My Account
× My Account Forum Buy Now

Last Epoch Forums

Removing RNG From Crafting

I want to raise the proposition that players would be better off in general if RNG were removed from crafting in ARPG’s and open it to discussion here. In fact, I have for awhile now. We’ve had a lot of productive and interesting conversations around topics like these in the past here on LE forums. One such conversation with Llama8 inspired this one. And after taking a big break to ponder the subject, I’ve finally decided to come back to it. Thanks ahead of time for indulging me. I hope you’ll enjoy this.

I should probably start by clarifying a few things: What I mean when I say “better off” has to do with some value judgments I’ve discussed on the forum in the past. These are a couple of metrics, but they’re ones I think are important for a few reasons. One is the rate of novel meaningful experiences. We all know what we mean when we refer to “the grind.” This is the game asking us to do the same or similar tasks over and over again to get different loot. We mean what we’re doing isn’t novel or meaningful. The activity in and of itself doesn’t compel us. Another is the quality of those experiences, IE what do we actually get out of them? This includes things like story payoffs, the amount of rewards we get for the time spent, etc. These are intertwined in my view because having to do the same content over again is inherently a low quality experience. It’s not no-value, but as I’ve said in the past, it’s lower value because it’s uninteresting and there’s a lot of other games with new and interesting experiences we could be spending our time on instead. With the amount of experiences available in life, let alone videogames, and a limited amount of time, I posit it would be a richer life to trade that time for more and better experiences. Hopefully we can at least agree on this, since this is important to being able to assess quality, which is what I’m attempting to do here.

Another thing I mean when I say “better off” is actually related to health. I start from a baseline assumption that anything that is compelling you to spend undue time on it comes with risks. We all know there are people that can’t get up from the poker table or the slot machine. This comes from what is called a “schedule of reinforcement.” To give you the short version, rewards or payoffs that are uncertain and become gradually less common over time cause you to want them more rather than less. Humans become more motivated in these situations for reasons stemming from the psychological will to succeed at these things when it’s a necessity in nature. That’s just how we are. In my view, when this is the main or only thing that compels you to do something that you would otherwise find boring, that’s inherently a low quality experience because it’s not rich enough on its own to entertain you.

Having clarified this, I’d submit that the RNG used in crafting systems in various ARPG’s do not provide value along these metrics, and mainly compel players to craft due to schedules of reinforcement.

To support this, let me discuss the activity of crafting that involves RNG for a moment: First off, what are we really doing when we craft an item? We’d have to agree that we have an end goal in mind, and that would be to make the best item possible. Now in the course of that we might spend resources we earned, pick the specific bonuses we want, reroll some things, etc. But despite any of the other things that might be involved, that is still what we’re doing. We’re trying to make the best item possible.

Now what does RNG add to that experience? It really only adds one thing, which is an arbitrary chance of failure. Since it’s not skill based, you can’t get good at it and thereby eliminate it. That’s why I say arbitrary.

So if we take it that we’re trying to make the best item possible, is failing an arbitrary number of times while crafting adding unique or meaningful experiences to the process? In other words, is the player getting to experience new things or learn more about how to succeed next time from these experiences? The answer to this would have to be “no.” Failure is failure, and once you’ve failed once, every following failure might be different goals denied, but they will all still be arbitrary in the way they were arrived at. This is simply an extra unnecessary step the game requires to get the item. You clicked the button and didn’t get the outcome desired, is what each of them amounts to. Does this convey some new or interesting meaning each time? Individually, clearly not. In the aggregate, they might tell us a little bit about the rates of success for certain bonuses to show up, but this is not a human experience. It requires a great deal of data that the human mind doesn’t passively experience. That’s the sort of thing you need a spreadsheet and a lot of attempts to actually see. You wouldn’t inherently notice how it’s going unless the outcomes you wanted were fairly close together. And even if they were, that information would only be useful if there were powerful and reliable ways to influence it as the player, which we’re taking for granted the player isn’t given; If they were, this RNG could be eliminated somehow. All of this being the case, we’re missing out on novel and meaningful experiences we could be having doing something else.

