Click here to go to the next staff post in this thread.   #1  
Old 01-13-2010, 07:18 PM
medierra's Avatar
medierra medierra is offline
Crate Employee - Designer
 
Join Date: Nov 2009
Posts: 4,020
Default On "Farming"

For my first post in “developer topics” I thought I’d go into some depth responding to a recent suggestion posted in the Grim Dawn section of the titanquest.net forums. This suggestion, posted by yerkyerk described how the behavior of “farming” could be curbed by linking drop rates to experience gain.

When I first read it, I thought it was a really fantastic concept. It especially appealed to me because the approach epitomized one of the most important ideals of good design: simplicity and unobtrusiveness. It is an elegantly simple way to seamlessly change gameplay behavior without introducing in some arbitrary, ostensibly awkward new mechanic.

If we wanted to discourage or remove farming behavior, most notably boss runs, this would perhaps be the perfect way to do it. However, after considering this for a moment, I wondered if that was a behavior we really wanted to discourage.

I'm not sure the answer is as clear cut as it might seem. First let us consider why farming is perceived negatively.

An obvious complaint may be that the behavior of repeatedly running down the same path to kill the same enemy over and over again certainly does not make sense in the context of a real, believable world. As much as we may want to create believable, immersive worlds though, we have to keep in mind that gameplay and fun are the foremost considerations. So, if farming is an unintended but fun gameplay dynamic that has organically emerged in this genre, shouldn’t that supersede believability?

To that point though, it could be said that farming is not, in fact fun. It is something that players feel forced to engage in if they want to acquire the best items in the most efficient way possible. Repeated runs on the same boss are monotonous and boring. Wouldn’t the game be so much better if players could spend their time focused on just playing through the game normally?

I was about to suggest that perhaps farming is fun to a certain portion of players, however, I then realized that farming is probably not fun for anyone. That doesn’t mean it isn’t enjoyable though. Wait what?! Confused? This is probably fodder for a whole other discussion topic and at some point I’ll have to write about what I see as the difference between fun and enjoyment in games. In brief, I take enjoyment of gameplay to mean that you’ve come away from the experience with a positive feeling. It was pleasurable in some way. That could mean that it was outright fun or it could be another positive feeling such as satisfaction or accomplishment. For many players, doing runs on bosses makes them feel as though they are “getting ahead” in the game. We’re hardwired by nature to feel good when we believe we are doing things to advance ourselves / promote our well-being. Of course, collecting loot in an ARPG is probably the total opposite of personal advancement but, our brains don’t seem to know that. We perceive the collection of loot and advancement of our character as personal gain for ourselves and this tricks us into feeling satisfied. When we believe we are doing work above and beyond what might normally be expected, it increases this satisfying sense of accomplishment.

I personally have a low tolerance for monotony but I do like knowing that if my character is getting behind on the equipment curve and progress through the game is getting too difficult (especially in hardcore) that I can turn to farming as a way to bring my equipment up to par. This, in turn, makes me feel like I am facing a much greater challenge. In hardcore D2, for example, in order to prepare for the passage from normal to nightmare and then from nightmare to hell, I found that I always needed to replay areas to stay safely ahead of the level curve and do boss runs to put together gear with enough defenses and damage that I’d have a good shot at surviving. If I was always just progressing though the game normally, I don’t think I’d feel like I was facing an epic challenge.

Farming also adds a different type of gameplay dynamic. I think it is good whenever a game allows you to engage in different types of activities to achieve different purposes. Normal progression leads you to the completion of the game while boss runs are an alternative means of pumping up your character at the expense of forward progress and enduring a little repetition. I often alternate between normal progression and boss runs when I play. Sometimes I don't have time to really sit down and seriously play for more then 15-20 minutes so I'll jump in and do a couple boss runs on the off-chance that I might hit the jackpot. If you remove boss runs, then you only have normal progression... it’s like you're removing a mini-game.

