Sometimes you've got perfectly calculated systems running in your game, but actually playing the game feels... off. Let's learn how to design purposeful break points, and avoid unintentional loopholes!
Shouldn't all those equations being made to run your game under the hood, be made WITH user experience in mind? I think if your adding a layer of systems/stats to your game, or expanding an existing one, it should be done so for the sake of the user experience. What other reason would there even be to add depth if not to improve user experience?
You could have brought up Earthbound (Mother 2) and the way it automatically fights enemies so low you will (most likely) one shot and then gives you minimal exp and rewards.
I like how Earthbound handled break points; when you were significantly stronger than the monsters, they A) ran away from you on the world map, rather than towards you and B) if you did engage them you either got a guaranteed "green," engagement OR instantly killed them without transferring to the battle screen at all.
I'm pretty sure the running away from you has nothing to do with power, but with whether you've defeated the boss of that area. The latter is pretty cool, though, and very simple math, too, I think. I believe that happens when the game calculates that you would just win without getting harmed if you simply mashed attack. Of course, that only really works in a few games.
Nixitur It was both power lvl and/or defeating the boss that decided whether enemies ran from you. The system for instant wins was a little more complicated then an autofight sim, but functioned basically like that.
Nixitur It's both. Enemies will run away after you kill the boss to make it easier to escape the dungeon. In the overworld, it's dependent on level. It makes a lot of sense mechanically, and just plain feels good to see yourself so strong that enemies literally flee in terror.
Skyrim is so awful for this. No damage is randomized so most damage upgrades are useless against most foes while some are substantial (3 hits to 2 hits, 2 hits to 1 hit). Also the armour system is garbage with each point being more valuable than the last because its flat %, and then everyone caps it out eventually so it stops mattering at all by mid game and everything is harmless until you fight monsters who can hurt you despite only doing 20% damage or whatever it is. One thing that helps a ton with breakpoints is small randomization consider a game where you always do the same damage, when fighting a 10 hp foe, the only numbers that matter are 1, 2, 4, 5 and 10. But if attacks do X plus or minus 1 damage than the numbers that matter are 0, 1, 2, 3, 4, 5, 6, 9, 10, 11. Another way to prevent breakpoints is with easier methods of doing a lesser result. XCOM uses this to great effect as a weapon upgrade of 4 damage to 6 damage might not change the number of shots to kill an 8 hp creature, but the 6 damage shot will leave it in range to be killed by a frag grenade (3 damage) that is a guaranteed hit. Xcom also uses +/- 1 damage and crits to eliminate breakpoints.
To be honest, I have the opposite take on this. Skyrim uses leveled enemies, which means that they're getting stronger as you do too (just slightly slower). This makes it feel like you're not improving at all. I agree that they should use your example of randomisation, but I think that they should tone down on the leveled enemies to have improvement that matters.
Definitely, (I highly suggest mods that remove the level scaling and add damage randomization btw). My main problem with skyrim progression is it basically doesn't exist until you reach a drastic point that completely trivializes everything. Also because the damage increasing perks are so substantial, as you level up you actually get a shrinking toolest to deal with encounters. At level 1, archery, all types of magic, one handed and two handed weapons are all viable paths. At higher levels your chosen skills have kept pace but everything else is now entirely useless. This is a worse problem in fallout 4 imo, which has really substantial damage scaling by perks on guns and actively punishes you for wanting a variety of weapons for a variety of situations.
I couldn't agree more. When I tried getting every skill to 100 (and I did, but I regret it), it was quite difficult leveling combat skills I didn't start with. Right now I'm trying a playthrough with such mods, definitely an improvement, albeit a lot more challenging.
Skyrim also has a problem with how enemies scale up, at least with destruction magic, you become impotent in the late game. Enemies scale up farther than your spells do once you hit that damage cap. If that was intentional to force you into using other tricks like alteration or illusion then maybe it gets a pass, but it doesn't feel intentional.
You can screw yourself over in Skyrim though if you do something funny with your build and hit a scaling level. And equipment can make a difference, though if you are just an amazing smith and enchanter, you can pretty much make equipment to play as any build and just change on the fly, thus still breaking everything. it is however also possible to go too fare into randomised damage. Xcom is great, but does anyone play with damage roulette?
A lot of this comes from doing the *wrong* math. The most obvious example comes from calculating damage per second, rather than time to kill. You actually see this a lot on things that reduce incoming damage. Let's say you have a skill that has 10 ranks, and each rank reduces the damage you take by 1 each time you get hit. There's a general intuition that, since each rank decreases damage by the same amount, each rank must be equally good. Now, imagine that enemies hit for 10 damage. If you have 100 hp, the 1st rank you take in this skill increases the number of hits it takes to kill you from 10 to 12. Fair enough. The 9th rank you take increases the number of hits from 50 to 100. The 10th rank increases the number from 100 to infinity/undefined (you're invulnerable). The damage you take is decreasing linearly, but doing the math that actually matters, we can see that the time to kill is increasing at a staggering rate. There's an obvious breakpoint at rank 10, but we can find that, and the other unintended trends, if we do the right math.
When I look at the Flowey fight in Undertale, it feals like the defense increases exponentially. You lose half of the bar after like two hits but can take over9000 until you die.
Yep, I get a 1% extra damage passive in many games due to equipment, skills, whatever. I never feel a difference and even forget they exist. So I dread skill trees or equipment that have tons of those small buffs, because I know it will take so much time until it becomes relevant, basically at endgame or post game.
I'd be really curious to hear from a dev on that type of game what they were thinking, actually. Sure, my fire-ball is now 5% stronger, and has plus 1% in radius. Whoop. The excitement is CLEARLY off the charts, right? Heck, you want the near same system be as near exciting as plausible? Have a 25% and 5% increase for five points. Bam, done, the player actually feels the difference.
Mattias Svensson Some of them think that its exciting for the player to minmax and theorycraft a lot to get a small or, at most, medium edge. Never a large bonus, that would break the flow they want to force upon you. In the end, its a type of mechanical railroading, they don't want to take options away, but they don't want you to use them , because they want to force you towards the "only right way to play". Like how DS3 very strongly discourages mage and hybrid builds because melee "is the only right way to play".
If given the choice I will definately take the option that changes how I play my character or the character interacts with the world over the incremental increase in whatever. I honestly don't understand the point of them in games aside from some sort of sense of obligation on the part of developers to make the player falsely get excited because the words "level up" have appeared on screen.
Edward Nigma Yep, I prefer how Nioh does it, skills cost a different number of points. You get melee ones which can be spent on the tree of any weapon, mostly in active skills. You can then customize which ones go to what command to make your own playstyle, while some skills are always available. And sometimes you get consumables that give extra points as a reward for a particularly difficult mission, which is pretty awesome. Ninja and onmyo skills give you consumables that recharge at a checkpoint. But you have loadout points for each one and many skills have 3 tiers, higher tiers costing less points to equip. So when you unlock a tier 3 Lightning Shot, for example, you can get two more to have a total of 6 shots or use tiers 2&3 to keep the same number and free up some points for something else. This system is really easy to code and can be huge for build variety.
In StarCraft 2, we have that sort of things. They called critical upgrades, and on of the things that you learn to keep track of to compose your strategy. For example, a zergling with +1 armor takes 3 hits to kill by a zealot, instead of just 2. Given the attack speed of the zealot, this thing completely changes the dynamic of this match-up, and protoss learn not to go out if the opponent has +1 armor. The +1 armor also allows zergling to resist a baneling hit with just one hitpoint left, a huuuge change in the dynamic of the match-up. So to complete your thing, in RTS, those breaking points can be (and often are) thought intentionnaly, both by players and designers, to mark turning points in a game, and also defining game stages. It encourages progression I think, since to counter a critical upgrade, you also need to tech up, which in StarCraft means building advanced building, which produce more high-tier units. It's a nice progression incentive.
One time I was playing StarCraft 1 (badly) and I sent a fleet of carriers against a terrain fleet. I had the the invincibility cheat on, but it still took a VERY long time to win because the enemy's battlecurisers had maxed out armor. The carrier drones have a lot of hits per second, but low damage, so a small damage reduction had a huge impact. Now whenever I think of linear damage reduction, I think of those battlecruisers.
Totally wanted to say that, props for saying it first! Also break point occurs not only with upgrades but with unit numbers to (4 stalkers to one shot workers) and in some engagement it's win or lose the game if you did or didn't estimate the ennemy force well enough (on the same level of micro).
So many hot fuzz references, oh my god. XD And that "feels" face at the end was just perfect. ...Anyway, I get the feeling that Warframe takes Break Points and runs the whole thing into the endzone. I'm like "Man, this weapon sucks, I'm barely killing enemies with it" and then tweaking a few Mods on it makes it suddenly one-shot just about everything. I'm at the point in the game now where I need to actively look for break points because the game is now suddenly REALLY freakin' challenging and if I'm not trying to break the game, I'm not winning anymore. XD
Aegix Drakan just came down to say exactly this. For gamers who love the game of searching through systems for the next game breaking combo, warframe is where it’s at. Really it’s their entire gameplay loop imo.... and it’s a lot of fun to finally find that crucial keystone piece and have a challenge that’s been grinding you down just evaporate under the weight of some new stat combo.
this is true until you realize that there's like 3 useable endgame builds for guns and all you need to do is figure out which one fits on your current gun based on it's base stats, then forma-grind the thing until you can fit them on there
That's only true for some weapons and concepts. Outside of the usual crit/status calculations, there's also armor stripper builds, H-cal ranks, and weapon specific weapon alterations. For example, most weapons cannot reliably sustain a Heavy Caliber of 8 or more, while the Lenz can support it at 10 and the Soma cannot run it at all. Likewise, Rakta Cernos cannot use fire rate mods, while Dread and Lenz absolutely need them. Penta cannot use multishot mods, while the Tonkor loves multishot. The Stug cannot use punchthrough. So on and so forth. Yes, there are initial baselines to how you optimize your damage, but many weapons require you tailor the baselines to the specific weapon.
Most bows use fire rate, due to the fact they have painfully slow inherent firing speeds due to charge time, and inherent punchthrough from projectile mechanics. The Cernos line is the exception, with a baseline 0.25s charge time, attempting to stack firing rate just makes it difficult to acquire ideal firing lines/heads without wasting time at full charge.
This is actually a pretty significant problem in Breath of the Wild. The way the armor boosts and weapon durability increases work, the way that a late-game Link fights can afford to be way less skillful than in the early game.
I personally prefer games that actually make me feel powerful once I've put in the time and effort. I don't mind late game challenges being introduced, but making me feel weaker late game is worse than making me feel more powerful. A perfect example is the original Everquest. Start any character/class combo. Level 1, you can rush out and kill anything with some work you run across in the starting zone. Of which you are fighting bats, rabbits, snakes, and little kobolds. Then over time you reach max level, go to the highest level zone in the game, and see that it is populated by the same exact looking bats, rabbits, snakes, and kobolds. The difference is there is literally no way you can solo anything in that zone anymore. Infact, you probably couldn't solo anything past level 6 anymore either, but now you need a huge party of people with the best min/max setup to take out the level 60 common bat monster that at level 1 you could beat with a few pokes of a rusty dagger. That is a much worse feeling as a player. Many game have done those horrible masochistic like progression systems and those are just truly abhorrent game design.
