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

Streaking: The Pros and Cons of RNG - Official Discussion Thread

Recommended Posts

1 hour ago, Arkade said:

Will the only possible outcomes be based on the lowest and highest mats used? So if I use white quality, blue quality and purple quality, a failure would be gray, a success white and an amazing purple? Or would there also be a chance of getting a blue?

It seems to roll for a result with the quality calculated just like TBlair posted above. So imagine we roll and get a purple result.

If we get a failure It is downgraded to green (or is it blue? I always get those things wrong. It is downgraded by 1) and loses its experimentation.
Sucess we maintain the purple quality and can continue to the experimentation part.
If we get amazing it is upgraded to the best quality item you used. If you used 1 orange mat you get a orange result.

It seem we actually have two rolls. One for quality that follows the math Blair showed above and another for the crafting result(failure/sucessful/amazing) which math is yet to be confirmed.

Edited by BarriaKarl

Share this post


Link to post
Share on other sites
49 minutes ago, BarriaKarl said:

It seems to roll for a result with the quality calculated just like TBlair posted above. So imagine we roll and get a purple result.

If we get a failure It is downgraded to green (or is it blue? I always get those things wrong. It is downgraded by 1) and loses its experimentation.
Sucess we maintain the purple quality and can continue to the experimentation part.
If we get amazing it is upgraded to the best quality item you used. If you used 1 orange mat you get a orange result.

It seem we actually have two rolls. One for quality that follows the math Blair showed above and another for the crafting result(failure/sucessful/amazing) which math is yet to be confirmed.

In my example, I was using white, blue and purple. So if I use 3 white iron, 3 blue iron and 3 purple iron, what will my results be for Fawed, Success and Amazing? Amazing I assume would be purple since that is the highest quality. The lowest quality is white, so I would expect a Flawed assembly to produce gray and a Success to produce white. 

Having the blue in there would make the assembly more difficult, but would it also give me a better chance of getting an Amazing?

Share this post


Link to post
Share on other sites
35 minutes ago, Arkade said:

In my example, I was using white, blue and purple. So if I use 3 white iron, 3 blue iron and 3 purple iron, what will my results be for Fawed, Success and Amazing? Amazing I assume would be purple since that is the highest quality. The lowest quality is white, so I would expect a Flawed assembly to produce gray and a Success to produce white. 

Having the blue in there would make the assembly more difficult, but would it also give me a better chance of getting an Amazing?

The quality is determined by calculating the probably giving each quality a weight. Blair last post suggests that white weights 7, blue 5, purple 3. So the math gives: 3x7 + 3x5 + 3x3 = 45. The chances become:

White: 21/45 = 47%
Blue: 15/45 = 33%
Purple:  9/45 = 20%

The first roll determines the quality using the chances above. The result will probably be white so lets assume that is what you got.

The second part is rolling for the crafting result, unfortunely TBlair didnt explain how this parts works so i cant show you the math. Lets assume all 3 results:

Flawed: This will downgrade your item by one grade (white -> poor/gray) and deactivates the experimentation part. This sucks.

Sucessful: You keep the white quality item and proceed to the experimentation. This is the okay result.

Amazing: This will boost your quality to match the best mat you used. We go from white to Purple! Jackpot!

 

All we know is that they are using the highest quality mat to calculate your crafting result so unless proven otherwise using the blue mixed with purple (like your example) doesnt makes it harder since they are only looking at the purple. Using 3 Purple + 6 Whites would be of the same difficulty as using 3 whites+3 blue+3purple, the same "purple difficulty".
 

Edited by BarriaKarl

Share this post


Link to post
Share on other sites

So what you're saying is

I need to face NW at a 3 degree angle parallel a Knotwood tree during a high tide while under the effects of a new moon while wearing the color purple.


etDenA9.png
Camaraderie ~ Loyalty ~ Honor ~ Maturity ~ Integrity ~ Duty

Share this post


Link to post
Share on other sites
2 hours ago, jetah said:

So what you're saying is

I need to face NW at a 3 degree angle parallel a Knotwood tree during a high tide while under the effects of a new moon while wearing the color purple.

