Scaling Enemies Without a Traditional Level System

[ZombieKidzRule]

Ignoring the significant amount of...off topic stuff in this thread, I have found a few nuggets of interesting points to ponder.

Spoiler: Slightly Off Topic Opinion - Feel Free to Ignore

What I personally find frustrating is when a post presents subjective opinion or perspective as objective, indisputable, pure fact. Which is unfortunate because just a few simple changes to wording can vastly improve the potential for the poster's message and point to be accepted and received by the audience. When you speak in a manner that insinuates that your personal opinion or perspective is an indisputable fact that the average person or any reasonable person would naturally agree with, then many folks will either disregard what you say, find that you lack credibility, or they will fall into the trap of arguing with you over the merits of your opinions couched as facts.

There have been so many times that I have typed responses to posts just like that and then deleted them. Why? Because the back and forth interactions of posting at each other is rarely productive. So I generally give myself a time out from the forum for a bit.

Anyway...one of the nuggets of interesting points that recent posts have prompted me to consider is that perhaps it doesn't make sense to allow the player to control their level ups. I know that many games don't actually do that, but I tend to prefer, as a player, to control and manage my level up...which can actually be an exploit.

Like how in Morrowind you can continue to increase your skills to become more powerful without increasing your character level. You can finish the game at Level 1.

I do like games that let me chose to level up when I want, instead of automatically triggering it. Especially for games where the level ups give you random benefits. This may be less common today, but in some of the older game I still enjoy playing, level ups were immediate and the benefits were random. Now, if the level up always yields the same benefits, then I don't mind as much.

But this got me thinking about scaling and reasonableness. Why even have those traditional level ups?

If a level up is supposed to reflect a characters gradual advancement and improvement based on experience, why would increasing level be optional or even sudden? I mean, regardless of how you examine the formula, meaning strictly XP based, or increasing skill based, number of enemies defeated based, etc., the underlying logic seems to be that your character is gradually improving based on time, experience, action, practice, whatever.

That would logically seem to be an automatic thing, but the effects would be small and gradual. I keep using a skill so I naturally get better at it. I defeat a bunch of enemies so I naturally improve certain things. I get stronger. I get hardier. My attack increases. My ability to defend increases. My ability to channel magic increases. Those seem like they would happen whether I want them to or not. They wouldn't just change in an instant. I defeat the 100th enemy and suddenly several stats improve. The changes would be modest and...well...gradual.

So perhaps if you want levels in a game, maybe it is appropriate to have those level increases and improvements triggered in a reasonable way or more drawn out over time. Small increases rather than big changes...gap...big changes.

And I sort of like the idea of enemies being what they are because of reason and logic and not level. Enemies don't get stronger just because a character comes along that is stronger. Perhaps enemies get stronger the same way characters get stronger. They get older, more experienced. That ogre that you might face early in the game might be deadly, but not as deadly a year later (in game) after it has more experience. It might not have a bunch more HP or increased natural attributes. But its experience makes it a more cunning enemy that can hit harder, defend better, etc.

In theory, you could completely do away with the concept of levels and just have character (and enemy) stats increase gradually through time and experience. Why can't enemies get more difficult, just by virtue of the fact that they have lived longer to get more experience themselves? If the enemies are humanoid, why wouldn't they potentially obtain their own better equipment, just like the player?

Now, this isn't a new idea, and some players might not like this because your starting numbers and statistics don't significantly change during the course of the game. And this probably wouldn't work for shorter games. This would probably be better for games that are measured in dozens and dozens of playable hours, rather than 8-10 hours.

But what about the complaint that some enemies will be very easy or some enemies might always be difficult? Well, that is also logical and reasonable. Realistically, if I decide to fight a grizzly bear today or fight that same bear a year from now, our encounter is going to be quite similar, unless what I bring to that encounter changes. Say instead of a knife and my pajamas, I bring an uzi in tactical armor.

Just like if I decide to attack a fluffy bunny (not the vorporal bunny from the Holy Grail). That fluffy bunny is going to be pretty easy. And because of that, I'm not going to get much in the way of experience for increasing my skills. In fact, that fluffy bunny would most likely actively try to avoid me or escape. I can fight all the bunnies I want, but it isn't going to do much for me. I have seen games do this by giving XP based on a comparison of the character versus the enemy. The more challenging, the more XP.

