to be standart socketing items, meaning that they provide the things that you usually look most for when socketing. They could be weaker than runes, but if you do not find the appropriate rune, you can use the gem instead (especially if you can unsocket it later on).
Another idea is to make them not be socketable at all, and be powerful cube recipe inputs instead.
At least make them be somewhat useful, since noone uses them in vanilla LOD. Those are just two ideas I wanted to give you.

At last, think about the cube recipe 3 gems -> higher quality gem. Is it really required? Perhaps it is better to remove it.

2.4.2 Runes

First of all, rebalance the rune drops. a ZOD rune every 82000 rune drops is not very funny. It should be actually findable.
Then, make the runes have strengths that represent their rarity in some kind. That shall not mean that you shall make ZOD runes have +1 to all attributes. A 10 times more rare rune can be perfectly balanced if its only 1.5 times stronger. People will look for it anyways. Remember the ED% / IAS jewels in 1.09? On the realms, how much more worth was a 40% ED / 15% IAS jewel than a 32% ED / 15% IAS jewel? How many times stronger was it actually? Keep that in mind.
Furthermore, make them be useful. If you feel the urge to socket each and every rune into some item because of its attributes and not because it is for a runeword, you have done a good job.
I personally made it that 5 dmg% runes make as much dmg% as the cruel affix makes as maximum and balanced all other runes accordingly. That might be a good point to start for you, too. It makes you want to socket six of them into a magic item with six-socket prefix and 40% IAS affix .

In my opinion, at least some of the runes should not be aquireable through buying. If you can buy all runes, whats the point in them dropping? At least the ... perhaps ten... highest runes should not be buyable. That means that you have to remove the upgrade recipes for them, too! Otherwise people will buy the lower runes and upgrade.

2.4.3 Jewels

Jewels are special because they have the same effects regardless of where you socket them. Because of this, IAS and enhanced damage on jewels is very problematic. People will, if they do not necessarily need the defense, use four-socket magic armors and socket ED% jewels in there. In 1.09, the ED% calculations were so that 100% ED in armors didnt really mean 100% more damage, but they worked like 100% damage from skills, meaning they only linear added with skill enhanced damage and strength / dexterity damage bonus, meaning that ED% in weapons was more effective than in armors. In 1.10, 100% enhanced damage in armors double your damage even when using skills. That means ED% in armors is stronger than in weapons. So be VERY careful with ED% on jewels and on armors in general. It is extremely powerful. Make jewels add to other things probably.

2.5 Caster Equipment

Caster do not need exactly those attributes that are usually most important: Physical damage on weapons and armor class on armor pieces. For casters, both is practically useless.
Because of this, you need to make other facts require these characters to try to get the elite versions of the caster weapons. For example, you could make +2 to sorceress skills make spawn only on elite orbs, or make elite orbs receive better automagics (more mana / life / fastcast / whatever). Be sure to make a difference between normal, exceptional and elite items here.

2.6 Class-Specific Items

Class-Specific Items are a very beloved topic. Many people seem to like to add them. But they lead to their very own problems.

First of all, it just sucks if you found that kickass unique for caster druids that would be perfect for your sorc, too, but sadly you are a sorceress. It also sucks to wade through hundreds of class specific items which you cannot use. So do not make too many of them.
Furthermore, it seems to be the idea of many people that class specific items should increase the strengths of a character class and so they make barbarian torsos that have higher defense than the ones of other classes. This is the wrong approach. Better: Make paladin armors that have higher defense than the armors of other characters. The reason is easy. The barbarian already has very high defense values through shout and iron skin. If you give him special armors that increase this defense even more, the difference to the paladin defense values increases even more, and it becomes impossible to play paladins without shields. Either barbarians get not hit any more at all (which isnt funny), or paladins get hit so often that they immediately die (they have less health usually, too).
Also, class specific items do not necessarily have to be better, but perhaps they are just different. They have less defense but an automagic that decreases the damage taken. They deal higher damages but are slower. They deal less damage but receive better bonuses from strength and dexterity. Something like that.

2.7 Charms

General rule here: The larger your inventory, the weaker the charms. If you have a 6*10 or larger inventory, you should quite for sure disable +skills on grand charms, otherwise you will get funny things like characters that run around with 20 +to lightning skills grand charms. Well, at least that would be possible. The difference between having no charms and having an inventory full of charms should be realizable, but not essential. If your characters have twice the hitpoints because of charms, something has gone wrong. Be careful not to make charms too strong, especially because most mods increase the inventory size. Charms should be something to balance characters weaknesses a bit, but not to be the backbone of the character strength. "Cannot be frozen" for example is an affix that doesnt belong into charms.
Make a small pool of basic attributes like mana, life, attack rating, strength, dexterity, resistances and so on that can spawn on charms and stick to only them. In my opinion, rare charms or socketed charms arent needed either, but thats a matter of taste I guess. But especially socketing makes charms be easily overpowered.

