1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Greetings Guest!!

    In order to combat SPAM on the forums, all users are required to have a minimum of 2 posts before they can submit links in any post or thread.

    Dismiss Notice

Tailor BOD's exp vs normal rate ...

Discussion in 'UHall' started by Lord Bishop, Jun 3, 2008.

  1. Lord Bishop

    Lord Bishop Seasoned Veteran
    Stratics Veteran

    Joined:
    May 15, 2008
    Messages:
    429
    Likes Received:
    80
    I finally started filling and turning in my cloth tailor BOD's with my 120 skill tailor now over 2 weeks I turned in about 150 and I was lucky if 20% of the bod's it gathered in return were exptional. I know 150 is not alot to make a study but is anyone eles seeing skewed exp vs normal results?
    Thanks Gil-galad
     
  2. Basara

    Basara UO Forum Moderator
    Moderator Professional Wiki Moderator Stratics Veteran Stratics Legend Campaign Supporter

    Joined:
    Jul 16, 2003
    Messages:
    8,354
    Likes Received:
    476
    I've suspected there is something wrong for quite some time...

    Instead of the 50% (60% at GM) for exceptional, it averages about 20-25%. I've even turned in 15 BODs before and gotten 15 normals back.

    Perhaps Jeremy can have someone look at the code; after all, the reason why we weren't getting Iron Plate smith LBODs, and getting tons of colored smith LBODs, prior to last fall, was a missing ")".

    Perhaps a similar issue is causing us to get so many normal BODs, and so few Tailor LBODs.
     
  3. Hildebrand

    Hildebrand Certifiable
    Stratics Veteran Stratics Legend

    Joined:
    Oct 12, 2005
    Messages:
    1,697
    Likes Received:
    1
    Too bad a 120 tailor doesn't have an even slightly higher chance at an excep deed or colored deed. I just hit 120 and it's a pitty.
    I'm almost hoping there's a bug, cause I can't get any great deeds as it is.
     
  4. ejpeters

    ejpeters Journeyman
    Stratics Veteran

    Joined:
    May 18, 2008
    Messages:
    284
    Likes Received:
    0
    I also think there is a problem. I found that Blacksmithy bods have a greater return than tailor bods do.
     
  5. Lord Bishop

    Lord Bishop Seasoned Veteran
    Stratics Veteran

    Joined:
    May 15, 2008
    Messages:
    429
    Likes Received:
    80
    Could a Red atleast comment on this please?
     
  6. Lord Bishop

    Lord Bishop Seasoned Veteran
    Stratics Veteran

    Joined:
    May 15, 2008
    Messages:
    429
    Likes Received:
    80
    Can we get a dev to look into this pretty please :popcorn:
     
  7. Crysta

    Crysta Babbling Loonie
    Stratics Veteran Alumni

    Joined:
    May 12, 2008
    Messages:
    2,418
    Likes Received:
    73
    Isn't this just the same misunderstanding people ALWAYS have when it comes to this game? It's 50% on EACH one turned in, not 50% OVERALL. This means you could roll that 50% that you'll get a colored on way less than 50% of what you turn in. It's all a matter of luck, and you people are just darn unlucky. :\
     
  8. Basara

    Basara UO Forum Moderator
    Moderator Professional Wiki Moderator Stratics Veteran Stratics Legend Campaign Supporter

    Joined:
    Jul 16, 2003
    Messages:
    8,354
    Likes Received:
    476
    You're talking to a math teaching major here, and it's apparent you're the one that doesn't understand the math.

    50% EACH ONE should AVERAGE OUT to 50% OVERALL, over time.

    IF one gets 100 BODs, at 50% chance each, one should have somewhere between 40 & 60 Exceptional BODs, most of the time.

    If one is consistantly getting 30% or less exceptionals, especially over large sampling amounts, there is something wrong with the "fairness" of the randomization process.

    Note: a 30% Exceptional rate, over a 100-pull example that is supposed to be 50/50, has a z score of 3.9 standard deviations. that SHOULD be an outlier, not the norm. it should occur about 0.002% of the time - yet for me, it's about 95% of the time.

    ***

    For example, since the Smith BOD changes, I kept a tally for the BODs I got with my 120 smith, to see if the fixes concerning the misplaced ")" actually fixed the issues we had before.

    Over time and Smith BOD requests (now have about 2200 tallied), the numbers have gotten closer and closer to the expected numbers, based on the chance for each type.

    While any given trip for multiple (20-100) new Smith BODs might be high or low for certain types, once you add them all together over the length of my sampling, the numbers come out to within 0.5% of the expected return after 2000+ samples.

