Rare Enemies/Drop chances

My friend was replaying Final Fantasy 4 the other day and there is a room in the final area with a rare spawn. This rare spawn and a few rare items that can drop one of which being 1 in 256 or something like that. He randomly go it on the 3rd try as there is an item that lets you fight the rarest enemy in the room. This item gets you some OP armor. This question here is should you have something so rare like this or should you have a super rare drop be more like 1 in 12 or 1 in 50 etc. We all love OP loot and working for in can be fun but at what point or you just being a dick to the player?

thats actually for balancing, the armor is very op so making the drop required to obtain it 1/256 is actually making somthing that only a perfectionist would go after. the armor isnt nessassary for completeing the game......

if any rare drop is an op piece of equipment its better to make the player earn it, a 1/12 or 1/50 is way to easy for op gear

1/12 or 1/50 could be fine if the monster that drops it is rare to find in encounters.

Because this games we make may only be played by a handful of people unless people really love it they will have no way of knowing it's there. Where is the happy medium between some people will find it and almost no one will find it? It's not like there will be a wiki or gamefaqs page for my game. In the case I gave I think the 1/256 is kinda understandable as I think the next best armor in the game has a def of like 25 and that armor have a def of like 100 with a ton of passive like ele damage reduction etc. I'm thinking having a pattern telling the player where the mats can be found and them being a pain in the ass may be a better way than a 1/256 drop from a 1/60 spawning mob in the middle of a dungeon may be a better way to do it.

The whole idea of "OP" random drops like this is that you don't need them to finish the game. They're secret bonuses.

If's it an enemy that is extremely rare to encounter, can only be encountered through limited/secret means, or is extremely powerful compared to other enemies, then the drop rate shouldn't be something like 1/256. But for a rare drop in a frequent encounter, 1/256 isn't a problem. The Might Sword in Lufia and the Fortress of Doom is a 1/256 drop, but I've gotten two in more than one playthrough largely because there's a lot of areas where Hydras (which drop the Might Sword) can be encountered.

Part of this is balance as something like the armor from FF4 way too good but I don't want to see the best weapon in FF12. You have to beat all the super bosses to get it and the only thing left in the game is final boss that is in on way as hard as the bosses you need to kill to get it. But this also speaks to something i'm trying to work on. I've always wanted to have a game that you can get gear that won't just be outclasses by gear from the next town. I don't mean just scaling gear because that is boring. I want something like some of the gear from Diablo 2. Some of that gear didn't have as high of stats but because of the buffs to damage or something else they were good into the end game.

Please don't speak for everyone. My preference is for games to be hard enough that I need strategy to get through them, so I feel rewarded for having used a good strategy. OP loot breaks this. If you want to include more powerful weapons / armour than the current tier available in shops, then firstly, keep it within reasonable limits. Your suggestion of a random drop with 100 DEF when the highest otherwise available is 25 sounds like it's way overtiered and would trivialise most of the game. Secondly, make it a reward for deliberately doing something challenging -- an optional boss, a side branch of a dungeon, following clues to a hiding-place, going out of your way to help an NPC. Not just random chance.

OP loot doesn't need to be game breaking, also that armor was in FF4 and you needed to kill something that was a rare drop with a 1/256 drop chance (I noted this above). Be that as it may I find games that give you armor that does something is better. Most games just give you armor with numbers and that's something I hate. Even say a sword that has a normal attack that attacks twice would be great or armor that makes you immune to one element but weak to another. Also I still want the player to work for it.

In Lufia: The Legend Returns, you can get a 200-Attack weapon on the first continent (where the highest-attack weapon you normally obtain, at the last dungeon of the continent, is 78 Attack) by beating a rather early boss who's meant to kick your butt. And even better than that, a generic enemy encountered in dungeons on the first continent can rarely drop a 480-Attack weapon.