2.8 Cube Recipes

Some words about other cube recipes that are item-related and for some reason seem to be quite popular:
1. The "Item + something -> unique / set version of the item" recipe
This is one of the recipes I really dislike. Okay, its fun to get the unique, but where is the point then? What is a windforce worth if its really easy to get it? Better make your uniques drop more often.
This recipe is even worse if the unique items are overpowered. I made myself an old Arcaine's Valor in Ancestral Recall (for 1.09 LOD). I always wanted to posess this item, i guess we all want. It was not too hard to get it. Once I posessed it, I was quite disappointed. The item was great, but the way I got it was lame. Remember: Uncommon items are more cool . But: This recipe might be quite okay if you use the second type of uniques I described or maybe if the mod is for singleplayer use.
2. The "ZOD + something -> kickass item" recipe
Yet another recipe that i dislike. The ZOD rune is there for a purpose, which is usually to make items indestructable. It is not meant to be used in cube recipes that give you charms with +2 to all skill levels. And if the ZOD rune is common enough through buying and / or usual drops, everyone will use these kickass items, which renders other items useless. Not quite what you wanted to achieve, is it? Spending hours making nice uniques, balancing rare and magic items, doing funny cube recipes, .... and noone uses them because of the kickass ZOD crafts.

For every cube recipe, check if it improves your mod (makes it better, funnier, more unique, more balanced, ....). If not, remove the recipe.

By the way, it might be a good idea to make recipes for resistance rings and amulets. Especially in the early stages, it might help if your characters can change the ring for a fireresist ring when fighting against monsters that deal very heavy fire damage.

2.9 Drops

General advice here: The harder your mod is, the better your drops should be, especially later on. If on hell, fifth act, 95% of the items are crap anyways because of their base type, that simply sucks. They could not have dropped as well, that makes no difference. But be careful: Casters rely on lowlevel items with high attributes, simply because they cannot wear the highlevel items like Ornate Plates.
The vanilla LOD drops are hopelessly unbalanced. If you use Excel or StarOffice / OpenOffice, you can use a sheet I made to help you rebalance this chaos. You can find it here.

Furthermore, this values might help you to determine how often an item will actually drop (originally posted by pmpch)

Approximate drop counts:

Area ~200 drops
Act ~2000 drops
Difficulty ~10000 drops
Game ~30000 drops

[Char]

3. Characters and Skills

First of all, Brother Laz has made a great post about skills, be sure to read it here. He writes more funny than I do anyways .
As he has covered most of the skills topic there, I will speak about numbers and values here mostly.

In general, you should make no skill that is really useful on hell on lvl 1. You should not make any skill that is not useful at least for certain builds on maximum level either. Increasing the level of the skill should really increase its effectiveness, too. That means that you should not just increase a value that isnt important anyways with the skilllevel (Amplify Damage: curse radius), but one that is important (Amplify Damage: 3% more damage taken per level, starting at 15% ... this is just an example).
You have to be careful though: Some skills are support-skills only (Life Tap for example), and will never be main skills. These skills ought to be better on low levels, since noone will ever max them out anyways.

What has been said about the affixes and skills in the affix section works the other way round, too. Don't make skills that render certain attributes useless. If a skill has +1000% enhanced attack rating, attack rating on items is basically useless since you hit always anyways. Always-hit on skills is an attribute that should be used only very rarely, too. Why does Immolation Arrow need always-hit?

Skills that have a negative side as well are just great. For example, I personally think that Berserk is one of the best skills in vanilla LOD because you are more vulnerable when using it. The War Cry nullifies this negative part, though (not too good).

Be careful to make all your characters be viable at all stages. For example, in vanilla LOD, the paladin is very strong at the beginning, but in comparison very weak at the end.

3.1 Active Skills

view Brother Laz's post here, that covers nearly everything here.

Some further comments:
Do NOT make area-effect skills so strong that they will be constantly used against single enemies. That is not the point of area-effect skills. The more enemies a skill can damage at a time, the less damage it should deal.

