Starman complain about "Portalarium Crafting Math" #6 with data.

Discussion in 'General Discussion' started by 2112Starman, Oct 10, 2016.

Thread Status:
Not open for further replies.
  1. yarnevk

    yarnevk Avatar

    Messages:
    402
    Likes Received:
    804
    Trophy Points:
    43
    People also complain about their D&D rolls, that it should not be possible to critically fail twice in a session, but they are very happy when they critically hit twice. Every one has stories about runs of bad luck and good luck. A proper test of randomness includes probability checks for runs in large samples, if runs do not occur then it clearly is not random.

    95% is the same odds as a d20 >1 hitting a 1 means 1:20 odds, hitting it twice means 1:400. The nature of randomness does not mean that it is
    guaranteed to happen once every twenty rolls or twice every four hundred. In fact if that does happen, then it is very likely to be a memory feedback system, that is deviating from true randomness.

    The confidence intervals are very large and swingy on small sample sets, you need sampling in the thousands to be confident it is working as intended.
    Only SOTA has sufficient sample size to test if the stats are confirming the predictions, players are an insufficient sample size. Sure you can complain about the design of the system relying on RNG, but on Portalarium can qualify the RNG as truly random.

    Since the perception is there the devs should be tracking all the rolls and keep a running stat with sufficient sample size to have a decent confidence interval on the results. That way when you do fail you can blame it on your bad luck and not the rolls.

    Knowing this game there are probably buried secrets that the chance only applies if the conditions are just right, phase of the moon, weather, time, hunger are all variables that the devs want to impact the game systems.
     
  2. 2112Starman

    2112Starman Avatar

    Messages:
    3,613
    Likes Received:
    7,989
    Trophy Points:
    165
    Blackjack is a 48% game for the player. Master crafting is a 25% game for the player.

    Casinos in Vegas are built and paid for by that 48% chance to the player.

    If Im playing blackjack at a casino and play the statistics right, I am expecting to with 48% of the time. However, if I am lucky I am hopping my chances are better and at times I bet higher. I know the longer I sit at that table, the higher chance I have of hitting that 48% which is not good for my chip count.

    If I was playing Blackjack with my real money and only had a 25% chance and after 300 hands (which is about my data set for crafting) and I only won 10% of the time, so a total of 30 hands when I should have won 75 statistically... I would call out the house for dealing from the bottom of the deck.

    (btw, yes, I can count cards ok, last time I was in Vegas I sat down at a table with $100 and walked away 6 hours later with $1800)
     
    lollie, Jivalax Azon and docdoom77 like this.
  3. Wintermute of CoF

    Wintermute of CoF Avatar

    Messages:
    1,372
    Likes Received:
    2,432
    Trophy Points:
    113
    Gender:
    Male
    Location:
    London, United Kingdom
    I'd actually be fine with a little more control over the bonuses (perhaps six to choose from instead of three) and getting a significant percentage of the materials back when something breaks.

    At the moment if I craft and masterwork/enchant a longbow that's 15 wood + 40 scrap + 40 animal hide + 2 gems + 20 silver ore + 20 gold ore + fuel (off the top of my head, correct me if I'm wrong). If it all goes tits up on the enchant I get back something like 2 scrap. It would be entirely reasonable to get more like a third or more of the raw materials back (perhaps depending on salvage proficiency).

    When you've gathered for a week and crafted 20 items to get (let's imagine) 5 +1's and then masterworked all of them and got just one with the bonus you actually wanted, then you take that one item to enchant it (at >90% chance) and you trash it and get back a few scraps. That, right there, is a game quitting moment.
     
    lollie, Jivalax Azon, FrostII and 4 others like this.
  4. Elwyn

    Elwyn Avatar

    Messages:
    3,617
    Likes Received:
    4,783
    Trophy Points:
    153
    Gender:
    Male
    Location:
    San Antonio, TX
    Leather: 2 animal hides, 1 curing salt, makes two leather

    Leather from scraps: 1 animal hide, 1 curing salt, 12 leather scrap, makes one leather

    So with scraps, I use twice as much curing salt, and end up with the same amount of leather as if I had never even used the scrap?

    MATH!
     
    lollie and Lord Baldrith like this.
  5. Kirran

    Kirran Avatar

    Messages:
    123
    Likes Received:
    255
    Trophy Points:
    18
    There's been another thread on the RNG problems on the Dev Feedback forums running for at least the past week. We speculated this could be the case and that Port introduced some type of Lore in the game that affected this. No one from the Dev team has either confirmed nor denied this happening, so it's a conspiracy theory to try and explain why so many of us see problems with crafting success. Picture Masterworking using a 'Day' component like Gold giving a higher success rate during the day, and a 'Night' component like Onyx having a lower success rate during the day. And so on. If this were the case, then fine. I can adjust to doing that. But if it was something hidden from us and we're supposed to discover it, then basically we'd have been getting lied to all this time.