That won't work when you commission another crafter to make components for your final combines using your mats. This is what generally happens below

 

Edited by corvax

Share this post


Link to post
Share on other sites
6 hours ago, jetah said:

So what you're saying is

I need to face NW at a 3 degree angle parallel a Knotwood tree during a high tide while under the effects of a new moon while wearing the color purple.

I think this is how crafting in Final Fantasy XI worked.


231e101d88.jpg

Share this post


Link to post
Share on other sites
16 hours ago, thomasblair said:

We are currently generating a weighting for the roll based on the quantity used! So that 4 white + 1 orange is going to be heavily weighted towards white. Heavily. Like 28 out of 29 weights. The more lower quality resource used the more it is going to pull the recipe to that quality.

So 4 epic and +1 orange is more 12 out of 13 weights towards epic. 

In this example, a 4:1 input quality ratio gives the player a 28no improvement :29improvement  ratio for successful combinations.

This tells us nothing about the chance of flawed assemblies.

Would (4 orange : 1 white) input combination yield a (28 amazing : 29 merely successful) improvement chance, or what ratio?

It seems that players can at least cut out the chance of amazing success just by waiting until they have uniform quality materials to combine.

Can I get some of my orange resources back by scrapping the flawed combined materials?

Can I stack the deck toward a less likely flawed combine of orange materials by doing the same recipe with white materials until I get a couple of flawed combines in a row?


I think the K-Mart of MMO's already exists!  And it ain't us!   :)

 

Share this post


Link to post
Share on other sites

Nice that you are trying to reduce the frustration of the players that stems from the way we perceive the world. Which is probably linked to our survival instincts: negative memories weight more than positive. Anyway...back to topic. Two questions come to my mind:

1) How many decks there are I.e. do all recipies have their own deck or are they somehow shared?

2) When does the deck reset? On success maybe?

What I'm getting at is that depending on the implementation a row of successes might then increase your chances of failure. Which means that the crafter will feel that he doesn't want to craft anymore since failure possibility has gone up. Or maybe he wants to craft something useless to burn through his failure cards. Or maybe he want to start by burning through his failure cards before slapping the prime materials on the workbench. Do you care to share with us the details on this implementation?

Share this post


Link to post
Share on other sites

Personally, I think that this is a really poor decision on ACE's part.

You're adding a bunch of complexity for no good reason.

I understand that people spend lots of time pissing and moaning when they get unlucky, however, given the crafting model that you are planning to use "one off" production is pretty meaningless.  And guess what...  The same people who are complaining about the current system are going to be complaining about this new system.

In the mean time, you've implemented a much more complex design which means that there's even more room for bugs.  

Speaking as someone who who suffered through the combine failures from long long ago - which also came about from an overly complex design - I'd prefer the you stick to the basics and try to get things right.

One thing that might make this design somewhat interesting is that there are real opportunities to exploit stuff.

I can see people doing a whole bunch of combines with crap materials hoping for a streak of bad luck.

If you screw up 2-3 times in a row THAT's the time to haul out your oranges...

Edited by narsille

WAZ6Fov.png

"The cinnabar is a lie"

Share this post


Link to post
Share on other sites
46 minutes ago, Aldahen said:

What I'm getting at is that depending on the implementation a row of successes might then increase your chances of failure. Which means that the crafter will feel that he doesn't want to craft anymore since failure possibility has gone up. Or maybe he wants to craft something useless to burn through his failure cards. Or maybe he want to start by burning through his failure cards before slapping the prime materials on the workbench. Do you care to share with us the details on this implementation?

I though about that but it is not so simple.