While it makes early battles a breeze, it doesn't negate the need for strategy in tough battles. That's just one person with a damage boost in a party of up to nine where you choose three per turn to act, and having members take turns using their abilities to inflict extra damage, buff/debuff, or heal the party is far more effective than having Wain smack the boss with the Alumina Blade every turn.

Addendum: Secret/"overpowered" items can also add their own elements of strategy. Using Lufia:TLR as an example again—if you know where and when to get it, you can get a gimmick weapon with 500 Attack that does scratch damage with normal attacks but uses its full power for IP attacks. You can give this to Aima—you lose one of the best characters for using normal attacks, but in turn get a character who can dish out extremely powerful IP attacks every now and then.

Huge Lufia fan myself here. I think that a weapon that is really strong doesn't really kill the game at all. It doesn't even have to be rare as long as the game is balanced in such a way that while it makes the game easier, it doesn't instant gib all the fights in the game.

I think combining a rare spawn and a rare drop is a bit much unless there is an easy way to find the rare spawn or a way to greatly increase the rate the rare drop drops. And you would likely want something in the game that at least hints of the existence of the item.

One game I played, there was a sidequest that had you hunt for a very rare enemy that only spawned on a very specific tile. So that game at least gave a hint of the existence.

I just worked the math and unless I made a mistake, to have a 90% chance of getting a 1-in-256 item at least once, you would need to battle 588 times.  10% of players who battle this many times still won't receive it!!  Even the most adamant completionists probably won't have fun battling the same (possibly hard-to-encounter) monster 600 times and hoping they luck out at some point.  (By comparison, 1-in-10 uncommon drops would require 22 tries to have a 90% chance of acquisition and 1-in-50 would require 114 tries for a 90% chance).

It's fine to make some of the coolest content require grinding, but we can do a lot better these days than relying on pure dumb luck.  It was harder back when memory capacity was limited (the entire concept of anything being 1-in-256 means it was the smallest/rarest they could make it with a single byte of entropy) - now that it's relatively cheap to just add hundreds of variables into any game we want to, there's no need to go with the most simple and memory-inexpensive (and player-unfriendly) way to enforce rarity.

um you did do your math wrong.......1 in 256 means that 1 battle out of 256 battles you will get the item, have no idea how you got the over double that to have a 90% chance

He's right. A 1-in-256 chance means that, in effect, you are rolling a 256-sided die every time you fight. Later rolls are independent of what's happened on previous rolls. If you roll 256 times, you won't get every side exactly once (if you don't realise this, try it with a six-sided die!) and so you won't necessarily get the one particular result you're looking for.

The best way to look at it is the chance of failure, which is 255/256 each battle. So after 588 battles, the chance of failing all of them is (255/256)^588 = 0.1001213. The chance of success is therefore very close to 90%.

Aoi Ninami pretty much explained it.  The actual calculation to get 588 battles can be done by ln(0.10) / ln(255/256) which equals 588.  The equation can then be reversed to (255/256)^588 = ~10% chance as Aoi illustrated.

Recognizing (and compensating for) the potential to experience such "bad beats" is an essential element in game design.

As was already said, 1/256 doesn't mean you will get it every 256 battles.

Like my luck goes, I rolled 20 20 sided dice (1/20). I could only fail if ALL of them were 1s.

ALL of them were 1.  My Game Master was nice and made it my special ability that 20s were fails for me and 1s were success.

While it is possible that if you rolled a 256 sided dice 2560000000000000000 times, you might roll each number about 10000000000000000 times; with lower numbers, this becomes less likely to be even due to each roll being independent.

As was said, roll a 6 sided dice 12 times. If you are right, you should get each number twice. This is very unlikely that you will get each number exactly twice.

Now, that isn't to say there is a way to make it work that way.

How I do 'random' drops is a bit different from the idea of 'true random'

Essentially, my ways are a modified rate. First way is this:  First try: 1/256.  Second try: 1/255, etc.  all the way to 1/10 where it stays there.  If they get the item, it resets to 1/256. At 1/10, every 10 tries moves it. So with a 1/256 drop rate, 246 + 90 (336) tries will get the item for you (I might be off slightly).  Still a lot of farming, but you could get it much earlier as well. And when I say try, I mean each time the roll is attempted whether you mean to roll it or not.