3.2 Passive Skills

Be careful not to make these too strong. Passive skills are always great. Making them too strong cannot be balanced by anything else.
For example, imagine Iron Skin granting 500% enhanced defense at level 25. Now, let us compare two highlevel characters, both godly equipped, 2000 defense from items.
The one, namely the Barbarian, has 12000+ defense (maybe he uses Shout, too). The other one, namely the Druid, has 2000 Defense because he has no or he uses no skill that enhances his defense. Now you got the choice between having barbarians that never die in close combat and having druids that constantly die. Both versions are not really satisfying.
Passive skills increase the difference between characters, and exactly this difference is what makes the game on the one hand interesting, and on the other hand is your worst enemy. If the difference between characters is too large, it becomes impossible to balance the difference so that the game stays fun for all characters (and is fair at the same time). So be careful not to make the difference between the character classes (in terms of defense, hipoints, damage, ...) to extreme.

3.3 Auras

Auras are passive skills you can choose from. In vanilla LOD, noone uses defensive auras. The reason: They are defensive. Offense is more important than defense. Especially if your active auras grant always-hit and 500% enhanced damage (Fanaticism) with the slight bonus of heavily increased attack speed.
Make your offensive auras be so weak that people will start using defensive ones, too. Another option is to turn defensive auras into passives (maybe not all, but some of them). Otherwise it doesn't matter what you do on your passive auras. Noone will use them anyways.

3.4 Skill System

Kindof my favourite topic because I personally think the vanilla LOD skill system is a complete mess (better than the Diablo1 one, but still a mess). But since not everyone has the time and the knowledge to change the skillsystem like I did in my mod, and I don't want just anyone to copy my skillsystem either , here are some ideas how the current skillsystem can be used. The skillsystem is still not as much of a mess than items like the Buriza are, it doesnt ruin entire character classes, so well, it can be used.
First of all, some thoughts about maximum skill levels. The breakpoint levels for skilldamage calculations are 8, 16, 22 and 28, so having a maximum skill level of 5 probably isnt that intelligent. Having a max skill level of 99 isnt too intelligent either on the other hand, since you have to set the additional damage for levels 29-99 all to the same value. A maximum level of 30-35 (with +skills items) might be just okay.

EDIT: The modding world changes, and new options appear. With the new d2mod system we can, using a specific plugin, change these levels to what we want, so nice skillsystems using maximum levels of 50 are not a problem at all .

When designing your skills, be sure that the one who uses it will usually have it at maximum level. But it would be nice to have at least some skills (not the active ones, probably) make sense at levels between 2 and 15, too. You could do this for example with some of the passives, or maybe with the resistance auras of the paladin. You could as well add passives that increase the strengths of your active skills only. Just some thoughts... The more variants (and different builds) you allow, the better (unless they are all practically the same, as Brother Laz stated in his post).

3.5 Lowlevel vs. Highlevel skills

Is there a reason why lowlevel skills should be less useful at the end of the game? I can't seem to find one. Well, the highlevel ones can be aquired later, but if they are way stronger than your lowlevel skills, noone will use your lowlevel skills on levels higher than lvl 1, and you could as well not have done them at all.


3.6. Character Stats

Here, the same has to be said as for passive skills: Do not make the differences too large. The barbarian has 33% more life than the paladin in unmodded LOD (4 HP / vitality vs. 3 HP / vitality). You should think about reducing that bonus. Here is the point where you can generally set the strengths and the differences between your characters, so be careful when doing so. Life is more important than mana for most characters. Having low mana values can be balanced with skills that have low mana costs. Having low life values cannot be balanced.
Personally, I also found it interesting to remove the mana / life per level. But that is a matter of taste.
Make every stat have a reason and a use. For example, I have seen a lot of caster characters that did not invest in energy at all, partially because they did not need the mana, partially because of the Stone of Jordan unique ring. Wow, great, so what is energy for? I don't see the point in that. So make mana costs either high enough or think about giving energy some other positive effect. I think its even possible to make items require spellpower to be equipped.

[Char]

4. Hirelings

Hirelings are there to protect you. They were, though, never meant to be the ultimate tank that just stands there and the caster you are just has to stand there and cast (this is lame). So make your hirelings so weak that they simply die if they are misused as tanks, especially the act2 ones.
Okay, there is a difference between the different hireling types. You should make this difference exist, too. The barbarian hires are probably the ones that are best used as blockers, but deal the fewest damage. The act2 hires are the ones that deal high physical damage and have auras, but die quite quick. The bonus of act1 hires is that they deal two or three types of damage (physical and elemental fire / cold), and are ranged, the bonus of the act3 mage is that they deal elemental damages and furthermore are ranged. But they die really fast.
Of course you can make this a bit different, but what is essential is that the more damage output a hireling has, the weaker should it be in other parts (ability to be used as tank for example).
And: Your hirelings should not be stronger than your character .

