Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
Hyriol

Unskilled Crafters > Skilled Crafters

Recommended Posts

Hey crafting nerds,

 

During testing this week, I noticed a quirk in the risk-based item experimentation system.  I can elaborate on any of this if ya'll would like, but here's the short-ish version:

 

1) Crafting experiments at 50% risk or higher give you a 50% bonus per experimentation point spent on your result.  So, a "success" with three points spent (37.5% risk) gives you an increase of 3.15%, but 4 points spent  (50% risk) increases by 6.32%.  This is a 1.05% vs. 1.58% per point spent on your result (+50% increase).  This holds true all the way up through "amazing success" 8.05% vs. 12.08% per point spent.

 

2) Increasing your crafting skills (which is simulated in Big World via potions) decreases your risk per point spent from 12.5% to 11.1%.  Ordinarily, this sounds like a pretty sweet deal, however....

 

3) Gray quality materials limit the number of experimentation points you can spend to a maximum of four (two in grade and two in durability).  A skilled crafter, then, spending all four points at once, will have an experimentation risk of 44.4%, which will not trigger the result bonus.  The unskilled crafter, however, will hit 50% risk with all four points, achieving a higher potential result when you factor in the risk bonus.

 

Bored yet?  Good, here's a video:

 

https://youtu.be/FPEsMUsaDZ0 (You'll probably want to hit 'pause' at the important bits)

 

It did take about 25 tries before I scored an "amazing success" while unskilled, but the result was better than the skilled version could *ever* achieve.  It also *only* holds true for gray quality materials, since maximum experimentation points is the limiting factor.

 

This could either be:

A) a bug, since a noob scoring better than an expert doesn't make much sense or...

B) an intended design, so that new crafters have a permanent "beginner's luck" to have some relevance until they start earning skills.

 

Does anyone out there have any data that will support (or refute) these observations?  Have I made an assumption somewhere that I shouldn't have? Questions?  Comments?  Fell asleep from all the carebear?

 

Have at it, gents and lady gents.

Edited by Hyriol

Share this post


Link to post
Share on other sites

So this only holds true for gray mats? Once you start using whites, greens, and blues, being a skilled crafter will be better than an unskilled one? If that's the case, then I'd be okay with it, let the beginners have their luck. Most Gray gear won't be quality gear anyways.


giphy.gif

You Can't Be A Genius, If You Aren't The Slightest Bit Insane.

Share this post


Link to post
Share on other sites

An easy fix would be to have a skill that lowers the bonus cap. So while the untrained crafter needs to hit 50% risk, the trained crafter only needs to hit 40%.

Or just raising the effect of a single point when raising a skill leaving the 50% cap still (like a reward for risk). So it will be not 1.05% per point base, but 1.2% and 1.8% for 50%+ risk.

Share this post


Link to post
Share on other sites
I would really like to fully understand the system.  That said I think they may want to look at a different approach than dealing with "simple" percentages. 

Full disclosure, I used to write gambling software and the crafting system feels a bit like a slot machine. I think some methods used to generate the desired "feel" could and should be used here. People really want to feel like they are winning, so the crafting system should support that.

 

Here are three (of many) ways you can approach a % game.

 

Method one is like the D&D 20 sided roll to hit.  You throw a die and add a modifier to that roll. With this model your % chance moves up directly, and at a certain point "misses" become impossible, and "hits" become inevitable. you apply an automatic sliding scale so that on a 20 or above you get a critical,  

 

Method two is pass/fail and generate results.  So this is the "roll for damage" part of D&D.  Once you pass the check, how good/bad you did you do is based on a second check. I think this is the system being described above to apply the growth for amazing success times risk.

 

Method three is more dynamic, in that the full range is only known after all modifications.  In this type of system you assign each type of result a range that you want it to occur in, then alter the specific range with modifiers.

 

For a practical application, suppose you wanted to start with the following chances

 

5% Critical Failure

5% Failure

10% Moderate Success

50% Success

25% Great Success

5% Amazing Success

 

Create a universe of 20,000 possible results, and assign each % 200 points in that universe.  5% becomes 1000 for example.

 

Then you can manipulate any of the possible results by simply expanding or shrinking the specific result size.  For example, say you have an ability that reduces your failures by 50%.  In that case you modify Critical Failure, and Failure by -500.  The universe shrinks from 20,000 to 19,000 and you then pick a result.

 

A system calculated like this allows for all sorts of flexibility. For example a "risk" of 20% could impact an increase the size of Critical Failure and failure by 1000 each, increasing the universe by 2000 to 22,000. (Not exactly 20%)  This also reduces the chances of an amazing success from 5% to 4.5%.

 

Skills could have a shifting effect, by causing for example a reduction in "Critical Failure" of 100, and increasing the chance of "Success" by the same 100, creating no change in the likelihood of an "Amazing Success".

 

I think the effect you are seeing here is the unintended consequence of not separating out all the possible variables without considering all the possible relationships.  By separating out each component a little bit more, the system could better cope with the possibilities. 

