[Discussion] Do we need more-focused crafting in PoE?

If you're going to respond, please at least skim through what I wrote in the spoiler tags.

-----

Let's address the 800-pound gorilla in the room. "Crafting" in this game is not crafting. In fact, it's far from it. There are multiple threads everyday complaining about RNG this, RNG that, and a large part of RNG is present in what we call PoE crafting. You cannot guarantee any sort of results in X amounts of orbs used. Right now, it's literally:

  • Gather as many orbs as you possibly can.
  • Start repeatedly spamming those orbs on your item.
  • Let random number generator decide if you're worthy enough to finally hit the jackpot.

You might as well be throwing coins into a Vegas slot machine and pulling the giant lever over and over. It'd give you about just as much opportunity to influence the outcome as crafting does in PoE. That is to say, none at all. There exists no tools that people can utilize to ensure an even halfway predictable craft. Yes, some people enjoy gambling. But below, I want to propose a less primitive crafting system. Perhaps we can keep the excitement of gambling, but also make crafting a more rewarding process.

So what's the problem?
There are generally two types of crafting orbs in PoE: (1) those that reroll the properties of an item completely (e.g. Jeweller's, Fusings, Chromatics, Alterations, Chaos), and (2) ones that add stats to an existing item (e.g. Transmutation, Alchemy, Augmentation, Regal, Exalted).

The problem with both of these types of orbs is that each time you use them, you literally have no idea what the orbs are going to do. The orbs could roll your item back to a state worse than when you first started. The orbs don't take into account what the item is beforehand, nor can the player influence/restrict the item to a specific set of results to roll to. The orbs draw from the entire pool of possible outcomes, making crafting really just like a slot machine = no user input.

What I'm suggesting is a completely new set of vendor recipes where the user has some control of the outcome when crafting an item -- whether the goal be socket colors, number of sockets, number of links, or adding certain modifiers to items.
How crafting is supposed to be done...D2 style
For those who crafted in D2, I'm sure you can remember the Hit Power, Blood, Caster, and Safety items. For those who haven't played D2, let me give you a quick description of how it worked, and why I think PoE can benefit form a similar crafting system:

'Hit Power', 'Blood', 'Caster', and 'Safety' are item formulas with partially-fixed stats, each suited towards a specific type of playstyle/build. For example, the 'Blood' formula guaranteed that, on an amulet, you would get 5-10% Movement speed, (1-4)% Life Leech, and +(10-20) to Maximum Life as fixed modifiers. The remaining modifiers are then randomly drawn from the mod pool. To craft a 'Blood' item, you would have to collect: a Magic Amulet, an Amn Rune, a Perfect Ruby, a Jewel, and after putting all of them into the Horadric Cube, you would get your crafted 'Blood' Amulet. Aside from 'Blood' amulets, there are other types of 'Blood' items you can craft, which require different recipe ingredients and yield slightly different results. Same with the 'Hit Power', 'Caster', and 'Safety' formulas.

Okay, no more D2 talk. The point that I wanted to make is that results should at least be semi-predictable when crafting. You should be able to look at your build, and decide which type of items would suit you best. After you've identified that, you can then proceed to improve your existing items by crafting -- crafting in a way that utilizes recipes tailored to your build. These crafting recipes should require you to combine ingredients in non-similar ways to achieve your result. Crafting shouldn't simply be spamming a single currency nonstop to get your desired result. In other words, a person who wants a ring with mana regeneration shouldn't be using the same recipe to craft as someone who wants a ring with attack speed.

I'll provide a few examples below.
Example 1 = I want to 6S an item!
Let's say I want to 6S this chest:


Sure, I would still have the ability to test my luck and spam Jeweller's Orbs. But what if... there was a more predictable alternative. Something like this vendor recipe:

  • Gather 6 of the same base item type, of any rarity (in this case, Assassin's Garb)
  • All 6 chests need to be at least 5-socket already, and have 20% quality. Visual example:
  • [dbLinkedItem id="54204442" inline="0"][dbLinkedItem id="54204443" inline="0"][dbLinkedItem id="54204444" inline="0"][dbLinkedItem id="54204445" inline="0"][dbLinkedItem id="54204446" inline="0"][dbLinkedItem id="54204447" inline="0"]
  • Give these 6 chests to vendor, and in exchange, the vendor has a 1 in 6 chance of giving you a consumable "Book of Jewellers". You may also trade 40 Jeweller's to the vendor at the same time to increase your chances to 1 in 3. In my case, the book would look like this:



The same concept could also be used when trying to 5L or 6L an item (e.g. collect 6x 5L items for a 6L). Some people will ask: "why not just a flat amount of, say 200, Jeweller's for a 6S or a flat amount of, say 800, Fusings for a 6L?" Well, the point of these recipes isn't to buy your way into a better item. It's to encourage people to look for different ingredients and give them a choice of influencing at crafting the result they want.