So, to wrap up these ramblings, I am wondering about not having a level system at all versus just having skills that increase and having very modest changes to attributes based on a variety of factors. With attributes changing both positively and negatively. And getting better equipment to make your characters more capable. Better defense. Better damage. And making a game where all of the systems work together in a logical and reasonable manner and making the player responsible for their own choices. But this does go to overall, consistent game design. If you don't want your players to be able to easily trounce your hardest enemies, then don't make a system where they can improve to that point. Don't make deity-like equipment they can find. Make it so the very best equipment in the game with the most improvements through experience still makes your hardest enemies challening. But that also won't be as appealing to some players. We wants the overpowering sword of ultimate destruction!

If the player wants to go to the deep, dark, scary, swamp of the Emerald Dragon when they first start out...ok...but they are probably going to get eaten and die. If they wait until later on, they will have a better chance of survival, but it still won't be a cake walk.

So enemies don't really scale...they modestly improve just like the player can.

One final thought, perhaps moving away from levels, while still maintaining challenging and enjoyable combat, would allow the player to focus more on the story that is being told and the role playing adventure that is unfolding. But if your game is centered mostly around combat, then NOT having leveling is probably not going to be very well received. But I could be wrong.

 

[Tai_MT]

@ZombieKidzRule

I just wanted to point one thing out.

You could, theoretically, "gain experience" from fighting those bunnies. But, it would have to be in a purely skill based system. Your aim could improve (bunnies are freakin' FAST!). Your stealth skill could improve (they are highly alert and hard to get close to!). Or, other similar skills could "gain experience" from fighting bunnies. I don't know how many people on these forums ever grew up in rural areas (I did), but rabbit hunting, even with guns, isn't something you can do with 100% effectiveness. Lures and traps are typically what is used. Long range rifles as well (even then, you can miss, and if you miss the first shot, good luck ever hitting the second as they bolt and hide very fast). Rabbits are quite the challenge to hunt.

But, if you treat rabbits in an RPG like every other enemy... it just runs right up to you, gets into melee range, and decides to bite it out with you... Well, yeah, "easy to kill" is an understatement.

The problem is just largely going to be that there's a "skill ceiling" in such acts. When you can effectively hunt a bunny 100% of the time with almost no effort... you get nothing for doing so.

But, from a gameplay perspective, that has the potential to rub players the wrong way. There's an argument to be made for, "Why should I bother fighting the Green Dragons and spend 20 minutes doing so for 15 XP a pop?" A dev would need to be very careful to continue to give proper incentive to continue engaging with the combat. Diminishing returns naturally removes that incentive for most players.

 

[gstv87]

I've always used the direct level-based scaling (base_stat * level_factor = value), but always wanted to try a more resource/experience-based system.
basically, a higher level enemy would be higher difficulty because of the available selections of skills they would have, while their *stats* would be largely the same as a lower tier.
I can do that manually by making each enemy have a number of skills available, but haven't been able to code a system to manage it yet.
it would always take you the same number of hits to defeat an enemy, but at higher levels that enemy might counter, parry or block your attacks, with [counter], [parry] and [block] being techniques that one would learn through leveling.
a fight would be divided in individual *exchanges* of blows, where a given exchange would result in the attacker landing the blow and doing damage, or it being deflected.
the higher the level, the higher the mastery, and the more number of techniques available per turn.

 

[ZombieKidzRule]

@Tai_MT That is a very good point and I 100% agree. After I posted that, I thought about the other skills that could be improved with such encounters. And I do tend to gravitate toward the improvement based on current “level” and complexity.

For instance, the higher a skill goes, the longer it takes to improve. It requires more use. And eventually, skills would max out. I hope to incorporate a reasonable and consistent system that utilizes this.

And I am leaning heavily toward reducing lower “level” encounters as the player progresses by things like the enemy actively avoids the player or hides or flees. This could be offset by having a ”Hunting“ skill that allows the player to flush out those types of enemies if they want. But the return for the player would eventually be negligible to none existent. Unless the enemy gives a resource.

This way it is entirely up to the player. They won’t be harassed by low level enemies that just take up time with little to no gain. Unless they want to. Or the player wanders into Bunnyville and gives them no choice but to defend their home.

@gstv87 This is very similar to what I want to do, although the ability to block, parry, deflect, dodge, etc. would be possible from the start. But it would be skill level based. So your skills start off lower and you are less effective at performing the skill. Through use, your skills improve and become more effective.

