Please reconsider the probability-based crafting system

[finelinebob]

I have posted a discussion about this elsewhere so I'll be brief (well, briefer than I am there), but it didn't strike me until I crunched the numbers.

 

Crafting in SWTOR requires luck; skill has nothing to do with it. Even if you get your skill to 400, even if you get your affection to 10k on all your companions, your RE chance never changes from 20% or 10%, depending on the item. To make matters worse, the schematic you get when you RE something can be absolutely worthless. Case in point? A recent implant RE of mine that produced an item with cunning and a secondary of shield chance. Tell me, how many operatives, snipers, scoundrels or gunslingers use shields?

 

All hard-core crafters want to maximize their game. Since we have no way of improving the quality of an item beyond the schematic's strict definitions, that means providing variety. How many crafters out there dream of having all the purple schematics for their craft? Have you ever wondered how long it would take? How much it would cost?

 

I did, and here are my calculations for Armstech ... just for green weapons you can RE to blues then purples. No augments. No barrels. No schematics that start as blues or oranges.

 

If you are neither fortunate or unfortunate, your success rate will actually come out to what your chance for success is. Ok, those of you who know a little about the Law of Probability are saying "Wait, a 10% chance still means it is possible to fail 1000 times in a row." That's true. There's also something called the Law of Large Numbers -- it's the one most people confuse for the Law of Probability most of the time. It basically says here that if I craft enough times, my success rate will approach my success chance.

 

At the rates we have to RE items (20% green weapons, 10% blue weapons), it will take 2600 crafted and RE'd green items of the 172 weapon schematics to get all possible blue schematics, and approximately 72,000 crafted and RE'd blue items to get all possible purple schematics. The Law of Large Numbers is a reasonable assumption here.

 

BioWare, that's the first problem. I'm trying to talk about crafting. I have yet to really about any sort of exercise of skill. This has all been an exercise in mathematics. Where's the crafting?

 

Let's talk investment of time. How long does it take to craft all those items? Try 1.5 years of sending out crew missions. Ok, it's true that you can send up to 5 companions out at once, but that still makes it 3 months, 24 hours a day, seven days a week of real time, not crew time.

 

Let's talk resources. First of all, common materials. You will need approximately 257,000 resources for those crafts, requiring 23,000 nodes harvest (average yield of 6 items per node) taking 3 weeks of real, at-the-keyboard can't-have-your-crew-do-it 24/7 time (figuring it at 400 resources per 30 minutes, about my average for manual harvesting). Why not send crew missions? We need to save money and crew time for other tasks, that's why.

 

Armstech uses fluxes as well. The most cost-effective way is to send out crew missions, but in this case the crew time spent is more costly than just buying from the vendor. Mark this down as 4500 units of various types of flux costing over 1 million credits.

 

Finally, all those blue crafted items require Investigation materials. This was the trickiest calculation for me, since time, cost and yield availability for missions varies so much. You can't just say, "Hey Corso, here's 1000 credits ... go to the store and buy some Phrik!" Missions for Phrik may be unavailable until you run and finish another mission and your mission list is randomized again, or the Rich Yield missions with the best bang for the buck may not be available. So, based on Abundant Yield missions, to craft all the blues you will need to RE will take approximately 16000 missions at a cost of 13.7 million credits. If you think that's a lot, you will have time to earn it ... it will take approximately 61 years of crew time to run them.

 

Yes, 61 years of crew time.

 

Or, with 5 companions at 10k affection running Rich Yield missions concurrently, you might be able to cut that down to 8 real years -- 24 hours a day, 7 days a week, 365.25 days a year.

 

Will this game even be running 8 years from now? Is the development team that hopeful that they would devise a system that would take a player 8 years to prepare for a grind that will take an additional 1.5 years?

 

Do you consider that to be a fun system of crafting?

 

My suggestion is this: Look at the other MMOs that have had truly skill-based systems of crafting. Copy what was/is best. Toss out the rest.

 

And toss out the notion that randomness is fairness. That randomness is challenging. That randomness is a "good idea", It's very possible, though not probable, that some poor crafter out there will never, ever RE an item from even a green to a blue because every single time they hit that RE button and right-click an item, they have an 80% chance to fail -- regardless of skill as measured by in-game metrics, regardless of companion affection, regardless of how knowledgeable they are about the game itself. Randomness doesn't make things "hard" or "challenging," it makes them futile and insulting.