 

Not sure why they wanted the final results of "Amazing Success" to be different based on "risk" in the first place though.  I would think of anything that should be impacted by skill. 

Edited by KrakkenSmacken

Share this post


Link to post
Share on other sites

Not everything is in yet with regards to crafting. That being said a skilled crafter is more likely to obtain a consistent result.

 

To make the best of the best will require a metric ton of materials and you should also consider playing the lottery. As an example, I will break down a long sword.

Names may be wrong but the basic list is the same.

 

Longsword

Weapon Blade: Long

3x Metal Bar

Weapon Grip

Pommel

1x Metal Bar

Hilt

1x Leather/Plank

Crossguard

1x Metal Bar

 

That requires 12x Amazing Success on Experimentation to get the absolute best result. Now unless they bungle crafting, a grey quality item a Skilled crafter should use less materials to produce the same result.

 

We don't know what the final results on crafting are going to be as they play with numbers and add more features. So things may change a lot between now and then. I am watching crafting closely to see if it will be a truly viable thing to be or just something someone in a guild picks up to easily supply everyone with.

Share this post


Link to post
Share on other sites

To make the best of the best will require a metric ton of materials and you should also consider playing the lottery. 

 

That requires 12x Amazing Success on Experimentation to get the absolute best result. Now unless they bungle crafting, a grey quality item a Skilled crafter should use less materials to produce the same result.

 

 

Did you do this math on higher quality base materials than green, and with all the bonus experimentation buffs possible in the training chain, specifically the ones that add more dots?

 

Also, unlike the lottery, we will be able to create BP's from our Amazing Success. So once you have one amazing success, on say the metal bar of your choice, that same success will be usable for potentially 100's of future metal bars. So in your example above, if your intent is the same metal bar type for the whole item, you can combine 5 of those into 1.

Share this post


Link to post
Share on other sites

Did you do this math on higher quality base materials than green, and with all the bonus experimentation buffs possible in the training chain, specifically the ones that add more dots?

 

Also, unlike the lottery, we will be able to create BP's from our Amazing Success. So once you have one amazing success, on say the metal bar of your choice, that same success will be usable for potentially 100's of future metal bars. So in your example above, if your intent is the same metal bar type for the whole item, you can combine 5 of those into 1.

 

Yes I did experiments with grades up to 'Orange' (obtained by killing a dev). There are more pips to fill with a higher grade (and I had to choose which to complete on 'Purple' quality if memory serves me). Quality of ingredients does not change the initial success rate (And for all we know this is inflated) but the success rate of Experiments is determined by your skill which varies by item.

 

As for blueprints, yes but I am pretty sure it will have limited runs of an unknown quantity. If it is unlimited (or even 100's lol).. do we even need to discuss crafting? Let alt accounts do it.

 

The simple facts are that until we have a more complete crafting system its mostly guessing.

Share this post


Link to post
Share on other sites

 

 

The simple facts are that until we have a more complete crafting system its mostly guessing.

 

Your right about that.  "Risk" is obviously not "Chance of failure", because I made a few things tonight that had 100% risk, and actually came off with Amazing Success.

 

The math is not... obvious.

Share this post


Link to post
Share on other sites

 

I would really like to fully understand the system.  That said I think they may want to look at a different approach than dealing with "simple" percentages. 
Full disclosure, I used to write gambling software and the crafting system feels a bit like a slot machine. I think some methods used to generate the desired "feel" could and should be used here. People really want to feel like they are winning, so the crafting system should support that.
 
Here are three (of many) ways you can approach a % game.
 
Method one is like the D&D 20 sided roll to hit.  You throw a die and add a modifier to that roll. With this model your % chance moves up directly, and at a certain point "misses" become impossible, and "hits" become inevitable. you apply an automatic sliding scale so that on a 20 or above you get a critical,  
 
Method two is pass/fail and generate results.  So this is the "roll for damage" part of D&D.  Once you pass the check, how good/bad you did you do is based on a second check. I think this is the system being described above to apply the growth for amazing success times risk.
 
Method three is more dynamic, in that the full range is only known after all modifications.  In this type of system you assign each type of result a range that you want it to occur in, then alter the specific range with modifiers.
 
For a practical application, suppose you wanted to start with the following chances
 
5% Critical Failure
5% Failure
10% Moderate Success
50% Success
25% Great Success
5% Amazing Success
 
Create a universe of 20,000 possible results, and assign each % 200 points in that universe.  5% becomes 1000 for example.
 
Then you can manipulate any of the possible results by simply expanding or shrinking the specific result size.  For example, say you have an ability that reduces your failures by 50%.  In that case you modify Critical Failure, and Failure by -500.  The universe shrinks from 20,000 to 19,000 and you then pick a result.
 
A system calculated like this allows for all sorts of flexibility. For example a "risk" of 20% could impact an increase the size of Critical Failure and failure by 1000 each, increasing the universe by 2000 to 22,000. (Not exactly 20%)  This also reduces the chances of an amazing success from 5% to 4.5%.
 
Skills could have a shifting effect, by causing for example a reduction in "Critical Failure" of 100, and increasing the chance of "Success" by the same 100, creating no change in the likelihood of an "Amazing Success".
 