T Blank Yeah, that’s also a problem. And that’s something BOTW actually does really well because of how it allows you to piecemeal gain power as a reward for completing tasks (weapon drops from difficult enemies, gear from shrines and side quests, armor boosts from collecting things, etc.) But, in the end, the game leaves you extremely powerful and basically invulnerable to even difficult monsters. It’s one thing to feel powerful. But you can feel powerful and understand that there are perils involved. BOTW does that very well by giving you the challenge straight up and then allowing you to grow and meet it. But there’s still a late-game problem with this design. That said, I don’t really think it’s fair to call it a major criticism because it only hits after you’ve put in maybe 30 or 40 hours into an incredible game.
Is that due to break points or due to simple progression? Yes, breath of the wild has a negative difficulty curve - no dispute there. But that boils down to giving the player steadly acces to stronger tools, the player having more experience and the enemies only slightly increasing in difficulty. Yes, a end game Link against an end game enemie is stronger (numerically) then a starting link against a starting Monster - but the change is gradual. Every heart helps - and yes the first extra hearth is more important then the last two. But there is no distinct break point. The master sword is a big boost - but that is intentional and is typically used as an axt and rock smasher as other weapons are numerically stronger and more reliable.
Adnan Ilyas Agreed. I find the early game much more rewarding and exciting than mid-to-late. Especially when it's so easy to sell gems and stuff to buy good armor and get Link out of his early-game vulnerability
I agree. It does take some of the edge off when you're eventually hunting Lynels for sport! But I feel it's also rewarding considering I was terrified to even engage them for the first 50 hours or so.
Examples like toe one in the video definitely, but it isn't always that simple. Say you make an RPG with powerful monsters that attack you when you get in range. So you map a break point where you become fast enough to kite certain monsters to infinity. Fair enough. What you didn't expect is that just getting the monsters into certain terrains can be worth most of your health. You can't graph break points you don't know are possible. And writing functions for every single possibility for say a card game can be more effort than you can spare.
I've been balancing my game by play testing it and always noticing these breakpoints and making sure they didn't hurt the gameplay, but I never knew there was an actual name for such a thing. Great video as always!
Personally I think Pokemon has a MASSIVE issue with Breakpoints. The hardest part of most Pokemon games is the beginning upto the second/third gym, but by then you've got access to mons (Most of which are at least second stage whilst the game's trainers still mostly use first stage Pokemon) and moves of most types, with said moves with enough power to OHKO anything they hit super effectively, meaning much of the Pokemon games is simply matching the mon you're facing to a move of a super effective type and watching their health bar plummet in one hit,
The thing that actually destroys pokemon is the fact that exp share is considered a standard feature that causes the player to naturally level up faster than the wild pokemon and trainers in the respective areas. That, and the fact that exp from fights doesn't decay or rise depending on the level gap between you and your opponent make for an rpg experience that is very fun if you're looking for something simple, but grossly unrewarding as a single player experience if you're looking for an engaging challenge. This is to say nothing about pvp which is one of the most splendidly complex games ever constructed, but the base games are very far from their JRPG roots.
I always turn exp. share off as soon as I get it. It is bizarre that it's activated by default when it completely breaks the pacing of the game that Game Freak worked hard to build. It is useful for if I need to grind quickly (I hate grinding so the extra exp is welcome) I gotta say, I love Pokemon's pvp metagame, I think it's incredibly fun, but the in-game always suffers from pacing issues for me.
I just don't understand why the exp share now goes to all of your pokemon, instead of it being a held item that only gives the pokemon that holds it exp. That was a good system, as you could now use a wild pokemon you got, as they'r generally very weak compared to the trainers in the area, making it difficult to train them in a reasonable amount of time.
This was a really really really good episode of Extra Credits. I think I've watched every main-series episode and this felt like a return to form. I'd never thought about this concept in practice. I've often wondered in a game, "is this supposed to be this easy?" and the concept of Break Points neatly answers that with, "no man OK! We might have missed that." this was really cool stuff. Great work guys. :D
It was for this very reason that in this game me and my friend were designing, we made the base health 21, because there's a new breakpoint for every number from 1 to 7, so there's a significant change if you alter the damage you do.
When I used to play Payday 2 (before the weapon overhaul) I remember most of the talk about weapon strategy was about hitting damage breakpoints. When choosing a gun and modding it to be optimal, you had to try to hit certain damage values that would let you, say, 1-headshot a normal SWAT or 2-bodyshot an armored cop. Any damage over each breakpoint wasn't important, so you could swap out some damage mods for other stats once you got to that point. I remember there was also this bit about "fake 40" where you needed to hit 40 damage for a breakpoint, but the game displayed 40 when it meant 39.95 or something (which wasn't enough). That was a weird bit of math that made the game a little more difficult to figure out. Also accuracy in that game was on some weird logarithmic scale where 18 was perfect accuracy and every 2 points was way better than the previous 2 points. Payday math was weird.
i think part of the experience that i love most about rougelikes and sandboxes is the exploration of possibilitiy spaces and find ridiculous builds. it's a fine line between break point optimization and exploits. the skyrim necromancer build with a 100+ undeads followers maybe gamebreaking, but is fun as hell.
I am so glad this came out right now... I'm currently working on a combat system for my role playing guild in wow and have spent the past 2 days trying to balance out how a bunch of different classes can do about the same amount of damage with different abilities. It's been a lot of work so far but i think it's time i put the excel ark aside and actually put this thing to the test if i can just convince my guild master to approve the system.
from what i can think you should look at systems that give you a reward instantly. like health. you can be busy 5 min to kill something but you will only experience the reward if the final health is gone. there is no reward for killing something halfway. (in classic health systems) buy systems can work the same. doesnt matter how much gold you have if you cant buy anything. that final gold might make the biggest difference.
I don't know you, but that beginning statement is the first time I've heard someone take that stance. As far as I have seen, every player is excited and happy that they can one shoot every enemy, in Dark Souls for example, when I can one shoot everything in the Undead Burg for the first time it feels like I've progressed, all the time spent there that felt like just traversing the land was actually my character improving, and that's rewarding, it makes me feel I'm moving forward
One thing you can do to soften up break points is to add some randomness to your game like crits and damage spread, so that (in the example the video gave) you can't one shot a mob at level one, you can with a crit at level two, you can with a high damage non-crit at level three and then you always one shot at four.
That's a big decision that leads to a completely different game. Its not a solution, it just means break points are less important to your game if youre already doing that anyways. And with an rng system, break points are harder to control, not easier. If there's even a 1% chance that you 1 hit kill the first boss, thousands of players are going to have a bad experience. If your formula is broken "lets just randomize that" is almost never the right answer.
That's why as player doing PVP I love special abilities in stat heavy games. They are can be combined in ways the devs never figured for frustrating (for my opponents) results. I used to do this particularly in WOW and just crush people in 49 pvp at lvl 42.
mixing and matching weapons with armor with character skills is why I still play Borderlands 2. that game seems to avoid the break point issue by dramatically ramping up difficulty in each new area, and shuttling the player along by limiting how many side quests are available. which is great! creating this sick melee build Zero is so great because that build is dependent on the quality of shield and the power of the other weapons. the game constantly getting harder means you are constantly hunting for new gear, which means constantly tinkering with your character. which means i am constantly coming back to that game
Another great example of this from the game Duelyst is that there is little difference between a minion with 1 health and a minion with 2 health, but *all* the difference in the world between 2 and 3 health. This is because duelyst heroes all start with a base 2 attack (like a permanent 2 power weapon to compare with hearthstone). 3 health is the breakpoint where a general cannot remove a minion instantly any more.
I think dark souls was one of the first games that made me really think about particular break points, and is a good example of them done right. By the time you have leveled up and learned an area well enough to traverse it quickly and kill all the Dudes, you have ascended the rewards in that area usually. It was one of the first games where I got actually thinking about how many hits an enemy takes to kill.
For a big example of this look at the balancing in Smash games where even making a move do not even 1 more damage even like 0.1 more damage makes an insane difference hence why teir lists are made, the character that has more very slight unbalance adds up and sometimes the Smash developers overlook certain minor exploits of the system that create things such as wave dashing, ledge canceling attacks, melee Fox and Falcos' amazing reflector or "shine", ladder combos for a large variety of characters some being much easier than others, rage making certain attacks way to rewarding on hit due to how much difference the addition knock back makes and also making some weak and or fast moves kill way sooner than even the development team expected, also look at how much stale moves can effect a match in Smash as a series to prevent spam...but then there's unintentional infinite combos, chai grabbing to death and "wobbling" with Ice Climbers in Brawl and Melee respectively....the difference of super armor vs heavy armor vs neither and much more is what makes certain match ups *very* different and for the most part very dynamic actually...
As an engineer I gotta say - the example of linear progression there is because its not balanced even mathmatically. You're increasing input linearly below a saturation limit so the increase has 0 value until a new saturation point is obtained. This is a case of bad understanding of math in systems.
There is visual feedback though. The player knows they are getting slightly stronger, even if it's only until level 4 (in the example shown in the video) that it actually makes a difference. I think being able to see the increase in your damage serves two purposes: A) It teaches you that you are getting stronger (yes, the first time playing an RPG you don't know that leveling up = you get stronger). B) It sets a goal for you and motivates you to push for that goal.
jonasn5 Thank you! I'm a mathematician, and I cringed everytime he said the math is not enough. It's not that the math isn't enough, it's that you aren't looking at the proper math!
You just re-explained his point to him. The point is that it isn't mathematically balanced and it is easy to see the "saturation" points in his example which is what he wants. The point is in a real situation these imbalances are usually much harder to see. In addition, changing a number to fix one instance frequently imbalances another instance. RTS, RPG, MOBA, and TCG games all have a very hard time with this. Edit: in theory the math alone is enough to balance a game but this is similar to telling a programmer that in theory a version of a program can be made that is without bugs. In theory it is possible, in reality the problem is so vast and complex that rarely can every issue be fixed and not spawn a new one unless the problem is of trivial difficulty in the first place such as his example.
I love it in D&D where that static upgrade of stats means a bizzare progression. Like think about Base Attack Bonus for worriors. You start with +1, at level 2 you have a +2 and at lvl 3 you have a +3. That means at lvl 2 you base attack becomes 100% better but leveling up to level 3 it has only grown by 50%. The same issues happens in D&D with HP. Getting to level 2 and suddenly you are about twice as compatent as you used to be.
This is all over the place in RTSes, especially games like AoE and StarCraft where damage and health values per unit are relatively small. Certain upgrades will be a top priority in certain matchups, as +1 armor allows Unit A to survive three hits from Unit B instead of only two. I gotta wonder: exactly how much of these "classic" break points are due to the fact that most player characters and enemies still do the same damage when they're only at 1 HP vs. 100 HP? Granted, if the enemy is being one-shot it doesn't really matter, but I feel like because the difference between 1 HP and 0 HP is so suddenly drastic (the classic "critical existence failure"), it's probably emphasizing math that wouldn't normally be emphasized.
