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

Deck style approaches are interesting, but I'd be curious if you're keeping track of the deck state in and around gaming sessions. If I logout, does my deck reshuffle itself as I log back in? Knowing that a deck style system is in play, it might be possible for very well versed crafters to game this system to their advantage. Knowing that you have a single card left in the deck to draw and what it is is significant and noteworthy. I assume you'd reshuffle the deck if a player "succeeds" but maybe my logic in that regard is flawed.

What would trigger a reshuffle?

  • Success
  • Failure
  • Logging off
  • Skilling up?
  • Getting down to having too few cards left in the deck (though this might increase streakiness...)
  • Other triggers?
Edited by Scree

Obsidian-ForumSignature.jpg

Share this post


Link to post
Share on other sites
19 hours ago, Scree said:

Deck style approaches are interesting, but I'd be curious if you're keeping track of the deck state in and around gaming sessions. If I logout, does my deck reshuffle itself as I log back in? Knowing that a deck style system is in play, it might be possible for very well versed crafters to game this system to their advantage. Knowing that you have a single card left in the deck to draw and what it is is significant and noteworthy. I assume you'd reshuffle the deck if a player "succeeds" but maybe my logic in that regard is flawed.

What would trigger a reshuffle?

  • Success
  • Failure
  • Logging off
  • Skilling up?
  • Other triggers?

Cards can be easily tracked by play, by building decks using seed control. It's pretty much the same technique I suspect they will use to generate entire worlds procedurally from a single number.

Link to unity function they will most likely use. https://docs.unity3d.com/ScriptReference/Random.InitState.html

If you have the same seed for the player, you only have to count the number of times the player has "rolled" the check before the specific card is picked.

Basically you say seed=.2309148321048.  This produced a deck of options with a specific pattern, every time you use that seed. Lets just take a 1-10 pattern and say that the seed I just used generates.

5-8-9-1-2-3-7-4-0-6

Then you place a count every time a "roll" is made,  of how many cards into that deck you have pulled. So on the fifth "roll", the player gets a 2.  whalla, you have both a full deck of as many possibilities as you like in random order, and a way to know where in that order you are, without needing to track each and every card.  

Once you are finished with a deck, pull a new random seed and set the count to zero.

It's a very elegant way of both generating semi-random results, and controlling distribution of results. The only real set back is that you can "count" this type of system like a real deck of cards.  If you know the deck size is 52, and you have pulled 26 fails in a row, you know now is the time to go for broke with that big ticket item.  

So to mitigate that, ACE needs to identify that risk, and re-shuffle/seed the deck when players do something to justify it.  For example they could have one deck for each quality level, so you can't load up poor quality fails and then jump to high quality with a greater chance of selecting the win from the remaining cards in the deck.

EDIT: Or who knows, maybe that becomes a real crafter personal skill like reaction time is for twitch players. That could be cool having in effect "card counting" as part of the crafter sub game. There is a reason Casinos refuse "back off" card counters if they notice them, because they can tilt the odds in their favor. I would not be opposed to that be built in expected behavior for "skilled" crafters. It adds a human choice and planning element to what is currently just RNG.

I do like very much that they talked about Skinner and Operant conditioning. The original punishment mechanic of total loss on failure was a major problem I am glad they are addressing directly. Call me a narcissist, but it's almost like they read a bunch of my combine failure complaint posts and felt they needed to address my concerns directly.

Thanks guys for the very informative walk through.  

 

Edited by KrakkenSmacken

Share this post


Link to post
Share on other sites

My opinion on RNG is that developers use it in place of depth. A easy out, or if you hate it that much, call it a cop out.

I have made a post about the deck type style of RNG awhile back. If you refuse to add depth at least you can force your RNG toward a smaller sample size of a truth table.

 

For the record: I would rather have something take 40 pieces of material and be 100% success rate then something take 10 pieces of material and be 25%.

Obviously thats a oversimplification but still, id take it 10/10 of the time. 

Edited by Vectious

CfWBSig.png

Share this post


Link to post
Share on other sites
1 minute ago, YouTubejasonwivart said:

great stuff looks very promising

Yeah that chipmunk butt looks great @davegreco :D

For the article, i'm not a crafter and i'll let this stuff for the people who like it, so can't say if it's good or not, but the 3 process now of failure sounds way better than loosing everything.


WnxjrJ9.gif

Thelanas Kar'Pal Membre fondateur de l'alliance Naerth en 2001 - Ex Shadowbane European Advisor Damnation/Carnage/Vindication/Corruption http://www.twitch.tv/gorwald/profile

Share this post


Link to post
Share on other sites

In the 5.4 update, crafting the assembly phase will now have 3 results:
Amazing Assembly – Improves the end product to the top level of quality of resource used in the combine, and moves the crafter to the experimentation phase.
Succesful Assembly – Assembles the object and moves the crafter to the experimentation phase.
Flawed Assembly – Assembles the object at one quality level below what the result would have been and skips the experimentation phase.