Of course, this is just my idea, and all of the numbers are arbitrary. The recipe could be something completely different, depending on what GGG comes up with. What I'm trying to get across is that perhaps there can be a more tailored crafting system that requires the player to gather various different items to achieve the crafting result they want, instead of mindlessly spamming a single type of orb.
Example 2 = I want to get 5 off-colors on my body armour!
Let's say one day I decide I want 4 reds, 1 green, and 1 blue on this piece of body armour:

Since, the dex requirement is so high, using Chromatics is likely to give me majority green sockets nearly all of the time. Sure, I would still have the ability to test my luck and spam Chromatic Orbs. But what if... there was an alternative way for the player would be able to influence the socket color outcome. Something like this vendor recipe:

  • For each color that a player wants on a particular item, an equivalent-colored gem of level 20 must be gathered. In this case, I want RRRRGB, so I would need:
  • [dbLinkedItem id="54204449" inline="0"][dbLinkedItem id="54204450" inline="0"][dbLinkedItem id="54204451" inline="0"][dbLinkedItem id="54204452" inline="0"][dbLinkedItem id="54204453" inline="0"][dbLinkedItem id="54204454" inline="0"]
  • Gather 2 of the same base item type, of any rarity (in this case, Cutthroat's Garb). The total combined socket colors between these 2 items must be at least the number of socket colors you want on your target item. So in this case, I have a total of 5 reds, 1 green, and 3 blues (just enough):
  • [dbLinkedItem id="54204455" inline="0"][dbLinkedItem id="54204456" inline="0"]
  • Give the above ingredients to vendor, and in exchange, there is a 1 in 3 chance you will receive a consumable "Chromatic Book". If you trade 100 Chromatic Orbs along with the recipe, you increase your chances to 1 in 2. In my case, the book would look like this:



Again, this is just my idea with arbitrary numbers. The recipe could be something completely different, depending on what GGG comes up with. What I'm trying to get across is that perhaps there can be a more tailored crafting system that requires the player to gather various different items to achieve the crafting result they want, instead of mindlessly spamming a single type of orb.
Example 3 = I want to add +maximum life to my amulet!
Alright, one more example. Pretend that I wanted to add the +maximum life mod to the following amulet:



It currently has 3 suffixes, and 1 prefix. Since an item can have up to 3 suffixes and 3 prefixes, I can still get a +maximum life mod on the amulet (see wiki for list of all affixes). Each prefix or suffix also has its own "tiers". Now, the normal way someone would do this would be to use an Exalted Orb on the item in hopes of getting the life mod. But again, what if there was a recipe that would enable the player to narrow down the outcome to a specific prefix/suffix? Something like this vendor recipe:

  • For each "tier" of your desired prefix or suffix, gather an equivalent amount of the same base item type (of any rarity). Each tier should be assigned to exactly 1 of the items, no more, no less. In my case, I'm crafting a Lapis Amulet, and since I want the +life mod, I need to collect 7 Lapis Amulets, which cover the full range of life prefixes that are able to spawn on amulets: 'Healthy', 'Sanguine', 'Stalwart', 'Stout', 'Robust', 'Rotund', and 'Virile', as shown:
  • Gather 1 Exalted Orb
  • .
  • Give the above ingredients plus your target craft to the vendor, and in exchange, there is a 1 in 4 chance you will receive your item with the desired prefix/suffix, but a roll that can land in any of the available tiers. You may also vendor 2 extra Exalted Orbs with the recipe, which will increase your odds to 1 in 2. In my case, my Dread Choker Lapis Amulet would receive a life mod, but the tier would be random (i.e. could be 'Healthy', 'Sanguine', 'Stalwart', 'Stout', 'Robust', 'Rotund', or 'Virile').

Once again, these are all just ideas. GGG's recipes could be completely different. But for the last time, I stress that crafting should at least be semi-controllable.
Conclusion:
Hopefully from the above examples, you can sort of get an idea of what I'm suggesting in terms of allowing players to be more in control of their crafting. At the moment, I believe we have a very indifferent, unspecific sort of "crafting" system. The player has very little input as to what specific results there will be. This is usually what causes the numerous rage threads where people exhaust their entire bank trying to improve their items, only to be left with an item that is usually worse when they first started. This almost always results in a sense of lost progress.

People like to gamble; but they like to craft even more.
-----

Thanks for reading -- looking forward to discussion!
✮ in-game @FTMFW
✮ twitch.tv/RRTSON
✮ [Shop] http://www.pathofexile.com/forum/view-thread/909223
✮ [Build of the Week] http://www.pathofexile.com/forum/view-thread/1034503
Last edited by rrtson on Mar 31, 2014, 7:54:17 AM
This thread has been automatically archived. Replies are disabled.
I really enjoy the crafting in this game...
Maybe I'm the only one?

Your ideas are neat though. I'm always up for more options and more control.
So I'd need 6 identicle 5s items plus 40 jewelers to get a 1 in 3 chance of a 6s? And you'd prefer that over what we have now?