In Suikoden: Teirkeries the game has you earn less and less experience in an area the higher level you are for that area. You always level up at 1000 experience but when you enter a new area you may start leveling up every 5-10 fights but after a while that amount increases due to no longer getting as much experience. This does help since you get 108 party members so you can level up new level 1 members quickly.
I've found a recent example in FighterZ. Practicing a combo that deals 50% of the opponents health is very important, as it allows you to "two touch" them, killing their character in two combos. Getting a combo with higher damage, however, isn't really useful, as you're still gonna have to two touch them.
For people that try to find break points in a game its always a good idea to figure out how the game uses its stats and skills. As an example, in diablo 3 socketing a high lvl red gem can lead to major power boost in the early game, however the combo of multipliers on set and legendary items with crit and already high base weapon damage makes the green gems far more powerful at max lvl.
I feel like Dark Souls avoids this trap well, you can grind all you like, but by the time you can start one-shotting, you're so overleveled to the point of all soul drops being borderline meaningless (Also shoutout to that Cave Story music at the end!)
That is true, though I do think it's countered by the fact that you have limited spell uses, attunement slots, and in the case of Dark Souls III, Mana points so that you can't use it 100% of the time, and then by the time you can, you're probably way overleveled, or a glass cannon.
Sort of. If you're fighting trash mobs throughout the level, you want to make sure you're being efficient, so you usually want to have just enough damage to do so efficiently, so the rest can be spent on other things, like defense. For example, in DS3, I needed to farm a ton of Blue Ears off some knights, using a pyro character. Assuming you use spell "X", you can boost your power by like 50%, but you only need a 30% power boost for X to 1 shot. So you grab just enough to hit that 30%, then you put everything else into other useful things, like defense, SP, increased item drop rate, increased soul gain, or, most importantly, improved fashion. It's a very classical breakpoint, and you can make similar examples with other situations (e.g. how much +atk stuff do I need to 1 or 2 shot some annoying mob, instead of 3 or 4 shotting it).
I think Darksouls have tons of breakpoints, but non of them matter long term. Getting a new weapon can help you drastically. the homing soul mass and soul spear spells are more damage then your old spells at the same speed of the fast but weak one. non of these matter though. If you cant dodge an incoming attack, it doesn't matter you have a +15 great axe. The spells are pointless if you cant get to a safe spot and cast them. The game have breakpoints, but it player skill that matter in the end.
I think that the damage example can be fixed by having multiple enemies with different amounts of health. For example: In the binding of Isaac, a +0.5 to damage doesn't change the amount of tears it takes to kill most enemies. However, you can now kill the dips (tiny poops) in one shot. It is also more useful against bosses where you can stack this bonus. This allows constant break points for a powerful experience.
Omg thank you, been designing an rpg for the past few years and my biggest problem was the power spike. After watching this video it gave me an great idea for a power spike I can add later in the development
To expand on the DnD example: In 5th Edition, the best example of a power spike is 5th level. I'm no game designer, but for any game designers who want a jumping off point on how to make a power spike, it seems like a pretty easy to understand example (considering even most incredibly casual players can see the power spike incoming at that level).
As a roleplayer at heart, I can say that I have run into my fair share of these breakpoints. How ever, I can not say that I have ever been mispleased by doing that extra bit of damage, even though it does not equal one less hit!
This episode could have been shown with the old Zerg versus Protoss matchup in starcraft 1. an unupgraded Protoss Zealot takes 3 hits to kill a Zergling. with one attack upgrade it becomes 2 Hits. A defense upgrade of the Zerg turns it back to 3 Hits. This is not just a balance point between 2 units, but the breakpoint of the whole matchup. It isn't just a breakpoint between 2 units, but the whole matchup hinges on these units. On the other hand it creates a strong theme of a race of upgrades or going to new kinds of units which makes these breakpoints obsolete.
Great video as always! Also the timing you released this video is so right for me because I'm currently loving Diablo 3 (4 years after when it was first released, yes) but what you're saying in the video is exactly what I feel when I play it. But the original skill and runes system creates endless possibilities for builds, plus there is also the absurd amount of increasing difficulties so I basically never got bored playing it, like hundreds of hours now! Like I said, awesome video!
I love "putting the work in" and getting a beefy feeling of noticeable bump as a reward, and don't like when everything is so linear that I feel that my planning and thought "never paid out".
These often become part of metagames in competitive strategy games. In Starcraft 2, the upgrades for weapon damage for units are usually +1 damage per attack. In a Protoss vs. Zerg match, if the Protoss player gets +1 Attack his Zealots can go from 3-shotting Zerglings to 2-shotting Zerglings, which means that a single Zealot can handily defeat 4 Zerglings instead of dying to them (they cost the same). This doesn't work in reverse though, where +1 Armour for the Zerg doesn't mean 4-shots instead of 3 - so the Zerg player needs to watch for that upgrade by the Protoss.
i think Chrono Cross on the PSOne did this quite well, for anyone who never played it, instead of standard levels the game used what are called "Star Levels" which only went up after beating certain bosses, the battles in between those bosses would increase your players stats, but only up to a certain point depending on your current Star Level, after awhile battles stopped giving you stat increases until you increased to your next Star Level, so the game kept you from feeling over/under powered through the game cause your characters progress is intentionally locked until more progress is made in the game, this was done very well in my opinion
I noticed a break point when I was playing BioShock 1. At a certain point in the game you will unlock the crossbow. According to the BioShock wiki, if you perform a headshot to a regular enemy it will have a 1000% damage output. Also, there's a upgrade you can get that reduces the chances of the bolts from being destroyed upon fire. Combine this with the Electro Bolt plasmid (Stuns Enemies) and Eve Saver Tonic it makes the player incredibly powerful. So, essentially the player can easily breeze through the game with little or no difficulties with this specific build and strategy.
This ties into an idea I've had regarding what I like to call "damage races." One example is this: say you have two weapons in your game, each does 100 damage per second. One of those weapons does all of that damage in one hit, and takes 1 second between each hit. The other hits 10 times in one second, each hit dealing 10 damage. Despite having the exact same DPS, the slower, more powerful weapon is better in every situation. See the Deagle and AWP in CS:GO for one example. There's a similar thing with offense as defense. Say a game gives you a choice between a big, high-damage two-handed sword and a lower-damage one-handed sword you can use with a shield. In many games you see that doing more damage and killing the enemy faster actually avoids more damage than the higher defense offered by the lower damage option, rendering the lower damage option obsolete. I would like to see an exploration of all of the different methods game designers use to overcome these balance challenges.
You know what game popped into my head when I heard your part on how designers can know when a break point will happen and design the game to then give trivial loot/rewards to signal that you must go on to the next area? Paper Mario! Paper Mario definitely does this when you level up to the point where the enemies in an area become super easy to beat and aren't a challenge anymore. You always need 100 star points (experience basically) to level up, but once an area becomes trivial to you, all enemy fights only give one star point. It's way too inefficient to try and train that way, when the next area can have fights give you 5 to 10 star points. Great example of knowing where your break points are. Especially since the numbers used in that game are all low values so each change in number is a bigger deal.
Vermintide 1 actually had a lot of these that were very intersting to observe, they were all based on Weapon Damage and difficulty level (since difficulty level effected enemy HP). The 1h Axe could only hit one enemy, but also instantly killed small and medium sized enemies. Suddenly though, on the highest difficulty, the 1h Axe can no longer one-hit medium sized enemies. When the game has you fighting 6-30 enemies at a time, going up from 1 hit to 2 hit is HUGE and felt terrible. Thus the 1h Axe was rarely used on this difficulty.
Thanks Extra Credits. I've seen you mention 'break points' in other videos, so it's great to see a video specifically about that. It's very interesting to learn about 'game feel'. It's something I'm aiming to focus on in the game I'm working towards making... but I'm only at the 'block breaker' module of my game course at the moment, haha. Oh well. All big things have small beginnings. Thanks for all your videos. Looking forward to your future content. Peace
The example that comes to mind off the top of my head is phantom particles moving the opposite direction of their momentum. A concept that still makes me think "wait.... what!?!"
The art in this one is a return to form. I'd forgotten that your videos used to be full of constant funny little jokes. Please reincorporate them back into your main artists' work
I’ve since dropped this project, but I ended up creating a negative break point on one of my RPG Maker games. Basically, by the midway point, the basic healing spell became pretty much worthless...and because I hadn’t anticipated this, the healer still would’ve had quite a few levels to go to learn the second level healing spell.
I remember in Legend of Dragoon, the first boss fight against Fruegel kind of depends on the player to not be very good with the attack items yet to play as intended, or at least expects as much. If you get really good at maximizing damage output with the attack items (their strength is determined by how good you are at rapid input sequences- which there is legitimate technique to), the fight would be too easy since you could kill him in like 2 turns. Game developers anticipated this and made it so that if you deal too much damage to him in one turn, he uses an item he otherwise doesn't use to double his offensive and defensive power (like a deliberate breakpoint countermeasure)- which makes the fight too hard, thus it mostly serves as a punishment/discouragement for the player trying to be too strong too early. Sure, you could whittle him down and then use a maximized attack item to finish him off before he can use the boost item, but that would require keeping track of all the damage you dealt up to that point with everyone and you'd be within a couple turns of finishing him off anyway, not to mention that successfully executed normal attack combos make your attacks stronger- a mechanic the game kind of depends on occurring at a certain rate for its difficulty curve, you can't really afford to miss opportunities to grow it by cutting your turn count short. Unless of course you grind, but that's not fun.
Celeste came out recently and the creator had a lecture about using level design to teach mechanics instead of tutorials. Really interesting lecture, I think you can make a decent episode out of that idea.
You could also smooth out the break point ramp, e.g. by introducing a "critical HP" stat to enemies at which they perform a lot worse in battle. (Too wounded to fight.) LV 1: Enemy takes two hits to kill. LV 2: Enemy still takes two hits to kill, but the first hit is enough to knock the enemy down to critical HP. LV 3: Enemy takes just one hit to kill. LV 4: Still one hit, but now the loot and EXP become worthless. Now there are more break points, but their relative impact to each other is less dramatic, and you get rid of those "didn't do anything" level-ups.
There is this tabletop rpg in Italy called The Last Torch that works a lot on breakpoints, given obviously that tabletop rpgs have much different feel from videogame rpgs. And it does it in a good way, in my opinion. Basically it's a fantasy roleplay game where the character can't level up, can gain few additional abilities that can allow them to do more stuff or empower them but overall they don't increase stats, health or such by playing normally. As such, the monsters have almost always the same health pool given how much they have to be strong. To give a comparison, an elven sorcerer has 6 hit points and a dagger deals 1d6 damage, and those health points do not change. This means a lot while deciding how and with what attack. For example, a wolf has 5 hit point and with a sword you deal 1d8 damage ( you roll an 8-sided dice to determine damage). There are few modifiers there, from -1 to +2 and the possibility to roll two dices to keep the highest depending on class, and that means that while wielding a sword you can oneshot wolves, even thought that is not certain. And unless the foe you are fighting is not a common wolf ( wolf chiefs exist that have 8 hp) the damage you deal has a lot of chance to kill it. Same applies to other weapons, from bows to daggers, that deal less damage but can still kill them on hit. Then there are two handed weapons which deal 1d10 damage, but occupy two hands and don't let you equip a shield. A shield is important in this game since it allows you to evade ranged attacks, so damage wise is not that good to deal few points of damage more only to be much more vulnerable. But having a 2-handed weapon dealing more damage allows you to kill immediately stronger enemies that otherwise would be much harder to kill, or impossible to do so. While facing trolls that have 20 hp you can defeat one in one turn if you manage to hit with two well-placed attacks together with your allies. The game has very quick combats with low health pools ( the strongest enemy in the game has 50 health points, which are MASSIVE compared to other foes but is not quite so intimidating), and since that applies to the player too it feels like always living on the edge and makes every fight not a trivial challenge.