There is no way for a row of sucesses to decrease your chance of sucess. Losing a bonus shouldnt be something players should feel sad about. Also remember that this is happening in the background and this boost will only become meaningful after a considerable number of failures. It isnt about failing once and getting a 10% crafting boost. No one should ever feel happy about failing and getting a bonus to the next try.

It seems possible to game the system and exploit the bonus after getting a failure streak. I just think it shouldnt be worth the materials lost if you try to create a failure streak. You probably cant fail using low end mats (Not enough for the boost to get big enough) so you will have to use some good resources for that. In the end it becomes you using dozens of purple mats to craft one orange item. Is it worth?

Good crafters might use the occasional (and probably rare) failure streak to try a harder recipe. Maybe they are lucky and the previous failure streak was a blessing in disguise or they fail again and fall deeper in despair.

Edited by BarriaKarl

Share this post


Link to post
Share on other sites
19 minutes ago, narsille said:

Personally, I think that this is a really poor decision on ACE's part.

You're adding a bunch of complexity for no good reason.

I understand that people spend lots of time pissing and moaning when they get unlucky, however, given the crafting model that you are planning to use "one off" production is pretty meaningless.  And guess what...  The same people who are complaining about the current system are going to be complaining about this new system.

In the mean time, you've implemented a much more complex design which means that there's even more room for bugs.  

Speaking as someone who who suffered through the combine failures from long long ago - which also came about from an overly complex design - I'd prefer the you stick to the basics and try to get things right.

One thing that might make this design somewhat interesting is that there are real opportunities to exploit stuff.

I can see people doing a whole bunch of combines with crap materials hoping for a streak of bad luck.

If you screw up 2-3 times in a row THAT's the time to haul out your oranges...

First, it's not really all that much more complicated.  They still had to build a function for seed, and a function to pull numbers.  All they really are storing is two values if they are doing it right. The players seed and count, and then altering both of the other functions to be based off that. It really is an easy and elegant solution.

The real question around the counting issue, is how many cards in a deck.  If they round up to the nearest %, so that each deck only has 100 numbers, counting will be a very viable, and probably quite entertaining. If they have 5 decimal places and a deck of 10,000,000 cards, not so much. A deck that size would effectively defeat the purpose of having a deck as periods of loss could still be extremely streaky.

I would prefer the 100 card model myself.  I think the meta game of counting could be really interesting and fun to mess around with, if the deck reshuffle triggers were known.

Also remember, when working with low quality materials, what ends up being a winning roll could very easily be a losing roll on higher mats.  So if you were counting on less difficult mats you would have the issue that you would have a bunch of false positive results with a much higher number of wins.

 

Edited by KrakkenSmacken

Share this post


Link to post
Share on other sites

There is no reason to go above 100 cards. This isnt rockets science.

So, if we take a deck of 100 we have the following:

OBS: This isnt about summing the previous chance plus bonus. MULTIPLY the original Chance by (1+Bonus).

F(failures) = Original Chance/(100 - Failures)

0 Failures: No Bonus
1 Failures: 1% Bonus
2 Failures: 2% Bonus
3 Failures: 3% Bonus
4 Failures: 4% Bonus
5 Failures: 5% Bonus
10 Failures: 11% Bonus
20 Failures: 13% Bonus

The previous formula bonuses were pretty low so lets remove two cards:

G(failures) = Original Chance/(100 - 2*Failures)

0 Failures: No Bonus
1 Failures: 2% Bonus
2 Failures: 4% Bonus
3 Failures: 6% Bonus
4 Failures: 8% Bonus
5 Failures: 11% Bonus
10 Failures: 25% Bonus
20 Failures: 66% Bonus

If you had a good chance of crafting the item you can get a significant bonus fast. A 75% can become 80% after the third failure. A 90% becomes 96%.

But the lower you original chance was the more useless the bonus becomes. A 50% becomes 55 only after 5 failures. A 20% need 10 failures to go to a miser 25%.