    ***

    Going through a sample size of 250 tailor BODs I have on hand (without sorting) from 5 runners over several weeks, it's quite obvious there is a disparity somewhere, and the "dice" are not "fair" in the Probablilty meanings of the words.

    Either there is some issue in the RNG, the numbers for the calculation, or both.
     
  9. Lord Bishop

    Lord Bishop Seasoned Veteran
    Stratics Veteran

    Joined:
    May 15, 2008
    Messages:
    429
    Likes Received:
    80
    I'm actually a math teacher and although you are talking colored vs non colored and cloth. This post is only about exceptional vs normal. Basara is right Crysta over a small sample you would expect between 40-60 % exceptional.
    :stir:
     
  10. Razz

    Razz Seasoned Veteran
    Stratics Veteran

    Joined:
    May 12, 2008
    Messages:
    347
    Likes Received:
    21
    Well I had a Math teacher in school. AND I can spell MATH :)

    Oddly enough I was trolling the boards looking for reasons why I was not getting any exceptional BoDs from NPC Tailors. I thought I may be doing something wrong considering I'm relatively new to BoDs. Apparently I'm not doing anything wrong......
     
  11. Knew there was a reason your posts usually made sense ...
    but

    something went "clunk" right around ....
    here:
    "the numbers come out to within 0.5% of the expected return after 2000+ samples."

    mmmm
    Earlier? :
    "not the norm. it should occur about 0.002% of the time - yet for me, it's about 95% of the time."

    BOTH statements are ....true? maybe ... IF you mismatched/labeled types ...
     
  12. Basara

    Basara UO Forum Moderator
    Moderator Professional Wiki Moderator Stratics Veteran Stratics Legend Campaign Supporter

    Joined:
    Jul 16, 2003
    Messages:
    8,354
    Likes Received:
    476
    Read closer, fayled.....

    The 0.5% discussion was separated from the main text of the discussion by asterix, because it was showing how the recently fixed Smith BODs were now working.

    The earlier comment was relation to the Tailor BODs that appear to be broken(the first and last parts of the post are about the tailor BODs, hence the need to separate the sections)
     
  13. Bing! thx!
    [​IMG]

    Much clearer (for me, math in tha hall and all :cool:) with the edited applied.
    Should be sufficient for any Dev inquiry, to find a misplaced paren, comma ... function ...

    Or have Q&A run an exhaustive test series ...

    It went *clunk* (without the edit) because the thread shifted from Tailor exceptionals
    to Smith "types" .... kinda hard to follow (for me) ... cause ... well cause I'm pretty spoilt
    when it comes to comparitive results analysis ...

    Before it hits my desk(attention) the charts and graphs and data and projections and results ...
    are compiled and >formated< and verified .... stamped: "Thus it is so" ...
    {wink}

    works pretty well ... for MY desk ...
    Now then .... as for Uhall ... [​IMG] hmmmm ...

    Probably "best" to single ONE dev out of "the group" (dev/QA .... someone with code access) to put our hypotheses to
    Known to be "best" ... not have -me-[​IMG] "do" the princple communication ({wink} agreed/given)

    as a >formated< method of presenting Data ... ideally ... the ONE, could create a downloadable tool (like the one way back, for an archery problem) that could collect and report a full range of selected variables. (controls input/operator errors)

    now ... since smith and tailor _formula for BODS_ are likely NOT identical (unknown IF there IS: 1 or 2) need to define which _same class_(excp,material, size, *shrug*)
    compares >as a range< TO the other (smith_tailor) and allow for the differences((compare ranges)) as to the rewards (runic classes)

    Thoughts?
     
  • About Us

    Stratics is the oldest continually running MMORPG Fansite on the Internet. Founded in 1997 Stratics has served the Ultima Online Community for 18 years. We strive to provide the most complete social experience for Ultima Online players.
  • Subscribe Now!

    Want to provide Continual Support? Subscribe and gain additional benefits as a patron of Stratics.com!
    Subscribe Now!

    Stratics Professional Accounts feature the following advantages:

    • Ad-Free Browsing of our Forums
    • Upload a custom Profile Cover
    • Unlimited media upload storage space
    • Use of the theme styler
    • Ability to collapse the sidebar
    • Premium background themes to choose from
    • Access to additional features of the Classifieds System
    • Ability to Customize Your User Title
    • No Post Delays
    • Additional Signature Allowances:
    • Special Professional Banner Display with your Account
    • PM Allowance Upgrade
    • Additional Thread Allowances
  • Support Us!

    Don't feel like subscribing? Donations to benefit the further development of Stratics and for purchase and inclusion of additional features are always welcome.

    Donate to us!