Making common rewards in low-risk areas be almost negligible sounds like a good solution for RPGs. You could make EXP go down significantly if the encounter is much lower level than the player, or give equipment specific upgrade materials that means you'll need to go on to more difficult areas to find what you need to keep upgrading. You could also tie permanent character upgrades to story progress, like expanding Zelda's heart container system into different stats that you put limited progress and exploration based points into, balanced in a way that someone investing heavily into one thing feels very powerful in that way, but not so much that someone skipping all optional content would have no reasonable chance against the next area.
There's a problem with that calculator suggestion. In many cases, a brute force solution like that will quickly balloon execution time. When you have start getting a number of combinations the trillions or higher, it starts to become extremely difficult to do this.
Oh Disgaea... this game is quite the intentional break point paradise. The game encourages you to play tons of New Game Plus playthroughs and to level your characters to level 9999. I try to play the main campaign by minimizing item world dives but you can seriously do some level grinding that trivializes most of the story levels. Fortunately, many of the other game play elements such as Geo Blocks/Panels for creative map/panel bonuses and penalties add layers of strategy that can somewhat help control for break points.
I actually kinda like some of those break points. I'm aiming for stats/builds that let me attack first all the time. or once you are strong enough to one-shot everything, you can do a few quick, easy, satisfying victory rounds of one-shotting whole zones before you move on.
Maybe worth mentioning how rng can sometimes "break" breakpoins. Like if that characters damage was under even +-5% rng (not a lot compared to some games) the a level 4 you would only have a ~50% chance to one hit kill.
You know, a perfect example of when this goes right is the Paper Mario games. If you get to a high enough level, smaller enemies won't give out star points.
Shadow Fiend got something like this, shadowraze damage got upped ages ago and until it was patched he was completely overpowered since he could kill entire camps in just two razes as opposed to the previous three (eg it went up from 90 damage to 100 and the neutral creeps had 200hp)
Hey, I’d be very appreciative if you would talk about a topic that may sound strange at first... Kirby: Star Allies and the importance of player choice. Depending on how many friends you make playing the game, this could be the easiest game in the franchise, or the hardest. This, to me, is pretty impressive, and should be shared.
a fact you could do an episode about: *perception is logarithmic* a jump from 1 to 2 is 100%, a great deal, but from 1000 to 1001 only 0.001%. players feel any change in detail like this and it's hardwired.
Of course, where at all possible, make your maths describe the payer experience. In your admittedly simplistic example, you should not have looked at the linear progression of damage, but rather the jerky and unbalanced regression of hits-to-kill, which you can formulate as enemy health divided by player damage, rounded up. If possible, maths is better than testing. I’m not saying testing isn't always important, but if you can’t find the reason why your system doesn’t play well, any fix you make will be of the sticky plaster variety, rather than a true repair of the underlying problem.
Damage formulas are never just enemies health - my damage stat. Player lvl, attack stat, weapon stat, weapon type, elemental dmg, equipment bonus, enemy type, enemy trait, player trait, any triangle system, weapon affinity, lvl of weapon, player stance, special resistances, environmental modification. Enemies defense rank, enemies armor value, ALL active buffs and debuffs for ALL OF THESE THINGS independently... Calculators only work if you can input the exact values for everything in every encounter in every area. Games are too dynamic for math to efficiently replace testing. And if there's any rng forget about it completely.
+Paige Wilson While there are rich combat systems, generally player damage is fairly predictable. Blind calculation generally won't help, but intelligent calculation can replace a lot of hours of testing different combinations and seeing similar results. The question to ask is not "Does every combination produce a sensible result?" (though if it's at all feasible to do so, that should be addressed in testing at some point) but rather "What needs to happen for a crazy result to come out?" If you're looking for jumps in time to kill, then you can work out the threshold damage values for those jumps, and work backwards through your damage equation, looking at best-case and worst-case values for the various terms, to arrive at ranges for the underlying player stats where jumps could happen. Approaching things from a theoretical perspective rather than an experimental one should always be faster for identifying instances of a known and understood problem - the playtester's job is to find the unknown problems, and the places where what happens in practice isn't what you predicted in theory. If your damage formula can produce unforeseeable outcomes, then you're going to have a hell of a time both in design and in testing; if its effects are constrained, then you can do a mathematical analysis to calculate possible outcomes, plausible outcomes, the underlying distribution of outcomes, the probability of a level X character managing to one-hit-kill any given monster, etc. If it's practical for playtesters to test every possible combination, then it's practical for an automated tester to do so in less time.
The difference is that play testers aren't testing every possible combination, they're testing the experience, and what players encounter during the typical experience. Humans can't plug in x to find y as fast as computers can and computers can never transcend the math to analyze and test the more fundamentals experience they formulate. so it's not a matter of which is more efficient at the same task, but that neither can be neglected. And specifically im saying that no formula will ever be able to simulate the real values a player will be encountering in a dynamic game world to a degree that player testing isn't a necessity. Not even if you decide on an acceptable kill time and work the numbers until your probability of encountering a faster time are minimized. His approach of a simple formula, and your approach of a complex one are both insufficient methods of reliably smoothing the player experience itself, in other words. It shouldn't be used as a justification for underutilize the solutions presented in the video. That's not something that can be overturned simply by making the formula more predictable using probability, but at this point we already agree anyways.
Yeah, I think the point we're converging on is that playtesting is crucial for picking up problems - blind spots in your mathematical analysis because you didn't think to ask the right question - but once you are aware of a potential problem, you can (and probably should) update your theoretical analysis to make sure you know all the circumstances where it could happen, rather than relying on playtesters to discover them all.
This reminds me of Zelda: ALttP. You get the master sword, started dealing double damage and tear through enemies. As soon as you get to the dark world, the enemies hit much harder and take even longer to kill despite having a stronger weapon. At this point the enemies in the light world might only exist to make sure the player actually feels more powerful then before since the dark world enemies never gave you a chance to use your regular sword on them.
Your easy example is easy solved: Put the few Strong Attacks in many Weak attacks. You have more area to increase attacks without generating breackpoints. Not 1 attack per 10 seconds, make 10 attacks per 10 seconds. So you havnt 100 -> 120 -> 140... You will have 10 -> 12 -> 14. That mean you will need 15 Attacks, than 13, than 11, than 10...
Another thing you can try if your combat systems makes more sense with few hit is to slightly randomize the damage. Let’s say every attack does the set damage +-30%. Then that 100 damage attack now does 77-130 will takes 2 hits for the 150 hp enemy. The 120 damage attack now does 92-156 and will still mostly take 2 hits but now will 11% of the time one shot an enemy. The 140 damage attack does 108-182 and one shots the enemy 57% of the time, and so on. Now you have a gradual procession even though there was only two outcomes in this particular scenario, 1 hit or 2 hit.
I'd love to see an episode on this sort of effect in the Borderlands games. Specifically in borderlands 2 with the Bee shield and the raid bosses forcing a total rework of future DLC.
As someone who frequently plays Pokemon Nuzlockes, I tend to rely on the break points found in most early-game areas to speed up grinding. For instance, Viridian Forest is a pretty good grinding spot up through level 10, and even a bit higher if you get a lot of Metapod encounters. Since those Metapods become exponentially longer fights the weaker your pokemon are, it's much faster if your initial hit takes out a fair swath of health. The break point is when the first hit deals 60% of its HP. At that point, Harden no longer prevents the second hit from KO'ing, since it will only reduce that second hit to 40%. Since Metapods tend to give about ten times their level in EXP, they are actually just as efficient as most of the trainers between Pewter City and Mt. Moon, with much less danger of losing a Pokemon to a lucky crit (RIP Leonardo, best Wortortle). Then again, abnegation is one of the core appeals for me, so maybe its just me.
I really like break points. I think it is really rewarding when you can one shot. Enemies that until a moment ago took twice the effort. Or when your mana regen out paces your spells so you can cast continuously. I think it is rewarding because it gives a boost in power that is not always game breaking but makes you feel just powerful enough to give you that rush.
Sometimes you've got perfectly calculated systems running in your game, but actually playing the game feels... off. Let's learn how to design purposeful break points, and avoid unintentional loopholes!
Ayy
lmao
Shouldn't all those equations being made to run your game under the hood, be made WITH user experience in mind?
I think if your adding a layer of systems/stats to your game, or expanding an existing one, it should be done so for the sake of the user experience. What other reason would there even be to add depth if not to improve user experience?
You could have brought up Earthbound (Mother 2) and the way it automatically fights enemies so low you will (most likely) one shot and then gives you minimal exp and rewards.
Extra Credits Earthbound had an exp to level up curve that I'm sure seemed really janky on paper but was amazing in actual execution.
"They're not necessarily bugs, they might even be legitimate game features."
A story as old as time itself.
Testing yourself against a skyrim giant "weeeeeeeeeeeeeeeeeeee"
Yep, I too like it when broken mechanics and system limitations are touted as "difficulty".:P
And then a picture of a fucking spider....except that spiders aren't bugs, so that was just a dick move...
@@PikaLink91 spiders are bugs, they're just not insects.
@@yonatanbeer3475 No, spiders are arachnids....
I like how Earthbound handled break points; when you were significantly stronger than the monsters, they A) ran away from you on the world map, rather than towards you and B) if you did engage them you either got a guaranteed "green," engagement OR instantly killed them without transferring to the battle screen at all.
Still one of my favorite JRPG mechanics.
I'm pretty sure the running away from you has nothing to do with power, but with whether you've defeated the boss of that area. The latter is pretty cool, though, and very simple math, too, I think. I believe that happens when the game calculates that you would just win without getting harmed if you simply mashed attack. Of course, that only really works in a few games.
Nixitur It was both power lvl and/or defeating the boss that decided whether enemies ran from you. The system for instant wins was a little more complicated then an autofight sim, but functioned basically like that.
Nixitur It's both. Enemies will run away after you kill the boss to make it easier to escape the dungeon. In the overworld, it's dependent on level. It makes a lot of sense mechanically, and just plain feels good to see yourself so strong that enemies literally flee in terror.
Persona 5 actually does that too. Think they were inspired by Earthbound?
Skyrim is so awful for this. No damage is randomized so most damage upgrades are useless against most foes while some are substantial (3 hits to 2 hits, 2 hits to 1 hit). Also the armour system is garbage with each point being more valuable than the last because its flat %, and then everyone caps it out eventually so it stops mattering at all by mid game and everything is harmless until you fight monsters who can hurt you despite only doing 20% damage or whatever it is.