Now over time, as we succeed in getting items we actually want, the next item we need that would be an improvement inherently becomes more unlikely to be crafted. The better or more unusual the bonuses we need, typically the less likely those will be as well. So we actually do have a decreasing rate of success built into games like this that require you to loot or craft with RNG involved. This is a schedule of reinforcement with a decreasing rate of success. So this is also a negative aspect to having this system.

One thing someone might then say in response to this is that the arbitrary failure is actually adding a sense of suspense, or otherwise delaying gratification, which makes the eventual success more satisfying. What I’d like to point out though is that what people are experiencing in this case is actually the relief from future potential failure, and mistaking that relief for satisfaction. Think about it: If the failures aren’t imparting a sense of fulfillment or fun, then what is creating that feeling in the body when one succeeds? Is it the item itself? If that’s the case, wouldn’t it be the same amount of fun if the item were completed the first time? So then it can’t be the reward being doled out that causes the good feeling. I’d argue for this reason that it has to be the relief from the previous failures and not having to worry about failing again that causes us to feel better when we get the bonuses we were hoping for out of crafting. This isn’t meaningful or interesting in and of itself. It’s just a mechanism in our psychology that is causing us to spend time on something that we otherwise could have been allowed to do with much less effort. This is mainly a way of padding an amount of content to last longer than it would have. We could instead delay gratification by simply playing more interesting content in the game.

Now another thing that might be said in regards to this is that the point is actually to never succeed or indeed, never reach a state of completion. This is difficult to accept: If the point is to never acquire a perfect or near-perfect item, then what is the player’s goal when grinding for that item? Does anyone truly do anything with the intention to never succeed? Moreover, if the goal is to compel the player to play indefinitely, even after they’ve run out of new and interesting things to do in the game itself, is that a reasonable amount of time to ask the player to play in order to see the content they clearly consider the goal of their efforts? To me, that doesn’t seem reasonable. But for the sake of argument, suppose that it is: In either case, even if we rule out success or completion entirely, we would still have to evaluate the quality of what we we’re doing in the interim on the criteria laid out earlier: Are the experiences the player is having while playing novel and meaningful, and is the RNG for loot or crafting compelling them to spend time playing that they otherwise wouldn’t have? If the answers are “no” to the first or “yes” to the second, then these are low quality experiences. Players would have more fun playing a game that was giving them novel meaningful interactions and that didn’t compel them psychologically with schedules of reinforcement, regardless of when the end goal arrives, or if it ever arrives at all.

Yet another thing that might be said in objection to this is simply that RNG makes loot or crafting more fun. I’ve thought about it quite a bit and the only context in which I think this is a strong argument is if you suppose that the goal of RNG is to provide a small variety of items each battle and have you examine to determine between them which of them is useful. This is the game that loot becomes. This seems to be true, because doing this requires knowledge and choice on the part of the player and impacts gameplay since the player can use the item. It can also be different every time. All of this together means these situations are novel and meaningful to players. It’s me getting off-track a bit, but I would push back on this a little by saying that loot is something we spend a disproportionately small amount of time on compared to the other activities in the game. This is especially the case by the time endgame rolls around since so few candidates for upgrades ever present themselves. Moreover, this argument doesn’t work for crafting, since in crafting we have the end goal in mind already: To make the best item possible. While this proposition is generally true of looting in the early to mid game, by the end game we’re also trying to pick up the best item possible, since incremental upgrades have already become exceedingly rare, or at least much rarer than earlier in the game. Maybe we can extend this fun by replaying the game as each class and build the game offers, but item selection during early and mid game is only a small part of the overall value of doing this. It’s not nothing. But I’d argue it’s not enough to outweigh the volume of lower quality experiences due to RNG that we have far more of later on at endgame.

Let’s have some fun while we’re on this subject also: If in defending this idea someone were to state that it is simply the excitement of RNG in general that makes things more fun, that would come with some logical problems. This would mainly be that it proves too much. (This is to say, the underlying principles of the statement could be used to reach other conclusions that are difficult to accept, so there must be problems with them.) Say I were to roll four six-sided dice until they turn up all identical, or all ones, before eating dinner every night. We would all agree that would be strange. It wouldn’t make my dinner taste any better, I don’t think. And certainly, I would end up being hungry for longer, which is not something most people are interested in prolonging. That’s kind of the point of eating. So, would an arbitrary chance of delay before any activity make it more enjoyable? By this example, probably not. I don’t think rolling dice would make the news more fun to watch. Maybe not seeing it for longer would be enjoyable, but I could just not turn it on in the first place. We see then that we can’t make a blanket statement about RNG being fun, since, as with most things, context definitely matters.