Edited May 29, 2012 by finelinebob

[finelinebob]

I just got a suggestion from a friend that makes sense for RE. Instead of having a purely random chance of getting a schematic from a table that includes useful AND useless items (such as items with cunning and shield chance), allow us to RE toward a specific build. Do we want a Hawkeye or Expert or Tempest item? Give us a pop-up or something that will allow us to choose the build we will receive if we successfully RE an item.

[anonnn]

That is truly a huge amount of time, if you want to learn all the RE schematics in your profession. However, have you considered that BioWare intended it to be a somewhat unreachable goal? It's become obvious to me after a certain amount of crafting and reflection, that BioWare intended to be exactly as it is. I think they've made the right design choices on the crafting system.

 

The reason is this: Rarity

 

In order for crafted items to be worth anything, they have to be rare. That requires them to need rare resources to craft, or be difficult to learn the schematic in the first place. It's the reason that Augmented and Artifact items sell for so much on the GTN, and if you reduce the ease of producing those items, crafting will be made far less fun because too many people will be running around in purple suits and no one will be buying the Prototype items.

[Mythlorr]

One way that I thought of to improve the current system is that every time you RE an item and don't get a new schematic, you have an increased chance to get a new schematic the next time you RE that item.

 

So if say you have a green you want to RE you currently have a 20% chance to get a new schematic. What i'm saying is that the next time you RE that same item you have an increased chance (say by 5% or 10% (I prefer 10%)) to get that new schematic, say 30%. And if you still don't get it and you are unlucky with more REing, eventually you will have a 100% chance to RE that item and you will need only one more to get that schematic.

 

So instead of thousands of items needed as shown above you will only need about 10 or so (based on a 10% increase every time and the crafter being very unlucky)

[finelinebob]

That is truly a huge amount of time, if you want to learn all the RE schematics in your profession. However, have you considered that BioWare intended it to be a somewhat unreachable goal? It's become obvious to me after a certain amount of crafting and reflection, that BioWare intended to be exactly as it is. I think they've made the right design choices on the crafting system.

 

The reason is this: Rarity

 

In order for crafted items to be worth anything, they have to be rare. That requires them to need rare resources to craft, or be difficult to learn the schematic in the first place. It's the reason that Augmented and Artifact items sell for so much on the GTN, and if you reduce the ease of producing those items, crafting will be made far less fun because too many people will be running around in purple suits and no one will be buying the Prototype items.

 

I'd agree with what you have to say -- that may have been BW's purpose. The problem is that they have failed at it.

 

If something in a game ought to be rare, it also ought to be clearly worthwhile. Other than, say, Biochem stims that (1) buff your primary more than an equivalent-level vendor stim, (2) last twice as long as vendor stims, (3) buff a secondary as well, and (4) persist beyond death (I think the advantage is clear here), how many people in the SWTOR community believe that purple crafted gear is better than custom gear with appropriate-level mods? Sure, the custom gear is going to have higher endurance and primaries, but when does the value of the secondaries make up for the drop you'll see in those primaries? I have yet to see any websites with any analysis.

 

For example, let's say I want to equip my gunslinger with two pistols that will let him hit harder and faster. I'll need alacrity. But should I favor power and maybe accuracy, or a crit/surge build? I'd like to be able to test it out on those new target dummies in my ship, but I need to RE up to the schematics. Going power/accuracy means REing off an Overkill schematic, but depending on the starting green item it may or may not have any alacrity. In any case. I'd best be looking for a purple schematic with secondaries of power/accuracy/alacrity ... if it exists. What are my chances in getting it in two consecutive crafts? If I remember the math right, that would be around one in one thousand. And that's just the one build. To max out the crit side, I need to RE off a Critical schematic of the same base model. I then need a crit/surge/alacrity build, if it exists, with the same chances of getting it quickly.

 

And I don't even know yet whether either is worth it compared to a customizable pistol I can get as an almost guaranteed drop in a flashpoint or heroic. Is it any wonder why there is no economy in this game, no demand for crafted goods?

 

I really wanted to have my commando in all purple gear leveling up. No reason I shouldn't have been able to, since I had crafters at 400 for each profession. But the amount of time it took to craft and RE to get the schematics, and the cost due to the vicious randomness of the failure chance, and the expense, led to me settling on blue gear a number of times. Or settling on non-optimal purple gear, because the 2 out of 3 or 3 out of 4 of the secondaries it boosted were adequate.

 

Rare is good. I like rare. But achieving something rare should be appropriately rewarded.