    This. Starman's example of 25% is easier to discount because you look at it and see that you're supposed to fail 75% of the time. But when I look at the Masterwork success % for tries 1 and 2, (or even Mining) the actual failure rates we get is a lot more than what they should be. If I have a 95% chance to mine the node then I'd expect it once every 20 attempts, and typically this is the case. HOWEVER, I often get 3 failures in a row. That probability is around 1/10000. I can see failing 3 nodes in a row every couple months. But if it happens 3x in a week then that's not statistically possible and there's something wrong.

    Masterwork and Enchanting build upon the base item and upon themselves. You can say 'well, you're making rare items and they're special, so there shouldn't be many of them and to keep them rare, you should fail a lot'. But keep the big picture in perspective. In the case of leather armor, I have to grind for Animal hides. I have to grind the Tanning. I have to grind making Yards of Leather. I have to grind making straps. I have to grind mining. I have to grind smelting. I have to pay for chunks of coal, curing salt, and Wax. I have to make each piece. Then I roll the dice to see if I blow up the piece I'm Masterworking and Enchanting. 2x Masterwork and 2x Enchanting is applying RNG x RNG x RNG x RNG.

    If I played 2 hours a night I could spend a week and a half grinding Animal Hides, Gold, Silver and Gems to create about 4 full sets of leather armor before I even attempt to start Masterworking and Enchanting, and on a failure of any of those 4 attempts I have a pile of leather scraps left. In my experience lately I'll end up with a set and a half, maybe less, out of 4. And that with 80% for each second attempt. I've had cases where each second attempt of the Masterworking at 80% has failed 6 times in a row.

    So yes, the perceived RNG problems are really, really frustrating. It's a bigger issue of the time sink required to be a crafter, but you have to fight the battles to win the war.
     
    Last edited: Oct 11, 2016
  6. Swan

    Swan Avatar

    Messages:
    90
    Likes Received:
    157
    Trophy Points:
    8
  7. 2112Starman

    2112Starman Avatar

    Messages:
    3,613
    Likes Received:
    7,989
    Trophy Points:
    165

    You are using a sample size of 100, I have mentioned a total sample size of around 200-300 chain pieces created.
     
    Jivalax Azon likes this.
  8. Swan

    Swan Avatar

    Messages:
    90
    Likes Received:
    157
    Trophy Points:
    8
    how many exactly you maded and how many was critted then? "200-300" isnt really accurate size to say.
    also i agree my sample size was not high enough to give accurate calculations, but sure is bigger than 10-30 what is in many posts i see.
    it is around normal to not crit single one in batch of 20 with 25% chance
     
  9. Kirran

    Kirran Avatar

    Messages:
    123
    Likes Received:
    255
    Trophy Points:
    18
    You do realize that 200 masterwork attempts is equal to 4000 silver ore? if Port wants me to test 200 masterwork attempts I'm more than happy to, if they want to set me up with all the raw materials to do it. But I'm going to want the base raw materials to make the armor first and then mastercraft it. May as well make it realistic. And while we're at it, I can record the number of Zanax I have to take waiting for the tanning, leathermaking and smelting progress bars to finish.

    You forget we're playing a game here. I wonder if Port ever tested to that extent, using a character on a live server.
     
    lollie, Jivalax Azon and LoneStranger like this.
  10. 2112Starman

    2112Starman Avatar

    Messages:
    3,613
    Likes Received:
    7,989
    Trophy Points:
    165
    Its not normal to crit at all with a 25% chance in 20 chances? I would hope to end up with 6-10
     
    Jivalax Azon likes this.
  11. AceBenSuited

    AceBenSuited Avatar

    Messages:
    63
    Likes Received:
    125
    Trophy Points:
    8
    I think the RNG is working as intended. Just because you have a 20% chance doesn't mean you should expect 20% of your crafts to be exceptional. Just because you have 4 fails in a row doesn't mean the next one is now, magically, 100% chance... it is still just a 20% chance. This is simply tested by just flipping a coin. You have 50% chance for heads or tails (just for giggles I flipped a quarter 7H-3T).

    I only have an issue with the result of a fail. Should get something more than just scrap on a fail, especially when using rare materials. At least give me one or two of my damn ingots back.
     