What we’d end up being stuck with is the argument “RNG and some amount of effort makes some things fun.” I think that this statement would be much more likely to be provable, especially since I was able to describe earlier at least one situation created by RNG looting that seems to be richer due to RNG. The problem is that you’re stuck with determining what amount of effort, what RNG and what kind of RNG-created scenario can then become more fun, as well as why. You’re left needing a lot of specific and hard-to-know information in order to know how this works case-by-case, a large amount of data, which is probably why you don’t hear it made very often when it comes to discussions around these types of mechanics. And what I suspect you would still find out after analyzing all of that information is that in a great majority of cases people are experiencing a sense of relief rather than actual gratification.

To give credit here, Llama8 objected to removing the RNG from crafting by saying that what we would be left with is an item editor. This is another objection I wanted to address because it is interesting. We might not think of it as such, but crafting with RNG, as we have in Last Epoch, is already an item editor. It’s just an item editor with more steps and delays built in. As we’ve discussed a bit here, those steps are not making the final item more interesting or useful. They are just making it take longer to acquire, by some random arbitrary amount of time. This means that the issue of it being an item editor is really a moot point. To argue taking them out would be bad, maybe what is meant is that it would just be too quick or simple to make the item you want so as to be unrewarding, since people like the sense of accomplishment they get from tasks that require time and effort, or some form of problem solving. What I would then argue is similar to what I’ve said in the past: That effort should track in some linear fashion with the amount of time played, so that we’re ensuring the player has a higher amount of unique and meaningful experiences across time, and so that we’re not imposing schedules of reinforcement on them in the process. Maybe we just make an item cost more resources, thus taking longer to get. Maybe we put in a bunch of components or bonuses you can only get by completing certain tasks or challenges. Maybe we also put so many possible bonuses in the game it becomes difficult for the community to theory-craft and test them all with each other, so that it requires more effort for the player to know which ones work the best for their particular build. All of this can contribute to an amount of time it takes the player before they reach their goal, that is not in any way governed by RNG, and that would satisfy the requirement for a sense of accomplishment. Of course, there are endless other ways of providing this sensation without using RNG. I’m sure there are even better ways of doing this that someone more creative than I am would come up with.

All of this is why I propose developers in the future look at removing RNG from crafting and focus on finding alternatives to give challenge and accomplishment to players using crafting systems. I believe it can be done, and would be better for everyone involved.

Thanks for hearing me out, as always. Looking forward to hearing your thoughts as well soon.

(Edit: Grammar / Typos)


Your post is a bit verbose. Having read it, I believe you are largely arguing in favour of two things; that the game should be as fun as possible while people play it, and that the developers should be willing to ‘trim the fat’ - that is to say, avoid artificially increasing playtime through the likes of repetitive gameplay and time-gating rewards.

Last Epoch is intended to be a live service game where (most) people play server-side (“online”) characters, see others in towns wearing MTX which is effectively a form of advertising - and I think we can safely presume that the % of people not actively playing that choose to support the game financially will be quite a bit lower than the % of those logging in regularly who choose to open their wallets.

EHG will, due to their ongoing costs, therefore very much have an incentive to generate recurring income through encouraging people to 1) play more, and, 2) play online when they do play. While I agree with your point regarding ‘low value’ activities and the existence of other games in principle, I do not believe it is reasonable to expect EHG to act on this. They are a for-profit company courting investors & selling shares in the company. Money is priority #1. (Naturally, they will never say this themselves.)

While you are perhaps correct that there is an ideal version of crafting that substitutes something else in place of RNG, you haven’t done much to ground this idea in examples and I can’t help thinking such an idea would be massively time-prohibitive to work on.