One thing that helps a ton with breakpoints is small randomization consider a game where you always do the same damage, when fighting a 10 hp foe, the only numbers that matter are 1, 2, 4, 5 and 10. But if attacks do X plus or minus 1 damage than the numbers that matter are 0, 1, 2, 3, 4, 5, 6, 9, 10, 11.
Another way to prevent breakpoints is with easier methods of doing a lesser result. XCOM uses this to great effect as a weapon upgrade of 4 damage to 6 damage might not change the number of shots to kill an 8 hp creature, but the 6 damage shot will leave it in range to be killed by a frag grenade (3 damage) that is a guaranteed hit. Xcom also uses +/- 1 damage and crits to eliminate breakpoints.
To be honest, I have the opposite take on this. Skyrim uses leveled enemies, which means that they're getting stronger as you do too (just slightly slower). This makes it feel like you're not improving at all. I agree that they should use your example of randomisation, but I think that they should tone down on the leveled enemies to have improvement that matters.
Definitely, (I highly suggest mods that remove the level scaling and add damage randomization btw). My main problem with skyrim progression is it basically doesn't exist until you reach a drastic point that completely trivializes everything. Also because the damage increasing perks are so substantial, as you level up you actually get a shrinking toolest to deal with encounters. At level 1, archery, all types of magic, one handed and two handed weapons are all viable paths. At higher levels your chosen skills have kept pace but everything else is now entirely useless. This is a worse problem in fallout 4 imo, which has really substantial damage scaling by perks on guns and actively punishes you for wanting a variety of weapons for a variety of situations.
I couldn't agree more. When I tried getting every skill to 100 (and I did, but I regret it), it was quite difficult leveling combat skills I didn't start with. Right now I'm trying a playthrough with such mods, definitely an improvement, albeit a lot more challenging.
Skyrim also has a problem with how enemies scale up, at least with destruction magic, you become impotent in the late game. Enemies scale up farther than your spells do once you hit that damage cap. If that was intentional to force you into using other tricks like alteration or illusion then maybe it gets a pass, but it doesn't feel intentional.
You can screw yourself over in Skyrim though if you do something funny with your build and hit a scaling level. And equipment can make a difference, though if you are just an amazing smith and enchanter, you can pretty much make equipment to play as any build and just change on the fly, thus still breaking everything.
it is however also possible to go too fare into randomised damage. Xcom is great, but does anyone play with damage roulette?
A lot of this comes from doing the *wrong* math. The most obvious example comes from calculating damage per second, rather than time to kill. You actually see this a lot on things that reduce incoming damage. Let's say you have a skill that has 10 ranks, and each rank reduces the damage you take by 1 each time you get hit. There's a general intuition that, since each rank decreases damage by the same amount, each rank must be equally good. Now, imagine that enemies hit for 10 damage. If you have 100 hp, the 1st rank you take in this skill increases the number of hits it takes to kill you from 10 to 12. Fair enough. The 9th rank you take increases the number of hits from 50 to 100. The 10th rank increases the number from 100 to infinity/undefined (you're invulnerable). The damage you take is decreasing linearly, but doing the math that actually matters, we can see that the time to kill is increasing at a staggering rate. There's an obvious breakpoint at rank 10, but we can find that, and the other unintended trends, if we do the right math.
Gandhi in civ1, when your break point is so epic that it makes Gandhi goes into a nuclear frenzy
El- Zee that’s more of an integer limit overflow bug than a break point
N.U.K.E.= Nuclear. Uranium. in Kansas. Energy.
This Allison Utterback is a Hot Fuzz fan? I can't tell.
cornetto for everybody!
Wot's wrong Danny? Never taken a shortcut before?
Who ISN'T a Hot Fuzz fan?
Plus Hot Fuzz references Break Point.
Jason Hardman Ah! That's the elusive missing link making the whole thing Hot Fuzz motif seem a little jarring to me. I get it now. Thanks!
It took me an embarrassing amount of time to realize that half of the images in here were Point Break references...
I didn't see Point Break so thx for telling me :D
Either Point Break references OR Hot Fuzz references WHICH itself referenced Point Break. One deep rabbit hole man...
No! Why would I fired my gun in the air and yelled "ahhh"?
Games: Where 10% of health are 50% of health.
Or to put it another way: The only hit point that matters is the last one.
"ABC Barbarian has a quadratically increased defense the lower hp he has?"
No more linear warriors, quadratic wizards. Quadratics for everyone!
When I look at the Flowey fight in Undertale, it feals like the defense increases exponentially. You lose half of the bar after like two hits but can take over9000 until you die.
Magic pixl, not that your set survivability has much to do with the video topic.
Actually yeah I think that was the case. Also you can only die to Photoshop Flowey's attacks, not the souls if memory serves.
Yep, I get a 1% extra damage passive in many games due to equipment, skills, whatever. I never feel a difference and even forget they exist. So I dread skill trees or equipment that have tons of those small buffs, because I know it will take so much time until it becomes relevant, basically at endgame or post game.
I'd be really curious to hear from a dev on that type of game what they were thinking, actually.
Sure, my fire-ball is now 5% stronger, and has plus 1% in radius. Whoop. The excitement is CLEARLY off the charts, right?
Heck, you want the near same system be as near exciting as plausible? Have a 25% and 5% increase for five points. Bam, done, the player actually feels the difference.
Mattias Svensson Some of them think that its exciting for the player to minmax and theorycraft a lot to get a small or, at most, medium edge. Never a large bonus, that would break the flow they want to force upon you.
In the end, its a type of mechanical railroading, they don't want to take options away, but they don't want you to use them , because they want to force you towards the "only right way to play". Like how DS3 very strongly discourages mage and hybrid builds because melee "is the only right way to play".
If given the choice I will definately take the option that changes how I play my character or the character interacts with the world over the incremental increase in whatever. I honestly don't understand the point of them in games aside from some sort of sense of obligation on the part of developers to make the player falsely get excited because the words "level up" have appeared on screen.
Edward Nigma Yep, I prefer how Nioh does it, skills cost a different number of points. You get melee ones which can be spent on the tree of any weapon, mostly in active skills. You can then customize which ones go to what command to make your own playstyle, while some skills are always available. And sometimes you get consumables that give extra points as a reward for a particularly difficult mission, which is pretty awesome.
Ninja and onmyo skills give you consumables that recharge at a checkpoint. But you have loadout points for each one and many skills have 3 tiers, higher tiers costing less points to equip. So when you unlock a tier 3 Lightning Shot, for example, you can get two more to have a total of 6 shots or use tiers 2&3 to keep the same number and free up some points for something else. This system is really easy to code and can be huge for build variety.
This is why i kinda dislike Path of Exile. The only exciting stuff are the major skills, the rest is just fodder along the way.
In StarCraft 2, we have that sort of things. They called critical upgrades, and on of the things that you learn to keep track of to compose your strategy. For example, a zergling with +1 armor takes 3 hits to kill by a zealot, instead of just 2. Given the attack speed of the zealot, this thing completely changes the dynamic of this match-up, and protoss learn not to go out if the opponent has +1 armor. The +1 armor also allows zergling to resist a baneling hit with just one hitpoint left, a huuuge change in the dynamic of the match-up.
So to complete your thing, in RTS, those breaking points can be (and often are) thought intentionnaly, both by players and designers, to mark turning points in a game, and also defining game stages. It encourages progression I think, since to counter a critical upgrade, you also need to tech up, which in StarCraft means building advanced building, which produce more high-tier units. It's a nice progression incentive.
One time I was playing StarCraft 1 (badly) and I sent a fleet of carriers against a terrain fleet. I had the the invincibility cheat on, but it still took a VERY long time to win because the enemy's battlecurisers had maxed out armor. The carrier drones have a lot of hits per second, but low damage, so a small damage reduction had a huge impact.
Now whenever I think of linear damage reduction, I think of those battlecruisers.
Totally wanted to say that, props for saying it first! Also break point occurs not only with upgrades but with unit numbers to (4 stalkers to one shot workers) and in some engagement it's win or lose the game if you did or didn't estimate the ennemy force well enough (on the same level of micro).
So many hot fuzz references, oh my god. XD And that "feels" face at the end was just perfect.
...Anyway, I get the feeling that Warframe takes Break Points and runs the whole thing into the endzone. I'm like "Man, this weapon sucks, I'm barely killing enemies with it" and then tweaking a few Mods on it makes it suddenly one-shot just about everything. I'm at the point in the game now where I need to actively look for break points because the game is now suddenly REALLY freakin' challenging and if I'm not trying to break the game, I'm not winning anymore. XD
Aegix Drakan just came down to say exactly this. For gamers who love the game of searching through systems for the next game breaking combo, warframe is where it’s at. Really it’s their entire gameplay loop imo.... and it’s a lot of fun to finally find that crucial keystone piece and have a challenge that’s been grinding you down just evaporate under the weight of some new stat combo.
this is true until you realize that there's like 3 useable endgame builds for guns and all you need to do is figure out which one fits on your current gun based on it's base stats, then forma-grind the thing until you can fit them on there
That's only true for some weapons and concepts. Outside of the usual crit/status calculations, there's also armor stripper builds, H-cal ranks, and weapon specific weapon alterations. For example, most weapons cannot reliably sustain a Heavy Caliber of 8 or more, while the Lenz can support it at 10 and the Soma cannot run it at all. Likewise, Rakta Cernos cannot use fire rate mods, while Dread and Lenz absolutely need them. Penta cannot use multishot mods, while the Tonkor loves multishot. The Stug cannot use punchthrough. So on and so forth. Yes, there are initial baselines to how you optimize your damage, but many weapons require you tailor the baselines to the specific weapon.
+AnemoneMeer So Dread needs Fire Rate, noted. I'm levelling that thing as soon as my Crit Soma is mastered.
Most bows use fire rate, due to the fact they have painfully slow inherent firing speeds due to charge time, and inherent punchthrough from projectile mechanics. The Cernos line is the exception, with a baseline 0.25s charge time, attempting to stack firing rate just makes it difficult to acquire ideal firing lines/heads without wasting time at full charge.
This is actually a pretty significant problem in Breath of the Wild. The way the armor boosts and weapon durability increases work, the way that a late-game Link fights can afford to be way less skillful than in the early game.
I personally prefer games that actually make me feel powerful once I've put in the time and effort. I don't mind late game challenges being introduced, but making me feel weaker late game is worse than making me feel more powerful. A perfect example is the original Everquest. Start any character/class combo. Level 1, you can rush out and kill anything with some work you run across in the starting zone. Of which you are fighting bats, rabbits, snakes, and little kobolds. Then over time you reach max level, go to the highest level zone in the game, and see that it is populated by the same exact looking bats, rabbits, snakes, and kobolds. The difference is there is literally no way you can solo anything in that zone anymore. Infact, you probably couldn't solo anything past level 6 anymore either, but now you need a huge party of people with the best min/max setup to take out the level 60 common bat monster that at level 1 you could beat with a few pokes of a rusty dagger.
That is a much worse feeling as a player. Many game have done those horrible masochistic like progression systems and those are just truly abhorrent game design.