I think the effect you are seeing here is the unintended consequence of not separating out all the possible variables without considering all the possible relationships.  By separating out each component a little bit more, the system could better cope with the possibilities. 
 
Not sure why they wanted the final results of "Amazing Success" to be different based on "risk" in the first place though.  I would think of anything that should be impacted by skill. 

 

You are my favorite person today.


https://en.wikipedia.org/wiki/Software_release_life_cycle#Pre-alpha <--this is where we are. If your complaint is that the game don't not works good, come back later.

Share this post


Link to post
Share on other sites

You are my favorite person today.

Thanks.

 

Like I said above, I used to build gambling products. There is an axiom in that industry when designing a system where loss is possible and frequent.

 

"It is more important to program the experience for the losers than the winners."

 

Winners get the win, so aside from blinking some "Yay You!!!" lights and making money noises (if it's a slot machine), it's pretty easy to make someone who is winning feel like they are winning. You should drag wins out a bit, and even if the result is a push, or in the case of a slot machine were you got part of your bet back but not all, make it seem/feel like a win if possible. The current rolling up of the results numbers does a really good job of that now, I would change the text color and item background every time it crosses a threshold using the same color language as materials, but other than that, winners take care of themselves.

 

What you want to do with losers is reduce the impact of the loss as much as possible by doing a couple of key things.  

 

1.) Don't show them the loss for very long.  Make it go by as quickly and painlessly as possible onto the next event.

2.) Don't highlight the loss, make it seem like the "normal" result, and move on.  That huge red failure and delay on the screen until you click something is probably the worst way of dealing with an event like that.

3.) Do give away extra chances.  What I mean by that, is when you have a high chance of a really crappy outcome (Payouts below 90%), replace some losses with free tries.  In the Crowfall world that would be adding "Inconclusive results" on experiments that do nothing or maybe take one dot total potential without blocking it's spot in the formula instead of all the dots attempted. On combines create "resources saved" result that lets you try again.

 

Crafting is going to be far more like a slot machine than I suspect ACE imagines, but I already see all the elements, from increasing your bet (different quality of materials), to prize tiers, (different quality of results), and all the process around "winning" as mentioned above. 

 

Because ALL high end crafters are going to consider almost everything but an "Amazing success" to make BP's out of as a "Loss", and the way they have the risk/reward set up there is only really two ways you want to do it. Experiment with only one dot at a time for "must work" and be decent items your not planning on making a BP for, and maximum risk to go for the Jackpot for anything your trying to make a BP out of.  

 

Given that assumption, players are going to need to be drawn to the process as much as the results, because crafting is going to involve a great deal of time running experiment after experiment looking for the "win", and ACE would do well to think long and hard about how they want unsuccessful attempts to "feel".
Edited by KrakkenSmacken

Share this post


Link to post
Share on other sites

I think this is a really important point for the developers, because simply by opening up the pre-alpha gameplay to people who are unfamiliar with the game development process, they have demonstrated how incredibly upset people can be when reality doesn't meet their expectations, regardless of how unrealistic those expectations were. I know that they've lost a couple of players because those players logged into the pre-alpha Big World expecting something closer to a late-stage beta test, but it seems like that's just human psychology writ large.

 

With crafting slated to be such a huge and integral part of the final game's DNA, I think it's important for losing to still feel good, because the game will need a large crafting base in order to have a functional economy, and in order for it to be large, the most dedicated participants can't be the only ones who are successful. It needs a low barrier to entry, and every part of it needs to feel good. Between that fact, the math you provided, and the fact that you explained it like I was 5 without seeming condescending, I think your analyses are important ones that everyone should be aware of. Or at least the developers. >_>


https://en.wikipedia.org/wiki/Software_release_life_cycle#Pre-alpha <--this is where we are. If your complaint is that the game don't not works good, come back later.

Share this post


Link to post
Share on other sites

And while I do understand that there are pieces missing that will be added later - like a search function in the recipe list and the ability to use a scroll wheel to navigate up and down - the core behavior of crafting needs a bit of fundamental tuning, as well. Being forced to click the Abort button after failing a craft check is problematic because it makes me feel the loss more keenly, and as obvious as that insight is in retrospect, it didn't occur to me at all.


https://en.wikipedia.org/wiki/Software_release_life_cycle#Pre-alpha <--this is where we are. If your complaint is that the game don't not works good, come back later.

Share this post


Link to post
Share on other sites

Update for anyone who was interested in the original thread:  

 

They seem to have patched out the per-point Risk reduction that the potions were providing.  As of the last test, 1 experimentation point was creating 12.5% risk across the board, regardless of potions or skill.

Share this post


Link to post
Share on other sites

 

 

2.) Don't highlight the loss, make it seem like the "normal" result, and move on.  That huge red failure and delay on the screen until you click something is probably the worst way of dealing with an event like that.

 

Yeah that's like rubbing salt into the wound  - "You Failed Looser !"

Edited by Count_Dirkoff

o8WHnLc.png


THE most active European guild. Join us

now!

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...