 

In the combat aspect of the game, that is the case. You get up to 50, you learn how to work your spec, you get together with equally-skilled people, and you succeed at Hard Modes and Nightmare Modes. And you are rewarded for it. Sure, there is a degree of randomness in what drops off of a boss, but there is no degree of randomness in whether something will drop. You kill the boss, it will. You do your dailies and weeklies, you earn comms towards purchasing the same rare gear if you cannot get a drop. There's even less chance in PVP ... you do enough warzones, just about no matter how bad you are, and you get the rare gear.

 

In the current crafting system, though, "rare' is not an achievement. "Rare" is winning the lottery. And, given the quality of the prize, it's not like you hit 5 numbers and the Powerball ... it's not even like you hit 3 and the Powerball. You basically have won a free ticket. It may lead to something good, it may be worthless. But BW let you win one, and that's supposed to make you feel like you've accomplished something. You, personally, have achieved nothing through skill or knowing how to craft. That simply isn't how the system works.

 

Crafting, as it stands, is BW's version of the Star Wars Floating Craps Game. It's a game of chance that follows you around wherever you go. Craps has this beat, though, since the odds vary when rolling dice depending on the point you need to make, and so knowing the odds can change your game. The odds never vary here. And you don't even get to see the pot before you roll ... you might win a bundle, you might win a bottle cap.

 

And, I guess, for some people that's fine. Given BW's reputation for quality games, given the promise of the IP here, given the apparent goals of making this the best MMO around from the development team, I expect a crafting system that gives me some measure of control. After all, the crafting system in KOTOR2 had more control than this game does.

[Autorch]

Excellent observations and post, bob. Seriously. I'm with you. I love to craft but it's killing me. Also, I'm a soloist as a player, maybe this'll change with 1.3, but you're right, one guy working alone has zero chance (effective) of coming up with anything that there'll be any true market demand for, when better stuff is so easily and deterministically gotten elsewhere either by brute force dailies or group content.

 

Here's the question I want to put to you though, and sincerely--you've stated the problem admirably, but do you have any real suggestions for a solution? Like really, rather than "do better", which is effectively what you're saying. Like game out a real system here. I'd definitely be interested in reading it.

 

For instance, maybe what should happen is that there is no more reverse engineering. Instead of "Reverse Engineering", you have "Disassemble". When you Disassemble, you break the item down into tokens that represent potential stats. So in your example of wanting some gun with alacrity and accuracy, let's say you find or make a pistol with those two attributes. On disassembly, it breaks down into some amount of tokens like "Quickness Crystals" and "Precision Polymers" (the amount of which per disassembly could be deterministic or we could RNG it a little, let's say).

 

Then, there's no reverse engineering. There's forward engineering. There's a whole new activity and UI for you to plug in these tokens into some kind of bench or machine on your ship--maybe it's a kind of mini game that you actually have to demonstrate some sort of, yes, skill in operating it in order to come out with the best results.

 

Let's say this theoretical machine allows you to select "Pistol" and then what happens is dependent on the number of Quickness Crystals and Precision Polymers that you plug into it or not. This plus your twiddling skill results in either a schematic, which is then built with regular materials, or maybe this system just comes up with finished products. Anyway, because you tokenized it the right way and twiddled it correctly, you get a schematic or weapon that's tilted toward what you were looking for. Then it becomes either your twiddling skill or your specialized knowledge of token recipes that gives you proprietary knowledge and makes your products special and different, potentially unique in your community.

 

Anyway, just throwing out ideas, but point being, what this thread needs is more ideas. You did an excellent job pointing out the problem, but wouldn't it be good to suggest doable solutions, is all I'm sayin'.

[finelinebob]

I like your ideas, Autorch. I hesitated to say too much at first because (1) I'm such a big fan of crafting in SWG but (2) I know most people can't handle it's complexity and interconnections plus (3) trying to introduce such complexity now, even if desired, would probably totally screw the system worse than it is. So, building upon what you and a few others have said to me:

 

Get rid of the RNG grind. Or, perhaps, modify it, by one or more or all of these methods:

• Create a "learning curve" that increases your RE chance as you increase in skill
  
• Bend that curve favorably the greater the difference is between a crafter's skill and the skill required to craft it (REing a skill 1 gun for a 400 crafter should be a given but a skill 400 gun a challenge)
  
• If an item is crafted by a companion with a crit, you should get an RE bonus if they crit on the craft
  
• If an item is crafted by a companion with high affection, you should get an RE bonus (since loyal workers do more diligent, careful work)
  
• Crafters know their work, so give us a chance to choose the schematic we get from an RE
  
• As you experiment and fail, you learn as well. Failed REs should result in a greater chance of future RE success
  

 

That would go some ways into creating a real value to your "skill" level and affection level of your crew, and it would reward perseverance more than dumb luck. It still doesn't address the problem of fixed values for schematics, of how it's currently impossible to actually make one particular weapon or piece of armor any better than anyone else. We already have crafting stations in our ships and you may have seen your crew toiling over it when assigned crew missions. Wouldn't it be great to be able to use it yourself? Along the lines that you suggest, and borrowing from the RE system that SWG developed, why not an RE enhancement system to improve your schematics?