  12. Curt

    Curt Avatar

    Messages:
    1,639
    Likes Received:
    2,356
    Trophy Points:
    125
    Gender:
    Male
    Location:
    Sweden
    Could give the fuel different tiers
    higher tier fuel => bonus on chance for exceptional (but also more expensive)
     
  13. mdsota152

    mdsota152 Avatar

    Messages:
    763
    Likes Received:
    2,047
    Trophy Points:
    105
    Gender:
    Male
    Location:
    Virginia
    The problem is you all are assuming it’s broken… but in reality it’s working exactly the way Port wants it to… we just don’t want to play the way they want us to. Port wants a high failure rate to waste lots of gold and resources to craft that one perfect item… they then expect us to sell that one item for an outrageous sum of money to someone, rinse and repeat… but the flaw in their logic is that nobody can really afford that one perfect item that you were finally able to make if you priced it based on ALL of the failures. NOBODY is willing to pay that kind of gold for anything either. In their minds having so few perfect items and a huge resource sink it perfect... but try telling and getting all the average crafters out there to understand that. Not going to happen.

    Everyone wants to buy things at a reasonable price… but we can’t price things like that due to the shear amount of resources that go into making items. If we could make that perfect +10 sword each time on the first try there’s no gold or resource sink for Port. Therefore their idea of an economy only works if we play THEIR way. ... just my two cents. :confused:

    What most players want to be able to do is craft items with limited resource waste and be able to sell it at a price other players are willing to buy it at. But that's not the game Port wants. So bring in the high resource sink crafting system. ****'s got to break...
     
    Aldo, lollie, Jivalax Azon and 7 others like this.
  14. Lord Baldrith

    Lord Baldrith Avatar

    Messages:
    2,167
    Likes Received:
    7,051
    Trophy Points:
    165
    Gender:
    Male
    Location:
    Wizards Rest
    Part of the problem with crafting is it's too easy, yet a grind. Creating a great item is all luck and no skill. If we could add some skill effects to our crafting and our harvesting, then we wouldn't be forced to rely on RNG which is generally unfair as a rule.

    I have long suggested giving us mini-games which would enable the crafter to create items with a higher chance of success based on his crafting skill and his actual skill in playing. It doesn't have to be an annoying take forever game on most stuff. Mostly easy, but allowing you to interact with your project enough to have a success possibility that is much higher than RNG.

    Obviously, production shouldn't require this type of thing...but crafting projects should shine when they are completed with great quality successes...I don't understand why our game has to be a click done boring RNG mashing uninteresting game no different than WoW besides the first time you learn how to craft the item and you have to physically drag the items on the table.

    Just my 2 cents ... hoping someone please notices that crafting needs a huge change for it to be fun and not a grind :)
     
  15. 2112Starman

    2112Starman Avatar

    Messages:
    3,613
    Likes Received:
    7,989
    Trophy Points:
    165
    I have to kinda agree with this.

    I think I would much rather have the components of crafting be rarer (say tin, nickle, gold, silver) so that they are harder to make and require less time rather then require vast amounts of time to farm a ton of ingredients and just have them utterly fail in bulk.

    example:

    100 common gold ingots = 20 enchanting attempts with terrible failure that ends up with one good item

    vs

    5 rare gold ingots = 1 enchant attempt with a very high chance.

    skill can be tied to how powerful the enchant is. skill 20 = +2 dex, skill 80 = +6 dex


    But its obviously to late for this.
     
  16. 2112Starman

    2112Starman Avatar

    Messages:
    3,613
    Likes Received:
    7,989
    Trophy Points:
    165
    This would only make sense in the case that at times if you craft 100 items with a 25% chance that you actually produce 40 instead of the norm which seems to be around 10.
     
  17. Wintermute of CoF

    Wintermute of CoF Avatar

    Messages:
    1,372
    Likes Received:
    2,432
    Trophy Points:
    113
    Gender:
    Male
    Location:
    London, United Kingdom
    Well we still get 'critical successes' when crafting ingots and other components, it's just that currently those successes have no meaning. I'd love to see grand mastering smelting mean something useful like this.
     
    lollie, Jivalax Azon and FrostII like this.
  18. Drocis the Devious

    Drocis the Devious Avatar

    Messages:
    18,188
    Likes Received:
    35,440
    Trophy Points:
    153
    Gender:
    Male
    You don't understand how statistics work.
     
  19. Gideon Thrax

    Gideon Thrax Avatar

    Messages:
    2,497
    Likes Received:
    6,771
    Trophy Points:
    165
    Gender:
    Male
    How's it work with a coin flip? Don't you have to flip a coin something like 4000 times just to get close to a 50% chance of hitting heads?

    Not trying to get in on the argument - just tossing out some cocktail logic. :D I do know I blew through 1000 silver ore so fast and destroyed so much bronze that it's now crystal clear to me that SotA wants me to wear leather.
     
  20. Raistlyn

    Raistlyn Avatar

    Messages:
    228
    Likes Received:
    479
    Trophy Points:
    18
    Ya I do agree the math seems to make mistakes in the game. I have had 4 fails in a row at 96% success. The odds of that happening, if it was truly random, are rather low.
     
Thread Status:
Not open for further replies.