Saying that crafting is already largely an item editor suggests you might not be quite interpreting the item editor arguments in the way they are intended. Crafting needs to be kept in check from a balance perspective, or else no loot would be compelling and the item hunt would be savagely curtailed. When people say removing the chance of failure would result in crafting being an item editor, that it would be quick and entail few steps is the point of the argument. ARPGs are, at their core, item hunts and making crafting guaranteed effectively removes the item hunt from the game.

Which leaves the question of why else would people play for long?


I don’t think that RNG in crafting is entirely there to provide “value”, it’s also there to reduce the speed at which you acquire your desired gear. If you take the other end of the scale, no RNG in crafting at all, you’d be able to get your desired t20 gear very easily (assuming that the affix drop rates weren’t adjusted to compensate for the lack of chance to fail in crafting).


I really appreciate your reply! You made a particularly interesting point I want to focus on and discuss a little bit, because this is really fundamental:

You certainly could describe many if not most ARPG’s as currently constituted as item hunts. That’s an astute observation. I guess something I failed to outline in any real detail is that I don’t feel this is necessarily good or has to be the case. Exploring, fighting and absorbing the story should be just as much core to the game as item acquisition. If the entertainment value of those other things has been expended, then the game has become mundane at that point.

Consider that when item acquisition becomes the primary or only goal, and the tasks you’re doing to acquire them are mundane (“the grind”,) that means the things you’re spending the lionshare of your time doing in the game aren’t entertaining. Remember that item acquisition isn’t a very big fraction of the time we spend playing the game.

This is ultimately why I try to point this out to other players; If more people recognized this, maybe the demand for more and better content from games would gradually increase. I believe people will pay for more content. It’s just up to developers to figure out how to deliver it to them.

1 Like

Right, and that makes sense. What I tried to point out though is that there’s other ways of doing that which don’t have to be governed by RNG. It doesn’t necessarily have to be easy to get all of the items you want, even if crafting them doesn’t involve random delays like these. You can challenge the player in a lot of other ways.

If you take away alot of the randomness most item acquistion will get even more grindy and most importantly predictable

Just for example’s sake, lets say all % chance or variation is removed, but you need 50 shards per affix tier upgrade (don’t have to be accurate numbers).

The time it takes to get an T20 rare item might be the same, but you will slowly grind towards you goal, but the endresult will be very very predictable.

This will take away all the moments of dopamine rush and excitment out of the loot hunt and basically turns the game into a braindead grind simulator.

Having things like loot and crafting to have high variance will make crafting and loot unpredictable and will give you this OH WOW moments, when you totally didn’t expect to make that T20 item from a T4 starting item. But on the flipside you have those T16 items running out of FP after 2 crafts.

But with all of this, the desired outcomes or even the totally unexpected outcomes will outweight the negative moments IMO.


Great thoughts here; Very interesting point, Heavy. This is actually a really good defense of the idea of having random loot. That being this:

I don’t think it necessarily applies to crafting, because we already know what bonuses we’re looking for and the act of crafting really just comes down to a pass / fail scenario, which I tried to describe earlier. The positive emotion you feel when succeeding at crafting is really relief from previous failures that we’re mistaking for satisfaction from accomplishment, since the outcome was arbitrary all along. But for loot specifically, it is a good point that getting a big unexpected upgrade every now and then does surprise the player and feel like an unexpected reward. That’s definitely something that RNG does create.

However, I’d still point out in regards to this that this is an experience that gradually becomes rarer as the game progresses, to the point that you can play for multiple hours without getting an item you really want to acquire. That’s why I don’t like the idea of a game’s entertainment value depending on it. Worse, if you’re playing a game where that item that drops is just a base item that you then have to craft in order to perfect, you’re actually funnelling those rewards into crafting that then is merely a pass / fail, multiplying the time it takes to actually get an upgrade. This is actually kind of a good argument for doing away with RNG in crafting, because at least then the rewards you got when looting would be valid and final before you used them in crafting.

It’s not that you’re surprised that you roll the max on a given tier, it’s when you manage to take a low tier item all the way up to a high tier item “beating the odds”. That’s the positive moment that RNG-based crafting can give you.