So yeah, i cant see people crafting things above their level by exploiting this bonus. Again, a smart and lucky crafter can exploit a 5% to his benefit but he wont be making miracles any time soon.

Share this post


Link to post
Share on other sites
1 hour ago, BarriaKarl said:

There is no reason to go above 100 cards. This isnt rockets science.

So, if we take a deck of 100 we have the following:

OBS: This isnt about summing the previous chance plus bonus. MULTIPLY the original Chance by (1+Bonus).

F(failures) = Original Chance/(100 - Failures)

0 Failures: No Bonus
1 Failures: 1% Bonus
2 Failures: 2% Bonus
3 Failures: 3% Bonus
4 Failures: 4% Bonus
5 Failures: 5% Bonus
10 Failures: 11% Bonus
20 Failures: 13% Bonus

The previous formula bonuses were pretty low so lets remove two cards:

G(failures) = Original Chance/(100 - 2*Failures)

0 Failures: No Bonus
1 Failures: 2% Bonus
2 Failures: 4% Bonus
3 Failures: 6% Bonus
4 Failures: 8% Bonus
5 Failures: 11% Bonus
10 Failures: 25% Bonus
20 Failures: 66% Bonus

If you had a good chance of crafting the item you can get a significant bonus fast. A 75% can become 80% after the third failure. A 90% becomes 96%.

But the lower you original chance was the more useless the bonus becomes. A 50% becomes 55 only after 5 failures. A 20% need 10 failures to go to a miser 25%.

So yeah, i cant see people crafting things above their level by exploiting this bonus. Again, a smart and lucky crafter can exploit a 5% to his benefit but he wont be making miracles any time soon.

Also this assumes two things.

Failure rate remains constant, which we know from difficulty changes in chances based on quality will alter the results of the same "roll".  For example a 75 on easy mats would be a success, while on hard mats it required a 85 or greater. As long as you don't see the actual number, and only the result, it makes it quite a bit harder to count.  Sort of like if the first half of a deck of blackjack had you shooting for 21, while the second half it changed to 31.

That the distribution is greatly skewed for front or back load in a specific "deck".  ACE could run a check on seed returns and distribution, and if they see a deck that is front or end loaded, simply toss out the seed and try a new one. Even if they don't do that it could take dozens of decks before you find one that has enough failures front loaded to make a difference on the end of the deck.

Edited by KrakkenSmacken

Share this post


Link to post
Share on other sites
3 hours ago, narsille said:

I understand that people spend lots of time pissing and moaning when they get unlucky, however, given the crafting model that you are planning to use "one off" production is pretty meaningless.

I have to disagree strongly with this point. One-offs are far more important, because they're what noobs will be doing. If they get turned off by that crafting experience they'll rage quit and never come back. The game's success depends on how well it can hook new players, and the current system is terrible at that. They'll never see the factory mass production phase of the game if they don't survive that gauntlet of one-off crafting for the first few days/weeks/months (however long it takes to unlock factories).

Edit: not necessarily saying I like the new system, but it's a step in the right direction IMO. The fact that you'll never lose everything, and get a flawed (but usable) product instead - that's huge and I love it. The somewhat unintuitive calculations that go into an amazing success? I'm not as much a fan of that. I like what that system is trying to achieve, but I want it to happen in a simple and very transparent way.

Edited by Avloren

Share this post


Link to post
Share on other sites
9 minutes ago, Avloren said:

I have to disagree strongly with this point. One-offs are far more important, because they're what noobs will be doing. If they get turned off by that crafting experience they'll rage quit and never come back. The game's success depends on how well it can hook new players, and the current system is terrible at that. They'll never see the factory mass production phase of the game if they don't survive that gauntlet of one-off crafting for the first few days/weeks/months (however long it takes to unlock factories).