• REing a blue or purple item gives you a choice for a new item or improving the current item
  
• REing an orange item gives you a chance to increase its rating or add a secondary
  
• Improving the current item would be impossible without integrating other "tech"
  
• that tech would come from the gray items you gather that are currently vendor trash
  
• useful items would be seemingly obvious -- junk "vibroknife parts" to improve a vibroknife
  
• effects would not necessarily be as one might predict ... junk from starter planets might be necessary for esoteric secondary stats, but Corellian junk might give a big boost to primary stats
  
• Improvements would be on a curve ... big leaps at first, additional REs of the same item providing a smaller and smaller refinement, diminishing returns producing a theoretical cap to improvement
  

That would take some know-how. You would need to research the available junk and the effects it had. You'd have to know what mobs drop it. You'd be able to shade your creations to favor the stats you think different specs would need. Your knowledge and effort spent would create unique items of quality with real market value. And people who want no part of it could still sell stacks of junk loot on the GTN for more than it's vendor value, because it now has real value. Even not wanting to be part of the system, you could benefit.

 

And it would all be optional. Unless you enjoyed crafting, unless you enjoyed the game of it and the investment into it, you could always stick to the default game, the default items. Craft them for yourself, craft them for friends -- those default items would sell as well under such a new system as they do now ... which is to say, they wouldn't. But I haven't seen anyone fighting a change in crafting and happy with the status quo say that they are in it to make money and that they ARE making money. Nor has any such person said they can out-craft a heroic quest, flashpoint or hard-mode quest drop. They can't, and right now neither can I.

 

The mini-game of crafting can continue as is, and those who want to take it to the next level can do so.

[finelinebob]

BTW ... I'm not saying copy the SWG RE system wholesale ... I don't like the idea of crafted mods stuck on an item.

 

What I am saying it that you choose one item, say the Tempest Nano-Optic Cardio System, a level 49 purple implant with 4 nice stats (+59 Endurance, +50 Cunning, +69 Critical, +29 Alacrity ... my Gunslinger would love it...) and 1 horrendous stat (+40 Shield Rating ... my Gunslinger is now crying ). Maybe the system allows you to use that useless stat as a pool for either improving the others or morphing it to something useful, like Surge. Maybe you have to keep it, but can still improve all the other stats. It all depends on the junk you integrate.

 

Once you start to modify one particular schematic, you own it. The only way to change the path you go down is to delete it. You cannot simultaneously have more than one variation on one particular schematic. That should simplify the coding of it, by forcing things into a template system and allowing you to modify values within the template. Sure, being able to come up with unknown combos would be fun, but to be realistic it would be a nightmare to code and a nightmare to track ... lots of opportunities for exploited goods if this were done.

 

If you delete it, you start from a fresh template, but you still need to RE up to that template.

 

Worried about making a mistake? Not sure what a particular piece of junk will do? No database yet on the web with known effects of junk in RE? Well, what do you think low-level green and blue schematics would be for? You don't experiment on your best work when you have no clue what the effects will be, you start with expendable materials so your rejects don't cost too much.

 

The thing I like about this is the research and experimentation potential. Even though your choices for any one schematic would be limited, you'd still have choice plus the chance to be truly good at what you do. You'd be able to develop builds based on your knowledge of different class specs, bolstering the secondary stats that work best. Your work would become known as you produced goods better than what drops through brute-force gaming, and people who wanted to maximize their game would beat a path to your door. That "Created by" tag on a crafted item would begin to mean something. If you could craft something people would want, then from my experience in SWG and other MMOs as a crafter, I know folks wouldn't just be giving you their money. They'd be harvesting mats for you. They'd be running Investigation and Underworld Trading and Treasure Hunting missions for you. They'd grind junk loot for you, so that you could experiment and improve a weapon they like and KNOW you could improve.