I have almost 1900h in LE with, about 8 chars at lvl 100 or almost and another handful of chars at about 50… if the campaign takes about 10h - I am roughly in for about 150h of story experience… the rest is entirely chasing levels and finding gear to improve my characters - either running via mono or dungeons…

That is WAY more than “isnt a very big fraction of the time spend”

I think that part of the discussion here is most definitely historical design of arpg games - the item chase and everything it includes plays a very big part of the genre… Your idea of moving this enjoyment to the other side of the game (story etc) isnt bad, but I have a feeling you are aiming for more R in the arpg type game which then changes the focus considerably from what LE is right now…

Nothing against it as its exactly why i like playing games like Baldurs Gate, Witcher and others, but I expect there are many people who like arpgs, that play as I do…


Beating the odds at an arbitrary step though. It’s not a valid sense of accomplishment. It’s just a relief that the game let you off the hook. We could have a more novel or meaningful experience if we instead of spent that time and mental energy doing something more entertaining.

You’re right, that is what I’m getting at. I suspect that also. What the issue really is with that is just that it’s a very small amount of content being held out of reach for a long time.

What I really meant by this is the fraction of the time we actually spend getting and comparing items, as opposed to all the other mundane repetative things we do to cause them to drop. The grind isn’t what’s entertaining to us, hence we’re spending a lot of time doing things we wouldn’t be interested in doing otherwise.


My first and most powerful thought is that you need to practice writing succinctly. There’s a lot of words here but a noticeable dearth of substance.

As for the idea itself, removing RNG from crafting makes it stop being crafting - instead, it becomes an item editor. The summary of what you took a lot of words to say is nothing more lofty than that you want to have a guarantee that eventually you will get a specific item you want by using the crafting system, and that’s simply a non-starter.

Specific responses:

You are using “arbitrary” incorrectly, as is often the case. You are using it to mean “I don’t like it” and “I don’t have as much control over it as I want to”, but what the word actually means is “based on random choice or personal whim, rather than any reason or system.” There is a reason, and there is a system. That you don’t like or don’t understand the reason does not make the reason arbitrary. That the system has elements of randomness to it does not make the system arbitrary.

All those “steps and delays built in” are exactly what makes it not an item editor.

They’re not supposed to. This criticism is irrelevant.

You could say that any achievement in a game is arbitrary though. Managed to do Julra t4? Well done, you’ve beaten a boss with arbitrarily high damage output!

And yet it still gives that endorphin rush to a lot of people.

But where the line between “meaningful” & “grind” is entirely dependant on the player, some people do find it “entertaining” & one would argue that in a loot-focussed game, the acquisition of loot is somewhat important.

Yes, but apparently we place higher value on things we have put more effort into obtaining, so if you finally get that item you want after a “long” grind, you’ll value it more.


Thanks for the response!

I guess part of this may be down to us all playing the games we play for different reasons, and looking to get different things from our time spent doing so. Often on the forums for ARPGs, for each post calling for more campaign content, there is at least one calling for the campaign to be made optional so that people can delve into the so-called endgame content right away. It would be interesting to hear which endgame activities in the genre you’ve liked most. I think one thing many of them share is a focus on randomisation, which some prefer for a less predictable experience through subsequent playthroughs.

In some ways I fear your points may conflict; you’re interested in story content being greater in both quantity and quality, but this would significantly increase the cost of development - likely resulting in us being charged more. But as you say - the market is rife with other games.

I do hope you won’t mind me asking, but have you played Guild Wars 2? The world ArenaNet created does a stellar job of feeling more ‘alive’ than most - and on top of the sizeable amount of levelling content with strong narrative elements, there is both a ‘personal story’ (not particularly standout, imo, but gives you some choices, which is nice) and a lot of what they call their Living World content.

I would really love to hear your thoughts on the game if you have tried it. I don’t mean to derail this thread - we could always take things to PMs or start a new thread if you’d prefer.

1 Like

Well yeah. If you drill deep enough into any anti-RNG argument, that’s all it comes down to. They don’t like that things exist in the game that they may never have. They want a guarantee that they can get everything they want.

Never really separated the grind from the item drops themselves - I suppose this is due to the endorphin loop or whatever the psy majors call it … but I suppose you are correct to some degree - the time actually looking at new drops is low by comparison to the grind.