Boss runs are also potentially one of the best ways to earn XP if you haven't out-leveled the boss you’re fighting. So the adjustment would probably have to be a little more complicated than just XP gain. In conclusion, the more I think about this, the less I feel that trying to do away with boss runs is a clearly positive goal.

Whatever the result, this suggestion was good in that it did get me thinking. It may be worthwhile to explore ways of making boss runs less monotonous because I do feel that, while there is an audience that derives satisfaction from boss runs, there is another segment that doesn’t like them but feels forced into doing them in the pursuit of loot. One idea would be to have a system where the drop rate for a boss decreases slightly each time you kill them but gradually returns to normal after enough time has passed. This would at least encourage players to farm different bosses instead of only doing runs on the final boss. I’m not sure how exactly we’d implement that though in a way that players couldn’t figure out some way to circumvent it. Another idea would be to set up loot tables so that each end-boss could only drop a certain portion of the unique items in the game. So, if you wanted different items, at some point you’d have to farm different bosses. That has some drawbacks of its own though…

Perhaps the best thing to do is nothing? ; p

What are your thoughts?
  #2  
Old 01-14-2010, 06:34 PM
yerkyerk's Avatar
yerkyerk yerkyerk is offline
Lord Advocate
 
Join Date: Jan 2010
Posts: 4,228
Default

First off, let me say congratulations to opening up the site! Was about time too

Anyway, on topic;
In the original post (on tq.net) I also stated that farming had to do with rebirth fountains and portals.
In D2 (yay, there we go again) I had less problems with farming, as I could do farming runs and access all locations quickly from a teleport pad. So I visited like a small dozen places each gaming session. It still was monotomous, but less so.

However, with the rebirth fountains in TQ, it became inefficient to visit several places in a single gaming session (or even hundreds of gaming sessions - took me a while to get my first legendary SBC's ). Since farming for good gear already takes a long, long time (which i find positive, mind you), I couldn't drag myself to doing it in a very inefficient way.
I was thankful for the Secret Passage though, as that was almost as easily accessible through a portal as through a rebirth fountain - and I often took that as my second and final farming location.

Ofcourse, with the xp=mf solution, there's the players who can't play long enough to benefit from it and they'll be penalized for not having tons of time, that's not really fair.
One possible solution to this problem is by quickly capping the mf, say that after an average of 20 minutes you already reached the max %mf you can get (or use a diminishing returns value where 2 hours of xp-farming hardly delivers more mf than 20 minutes - much like in D2, where going pass the 200%mf find was almost useless).
The other part is to offset the %mf formulae against the time won by doing quick - so farming bosses or just farming the game is both as effective.

As for boss-exclusive droppings*, I already thought that was the case in TQ?


EDIT: I hope the order of posts will be reversed, as this change is too much for my little mind to fathom..

Last edited by yerkyerk; 01-14-2010 at 06:37 PM.
  #3  
Old 01-14-2010, 06:42 PM
Renevent's Avatar
Renevent Renevent is offline
Lord Advocate
 
Join Date: Jan 2010
Posts: 4,655
Default

My personal opinion is that killing bosses and collecting the best loot from their dead corpses is the best part of these types of games. I think where it becomes a drag is you start so close to the boss. Basically you can just run for 20-30 seconds, kill the boss, collect loot, then restart the game and repeat ad-nauseam.

I think there's something that can be learned from newer games that have expanded on the ARPG formula...namely MMORPG's. I know some people might get riled up with my even mentioning this game, but let's talk about World of Warcraft. Basically, you can run bosses in that game over and over too...and people do...the difference is you have to complete some challenges prior to slaying it...mainly completing a difficult dungeon.

I think this keeps things a bit more fresh for two reasons:

1) The amount of time per run is increased, and the player is involved in more than simply killing one enemy over and over and over and over every 30 seconds.

2) Greater risk which means the reward is that much sweeter

Also, because the time investment is so much greater, the reward can also be scaled up accordingly. There's nothing more disappointing than killing Typhoon and seeing nothing but yellows. I realize why it has to be like that, though, as if the loot % quality was high a couple hours of farming would be out of whack balance wise.