And any enemy could have the same skills, if reasonable. A Hobgoblin might be able to do the same actions as any other humanoid, but not necessarily a Gelatinous Cube.

But for me, some of these skills would be both automatic/passive and selectable. You don’t have to choose to block, the check is automatic up to a certain number of times per round. But if you choose to block, that is essentially all you are doing that round so you can block more times and it is more effective. You might also be able to counter while choosing to block.

But I also like the idea of new skills/powers/spells/whatever that enemies might get through passage of time. They improve through experience, just like the player. So a younger wolf might know Snarl and Bite, but an older wolf might also know Cunning and Rend.

 

[gstv87]

So a younger wolf might know Snarl and Bite, but an older wolf might also know Cunning and Rend.

exactly that is how I envisioned it to begin with.
a low-level wolf might outright run away instead of fighting!
and I wasn't thinking of that at the moment... I was calculating how much pelt one would get from killing pelted animals, such as wolves or bears... and I was like "Wait... why is it more worth to kill a bear than a wolf? A bear might tear you to pieces!"
and from that inner debate is that I came up with "Oh, the bigger the animal, the more life experience"

 

[wilpuri]

Way too long posts to read, but ES: Oblivion really turned me off from level scaling. It was really easy to level up wrong, and the world around jus got harder and harder. In the end common highway bandits had extra rare gear. It was ridiculous.
I also don’t like the ”same, but different” approach that same enemies are harder in some places. Some really good games like Elden Ring do this. The very same mobs are suddenly harder if you travel around the lake.

What I prefer is, enemies have certain locked stats, but the gear can vary. Different areas have different enemies, and the further you go the harder the enemies get. When you come back to the beginning you can have the (power fantasy) perspective to the progress when you mow down the enemies there.

 

[eomereolsson]

Way too long posts to read, but ES: Oblivion really turned me off from level scaling.

Hard agree. Oblivion was the first game when little me realized that scaling enemies was a thing that games could do and I hated the concept ever since. It just makes me feel all my progression just does not matter.

Because my dumb monkey brain likes watching Small number go Big.

I guess my monkey brain is just built different then yours.

Let me illustrate my point with two examples:

I once DMed a D&D adventure for a group of fairly high level characters. I also had one encounter where the players faced a large group of goblins. For context goblins in D&D are fairly dangerous enemies on the first couple of levels, but quickly become irrelevant after that.
When designing that encounter I was deadly afraid. Would my players be bored because of the easy enemies? Would they be annoyed for me wasting their time? But the encounter made sense for the story and so I stuck to my guns and ran it.
And much to my surprise it was the most fun I had seen my players have in an encounter for a long time. They very much enjoyed just mowing down once dangerous enemies left and right.

The second example is Valheim. I recently played through most of the available content with two friends and again all of us very much enjoyed the feeling of seeing once dangerous enemies which we only could take down when teamed up reduced to minor annoyances.
And at least one of us very much enjoyed sneaking into zones we were underprepared for, grabbing what resources they could and running from everything that moves.

I feel we as game designers just tend to worry too much about "but what if it is too easy?" and fail to realize that more relaxed periods and the sense of progression are also very much important for a good game.

And at last I wanted to adress some points you made regarding no progression:

Super easy for me to do.

I don't know about your experience, but indie/solo game development is darn hard and time consuming. Every one of my projects took/takes way longer than I anticipated even when they are "just simple projects to get my feet wet". So when an easier option leads to equally good (or as I argue even better) gameplay: TAKE IT!

Inevitably leads to the classic unrealistic scenarios where huge, deadly enemies that once ripped the player to shreds now die from but a single poke.

I am not denying that this CAN be a problem, but you are in charge of your game's narrative. It is by no means inevitable. Like no one questions why the small town heroes equipped by ye olde blacksmith struggle to take down the same enemies that the kingdom's mightiest warriors equipped with legendary magical gear just breeze through.

Nonlinear progression and/or its resulting over-leveling could mean the player never experiences an encounter as intended, even once.

I made this point already some other thread quite a while ago:
That's game design for you. The only way your content is experienced "as intended" is to write a book. The moment you give the player agency through input controls, they will do and experience things differently than you intended.
Come to terms with it and take solace in the fact that at least most of your players will probably play your game somewhere along the intended lines. But you just can't ensure it.