T Blank
Yeah, that’s also a problem. And that’s something BOTW actually does really well because of how it allows you to piecemeal gain power as a reward for completing tasks (weapon drops from difficult enemies, gear from shrines and side quests, armor boosts from collecting things, etc.) But, in the end, the game leaves you extremely powerful and basically invulnerable to even difficult monsters. It’s one thing to feel powerful. But you can feel powerful and understand that there are perils involved. BOTW does that very well by giving you the challenge straight up and then allowing you to grow and meet it. But there’s still a late-game problem with this design. That said, I don’t really think it’s fair to call it a major criticism because it only hits after you’ve put in maybe 30 or 40 hours into an incredible game.
Is that due to break points or due to simple progression? Yes, breath of the wild has a negative difficulty curve - no dispute there. But that boils down to giving the player steadly acces to stronger tools, the player having more experience and the enemies only slightly increasing in difficulty. Yes, a end game Link against an end game enemie is stronger (numerically) then a starting link against a starting Monster - but the change is gradual. Every heart helps - and yes the first extra hearth is more important then the last two. But there is no distinct break point. The master sword is a big boost - but that is intentional and is typically used as an axt and rock smasher as other weapons are numerically stronger and more reliable.
Adnan Ilyas
Agreed. I find the early game much more rewarding and exciting than mid-to-late. Especially when it's so easy to sell gems and stuff to buy good armor and get Link out of his early-game vulnerability
I agree. It does take some of the edge off when you're eventually hunting Lynels for sport! But I feel it's also rewarding considering I was terrified to even engage them for the first 50 hours or so.
Break points should be graphed out too. Functions like div, floor can be very useful for that.
Examples like toe one in the video definitely, but it isn't always that simple.
Say you make an RPG with powerful monsters that attack you when you get in range.
So you map a break point where you become fast enough to kite certain monsters to infinity. Fair enough. What you didn't expect is that just getting the monsters into certain terrains can be worth most of your health.
You can't graph break points you don't know are possible. And writing functions for every single possibility for say a card game can be more effort than you can spare.
WHERE ARE YOUR PRECIOUS SPREADSHEETS NOW, EMMONS!?
I've been balancing my game by play testing it and always noticing these breakpoints and making sure they didn't hurt the gameplay, but I never knew there was an actual name for such a thing. Great video as always!
0/10 not enough Point Break jokes
Hmm, this might explain why I find Pokemon games to always suddenly get a lot easier at around level 25-30.
Personally I think Pokemon has a MASSIVE issue with Breakpoints. The hardest part of most Pokemon games is the beginning upto the second/third gym, but by then you've got access to mons (Most of which are at least second stage whilst the game's trainers still mostly use first stage Pokemon) and moves of most types, with said moves with enough power to OHKO anything they hit super effectively, meaning much of the Pokemon games is simply matching the mon you're facing to a move of a super effective type and watching their health bar plummet in one hit,
The thing that actually destroys pokemon is the fact that exp share is considered a standard feature that causes the player to naturally level up faster than the wild pokemon and trainers in the respective areas.
That, and the fact that exp from fights doesn't decay or rise depending on the level gap between you and your opponent make for an rpg experience that is very fun if you're looking for something simple, but grossly unrewarding as a single player experience if you're looking for an engaging challenge. This is to say nothing about pvp which is one of the most splendidly complex games ever constructed, but the base games are very far from their JRPG roots.
I always turn exp. share off as soon as I get it. It is bizarre that it's activated by default when it completely breaks the pacing of the game that Game Freak worked hard to build. It is useful for if I need to grind quickly (I hate grinding so the extra exp is welcome)
I gotta say, I love Pokemon's pvp metagame, I think it's incredibly fun, but the in-game always suffers from pacing issues for me.
I just don't understand why the exp share now goes to all of your pokemon, instead of it being a held item that only gives the pokemon that holds it exp. That was a good system, as you could now use a wild pokemon you got, as they'r generally very weak compared to the trainers in the area, making it difficult to train them in a reasonable amount of time.
Love these technical videos, Extra Credits! The ones that really get you thinking about problems you haven't thought about before!
I know I have encountered this in many games, but I can't recall which ones. I just know I felt an overwhelming "Oh yeah, THAT!" while watching.
This was a really really really good episode of Extra Credits. I think I've watched every main-series episode and this felt like a return to form. I'd never thought about this concept in practice. I've often wondered in a game, "is this supposed to be this easy?" and the concept of Break Points neatly answers that with, "no man OK! We might have missed that." this was really cool stuff. Great work guys. :D
It was for this very reason that in this game me and my friend were designing, we made the base health 21, because there's a new breakpoint for every number from 1 to 7, so there's a significant change if you alter the damage you do.
When I used to play Payday 2 (before the weapon overhaul) I remember most of the talk about weapon strategy was about hitting damage breakpoints. When choosing a gun and modding it to be optimal, you had to try to hit certain damage values that would let you, say, 1-headshot a normal SWAT or 2-bodyshot an armored cop. Any damage over each breakpoint wasn't important, so you could swap out some damage mods for other stats once you got to that point.
I remember there was also this bit about "fake 40" where you needed to hit 40 damage for a breakpoint, but the game displayed 40 when it meant 39.95 or something (which wasn't enough). That was a weird bit of math that made the game a little more difficult to figure out.
Also accuracy in that game was on some weird logarithmic scale where 18 was perfect accuracy and every 2 points was way better than the previous 2 points. Payday math was weird.
i think part of the experience that i love most about rougelikes and sandboxes is the exploration of possibilitiy spaces and find ridiculous builds. it's a fine line between break point optimization and exploits. the skyrim necromancer build with a 100+ undeads followers maybe gamebreaking, but is fun as hell.
Love the "Hot Fuzz" reference(s) :)
I am so glad this came out right now...
I'm currently working on a combat system for my role playing guild in wow and have spent the past 2 days trying to balance out how a bunch of different classes can do about the same amount of damage with different abilities.
It's been a lot of work so far but i think it's time i put the excel ark aside and actually put this thing to the test if i can just convince my guild master to approve the system.
Skyrim's break point: getting lvl 100 enchanting. After that, nothing else matters anymore.
Skyrims break point: getting a bow.
from what i can think you should look at systems that give you a reward instantly. like health. you can be busy 5 min to kill something but you will only experience the reward if the final health is gone. there is no reward for killing something halfway. (in classic health systems) buy systems can work the same. doesnt matter how much gold you have if you cant buy anything. that final gold might make the biggest difference.
I don't know you, but that beginning statement is the first time I've heard someone take that stance. As far as I have seen, every player is excited and happy that they can one shoot every enemy, in Dark Souls for example, when I can one shoot everything in the Undead Burg for the first time it feels like I've progressed, all the time spent there that felt like just traversing the land was actually my character improving, and that's rewarding, it makes me feel I'm moving forward
One thing you can do to soften up break points is to add some randomness to your game like crits and damage spread, so that (in the example the video gave) you can't one shot a mob at level one, you can with a crit at level two, you can with a high damage non-crit at level three and then you always one shot at four.
That's a big decision that leads to a completely different game. Its not a solution, it just means break points are less important to your game if youre already doing that anyways. And with an rng system, break points are harder to control, not easier. If there's even a 1% chance that you 1 hit kill the first boss, thousands of players are going to have a bad experience. If your formula is broken "lets just randomize that" is almost never the right answer.
I was going to say the same thing, there's a reason D&D has dice.
In Biology, technically 1+1=3 !
Sometimes 1 = 2
Or 1 = 4
In Chemistry or physics it can often be 11.
No, nowhere in nature does 1 + 1 = 3. you get 1+1 + food +energy + water + time = 3
Also in maths you get 1+2+3+4+..... = -1/12
NOW Thats a break point
That one took me a moment. My mind is way too innocent...
That's why as player doing PVP I love special abilities in stat heavy games. They are can be combined in ways the devs never figured for frustrating (for my opponents) results. I used to do this particularly in WOW and just crush people in 49 pvp at lvl 42.
mixing and matching weapons with armor with character skills is why I still play Borderlands 2. that game seems to avoid the break point issue by dramatically ramping up difficulty in each new area, and shuttling the player along by limiting how many side quests are available. which is great! creating this sick melee build Zero is so great because that build is dependent on the quality of shield and the power of the other weapons. the game constantly getting harder means you are constantly hunting for new gear, which means constantly tinkering with your character. which means i am constantly coming back to that game
Looks like someone forgot to give Allison the memo: Hecks should always cause Sneks.
Another great example of this from the game Duelyst is that there is little difference between a minion with 1 health and a minion with 2 health, but *all* the difference in the world between 2 and 3 health. This is because duelyst heroes all start with a base 2 attack (like a permanent 2 power weapon to compare with hearthstone). 3 health is the breakpoint where a general cannot remove a minion instantly any more.
The hot fuzz bits were on point big props to the artist and editor for that
I think dark souls was one of the first games that made me really think about particular break points, and is a good example of them done right. By the time you have leveled up and learned an area well enough to traverse it quickly and kill all the Dudes, you have ascended the rewards in that area usually. It was one of the first games where I got actually thinking about how many hits an enemy takes to kill.
When does 1 + 1 =/= 2? In a field of characteristic 2
For a big example of this look at the balancing in Smash games where even making a move do not even 1 more damage even like 0.1 more damage makes an insane difference hence why teir lists are made, the character that has more very slight unbalance adds up and sometimes the Smash developers overlook certain minor exploits of the system that create things such as wave dashing, ledge canceling attacks, melee Fox and Falcos' amazing reflector or "shine", ladder combos for a large variety of characters some being much easier than others, rage making certain attacks way to rewarding on hit due to how much difference the addition knock back makes and also making some weak and or fast moves kill way sooner than even the development team expected, also look at how much stale moves can effect a match in Smash as a series to prevent spam...but then there's unintentional infinite combos, chai grabbing to death and "wobbling" with Ice Climbers in Brawl and Melee respectively....the difference of super armor vs heavy armor vs neither and much more is what makes certain match ups *very* different and for the most part very dynamic actually...
I've got to say, That music track you put at the end of this episode led me on to play one of the best games I've ever played.
As an engineer I gotta say - the example of linear progression there is because its not balanced even mathmatically. You're increasing input linearly below a saturation limit so the increase has 0 value until a new saturation point is obtained. This is a case of bad understanding of math in systems.
There is visual feedback though. The player knows they are getting slightly stronger, even if it's only until level 4 (in the example shown in the video) that it actually makes a difference. I think being able to see the increase in your damage serves two purposes: A) It teaches you that you are getting stronger (yes, the first time playing an RPG you don't know that leveling up = you get stronger). B) It sets a goal for you and motivates you to push for that goal.
yomojoflo Only if the game tells the player all the numbers and how they are used.
I agree
jonasn5 Thank you! I'm a mathematician, and I cringed everytime he said the math is not enough. It's not that the math isn't enough, it's that you aren't looking at the proper math!
You just re-explained his point to him. The point is that it isn't mathematically balanced and it is easy to see the "saturation" points in his example which is what he wants.
The point is in a real situation these imbalances are usually much harder to see. In addition, changing a number to fix one instance frequently imbalances another instance. RTS, RPG, MOBA, and TCG games all have a very hard time with this.
Edit: in theory the math alone is enough to balance a game but this is similar to telling a programmer that in theory a version of a program can be made that is without bugs. In theory it is possible, in reality the problem is so vast and complex that rarely can every issue be fixed and not spawn a new one unless the problem is of trivial difficulty in the first place such as his example.