Oh, and just to be clear, I am not suggesting 4 hour raids requiring the perfect team of 5 specific classes

I don't have that kind of time either, and I would absolutely hate that kind of thing.

Another idea that might be worth consideration is something that HG:L did. I don't know how many people played it, but in the stone henge area they had an interesting mechanic for fighting the final boss (and most lucrative boss). Basically you had to run around different areas and kill his generals and take their heads, after collecting all 4 you could open up the portal to his dungeon and eventually fight your way to him.

While HG:L was a seriously flawed (in many ways) game, I thought that game play mechanic was very cool. It also had a secondary effect...which is encouraged cooperation in multiplayer. People would try to find others who had that extra head they were missing or find people to maximize the amount of runs they could do. The nice thing is this didn't hamper the lone ranger either, as it was all completable solo as well.

Last edited by Renevent; 01-14-2010 at 07:24 PM.
  #4  
Old 01-14-2010, 06:44 PM
decius decius is offline
Seeker
 
Join Date: Jan 2010
Posts: 1
Default

Followed from the tq.net forum.

I don't like the aspect of games where I have to do the same thing n times in a row to proceed. I think it would be far better to have a large world, all of varied difficulty, and allow the player to decide where in the world he goes.

Don't fall into the Morrowind trap, and make the entire world of uniform challenge, scaled to the player. Don't make areas inaccessible to 'low level' characters, either; give everyone enough rope to hang themselves, and let them use it.

As far as top-tier equipment goes, every single piece should be interesting. Give each 'set' or higher quality bit of stuff at least a paragraph of backstory, explaining the benefits and drawbacks of that piece, then base the stats of the equipment on that story. (Don't neccesarily make the history player-readable. It's for your benefit.)

And finally, don't make mechanical changes to gameplay based on the playstyle of the player. Ideally, have an option to save the entire state of the current game, and continue from exactly the same point at a later time. If that is impractical, minimize the amount of replay needed to get to the same point.
  Click here to go to the next staff post in this thread.   #5  
Old 01-14-2010, 07:44 PM
medierra's Avatar
medierra medierra is offline
Crate Employee - Designer
 
Join Date: Nov 2009
Posts: 4,020
Default

Quote:
Originally Posted by decius View Post
Followed from the tq.net forum.

I don't like the aspect of games where I have to do the same thing n times in a row to proceed. I think it would be far better to have a large world, all of varied difficulty, and allow the player to decide where in the world he goes.

Don't fall into the Morrowind trap, and make the entire world of uniform challenge, scaled to the player. Don't make areas inaccessible to 'low level' characters, either; give everyone enough rope to hang themselves, and let them use it.
Don't worry, we'll give you plenty of rope. We have a hybrid scaling system where enemies within an area will scale to provide the right level of challenge but only within certain level limits. If you run ahead in the game without leveling, you will easily be able to find enemies far enough above you in level to handily punch a large hole in your face. On the other hand, if you linger around the starting area too long, eventually you'll exceed the level limit of enemies that can spawn there and you'll want to move on.

This issue seems like something players are concerned about, so perhaps int he near future I'll write about it in more detail.
  #6  
Old 01-14-2010, 10:15 PM
Munderbunny Munderbunny is offline
Initiate
 
Join Date: Jan 2010
Posts: 14
Default

Congrats on website! (Read the backstory, and man, the premise is pretty awesome).

On topic: what I personally like about the exp-to-loot gain is not that it discourages farming (it doesn't as far as I can tell), but discourages farming backwards. Obviously this encourages players to move forward with newer characters, causing them to progress through the game, and get better loot and experience. It also means, when they roll a new toon, they won't want to take their high-level, already-beaten-the-game-90-times toon to farm up some gear to mule over; they'll want to fight those bosses fresh with the new toon, as that's how they'll get the best loot. And, because they won't be overleveled, it'll still be challenging and exciting. If you give players a reset on the penalty, it'll actually encourage them to go back and fight those boring fights even more so than no penalty at all, because it's a virtual reward.