 

How are the amazing assembly chances calculated?

The fact that you only take the highest grade mat quality seems worrisome to me. Trying to craft using 1 orange + 4 whites has the same chance of upgrading to orange as using 1 Orange + 4 Purple?

Are we doing only one roll? Either failure, sucessful or amazing?

Taking only the highest grade seems a little simplistic. It is a step forward though.

Maybe using the highest grade mat quality inversely proportional chance rolling for failure or sucess. After that rolling again for a upgrade making the chance proportional for the quality of ALL items used.

So a sucessfull crafting (not failure) using all orange items has a 100% chance of being orange. Using 1 orange + 4 Purple has 11% of upgrading to orange, 89% of purple. 1 Orange + 1 Purple + 3 rare goes 8, 17, 75%.

EDIT.: Hmmm... Maybe you could use a similar pattern for the first roll (sucess/failure chance) making it dependent on the quality of all mats used. It would make the chance calculus harder to understand for the players but the increased control could make up for it.

The math above is just an example but i think you get the idea.

PS.: Can someone guess the pattern i used?

Share this post


Link to post
Share on other sites

Thank you for writing this. I don't agree with it but it is appreciated that you took the time to explain your thoughts on it.

Personally, I think heavy handed use of RNG can become exploitative and abusive. It works because it manipulates the addictive and protective mechanisms of the brain, not because it is enjoyable or because people want to be used in that way. Yes, you can exploit the brain to get what you want but that doesn't mean you should, and I would definitely not attempt to justify it as a good thing. Most sane people will realize that they are being manipulated and begin to resent you for it. I would proceed with caution.

Share this post


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

How are the amazing assembly chances calculated?

The fact that you only take the highest grade mat quality seems worrisome to me. Trying to craft using 1 orange + 4 whites has the same chance of upgrading to orange as using 1 Orange + 4 Purple?

Are we doing only one roll? Either failure, sucessful or amazing?

Taking only the highest grade seems a little simplistic. It is a step forward though.

Maybe using the highest grade mat quality inversely proportional chance rolling for failure or sucess. After that rolling again for a upgrade making the chance proportional for the quality of ALL items used.

So a sucessfull crafting (not failure) using all orange items has a 100% chance of being orange. Using 1 orange + 4 Purple has 11% of upgrading to orange, 89% of purple. 1 Orange + 1 Purple + 3 rare goes 8, 17, 75%.

EDIT.: Hmmm... Maybe you could use a similar pattern for the first roll (sucess/failure chance) making it dependent on the quality of all mats used. It would make the chance calculus harder to understand for the players but the increased control could make up for it.

The math above is just an example but i think you get the idea.

PS.: Can someone guess the pattern i used?

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. 


Thomas Blair
ArtCraft Entertainment, Inc.
Follow us on Twitter 
@CrowfallGame | Like us on Facebook
 

 

Share this post


Link to post
Share on other sites
13 minutes 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. 

I assume you use something similar to the sucess chance? Not failing using a 4 white + 1 orange is easier than 4 purple + 1 orange? If yes, i suggest explaining it a little more in the article since just saying "calculated using the highest quality mat" sounds like everything else is ignored.

Share this post


Link to post
Share on other sites
18 minutes 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. 

So a probability distribution type thing, where 4 epic 1 orange vs 3 epic 2 orange vs 1 reg 3 epic 1 orange each gets a particular probability distribution?


Mic MWH, Member of Mithril Warhammers since 2003,


Hammers High! http://www.mithrilwarhammers.com

Share this post


Link to post
Share on other sites
18 minutes ago, McTan said:

So a probability distribution type thing, where 4 epic 1 orange vs 3 epic 2 orange vs 1 reg 3 epic 1 orange each gets a particular probability distribution?

Seems like white weights 7, purple 3 and orange 1. Rare (Green) is probably 5.

Share this post


Link to post
Share on other sites

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?

Share this post


Link to post
Share on other sites

This new system actually decreases incentive to try and "bootstrap" a Craft or "Amazing Assembly" as it is termed here.
Not only are the chances quite low according to what's stated here but you would also be Increasing the Difficulty of the Craft.


ScornofLife2.jpg

Share this post


Link to post
Share on other sites
25 minutes ago, Scorn said:

This new system actually decreases incentive to try and "bootstrap" a Craft or "Amazing Assembly" as it is termed here.
Not only are the chances quite low according to what's stated here but you would also be Increasing the Difficulty of the Craft.

That is offset though by being able to potentially grind out dozens of higher quality items with lower quality materials once manufacturing comes into play. 

Only that first try is going to get the fail roll, everything else will just work.

 

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...