I love it in D&D where that static upgrade of stats means a bizzare progression.
Like think about Base Attack Bonus for worriors. You start with +1, at level 2 you have a +2 and at lvl 3 you have a +3. That means at lvl 2 you base attack becomes 100% better but leveling up to level 3 it has only grown by 50%.
The same issues happens in D&D with HP. Getting to level 2 and suddenly you are about twice as compatent as you used to be.
This is all over the place in RTSes, especially games like AoE and StarCraft where damage and health values per unit are relatively small. Certain upgrades will be a top priority in certain matchups, as +1 armor allows Unit A to survive three hits from Unit B instead of only two.
I gotta wonder: exactly how much of these "classic" break points are due to the fact that most player characters and enemies still do the same damage when they're only at 1 HP vs. 100 HP? Granted, if the enemy is being one-shot it doesn't really matter, but I feel like because the difference between 1 HP and 0 HP is so suddenly drastic (the classic "critical existence failure"), it's probably emphasizing math that wouldn't normally be emphasized.
In Suikoden: Teirkeries the game has you earn less and less experience in an area the higher level you are for that area. You always level up at 1000 experience but when you enter a new area you may start leveling up every 5-10 fights but after a while that amount increases due to no longer getting as much experience. This does help since you get 108 party members so you can level up new level 1 members quickly.
Stellar pick for ending music cave story is a true classic. Thanks for the quality content, I hope you have a good day :)
I've found a recent example in FighterZ. Practicing a combo that deals 50% of the opponents health is very important, as it allows you to "two touch" them, killing their character in two combos. Getting a combo with higher damage, however, isn't really useful, as you're still gonna have to two touch them.
For people that try to find break points in a game its always a good idea to figure out how the game uses its stats and skills. As an example, in diablo 3 socketing a high lvl red gem can lead to major power boost in the early game, however the combo of multipliers on set and legendary items with crit and already high base weapon damage makes the green gems far more powerful at max lvl.
I like the jam at the end. I don't know when they changed to that, but I just now noticed it.
They put a different end song in every video, so in the next one they'll use another one
It sounds like a remix from Cave Story
Checked the description. It is Cave Story
I feel like Dark Souls avoids this trap well, you can grind all you like, but by the time you can start one-shotting, you're so overleveled to the point of all soul drops being borderline meaningless (Also shoutout to that Cave Story music at the end!)
That is true, though I do think it's countered by the fact that you have limited spell uses, attunement slots, and in the case of Dark Souls III, Mana points so that you can't use it 100% of the time, and then by the time you can, you're probably way overleveled, or a glass cannon.
Sort of.
If you're fighting trash mobs throughout the level, you want to make sure you're being efficient, so you usually want to have just enough damage to do so efficiently, so the rest can be spent on other things, like defense.
For example, in DS3, I needed to farm a ton of Blue Ears off some knights, using a pyro character. Assuming you use spell "X", you can boost your power by like 50%, but you only need a 30% power boost for X to 1 shot. So you grab just enough to hit that 30%, then you put everything else into other useful things, like defense, SP, increased item drop rate, increased soul gain, or, most importantly, improved fashion.
It's a very classical breakpoint, and you can make similar examples with other situations (e.g. how much +atk stuff do I need to 1 or 2 shot some annoying mob, instead of 3 or 4 shotting it).
I think Darksouls have tons of breakpoints, but non of them matter long term. Getting a new weapon can help you drastically. the homing soul mass and soul spear spells are more damage then your old spells at the same speed of the fast but weak one. non of these matter though. If you cant dodge an incoming attack, it doesn't matter you have a +15 great axe. The spells are pointless if you cant get to a safe spot and cast them. The game have breakpoints, but it player skill that matter in the end.
I think that the damage example can be fixed by having multiple enemies with different amounts of health. For example: In the binding of Isaac, a +0.5 to damage doesn't change the amount of tears it takes to kill most enemies. However, you can now kill the dips (tiny poops) in one shot. It is also more useful against bosses where you can stack this bonus. This allows constant break points for a powerful experience.
I appreciate all the Hot Fuzz references here. ^^ Great vid.
Okay the start of this episode got a smile and a chuckle out me, I wasn't expecting that sudden change in tone lmao
Omg thank you, been designing an rpg for the past few years and my biggest problem was the power spike. After watching this video it gave me an great idea for a power spike I can add later in the development
To expand on the DnD example: In 5th Edition, the best example of a power spike is 5th level. I'm no game designer, but for any game designers who want a jumping off point on how to make a power spike, it seems like a pretty easy to understand example (considering even most incredibly casual players can see the power spike incoming at that level).
As a roleplayer at heart, I can say that I have run into my fair share of these breakpoints.
How ever, I can not say that I have ever been mispleased by doing that extra bit of damage, even though it does not equal one less hit!
This episode could have been shown with the old Zerg versus Protoss matchup in starcraft 1.
an unupgraded Protoss Zealot takes 3 hits to kill a Zergling. with one attack upgrade it becomes 2 Hits. A defense upgrade of the Zerg turns it back to 3 Hits. This is not just a balance point between 2 units, but the breakpoint of the whole matchup. It isn't just a breakpoint between 2 units, but the whole matchup hinges on these units. On the other hand it creates a strong theme of a race of upgrades or going to new kinds of units which makes these breakpoints obsolete.
Great video as always! Also the timing you released this video is so right for me because I'm currently loving Diablo 3 (4 years after when it was first released, yes) but what you're saying in the video is exactly what I feel when I play it. But the original skill and runes system creates endless possibilities for builds, plus there is also the absurd amount of increasing difficulties so I basically never got bored playing it, like hundreds of hours now! Like I said, awesome video!
I love "putting the work in" and getting a beefy feeling of noticeable bump as a reward, and don't like when everything is so linear that I feel that my planning and thought "never paid out".
This episode’s art and visual aides were great
There were several moments where I could tell a pop culture reference was going over my head.
These often become part of metagames in competitive strategy games. In Starcraft 2, the upgrades for weapon damage for units are usually +1 damage per attack. In a Protoss vs. Zerg match, if the Protoss player gets +1 Attack his Zealots can go from 3-shotting Zerglings to 2-shotting Zerglings, which means that a single Zealot can handily defeat 4 Zerglings instead of dying to them (they cost the same).
This doesn't work in reverse though, where +1 Armour for the Zerg doesn't mean 4-shots instead of 3 - so the Zerg player needs to watch for that upgrade by the Protoss.
i think Chrono Cross on the PSOne did this quite well, for anyone who never played it, instead of standard levels the game used what are called "Star Levels" which only went up after beating certain bosses, the battles in between those bosses would increase your players stats, but only up to a certain point depending on your current Star Level, after awhile battles stopped giving you stat increases until you increased to your next Star Level, so the game kept you from feeling over/under powered through the game cause your characters progress is intentionally locked until more progress is made in the game, this was done very well in my opinion
test early test often (it's cheaper that way in the long run)
I noticed a break point when I was playing BioShock 1. At a certain point in the game you will unlock the crossbow. According to the BioShock wiki, if you perform a headshot to a regular enemy it will have a 1000% damage output. Also, there's a upgrade you can get that reduces the chances of the bolts from being destroyed upon fire. Combine this with the Electro Bolt plasmid (Stuns Enemies) and Eve Saver Tonic it makes the player incredibly powerful. So, essentially the player can easily breeze through the game with little or no difficulties with this specific build and strategy.
This ties into an idea I've had regarding what I like to call "damage races." One example is this: say you have two weapons in your game, each does 100 damage per second. One of those weapons does all of that damage in one hit, and takes 1 second between each hit. The other hits 10 times in one second, each hit dealing 10 damage. Despite having the exact same DPS, the slower, more powerful weapon is better in every situation. See the Deagle and AWP in CS:GO for one example. There's a similar thing with offense as defense. Say a game gives you a choice between a big, high-damage two-handed sword and a lower-damage one-handed sword you can use with a shield. In many games you see that doing more damage and killing the enemy faster actually avoids more damage than the higher defense offered by the lower damage option, rendering the lower damage option obsolete. I would like to see an exploration of all of the different methods game designers use to overcome these balance challenges.
You know what game popped into my head when I heard your part on how designers can know when a break point will happen and design the game to then give trivial loot/rewards to signal that you must go on to the next area? Paper Mario!
Paper Mario definitely does this when you level up to the point where the enemies in an area become super easy to beat and aren't a challenge anymore. You always need 100 star points (experience basically) to level up, but once an area becomes trivial to you, all enemy fights only give one star point. It's way too inefficient to try and train that way, when the next area can have fights give you 5 to 10 star points. Great example of knowing where your break points are. Especially since the numbers used in that game are all low values so each change in number is a bigger deal.
Vermintide 1 actually had a lot of these that were very intersting to observe, they were all based on Weapon Damage and difficulty level (since difficulty level effected enemy HP).
The 1h Axe could only hit one enemy, but also instantly killed small and medium sized enemies. Suddenly though, on the highest difficulty, the 1h Axe can no longer one-hit medium sized enemies. When the game has you fighting 6-30 enemies at a time, going up from 1 hit to 2 hit is HUGE and felt terrible. Thus the 1h Axe was rarely used on this difficulty.
Thanks Extra Credits. I've seen you mention 'break points' in other videos, so it's great to see a video specifically about that. It's very interesting to learn about 'game feel'. It's something I'm aiming to focus on in the game I'm working towards making... but I'm only at the 'block breaker' module of my game course at the moment, haha. Oh well. All big things have small beginnings. Thanks for all your videos. Looking forward to your future content.
Peace
I love you guys for throwing in a reference to Disgaea in there.
The example that comes to mind off the top of my head is phantom particles moving the opposite direction of their momentum. A concept that still makes me think "wait.... what!?!"
The art in this one is a return to form. I'd forgotten that your videos used to be full of constant funny little jokes. Please reincorporate them back into your main artists' work
I’ve since dropped this project, but I ended up creating a negative break point on one of my RPG Maker games. Basically, by the midway point, the basic healing spell became pretty much worthless...and because I hadn’t anticipated this, the healer still would’ve had quite a few levels to go to learn the second level healing spell.
Well now I need to watch Point Break again. THANKS A LOT.
No, really. I love Point Break. Thanks for getting me to watch it again.
I remember in Legend of Dragoon, the first boss fight against Fruegel kind of depends on the player to not be very good with the attack items yet to play as intended, or at least expects as much. If you get really good at maximizing damage output with the attack items (their strength is determined by how good you are at rapid input sequences- which there is legitimate technique to), the fight would be too easy since you could kill him in like 2 turns. Game developers anticipated this and made it so that if you deal too much damage to him in one turn, he uses an item he otherwise doesn't use to double his offensive and defensive power (like a deliberate breakpoint countermeasure)- which makes the fight too hard, thus it mostly serves as a punishment/discouragement for the player trying to be too strong too early. Sure, you could whittle him down and then use a maximized attack item to finish him off before he can use the boost item, but that would require keeping track of all the damage you dealt up to that point with everyone and you'd be within a couple turns of finishing him off anyway, not to mention that successfully executed normal attack combos make your attacks stronger- a mechanic the game kind of depends on occurring at a certain rate for its difficulty curve, you can't really afford to miss opportunities to grow it by cutting your turn count short. Unless of course you grind, but that's not fun.