The problem is, players will, against all reason, farm like compulsive gamblers for better loot. While there are a few true, OCD farmers for whom this is the entirety of the gameplay experience, for the rest of us, the game suddenly becomes unfun, and before you know it, you're burnt out.

Surely you must have recognized this on some level, otherwise why have a loot penalty for repeating a boss chest in TQ? It keeps players moving forward, and that's what players want--they want to get better gear by moving forward. An exp system doesn't need to have a hard line drawn; it can be more gradual and allow the player a few decent runs before the slope begins to get steep, meaning players who have been falling behind, or have made poor skill decisions, can still farm up. That's a perfectly reasonable and common ARPG thing to do.

And, players who want to gain an advantage still have the hand-me-down edge with better gear from later sections of the game that can help them through those tough boss fights more easily.

And, best of all, OCD-like farmers will be even happier with the new system, as it means they get to make spreadsheets to figure out what the prime farming sweet-spots are for their desired items--I can already see post number 42 in the thread Top-25 Optimal Farming Runs--"Oh ho ho, DarkAvengerCrow, you may be correct that at level 48 a hero has a 38.229% less chance to get a Golden Scepter of Eternal Glory than a level 33 hero, but a level 48 Ovinomancer has the Sheepination skill, which lets them get past the Dread Guardians 3 minutes faster than a level 33 could, which means a 62.765% reduction to your posted farming run time, which actually increases your hourly chance to get the GSoEG by 24.536%!!!"
  Click here to go to the next staff post in this thread.   #7  
Old 01-14-2010, 11:28 PM
medierra's Avatar
medierra medierra is offline
Crate Employee - Designer
 
Join Date: Nov 2009
Posts: 4,020
Default

I'm not sure the proposal of tying exp-gain to drop rate will accomplish what you're talking about though if I understand you correctly. So long as you're fighting a boss that gives you decent exp, farming would probably be just as effective as it is now. Typically when I'm farming a boss, it is a boss that will give me good XP because A) I want to also be gaining XP while I farm otherwise I'm collecting loot at the expense of char progression, which is already penalty enough in my mind and B) if you're farming lower level bosses, you're much less likely to get good loot that is going to be useful to your current character since they will tend to drop lower level items.

The main purpose of the high initial drop rate on first time opening of boss chests and then diminished rate thereafter was to ensure that people had a high chance of getting a good reward the first time defeating a boss. After that it went down to what we felt was a more typical drop rate otherwise people could quickly farm every item in the game. Of course, since the probability settings on the drop system in TQ are terrible, people often didn't get a very good reward. We've vastly improved the way we handle drops in GD.

Are you trying to prevent people from farming lower-level bosses to collect gear for new characters? Perhaps I am confused. I interpreted yerk's suggestion to mean that drop rate would be tied to the rate at which players were earning XP. Perhaps you're talking more about a system where drop rate is tied to the disparity between player and enemy level?
  #8  
Old 01-15-2010, 04:22 AM
Munderbunny Munderbunny is offline
Initiate
 
Join Date: Jan 2010
Posts: 14
Default

Quote:
Originally Posted by medierra View Post
Are you trying to prevent people from farming lower-level bosses to collect gear for new characters? Perhaps I am confused. I interpreted yerk's suggestion to mean that drop rate would be tied to the rate at which players were earning XP. Perhaps you're talking more about a system where drop rate is tied to the disparity between player and enemy level?
Yes, the latter? This is how I understood it: the more exp you gain from the monster at a kill (the percentage of your the exp it takes for you to go from your current level to the next), the more likely you are to get a good drop. So, if I'm a level 60 and I kill Typhon lvl 66, I should get a bigger chance for primo-drop than if I'm level 75. But, it has to be the exp gain from the monster, not the monster level alone--don't want to get the same boost when killing a level 66 Maenad champion; the exp is always--ideally--tied directly to the difficulty of the monster, so having the relational gain seems like the best way to give greater reward for overcoming greater challenge.