4.1 Stats

It is a good idea to make not every hireling be able to wear everything just because he won't have the strength required for that. Its furthermore a good idea to make hirelings have high life amounts, but make them TAKE damage as well. This way, you will need to heal them, and will be able to heal them in case they are under heavy fire (and they wont die in two hits just because they are unlucky and got hit twice in one second although they have kickass defense and kickass block rating). In this approach, it might be a good idea to reduce the regeneration rate as well.

4.2 Skills

Auras are your worst enemy. If your hireling is just there to enhance you via its kickass Might aura (vanilla Multiple Shot Amazon), something has gone wrong. At least in my eyes. It quickly renders some skills useless too. If your hirelings can get lvl 35 Defiance auras, whats the point of a Defiance-Paladin then? Make the auras be able to cure the little weaknesses of your character, but not more. For example, if your highlevel aura gives + 45% attack rating, thats quite nice, but not kickass. It doesnt make your character kill very much faster, and it doesnt render the non-aura hirelings useless. Think about making special auras for hirelings only that are weaker than the original Paladin auras so you don't have to extreme steps between levels (lvl 30 hireling: lvl 1 Defiance aura, lvl 31 hireling: lvl 2 Defiance aura).
About other skills, think about what the hireling will be used for and give the appropriate skills. If the cold-damage based hireling will be used to kill those cold immune bosses, give it single-enemy damage skills like Ice Blast, but not unes like Blizzard.

4.3 Resurrection costs

If your hirelings die too often, you got a problem with the resurrection costs since these are hardcoded, and especially around lvl 60 are very evil (they have reached their maximum of 50.000 gold then). So you either make your hirelings die less often if cared about (make your characters run away just because their hireling could die) or you increase the general amount of gold characters usually have if your hirelings die too often. But think about this, too.

[Char]

5. Monster Modding

This is an interesting part I actually never dealt with. Well, at least I never dealt with the part you think monster modding is: Adding new monsters, changing monster skills and so on. I dealt with the other part, which deals with monster damage, attack rating, defense, speed and so on. This is monster modding, too.
Monster modding is the art of balancing the environment characters run around in to the strengths of the characters. To make the character strengths comparable, the level system helps you: Always balance the monsters to characters of the same level. This implies that different characters are about equally strong on different levels.

5.1 Adding Monsters / Monster Graphics

This is basically just to make everything look anew. Adding monsters does not change the difficulty. Adding monsters and changing monster graphics is just a cosmetic feature, because what do monsters differ in? Basically just damage, hitpoints, attack rating and defense rating. Even if they use skills they mostly just differ in skill damage and perhaps stun length. That is, of course, unless you modify monsters as heavily as in Dark Alliance for example.

5.2 Monster Levels

Make a plan: Where should a character have which level? On normal Difficulty, or at least until level 25, set the monster levels accordingly because of the experience formula. Later on, monster levels may be a bit higher than the character levels. This part is definitely the first one you should do, because with the new monlvl.txt file, you would have to do all the balance again if you changed the monster levels later on.
Bosses like the act bosses or sub-bosses like Hephasto or The Summoner may have higher levels of course, but be careful not to make them too high especially for lower levels. Because of the experience formula, you get less experience if the monster you kill is more than five levels higher than you until you are level 25.

5.3 Monster Stats

In 1.10, we have the positive effect of having both monstats.txt and monlvl.txt, which makes life way easier. Use monstats.txt to set the specific stats of each monster (more or less HP / damage / AR / experience given than the usual monster from that level) and use monlvl.txt to balance the "average" monster of each level.

5.3.1 Experience Given

After you have decided where your character should be which level and set the monster levels, modify the experience the monsters give so that your characters actually are those levels when they just walk through your mod once. That is, of course, if you want them to walk through your mod once, but I currently do not see any reason why anyone would want mod players to play each level twice or thrice.