Celeste came out recently and the creator had a lecture about using level design to teach mechanics instead of tutorials. Really interesting lecture, I think you can make a decent episode out of that idea.
You could also smooth out the break point ramp, e.g. by introducing a "critical HP" stat to enemies at which they perform a lot worse in battle. (Too wounded to fight.)
LV 1: Enemy takes two hits to kill.
LV 2: Enemy still takes two hits to kill, but the first hit is enough to knock the enemy down to critical HP.
LV 3: Enemy takes just one hit to kill.
LV 4: Still one hit, but now the loot and EXP become worthless.
Now there are more break points, but their relative impact to each other is less dramatic, and you get rid of those "didn't do anything" level-ups.
There is this tabletop rpg in Italy called The Last Torch that works a lot on breakpoints, given obviously that tabletop rpgs have much different feel from videogame rpgs. And it does it in a good way, in my opinion.
Basically it's a fantasy roleplay game where the character can't level up, can gain few additional abilities that can allow them to do more stuff or empower them but overall they don't increase stats, health or such by playing normally. As such, the monsters have almost always the same health pool given how much they have to be strong. To give a comparison, an elven sorcerer has 6 hit points and a dagger deals 1d6 damage, and those health points do not change.
This means a lot while deciding how and with what attack. For example, a wolf has 5 hit point and with a sword you deal 1d8 damage ( you roll an 8-sided dice to determine damage). There are few modifiers there, from -1 to +2 and the possibility to roll two dices to keep the highest depending on class, and that means that while wielding a sword you can oneshot wolves, even thought that is not certain. And unless the foe you are fighting is not a common wolf ( wolf chiefs exist that have 8 hp) the damage you deal has a lot of chance to kill it. Same applies to other weapons, from bows to daggers, that deal less damage but can still kill them on hit.
Then there are two handed weapons which deal 1d10 damage, but occupy two hands and don't let you equip a shield. A shield is important in this game since it allows you to evade ranged attacks, so damage wise is not that good to deal few points of damage more only to be much more vulnerable. But having a 2-handed weapon dealing more damage allows you to kill immediately stronger enemies that otherwise would be much harder to kill, or impossible to do so. While facing trolls that have 20 hp you can defeat one in one turn if you manage to hit with two well-placed attacks together with your allies.
The game has very quick combats with low health pools ( the strongest enemy in the game has 50 health points, which are MASSIVE compared to other foes but is not quite so intimidating), and since that applies to the player too it feels like always living on the edge and makes every fight not a trivial challenge.
Making common rewards in low-risk areas be almost negligible sounds like a good solution for RPGs. You could make EXP go down significantly if the encounter is much lower level than the player, or give equipment specific upgrade materials that means you'll need to go on to more difficult areas to find what you need to keep upgrading.
You could also tie permanent character upgrades to story progress, like expanding Zelda's heart container system into different stats that you put limited progress and exploration based points into, balanced in a way that someone investing heavily into one thing feels very powerful in that way, but not so much that someone skipping all optional content would have no reasonable chance against the next area.
There's a problem with that calculator suggestion. In many cases, a brute force solution like that will quickly balloon execution time. When you have start getting a number of combinations the trillions or higher, it starts to become extremely difficult to do this.
Oh Disgaea... this game is quite the intentional break point paradise. The game encourages you to play tons of New Game Plus playthroughs and to level your characters to level 9999.
I try to play the main campaign by minimizing item world dives but you can seriously do some level grinding that trivializes most of the story levels. Fortunately, many of the other game play elements such as Geo Blocks/Panels for creative map/panel bonuses and penalties add layers of strategy that can somewhat help control for break points.
I actually kinda like some of those break points. I'm aiming for stats/builds that let me attack first all the time. or once you are strong enough to one-shot everything, you can do a few quick, easy, satisfying victory rounds of one-shotting whole zones before you move on.
Nice Cave Story theme at the end card. Not sure if it's at all remixed, but I recognized it immediately :)
I can tell Allison had a lot of fun with this
Maybe worth mentioning how rng can sometimes "break" breakpoins.
Like if that characters damage was under even +-5% rng (not a lot compared to some games) the a level 4 you would only have a ~50% chance to one hit kill.
While there would still be situations where you one hit at a level lower or two hit at a higher level than expected (albeit rarely)
You know, a perfect example of when this goes right is the Paper Mario games. If you get to a high enough level, smaller enemies won't give out star points.
Shadow Fiend got something like this, shadowraze damage got upped ages ago and until it was patched he was completely overpowered since he could kill entire camps in just two razes as opposed to the previous three (eg it went up from 90 damage to 100 and the neutral creeps had 200hp)
Hey, I’d be very appreciative if you would talk about a topic that may sound strange at first...
Kirby: Star Allies and the importance of player choice. Depending on how many friends you
make playing the game, this could be the easiest game in the franchise, or the hardest. This, to me, is pretty impressive, and should be shared.
a fact you could do an episode about: *perception is logarithmic*
a jump from 1 to 2 is 100%, a great deal, but from 1000 to 1001 only 0.001%. players feel any change in detail like this and it's hardwired.
not only a great episode but those references were... on point!
Of course, where at all possible, make your maths describe the payer experience. In your admittedly simplistic example, you should not have looked at the linear progression of damage, but rather the jerky and unbalanced regression of hits-to-kill, which you can formulate as enemy health divided by player damage, rounded up.
If possible, maths is better than testing. I’m not saying testing isn't always important, but if you can’t find the reason why your system doesn’t play well, any fix you make will be of the sticky plaster variety, rather than a true repair of the underlying problem.
Damage formulas are never just enemies health - my damage stat.
Player lvl, attack stat, weapon stat, weapon type, elemental dmg, equipment bonus, enemy type, enemy trait, player trait, any triangle system, weapon affinity, lvl of weapon, player stance, special resistances, environmental modification. Enemies defense rank, enemies armor value, ALL active buffs and debuffs for ALL OF THESE THINGS independently...
Calculators only work if you can input the exact values for everything in every encounter in every area. Games are too dynamic for math to efficiently replace testing. And if there's any rng forget about it completely.
+Paige Wilson
While there are rich combat systems, generally player damage is fairly predictable. Blind calculation generally won't help, but intelligent calculation can replace a lot of hours of testing different combinations and seeing similar results.
The question to ask is not "Does every combination produce a sensible result?" (though if it's at all feasible to do so, that should be addressed in testing at some point) but rather "What needs to happen for a crazy result to come out?" If you're looking for jumps in time to kill, then you can work out the threshold damage values for those jumps, and work backwards through your damage equation, looking at best-case and worst-case values for the various terms, to arrive at ranges for the underlying player stats where jumps could happen.
Approaching things from a theoretical perspective rather than an experimental one should always be faster for identifying instances of a known and understood problem - the playtester's job is to find the unknown problems, and the places where what happens in practice isn't what you predicted in theory.
If your damage formula can produce unforeseeable outcomes, then you're going to have a hell of a time both in design and in testing; if its effects are constrained, then you can do a mathematical analysis to calculate possible outcomes, plausible outcomes, the underlying distribution of outcomes, the probability of a level X character managing to one-hit-kill any given monster, etc.
If it's practical for playtesters to test every possible combination, then it's practical for an automated tester to do so in less time.
The difference is that play testers aren't testing every possible combination, they're testing the experience, and what players encounter during the typical experience. Humans can't plug in x to find y as fast as computers can and computers can never transcend the math to analyze and test the more fundamentals experience they formulate. so it's not a matter of which is more efficient at the same task, but that neither can be neglected. And specifically im saying that no formula will ever be able to simulate the real values a player will be encountering in a dynamic game world to a degree that player testing isn't a necessity. Not even if you decide on an acceptable kill time and work the numbers until your probability of encountering a faster time are minimized.
His approach of a simple formula, and your approach of a complex one are both insufficient methods of reliably smoothing the player experience itself, in other words. It shouldn't be used as a justification for underutilize the solutions presented in the video. That's not something that can be overturned simply by making the formula more predictable using probability, but at this point we already agree anyways.
Yeah, I think the point we're converging on is that playtesting is crucial for picking up problems - blind spots in your mathematical analysis because you didn't think to ask the right question - but once you are aware of a potential problem, you can (and probably should) update your theoretical analysis to make sure you know all the circumstances where it could happen, rather than relying on playtesters to discover them all.
That block with a hole in its chest is my spirit animal.
This reminds me of Zelda: ALttP. You get the master sword, started dealing double damage and tear through enemies. As soon as you get to the dark world, the enemies hit much harder and take even longer to kill despite having a stronger weapon.
At this point the enemies in the light world might only exist to make sure the player actually feels more powerful then before since the dark world enemies never gave you a chance to use your regular sword on them.
Your easy example is easy solved:
Put the few Strong Attacks in many Weak attacks. You have more area to increase attacks without generating breackpoints.
Not 1 attack per 10 seconds, make 10 attacks per 10 seconds. So you havnt 100 -> 120 -> 140... You will have 10 -> 12 -> 14. That mean you will need 15 Attacks, than 13, than 11, than 10...
Another thing you can try if your combat systems makes more sense with few hit is to slightly randomize the damage. Let’s say every attack does the set damage +-30%.
Then that 100 damage attack now does 77-130 will takes 2 hits for the 150 hp enemy. The 120 damage attack now does 92-156 and will still mostly take 2 hits but now will 11% of the time one shot an enemy. The 140 damage attack does 108-182 and one shots the enemy 57% of the time, and so on.
Now you have a gradual procession even though there was only two outcomes in this particular scenario, 1 hit or 2 hit.
I'd love to see an episode on this sort of effect in the Borderlands games. Specifically in borderlands 2 with the Bee shield and the raid bosses forcing a total rework of future DLC.
Finding these break points are all for the greater good
OMG another EC episode. I waited too long for this.
Great timing on this topic! I've just started play-testing a pirate ship based card game. Thanks for the info!
As someone who frequently plays Pokemon Nuzlockes, I tend to rely on the break points found in most early-game areas to speed up grinding. For instance, Viridian Forest is a pretty good grinding spot up through level 10, and even a bit higher if you get a lot of Metapod encounters. Since those Metapods become exponentially longer fights the weaker your pokemon are, it's much faster if your initial hit takes out a fair swath of health. The break point is when the first hit deals 60% of its HP. At that point, Harden no longer prevents the second hit from KO'ing, since it will only reduce that second hit to 40%. Since Metapods tend to give about ten times their level in EXP, they are actually just as efficient as most of the trainers between Pewter City and Mt. Moon, with much less danger of losing a Pokemon to a lucky crit (RIP Leonardo, best Wortortle).
Then again, abnegation is one of the core appeals for me, so maybe its just me.
I really like break points. I think it is really rewarding when you can one shot. Enemies that until a moment ago took twice the effort. Or when your mana regen out paces your spells so you can cast continuously. I think it is rewarding because it gives a boost in power that is not always game breaking but makes you feel just powerful enough to give you that rush.
Wooohooo!As much I appreciate Extra History and watch it, Ilove your Game Design videos even more!