Another way I do this is: 1/256 the entire way. However, if you have tried 512 times (double 256), it will guarantee a drop.

Yet another way is a 'luck' bar. Each time you fail at a drop, you fill up the bar based on the chance of the drop. (So a 1/256 would fill more than 1/20). Once the bar is full, you can activate a skill to increase your drop rate for that battle to 100%.

There are more complicated methods as well. Like the Agarest series (as well as other IF games), you can overkill (essentially instead of just killing it, you fill up a second bar of 'health' where if you fill that bar, it is an overkill) in order to guarantee a rare drop. Often, these drops can only be gained through an overkill.

You also have enemies that have to be killed under a specific condition. Take Lufia: Must be killed before you are supposed to be strong enough to do so.

Other conditions include: Must be killed while in stun state, must be killed with ice, must be killed on turn 5, must be killed after enrage.

There are plenty of ways to make it so that a loot is hard to obtain without relying on luck.

well i know every one is gonna hate my game cuase the drop rate for rare drops are 1/755 but thats due to being able to get stat stones from every enemy, for gear its 1/256-1/670, based on how far you are in the game and if that piece of gear is a game breaking piece of gear (and for my game a game breaker doesnt mean godly stats) for instance the second dungeon is gonna be full of poisen ailment enemies, you can buy poisen charms but those are gonna be very expensive, if you also want to upgrade armor and weapons aswell, so i added a rare drop to an enemy in the area that called the purity glove, this item is equipible by every character, and prevents poisen, paralyze, and instant death, now i have the drop rate set to 1/315 on an enemy that is in every battle arrangement except 3, but i'm guessing you guys would think that this drop rate was too difficult to obtain

also the FF4 enemy that 1/256 isnt as bad as you think considering that the battle with it there are 6 copies of it so the chance is actually 6/256

That wouldn't be a reason for us to hate your game. If it is a common enemy, it isn't really that bad.

Also, no, it is not 6/256. Please, listen to what we are saying.  Your logic is like this: If you flip a coin twice, it will always give you head once and tails once.  That isn't true. It could give you tails several times in a row. Please take a statistics class if you are having trouble understanding this.

Take for example this scenario: For a certain mmo, you have a super low drop rate with timed spawns where you must use a specific class and weapon in order to get a drop. You then need to collect 8 or so of these drops for one of the best items in the game. (I am not naming the mmo).

Honestly, frustrating. People might farm for months without ever getting it. People get kicked out of parties for not having it or comparable gear that is harder to obtain. Other people get it first try.

Bad way to do it.

Better way to do it is an mmo that had a super low drop rate, but it was for an optional side-grade from a relatively common enemy. You generally got it within a few days at most.

you competely mis read what i wrote about FF4 in that post please reread it

If you are saying that, then if there were 256 copies of it, it would be 256/256. That is NOT the case. I didn't misread.

Here is how you can figure out the number:    1 - ((255/256) ^ 6)  = chance that at least one will be successful (item will drop).

I am using parenthesis here to make sure it is easy to see how you calculate it properly.

That means you have about a 0.02321 (2.321%) chance.  6/256 is about a 0.02344 (2.344%) chance. NOT the same.  (I might be slightly off as I am rounding and doing this really quick, but it should give you an idea).

I want you to roll Six D6 (D6 = Six Sided Dice/Die) at the same time. If you are correct, you will get one of each number every time you roll. Again, please take a statistics class if you are having trouble understanding this concept. This is a rather important concept to at least understand and it might be easier to understand if you take a class on it.

once again misreading.........i specificly state thats its not as bad as you are saying becuase there are 6 of the enemy there, so its not a 1/256 chance in the fight to get the item, but go on continue to try and make me look stupid when you are odviously failing to see simplicity of what i said