 

Geez, if you could experiment on schematics that require exotic mats like BCA, how much do you want to bet there are folks who would run hard-modes for you to be able to craft and improve that item? If the possibility for success were purely random? Not one frickin' chance. Under a system of discovered/known effects where outcomes could be determined and risk of failure minimized (but not ruled out completely)? That would be a game-changer, and for the good for everyone.

Edited May 29, 2012 by finelinebob

[finelinebob]

And now for something completely different...

 

A new type of augment.

 

Above I argued for essentially implementing schematic templates that would constrain what you could improve. I did this to suggest something I think would be a reasonable change to implement in the code. Here, I have a suggestion for actually introducing new stats to an existing item.

 

These would be one-shot, "consumable" crafted modifications to existing items. They could potentially be unbalancing to combat, so they would have to be limited ... they might wear out and need replacing, losing power over time. Trying to remove or replace them might create a chance to destroy the item, with a smaller chance to destroy when putting them on. They would have to be expensive relative to the level of the item, as one means of making them "rare" or worth the effort. They would be along the lines of "illegal mods", so that it would be appropriate to have Slicing crew missions to gather needed materials for them and perhaps rare missions to have appear in your Crew Management list. Perhaps, instead, they appear in Slicing harvest nodes in Flashpoints, Hard-Modes, Operations ... even on Warzone fields, hehe. (Would you stop defending a node if a slice node popped up on the map? If you knew that harvesting that node might get you a mat worth a million credits?)

 

What sorts of effects could they have?

• DOTs
  
• HOTs (for those Mercs and Commandos who shoot friendlies to heal them)
  
• Roots & Snares
  
• Armor Breaks
  
• Interrupts
  

... to name a few. All effects that could be a small bonus to damage or a small combat effect. Low additional damage, chance for targets to resist breaks, snares or interrupts, effects that only proc a very low percentage of the time, so as not to be like giving a class a completely new special it never had with no cooldown or other cost.

 

Optionally, it could incorporate the Junk Loot system again. You want a fire DOT? Find some junk incendiary grenades. Root a target? Maybe a junk ion cell that mimics a sorc's lightning root. Junk bacta ampules for a HOT. Some sort of junk brain scanner or neural interface for an interrupt. Junk keen-edged blades to put a bleed DOT on that vibroknife for your shiv.

 

I just realized, rereading this post, that I'm talking all damage. Perhaps the opposite -- things to cancel such effects -- could be done for wearable items. Resist to DOTs and Armor Breaks for armor; Bonuses for Healing, resists to Interrupts or Roots for earpieces and implants. I got my head in Armstech right now, so maybe an Armormech or Synthweaver might have some suggestions on defensive boosts that could proc to counter enemy specials.

 

It would be expensive to get, hard to gather materials for, risky to implement, and minimal in effect. But still, it would have an effect at times, and how many hard-core gamers aren't looking for the slightest edge they can get?

 

It also would be totally new code. Unlike my suggestions above, which may be possible through modification of the existing system, this would be new ... that's why I'm putting it out separately. As mentioned, I'd worrying about it being unbalancing as well to the combat system, so I wouldn't want to consider it a core suggestion.

 

... I hope those responses answer your question, Autorch! Thanks for pushing me, thanks for the ideas, and thanks to the others I've talked to that have spurred this "belch" of thinking.

Edited May 30, 2012 by finelinebob

[Deewe]

There's a need for randomness in crafting but BioWare set it in the wrong place and to the wrong amount

 

If they want to improve the crafting system they have to think SWG, Vanguard, EQ II.

The good thing in these games is the crafter was in control of the output.

[finelinebob]

There's a need for randomness in crafting but BioWare set it in the wrong place and to the wrong amount

 

If they want to improve the crafting system they have to think SWG, Vanguard, EQ II.

The good thing in these games is the crafter was in control of the output.

 

ZOMG someone who liked Vanguard crafting, too

[Deewe]

ZOMG someone who liked Vanguard crafting, too

Well there crafting and... Crafting

[finelinebob]

This popped up on the other thread I mentioned in the OP -- a possible mechanic for increased RE chance.

 

The suggestion was (as also mentioned above) that the more you fail at an RE, the better your chance for success on the next. The poster noted that the devs have talked about having too many variables to track in game as it is and how loath they are to introduce too many new ones. Well, here's a mechanic with one new variable.