Different strokes for different folks I guess, but I don't see how this takes the randomness out, if anything I think it exasperates it.

I advocate for more of a Monster Hunter approach. Different mobs have more chance to drop certain crafting items. When you want to craft something, you seek out specific mobs that have an increased chance of dropping components that you need to craft whatever it is. It would require more complex recipes though, and for that reason I'm not sure if it would ever be considered.

Or a more simple alternative is slightly increasing the rates for different items to suit the monsters that drop them. For example, if you want summoner gear, go farm necromancers, they should have a higher chance to drop summoner gear than a bear. Want a bear trap gem? Go farm the bears instead.

It couldn't be made too reliable, there would still have to be a decent element of luck to it, but at least people might feel like they were working towards something in specific rather than just totally random farming.
Last edited by MonstaMunch on Mar 31, 2014, 7:25:24 AM
"
MonstaMunch wrote:
So I'd need 6 identicle 5s items plus 40 jewelers to get a 1 in 3 chance of a 6s? And you'd prefer that over what we have now?


I would actually. It's a lot less mind-numbing, and the items require players to go searching for uncommonly looted pieces of equipment. But once you have the pieces, your likelihood of attaining your goal is greatly improved. Also, the recipe would allow you to keep your existing chest piece as is, without destroying any links/sockets that it might already have.

And again, the numbers are arbitrary. The recipe is completely made-up and can be changed/adjusted. It's simply a demonstration of concept.

"
MonstaMunch wrote:
Different strokes for different folks I guess, but I don't see how this takes the randomness out, if anything I think it exasperates it.

I advocate for more of a Monster Hunter approach. Different mobs have more chance to drop certain crafting items. When you want to craft something, you seek out specific mobs that have an increased chance of dropping components that you need to craft whatever it is. It would require more complex recipes though, and for that reason I'm not sure if it would ever be considered.

Or a more simple alternative is slightly increasing the rates for different items to suit the monsters that drop them. For example, if you want summoner gear, go farm necromancers, they should have a higher chance to drop summoner gear than a bear. Want a bear trap gem? Go farm the bears instead.

It couldn't be made too reliable, there would still have to be a decent element of luck to it, but at least people might feel like they were working towards something in specific rather than just totally random farming.


That's definitely also another approach to improving loot acquisition. Certain mobs would drop specific items so you actually have motivation to scope out certain areas for loot you need. The two could coexist though. Give players more control in crafting and diversify loot drops between different monsters.
✮ in-game @FTMFW
✮ twitch.tv/RRTSON
✮ [Shop] http://www.pathofexile.com/forum/view-thread/909223
✮ [Build of the Week] http://www.pathofexile.com/forum/view-thread/1034503
Last edited by rrtson on Mar 31, 2014, 7:36:36 AM
"
rrtson wrote:
That's definitely also another approach to improving loot acquisition. Certain mobs would drop specific items so you actually have motivation to scope out certain areas for loot you need.


I think the biggest problem with my suggestion is that it's ripped from games that use a fundamentally different loot system; Top end items in games that use the system I referred to are generally only obtainable through recipes which require combining a bunch of base items.

The mobs are strategically scattered so that in order to produce a really strong item, you'll have to farm a series of increasingly strong enemies over a variety of areas and over a decent period of time. I don't know how well/badly that concept would translate to PoE, I think it would require a huge amount of planning and experimentation.
I would really (and I mean really) like to see a crafting system where you gather materials from monsters (loot too) and if the monsters are rare or sth they give rare loot (for example Tough Sap Wood(used in bow crafting)). But that won't happen for many reasons.
"
Evgenit wrote:
But that won't happen for many reasons.


Can you name a few? I'm not saying there aren't any, but I don't think any of them are insurmountable, other than the possibility that it may simply be something they aren't interested in implementing.
I really like the concept but if there is no way to achieve a 1-in-1 chance, I don't think I would bother using any of those recipe.
"
MonstaMunch wrote:
"
Evgenit wrote:
But that won't happen for many reasons.


Can you name a few? I'm not saying there aren't any, but I don't think any of them are insurmountable, other than the possibility that it may simply be something they aren't interested in implementing.


1. As you said yourself they aren't interested in implementing this unless many people support it which I doubt.
2. It requires a lot of work on their side. This isn't just another map or anything, this is an entire new crafting system.
Actually, these are the only reasons I can think of.
If you support my idea you can sign this petition!
@OP: Actually not horrible ideas. Might be overpowered, but that would just mean some tweaking. (In my original of this post I didn't really read the thread. Sorry about that.)
When Stephen Colbert was killed by HYDRA's Project Insight in 2014, the comedy world lost a hero. Since his life model decoy isn't up to the task, please do not mistake my performance as political discussion. I'm just doing what Steve would have wanted.
Last edited by ScrotieMcB on Mar 31, 2014, 3:29:05 PM

Report Forum Post

Report Account:

Report Type

Additional Info