Edit: not necessarily saying I like the new system, but it's a step in the right direction IMO. The fact that you'll never lose everything, and get a flawed (but usable) product instead - that's huge and I love it. The somewhat unintuitive calculations that go into an amazing success? I'm not as much a fan of that. I like what that system is trying to achieve, but I want it to happen in a simple and very transparent way.

Yea let's take one win at a time. 

That whole risk bonus thing on Amazing success coupled with manufacturing is going to make going for broke every time the only mathematical reasonable choice. I suspect they will have to see that happening in the crafting logs before they re-think that mechanic.

Share this post


Link to post
Share on other sites

The most important change to your system was to make failure non-destructive. If a crafter has a 75% - or even a 90% - chance of success, but failure destroys hard-earned components, that's too much of a penalty. Here are some alternatives:

  1. As you've specified, the quality of the resulting item depends on a mixture of crafting skill and luck.
  2. Failure results in the item not being crafted, but does not destroy the components. To make this work, there needs to be a cost to each attempt - either catalysts that cost money and *are* used up in an unsuccessful attempt, crafting taking a fair amount of time, etc. Failure is sad, but players know they can keep trying without having to spend hours gathering new matericals.
  3. Failure damages or destroys one of the components, but not the others. Players just have to gather and remake the destroyed component.
  4. Failure destroys some or all of the components, but yields new materials, e.g. crafting the silver candlestick fails, but the flawed creation can be melted down, and the silver reused.
  5. No chance of failure, but crafting takes much longer at a lower skill level, or with more advanced materials, or if the player is unlucky.
  6. Something amazing happens on unusually good luck. The crafted item gains a special attribute, or the process also grants a rare material.

I'm sure you can think of a few more ways to make it so that failure is not rage-inducing.

The deck system is interesting, but as suggested, prone to gaming, especially if the deck is used until exhausted ("fail five times in a row? Gee, the next five attempts will all be successful.) I prefer increased chance of success after several failures over the deck system. Blizzard calls it a "bad luck protection" system.

 

Share this post


Link to post
Share on other sites
5 hours ago, narsille said:

One thing that might make this design somewhat interesting is that there are real opportunities to exploit stuff.

I can see people doing a whole bunch of combines with crap materials hoping for a streak of bad luck.

If you screw up 2-3 times in a row THAT's the time to haul out your oranges...

Like fail stacking in BDO.  Get the crap gear out to roll on, once above x fail stacks you try and upgrade your already super upgraded weapon to the next tier and hope it works (if it fails above +15/20 you lose a tier).  Or you get the other out come which happened to me in BDO, all my armor went to +15/20 while trying to fail stack enough stacks to get my sword to Tri (+18/20).  Except that was actually beneficial to me as it upgraded my main armor, in this sense you would be making a ton of "trash" armor lol.  


 

Formerly known as - AmazingTacoBurito

Share this post


Link to post
Share on other sites

I know the article/breakdown touched on it, but adding a random factor to your game/system isn't at all bad. Giving it too much weight is what's bad.

A blatant example is a coin toss: heads results in you get 200 million gold, tails results in your account being deleted and you have to buy another copy/license of the game to play again. In this, the existence of a coin toss isn't bad. The the degree to which the sheer outcome of a random coin toss is affecting my experience, is. If the coin toss was to determine whether or not a given goblin in a group of 10 tries to flank me or tries to distract/engage me, that's much less ridiculous. That's actually a useful random thing. Why? Because I don't want to see a goblin with a purple hat, and go "Oh, all purple hat goblins try to flank me... I know exactly how to deal with this goblin, and it's exactly the same as the way I deal with every purple-hatted goblin in existence." So, when I don't want something to be the same, but I don't want the developers to just have to determine what's what up front, a random factor is excellent!