5.3.2 Hitpoints

This is an interesting point, too. Many people seem to think that increasing the hitpoints of monsters makes the mod more difficult. Well, actually it does. You get crowded more often, you need more time to kill monsters, although some characters are hit less heavy by this than others (Tweaker Sorceress, classic-style Zealot with lots of crushing blow). But the main effect is not that it is harder, but just that it takes more time to play through the mod. This is probably not what you want. It is better to modify damage, chance to hit and so on to make your mod harder. Make your mod more dangerous to make your mod harder rather than making it take more time to finish it.
A good idea is to balance monster hitpoints to the average damage of your characters. Determine the damage dealt by your characters. Then, multiply this by the average number of hits you want a character to require to kill a monster and voila, you got the hitpoints for your standard monster.

5.3.3 Armor Classes

This is a thing that only effects characters that use attack rating, most caster-style characters are completely unaffected by this. Make your monsters hard, but not impossible to hit, but remember that the characters have different attack ratings and will always have, and even those with lower AR have to have a chance to hit monsters. And 50% chance to hit is not really a chance, it just sucks. But if you have followed my hints, your characters should have not too extremely different attack ratings, and there is a lot of space to balance even a difference in AR from 1k to 10k (make 10k be 95% chance to hit, 1k will then be 77% or so). But make the monster armor classes high enough that attack rating actually matters (like in the good old Diablo1 times...). What's the point in the whole attack rating system if you always hit anyways?

5.3.4 Attack Ratings

The other way round than armor classes....
In general, the attack ratings in vanilla LOD are way too high. About one third is enough. The reason: Because of the way the defense vs. attack rating formula works, too high attack ratings make defense useless. Having 1000 defense compared to 500 defense is the difference between 65% chance of getting hit and 50% chance of getting hit, so... who cares? Let's all ignore defense entirely. Like it was done in Diablo2 Version 1.09 on the realms from what i remember. If you divide all attack ratings by 3, the defense between 1000 and 500 defense might become the difference between 20% and 35% being hit... worth a shot, no? In Diablo1, the chance of getting hit in general was lower, and defense was the one of the main attributes fighters would look out for. If you ask me, I like that much better than having defense that is entirely useless. Not just to satisfy me, but because others think so, too, try to balance monster attack ratings so that defense is useful. If you want to make monsters more dangerous, better increase their damage and speed.
Lifeleech on characters is closely related to this as I stated in the items section, too high values of lifeleech render defense useless.

5.3.5 Damage

This is the main attribute that makes monsters Dangerous. Remember that changing this in monlvl.txt also increases the elemental damage monsters deal.
Generally, set the damage your monsters deal high enough to make uncareful characters die. The worst thing you can do though is to increase it so much that characters start suffering from one-hit kills. That just sucks and is not very motivating when playing the mod. Paladins and Asassins are the character classes that are in the highest danger of getting caught by onehit kills, so probably balance the damage towards them. If a ranged (caster) character that didnt increase vitality dies in a one-hit kill, that is not so bad btw. They are not meant to stand in close-combat and could have dodged the missile / ran away before the close-combat hit stroke them.

5.4 Monster Skills

Basically, if you give your monster skills like Meteor, Volcano or Chain Lightning, be told that this mostly only changes the graphics. The damage could be dealt using other skills as well. Well, some skills are more effective against ranged fighters, but that's about it. If you really want to make a difference, you need to start with other things like giving monsters curses, auras, hits that reduce your stats or making monsters attack the main character mainly (and not the minions). Everything else is mostly cosmetic (the positive side of that: it's hard to do anything wrong balance-wise if its only cosmetic anyways ).

5.5 Champions and Bosses

Champions and bosses should be stronger, but not impossible to kill. On the other hand, fighting against normal monsters should (in my opinion, and in the one of many others too, i guess) not be just undangerous clicking around. Because of this, you might have to decrease the difference in strength between normal monsters and champs / bosses. Some of the champion bonuses are hardcoded (for all the bonuses look here), but you can change several in monumod.txt (there is a strange table embedded in that file, where you can set a lot of funny things, very useful...).

5.6 Special Bosses and Act Bosses

Same as for normal bosses: They should be killable. If you have balanced your normal monsters to be funny and strong enough, that might mean that you have to reduce the strengths of the bosses.
Furthermore, these bosses are at fixed positions each times, so you can make runs to them, either for experience (the Baal minionsseem to give extreme amounts of experience), or for items. Ask yourself if you want it to be that way, or if it is better to make characters perfom what Joel called "act runs", meaning that they just clear entire acts in order to find the items and experience they are looking for. To do this, lower the Boss drops, make the good Boss drops be quest drops and / or increase the drops of usual enemies.