Removing this grind part of the process - no idea how to do this without making this a quest driven game like a more traditional RPG… which doesnt appeal in an action-rpg game tbh…

1 Like

I can see why you vote against RNG mechanics in crafting in games general. But humans are conditioned in a particular way.

The first games childs play often involve rng, be it rolling dice (starting with colors) or picking cards. Later game mechanics get more complex. But 99% of tabletop games (chess is an example where there’s no rng involved) are based on rng.

When pc games were created they copied these rng elements. There’s rng everywhere to simulate real life experiences.

One of the most predictable real life mechanics is going to work. In general people that have a creative job with unpredictable elements that prevent work from being monotonous are happier in their job than people that are doing the very same thing every day for years until the end of their life.

Playing games = random elements = ups and downs = emotions = memorable = exciting = fun

Doing work = predictable = normalized emotions = trivial = boring = not fun

The main thing I project into your post is that you want to turn playing into something that has more in common with work.

Because this will be the end result if you take RNG out of crafting. You would create a work simulator where you have to grind for hours for the predictable outcome.

I’m currently fine with the crafting system. It’s also very predictable while still having rng involved. No rng would need to put more rng into the resource hunt for crafting. Taking out the rng of the resource hunt would mean to make it grindy.

Finding loot is also random. Yet, people don’t complain too much. There could also be just progress bars that reward you with a specific item after hours of i.e. killing a specific mob type.

A core element if playing games is rng because it adds replayability. The simplest game will be replayable by adding dice rolls. Just image Patchesi where there are no dice and everybody would move the exact same amount of fields…


That’s a separate issue but this is actually true; In some ways, higher tier loot does rob you of the entertainment value of bosses and enemies, since it trivializes your interactions with them over time. This is why some people really enjoy the Dark Souls series, because it stays deadly to the player and is overt about that being the point.

So does a roulette or a slot machine. (Poker I realize is not the best example, because there is a mind game to that which is more of a game than these.) Yet, you haven’t really demonstrated skill or knowledge when winning at them. You also don’t exercise any joy in the activity prior to the outcome. It’s just a ball landing in a slot or pulling a lever. The compulsion is merely from the psychological mechanism that these are acting on to get you to do them again.

I mean something very specific by ‘novel’ and ‘meaningful’ in regards to the content in the game and the way the player experiences it. Novel is obviously unique, IE not repetative. Meaingful would be impacting the way we play in some fashion. A great example of this is when we gain knowledge that changes the way we do things. Fighting games are almost entirely knowledge based: How well you read a situation, know a match up, etc. Interactions in fighting games are meaningful because they teach us how to problem solve better. I guess put simply, it engages the player in some way. Those are meaningful interactions.

Well yeah, that’s precisely the problem; We’re over-valuing a small amount of content that compels us to do tasks we don’t find entertaining in and of themselves. That time could be spent doing something else, like playing a different game or new content. If we’re compelled in this way, we’re actually missing out.

To me it’s more that whether or not those rewards ever come is a separate issue to whether or not you’re having novel or meaningful experiences doing the tasks that lead to them. If what you’re doing isn’t fun before an item drops, then there you go: What you’re doing isn’t fun. The fact that the game can lead you on for hours at a time before that happens is just part of the value assessment.

Now, having said that, having it take a reasonable amount of time to unlock every item you want in a game would actually be good in that this is content people have paid for and want to see. The alternative is being compelled to play for an indefinite amount of time to see a handful of rewards that have been held out of reach arbitrarily. I don’t think it’s reasonable to compel people to keep playing in that way.

I think you would still have a grind, but you would just make it shorter and the progress less unpredictable. Maybe you combine random loot with a preditable crafting process that makes the over-all amount of effort not hundreds of hours to finish one character.

The way I assess the value of my time is whether or not one playthrough to the perfect or near-perfect state of the character takes longer than a week of work. If I spend more than 40 hours getting there, the game is asking to be played like a job, without being as rewarding as one. Unless what I’m doing during that amount of time is really interesting, I can’t see that as a good exchange for my time.

Really enjoyed this discussion so far, by the way. Wish we could do this more often.

1 Like