As a general rule in games, the best use of randomness is to make the player deal with a variety of circumstances in an interesting/"mysterious" capacity. This moment of the day there's a strong wind in this area. Another time there isn't. That affects the way in which you, the player, interact with and make choices about combat. Maybe you forego ranged attacks because the wind's hard to deal with? Maybe you get upwind and cast fire spells, letting the wind gusts spread the fire across the part of your battlefield that your foes occupy? Who knows. But the numbers aren't determining what your character is doing. They're determining what you're confronted with.

That being said, sometimes it's not so bad to have a sprinkling of little bonuses here and there that are just sort of jackpotish, direct random bonuses. But they should be rather small. Like... Critical hits get pretty out of hand in games. A critical hit, as another example, is way better as something that affects the factors of combat, rather than something that just grants you +X damage when it happens. If it, for example, rends armor (grants you +20% damage for the next 5 Piercing hits or something), that's a guarantee of more damage than you would've had, but it's more of a factor change. How exactly you manage that bonus from here on out is up to you. You can just attack 5 times, or hit with dots, or hit with super big attacks, etc. The random roll didn't determine what actively happens. It only determined a shift in circumstances.


This post brought to you by...
Lephys. Because everything's better with a smile facepalm.

Share this post


Link to post
Share on other sites

I would also like to reccomend that you guys dont hide how the result is determined. I think every crafter would like if you showed the possible outcome updated real time based on the mats he is currently using.

"Hmmm if i swap that green for this blue the chances of a blue quality result increases by X while the failure chance increased only a little. What happens if i do the same with another green?"

All this would happen before the crafter even tried crafting anything. Just selecting the mats becomes fun and interactive.

Perhaps you could also show the amazing assembly chance together with the sucess chance.

Edited by BarriaKarl

Share this post


Link to post
Share on other sites

Moving this over to this thread since this one killed the other one. I will be asking questions that have been answered in the article from Blair, but stay with me I have reason. I commonly play devils advocate to explore all angles.

My main question is: What is the purpose of RNG for crafting in Crowfall?

  • Is it to sink items out of the economy?
  • Add difficulty/feeling of progression?
  • Add a flavor of randomness?
  • Something not yet stated?

Currently the proposed system, based on the article, is going to feature RNG at the lowest level of item crafting. The system will probably use a safety net (cards) that will slowly increase the chance the player succeeds if they fail. This sounds very similar to what a lot of Korean designers place into their MMOs, example would be Black Deserts item leveling. However, those systems are much more grindy and harsh (players losing lots of work in one attempt). They are also not as popular with Western players.

The fear I have is not about the RNG, just where the RNG exists.

The lowest level of crafting should be inviting to players. The goal isn't to frustrate a new player, even if there is a chance all their items assemble. At the basic level, items should be 100%. I would go on to say that mos things in the first, say 2 tiers, of crafting should be 100%. They are base foundation for the rest of the crafting process. If you follow csikszentmihalyi's state of flow the player will be rewarded for low skill and low difficulty (because they are new). As the player advances at crafting we throw in % chance to scale the difficulty. We also add more variety in components asking them to learn more scaling the skill required. This is where you begin the line going up the middle of flow.

figure1.png

The concern with the current proposal of RNG it will cause the line to rise and fall outside of the flow.You could be super successful, thus only one axis moves. You could super unsuccessful and it moves a different direction. The idea of cards would help this by correcting it. If it was too low in one area the system would help bring it back into flow. For mid to high tier of crafting that makes the most sense.

My primary suggestion is to create a foundation that has no RNG present, sans for extra benefits (look below). Then as the difficulty curve rises so does the chance of failure. You teach the players about crafting, you teach the players about potential risks vs rewards, then you allow them to explore possibilities.

 

E9NPfNB.jpg

 

Thoughts?

TL;DR: Possibly add a base layer that has no RNG and introduce RNG as a secondary layer that evolves into the difficulty curve. The card system would be beneficial at keeping players within the ideal flow of crafting difficulty vs skill.

Edited by KanashiGD

Game Designer | KanashiGD.com | @KanashiGD

Elf of Hy'shen Avari

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...