[Char]

6. Level Modding

Okay, first of all, level modding is mostly cosmetic just like adding monsters or changing graphics. NO WAIT, dear level modders, do not start to beat the crap out of me yet. I said, mostly. And as I stated somewhere in the beginning, cosmetic changes are actually a great thing to do.
Okay, there are of course some level edits that change the balance. For example, an act1 outdoor area plays completely different than the Maggot Lair or the Arcane Sanctuary (ever had fun with a hammerdin in the Maggot Lair?). But you have to be really creative if you want to invent whole level designs.

6.1 Act Layout / Level Distribution

The main route that leads through your levels is the one that characters will use. When did you last visit the Hole level 2? Do not know what that is? One of the underground levels in act 1, but I don't know which either. Because of this, your balance should be done considering the main route, there are lots of players who just want to get to the next act. This means that there should not be too many side levels you don't need to visit, cause otherwise a player that does visit them all will be too strong / too high in level, so your main route should cover most of the levels of an act.
You could also make the fun of giving two alternative routes or something like that, or, in the best case, a non-linear approach to level distribution, but I have no idea how to do that.

6.2 Special Levels (Cow Level)

Special levels should serve a certain purpose, which is quite for sure not to give insane experience and great drops, because this makes people do things like cow runs all the time. I don't think that was the idea.
The purpose they deal for might be a special boss you can kill at the end as well as giving you some more monsters to kill if you are still too weak to compete with the end boss. Also, they could be there for side quests or something like that, or, like the cow level was intended to be, just for fun. They could be an area where your characters go when act5 has become too easy (kindof the place for the masters), or, like in Ancestral Recall, a place where certain items drop (elixirs in that case). Just some ideas, I am sure there are lots of more options. But prevent people from playing in those areas exclusively, that's not fun. The usual acts should still be the usual place to run around in. Because of this, the drops and experience received there should not be better than in the usual levels.

[Char]

7. Making YOUR Mod

Okay, you now have read all this, and you are now (at least I hope so) a technically good mod. This is perhaps not enough. The trick in modmaking is to do some things that make your mod unique, to be creative, to be funny, ....

Basically the trick is to NOT follow this guide into every detail. You have to go your own way at some parts. This guide was made to help you so that you do not loose the general balance while doing this. This guide is not a dogma or something like that. Be yourself, and do some things that are not in this guide. But if you keep in mind the hints i gave you here, that should hopefully not lead to a balancing desaster .

You don't have to cover every single topic I covered in this guide either. You could do a great mod that only deal with items. You could do a great mod that only deals with skills. You could do a great mod that only deals with level layout. And perhaps the only thing you thought that sucked about Diablo2 was the lack of good cube recipes.

Finally, it is your mod, and it has to satisfy you. If it satisfies others, too, release it to the public .

[Char]

8. Misc Stuff

Some things that did not fit in the above categories and some space to put more information into later (if I happen to get even more ideas).

8.1 Hunting Bugs

Bugs are your enemy, and if you do a larger mod, you will probably never catch them all. Furthermore, there are already enough bugs in unmodded LOD.
But if you have identified a bug though, write it down, and try to fix it. If you cannot find it immediately, this technique has proved to be quite good.

1. Remove every modded files
2. Insert them one by one (if that works, sometimes some files need other files to work correctly).
3. If the bug re-appears, you can be quite sure that it is in the file you inserted last. Remove that file again.
4. Insert the modifications you did on that file partially and see when the bug re-appears. If it re-appears, remove the last modifications you re-inserted
5. Insert smaller and smaller parts of the modifications you do. Doing this, you get closer and closer to the bug and can, at the end, identify it.
6. Remove the bug .

8.2 Tools of the Satan

Tools of the Satan are things that you should be VERY careful with cause they are easily imbalancing your mod or for other reasons just a bad idea. This doesnt mean you should under all circumstances prevent to use them, just be VERY careful with them. Tools of the Satan include:

- Overpowered uniques and set items
Renders all other items useless.

- Too rare items
If you cannot find them, whats the use of it? Mostly they are overpowered, too, and you are god if you find them, and have problems if you do not find them.

- Essential items
If you need an item because you cannot fight / survive otherways, you are doomed if you do NOT find the item.

- Rejuvenation potions that are too common or, even worse, aquireable from vendors:
You can survive nearly everything except for onehit kills with rejuvenation potions.

- Insane monster HP
Primarily makes venturing through a mod take longer, but is not more fascinating

- Leech, meaning especially lifeleech
Makes fighting easy monsters even more easy. Doesnt help too much against highlevel monsters. Furthermore makes defense less valueable.