I did not understand the suggestion to be like based on exp per minute or something--I definitely don't want the best farming to be AoE-rolling Satyrs. :O

And, just to be clear, I'm not down on farming; I think it's fine and fun (within reason). I just think players become item junkies (have been there myself), getting skittish with the game when it starts to get a little difficult, and stopping in their tracks to begin farming like crazy. While it's smart to farm up some if you're struggling, I don't think it should be to the exclusion of improving the hero's skills (and the player's). Overcoming obstacles through more than a shiny breastplate is far more rewarding if you can get there. However, if you can just kill the same mini-boss over and over again, stock up on a bunch of loot, what follows will be less exciting, and every subsequent drop less interesting.

Dunno if you've tried Torchlight, but there's this thing they have, where you can buy these dungeon scrolls that teleport you to one of their many cut-and-paste dungeons. The monsters are much higher level than you, it's tougher, the drops are better, the exp gain is better. You get to the end of it at some point and are teleported back to the city. I did two of those in a row. I went back to the main quest, and I was grossly overlevelled and overequipped for the game. The main quest became too boring to play, yet I didn't want to spend anymore time pointlessly spelunking for levels and loot. It was a stalemate. I don't know if their dungeon scroll thing was just messed up, and they weren't supposed to be that hard or rewarding, but that's exactly the kind of thing I think needs to be avoided: encouraging players to stop making forward progress.

The exp/droprate seems like such a clever solution to that. I've never heard of anything like that, but I like it instantly. So, I'm voting yes, if something like that were even doable within reason.
  #9  
Old 01-15-2010, 11:24 AM
jiaco's Avatar
jiaco jiaco is offline
Super Moderator
 
Join Date: Jan 2010
Posts: 6,164
Default

Quote:
Originally Posted by medierra View Post
Of course, since the probability settings on the drop system in TQ are terrible, people often didn't get a very good reward. We've vastly improved the way we handle drops in GD.
This is the most important aspect of farming, IMO, that needs to be fixed. People that re-play the game ALOT, will farm. But when a first timer in the Act I Labyrinth in TQ, is looking for SBCs and gets crap drops time and time and time again because the random number never rolled right is just nuts. I have played with these weights and such in the loot tables quite a bit and it seems like you developers had to push the crappy table weight to extreme values to limit the incessant triggering of the good loot as if I just take each loot record and divide the largest weight by 10, all of a sudden there are blues all over Helos. But if you could somehow put a more toon-dependent equation into the formula that dictates loot (and remove the dependence on the initial random roll at the same time) than that would make the best re-play value as then a new toon could just skip the trash mbs and head straight to the loot-able areas in hopes of finding something worth the tough fight.

While the exp-to-drop-rate equation could be cool, it almost seems like the drop-rate should depend on a player OA/DA vs monster OA/DA or some such stat which takes into account gear. As it seems that in an attempt to thwart people going back in game with their level 75 toon to farm bosses, you have created a loophole where players can just take a level 18 toon twinked to high heaven and get better farming results.
  Click here to go to the next staff post in this thread.   #10  
Old 01-15-2010, 11:45 AM
medierra's Avatar
medierra medierra is offline
Crate Employee - Designer
 
Join Date: Nov 2009
Posts: 4,020
Default

I think a completely different solution that might work better would be to create special "challenge" areas or dungeons where the enemies are tougher than normal but also have significantly higher chances of dropping good items. Then basically balance the difficulty and drop rates such that it is more effect to farm an entire area like that then to just repeatedly kill the same boss. In my opinion this seems like a better direction since it isn't adding a new "rule" to the underlying game mechanics that might seem to penalize certain players. Instead it seems like a universally positive addition that, at the same time, would help to reduce the monotony of performing runs on an individual boss. I also still like the idea of making it so that each area has a different range of possible uber items that can drop.
Closed Thread

Tags
boss runs, crafting, difficulty, drop rate, equipment, farming, game design, loot, loot runs, spawning

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


Grim Dawn ©2018 Crate Entertainment, LLC.
vBulletin® 3.8.4 ©2000-2014, Jelsoft Enterprises Ltd.