 

What can be used is a simple "fail counter" variable, an integer count of how many consecutive fails you have. It starts at 1 (I'll explain why not 0 below). You fail an RE, then the mechanism that now simply does nothing extra (instead of rolling for a new schematic for you on a success) now adds 1 to that counter. It keeps adding 1 every time that branch of code is called. When you succeed at an RE, add a line of code to the part of the routine that rolls for a new schematic for you, one that reset the counter to 1.

 

Why start at 1? Because right now, your RE success chance roll is probably a randomly-generated number between 0.0 and 1.0. When that result is compared against 0.2 (20% success chance) or 0.1 (10% success chance), if the roll is equal to or less than the comparison value, then you succeed; otherwise, you fail. Now, if you are starting at 0 consecutive fails, the fail counter has a value of 1 (that's why they call 1 the "multiplicative identity" ... 4th grade math ftw). Multiply that times your success roll and the roll is unaffected. What if you have failed once? Your counter is now 2, and your RE roll is effectively doubled.

 

(well, ok, someone is bound to point this out ... if you want a low result, multiplying is not the way to go; dividing is, or multiplying by the inverse. It's just that "dividicative", or whatever the word might be, is not as cool as "multiplicative")

 

If a linear increase in chance is too much, then I'm sure they could figure out some factor or function to modify the fail counter so that the effect isn't so pronounced. They could even "apply a curve" to it, so that the fail counter essentially moved further and further along a "diminishing returns" curve, with the cap value being 10 and an almost guarantee, eventually, for success.

 

It was also suggested that some people would want to "game" the system; to RE low-level green items in order to pump up their fail counter, then try a high-level RE. That certainly would happen. But two things could easily disrupt it -- one a natural consequence of the system, and the other an additional check,

 

The first counter is simply the fact that as you fail at REing an item, your chances get better. Let's say you decide to pump up your RE chance with 3 preliminary REs of low items. With improved odds to RE successfully, the more you prime the pump, the more likely you are to succeed earlier than planned and reset your fail counter to 1.

 

The second check would be this: REing an item with "No Research Available" would result in no increase to the fail counter. So, with as many as 3 possible blue schematics from a green item, it wouldn't take many early successes before that low-level schematic was useless for gaming the system. This would also be required to stop REing looted green and blue items in order to pump up the fail counter as well.

 

So, for those who want the system to include a feature analogous to crafters learning from their mistakes, namely an increased chance to RE when you fail, there's a concrete suggestion with even some ideas on how to code it in principle. Since it only introduces one new variable, it's hardly all that onerous, and the checks that can be placed on the system are reasonable and quite adequate compared to the most heavy-handed means of doing this: tracking each individual schematic with its own fail counter. That would be tens of thousands of additional variables.

Edited May 31, 2012 by finelinebob

[Darth_Halford]

just for the sake of curiosity, what would happen if they simply increased the chances learning new formula?

[finelinebob]

just for the sake of curiosity, what would happen if they simply increased the chances learning new formula?

 

Even if you had a 99% chance of success, there is always a 1% chance of failure. Extreme example, true. Improbable event, true. But you can still roll 1000 fails in a row. Or think about flipping a coin, 50/50. Much easier to see from that analogy how a fixed chance can lead to frustration, especially set low. There's no way they'll ever set the rate, as a flat rate, that high, IMO. Right now, they're not even giving you your companion's crit bonus, if they crafted an item or if you're REing an augment slot item.

 

The other problem with the current system, again, is that even when you crit, the new schematic you learn is random amongst up to 3 (green items, not mods or meds), 4 (redoubt) or 5 (critical, overkill) choices. And the devs were ... well, there simply is no other word ... stupid enough to include useless, pointless combinations of skills on some items (my long-time example, cunning items with shield chance), That means that after flailing around trying to get an RE success, you have a chance at getting pure crap.

 

It's kinda funny how part of the "story" aspects of this game, built-in RP if you will, happen in combat. You ever hear the little "war cries" different classes of characters make with certain specials? All the people I know who get into crafting like to think they are actually crafting something: that they have some control, that they are exercising some skill, using some knowledge. There is nothing skillful hinted at with crafting as is.

 

What would a master crafter do? They'd be better at things than lower skilled crafters, with simple items being easy to create and modify (boost to RE when a crafter's skill level is much higher than the skill required for the item). They would learn from their mistakes (increased RE chance on failure). They would create unique items that show their expertise and interests, and continue to refine them over time (RE to improve a schematic, raising the stats of items produced). Just raising the success rate might shut some people up (not me), but it still wouldn't change it from an RNG grind. It still wouldn't be crafting.