- Too strong hirelings / summons
Makes it have no meaning what damage the enemies deal for ranged fighters, and just makes the hitpoints and resistances of enemies be important for them.

- Regeneration (if set too high)
On monsters: Makes it especially hard to kill hard monsters, doesnt affect the difficulty against weak monsters. Makes the difference between characters that deal high damage and those that do not even more extreme.
On hirelings: Makes them not die at all against weak monsters, but quite often against strong monsters (if they die at all).

- Runes with too strong powers (like +to all skill levels)
Render the item enchantments practically useless.

- Very good highlevel skills
Render the lowlevel skills useless.

- The Cow Level
The ultimate area. Only one monster type, high monster levels, high experience, good drops, high monster density. A paradise especially for area-effect fighters.

[Char]

Appendix A: Formulae

Leech

Leeched Mana / Life = LeechValue * (PhyDam * ((100 - PhyRes) / 100) - PhyDamReduce) * (Drain / 100) / DiffMod

With:
PhyDam = physical damage dealt (before reduced by physical resistances
PhyRes = physical resistance of defender
PhyDamReduce = physical damage reduce (Damage reduced by X) of the defender
Drain = drain value found in monstats.txt (dependant on difficulty)
DiffMod = difficulty modifier, found in difficultylevels.txt (LifeStealDivisor / ManaStealDivisor), must be an integer


Attack Rating Formula

totalAR = absAR + (absAR * %AR)
absAR = 5 * (dex - 7) + absARBoni + ToHitFactor

ToHitFactor is found in charstats.txt



Chance to Hit / Chance to get hit

CtH = 2 * 100 * (AR / (AR + DR)) * (alvl / (alvl + dlvl))

AR = Attack Rating (attacker)
DR = Defense Rating (defender)
alvl = attacker level
dlvl = defender level

Maximum chance to hit is 95%, minimum 5%


Chance to Block

CtB = (dex - 15) * ToBlock / clvl / 2
ToBlock = ShieldBlock + ChrBonus + OtherBonuses

with:
dex = dexterity value of the blocker
clvl = character level of the blocker
ShieldBlock = entry in "block" field in armor.txt
ChrBonus = characterclass specific bonus, found in charstats.txt (BlockFactor)
OtherBonuses = block bonus percentages from affixes on items or skills (Holy Shield)


Crushing Blow

Crushing blow damages by a percentage. This percentage is calculated as follows:

CB_percentage = 100% / (2 * (players +1)) * ((100- PhyRes) / 100) / monstermodifier / rangedmodifier

with:
players = number of players in the game
PhyRes = physical resistance of the defender
monstermodifier = 1 if defender is normal monster, champion or unique monster, 2 if defender is superunique or act boss, 10 if defender is a player
rangedmodifier = 1 if not-ranged attack, 2 if ranged attack (bow, thrown javelin)


Open Wounds

The damage of open wounds is a function of the character level. The damage is dealt over eight seconds.

OpenWoundsDamage(clvl) = (9 * BaseDam(clvl) + 40) * 25 / 32 / modifier

BaseDam(clvl) =
-> clvl - 1 , if 0 < clvl <= 15
-> (clvl - 15) * 2 + 14 , if 15 < clvl <= 30
-> (clvl - 30) * 3 + 44 , if 30 < clvl <= 45
-> (clvl - 45) * 4 + 89 , if 45 < clvl <= 60
-> (clvl - 60) * 5 + 149 , if 60 < clvl

modifier is:
1 usually
2 if attacked one is boss or minion
4 if attacked one is a player
8 if attacked one is a player and attacker attacks with a ranged weapon


Hit blinds target

chance = 30% + (4 * blindbonus + alvl - dlvl) * 5%

This is divided by 3 if the attacker attacks with a ranged weapon.

If the check is successful, a random level 1-20 "Dim Vision" curse is laid upon the target.

"Hit causes monster to flee" is stronger than "hit blinds target". If a you have both attributes on your equipment, the chance to blind target is lower than displayed this way. The "hit causes monster to flee" is processed before "hit blinds target", and if its successful, "hit blinds target" cannot strike the monster anymore.


Hit freezes target

Chance = 30% + (Bonus * 4 + alvl - dlvl) * 5%

This is divided by 3 if the attacker attacks with a ranged weapon.

Duration = (rnd(chance) * 2 + 25) / DiffDivisor

with
rnd(chance) = random integer value between 1 and chance
DiffDivisor = MonsterFreezeDivisor from difficultylevels.txt (must be an integer)


Level Requirement of Crafted Items

LevelRequirement = min{98, LvlReqAffix + 3 * AffixCount + 10} + LevelreqBonus

LvlReqAffix = max level requirement of random affixes
AffixCount = number of random affixes
LevelreqBonus = extra level requirements from the levelreq property (if added to your items)



Most of this data is copied from the forums on Diablo2.de. If you are german, you can read the post here.

[Char]

Appendix B: Other good guides

I will put links to other guides and interesting posts here soon.
Your guide is not here but is really vitally important? PM me.

[ChaoticDeath]

Very, very nice stuff Char. Worthy of a sticky pin, I think .

[sPoT]

Yeah! Kickin' a**! Big part of hard work . I will help me a lot in modding

[kingpin]

Yes, this guide will be very useful for modders. Good work here Char

[Brother Laz]

It's here! w00t!

And now, Char, you see why I didn't really have anything to add to it... =)

One slight nit: curses should be sort of useful with only one point. They should not be r0x0rz, but they should at least do something. Remember that damage is low in the early game, and both amplify damage and weaken may end up doing absolutely nothing if the bonus is too low at level 1.

Also, a revive user needs amp and LR, and needing 20 in both skills to make them work at all would render that build impossible.

Also, if you need 20 points in life tap to make it usable, it is ruined. It is only good for helping your mercenary from time to time, and if you need 20 points to do so, no one will ever use it.

[Char]

you got some points there. got to add that.

i didnt really think about it too much since with the skillsystem i use, lvl 10 skills are rather cheap to get compared to lvl 20 ones, and i simply made the support skills do more at lower levels. this way, i can balance the costs for each skill to make it useful individually.

[sPoT]

And it will be very nice, if you add some notes about sense of changing tiles, overlays and such things. Generally: it's the best guide I ever seen , and the second thing: can I write here some issues?

[onyx]

[quote=sPoT";p="160318"]
and the second thing: can I write here some issues?
[/quote]

No! Your issues you post in a General Mod Making thread, not here. This is a guide, and not a "Help me" post.

[kingpin]

No! Your issues you post in a General Mod Making thread, not here. This is a guide, and not a "Help me" post.

I would say, if the issue is related to somethings that are missing in the guide he would post it here. But, if its mod related questions he would post it in GMM

[sPoT]

[quote=kingpin";p="160336"]I would say, if the issue is related to somethings that are missing in the guide he would post it here.[/quote]

Yeah. I meant this .

[mouse]

thank you char, VERY informative and well thought out.

[pmpch]

[quote=Char";p="160081"]Formulas are your best friend.[/quote]

A massive for that statement.


Edit: Typo Corrected

[Char]

corrected. thanks.

[Lord_Drekas]

Nice work char, most of what you wrote i had already been thinking about, its one of the reasons my mod is taking forever, i stop and think about something for as much as a week or two before i even change it or test.
(no sense in changing things just becasue you can) i agree with that statement totally. your points are excellent like

1) why make it as easy or easier then vanilia d2 but at the same time who really wants to play a version where it takes you ten minutes to kill a sinlge monster or fight your through 3000 monsters in one area heh.

2) balancing , while you said it isnt that hard, id have to say will it isnt hard to balance , the balance aspect should and does take the most time to plan out if you actually want to it to be truly balanced item/skill/game play wise, its not hard just requires a a lot of planning/thinking/experimenting

3) and the last thing as you pointed we all played d2 and loved it or we wouldnt be modding for it , i agree 100% with that. so in my opinion why make a d2 mod that completey changes everything just because you can, new animations, new items , new cube reciepes new maps, tweaking the skills and balance issues is what itss about but if it doesnt have a similiar game play feel as d2 then why do it we loved d2, why change things that will make it totally un d2 like.

L D

[TheMadModder]

Wonderful modding guide you have written!!! I reread it when I come up against a wall when thinking of things. I wish someone had written something like this a couple of years ago.

I have a couple of questions about a couple of comments you made and as you requested have posted here in the thread.

2.3.1 Automagics / Staffmods
"By the way, automagic.txt is your best friend cause it makes a lot of softcoded workarounds possible. "

I had totally forgotten about this file and was wondering what you meant by this.

5.3.5 Damage
"Remember that changing this in monlvl.txt also increases the elemental damage monsters deal. "

I was under the impression that monlvl.txt only affected the stats that are listed in this file(monlvl.txt). Also are these elemental damages you speak of are they the ones that are given at the end of the monstats.txt file (El1-3type)?