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

[News] FoF: Fiddly Bits

Discussion in 'UHall' started by Larisa, Feb 29, 2008.

  1. Larisa

    Larisa Reporting Manager
    Editor Reporter Moderator Professional Wiki Moderator Stratics Veteran Stratics Legend Editor Fallen Lords

    Joined:
    Sep 8, 2006
    Messages:
    4,345
    Likes Received:
    1,958
    Posted to <a target="_blank" href=http://www.uoherald.com/news/>uoherald.com's</a> <a target="_blank" href=http://www.uo.com/fof/fiveonfriday99.html>Five on Friday:</a> <blockquote><span class=osi>

    [​IMG]

    <font size=+1>FoF: Fiddly Bits
    </font> Jeremy Dalberg
    29 Feb 2008 13:39:03


    Join us this week as we dig into accuracy calculations, enhancing, perfection, and more! And don't forget - the Jacksonville Town Hall is tomorrow! </p>

    "Why do I seem to miss so often when using Archery?"</p>

    I've seen several variations on this question, so Leurocian scoured the accuracy code for me, and we did not find anything that would specifically affect Archery accuracy (with one exception, covered below.) Accuracy in general is a fairly complex topic with many variables, so it's a tricky thing to test at the best of times. Here's an explanation of the different factors - if anyone sees any results that aren't consistent with this, please let us know!</p>

    Attacker's stats: Weapon skill and Hit Chance Increase are the stats the formula uses on the attacker's side. HCI is increased by HCI items and decreased by the defender's Hit Lower Attack, if any.</p>

    Defender's stats: Weapon skill and Defense Chance Increase are the stats the formula uses on the defender's side. DCI is increased by DCI items and decreased by the attacker's Hit Lower Defense, if any. (Exception below.)</p>

    Barehanded defense: If the defender is not using a weapon, "weapon skill" will be EITHER Wrestling skill OR (Anatomy + Evaluating Intelligence + 20) / 2; whichever is greater (which means that if you have none of these skills and are barehanded, you'll have a functional weapon skill of 10.)</p>

    Mage Weapons and Use Best Weapon Skill: UBWS uses whichever weapon skill is highest, and Mage Weapons use Magery (with whatever penalties listed on the weapon) assuming it's higher than the appropriate weapon skill for that weapon.</p>

    The minimum chance to hit is always at least 2%. If both attacker and defender have the same weapon skill and HCI/DCI, the chance to hit is 50%.</p>

    But wait, there's more!</p>

    Attacks can be parried or evaded - those are completely different calculations that happen after the hit chance is calculated. If you're trying to track your accuracy, make sure you're taking parried or evaded attacks into account.</p>

    Archery has one more caveat - if you move, it resets the swing cycle. There's only one place where this might break the swing - between the "swing" and the "hit" calculations - and we couldn't make it break in a few minutes of testing, but it's possible that a move could break the swing and it would look exactly like a miss, when in fact the attack was simply aborted. Archers, hold still!</p>

    The actual formula is conveniently listed in the question below!</p>

    "The formula for Hit Chance on Stratics is: Hit Chance% = ( ( [Attacker's Combat Ability + 20] * [100% + Attacker's Hit Chance Increase] ) divided by ( [Defender's Combat Ability + 20] * [100% + Defender's Defense Chance Increase] * 2 ) ) * 100</p>

    Does HLD lower the defender's combat ability by 25%? Does HLA lower the attacker's combat ability by 25%? Or are both applied to the DCI and HCI portion of the calculation?"</p>

    HLA/HLD are directly applied to the HCI/DCI portion of the equation - when the HLA/HLD effect goes off, it reduces HCI or DCI by 25 (and yes, it can make that value negative.)</p>

    "Is the change of colored ore spawning a straight 100% calculation or is it 50/50 to be colored and then a second roll for the type of ore?"</p>

    Each ore type has a different percentage chance to be chosen, but it is a single random roll.</p>

    "Is there any difference in the success chance between enhancing wooden items or leather and metal?"</p>

    Not necessarily - success chance is actually a function of the intensity of the properties you're changing and the number of properties on the item, not which skill you're using. Of course enhancing armor with gold, which affects six properties (bonuses to four resists, luck, and lower requirements) is, in general, more likely to fail than enhancing a weapon with bloodwood, which only affects two (hit point regeneration and hit life leech,) but if that weapon already has near-max intensity on either of those properties, or has several other properties, it may have a smaller chance of success.</p>

    In short, while determining the actual success chance of enhancing is complicated and depends on several variables, all the crafting skills use the same formula.</p>

    "Why does perfection, step 9, always return the same or less damage than step 8? I have verified this over a run of 200 honors, and step 9 is ALWAYS less than or equal to step 8."</p>

    There are a couple of things happening here - first of all, the size of the damage bonus increase is dependant on your Bushido skill. Second, it's capped at 100%. Third, it's a percentage boost to your damage, so the total damage is dependant on the base damage of your weapon, which is variable.</p>

    With Legendary Bushido, you'll be at a 96% damage bonus at step 8, and you'll hit the cap of 100% bonus at step 9 - with only a 4% increase, the extra damage will likely get lost in rounding even if your base swing is the same (decimals get truncated, so the base damage would have to be at least 25 to see even a 1-point increase.) And weapons have a variable damage per swing - it's a random roll within a range - so it's absolutely possible for your actual damage to drop.</p>

    Playguide Update Roundup:</p>

    No Playguide updates this week, but more are on the way once I'm back in the office next week.</p>

    What's To Come:</p>

    A few more confirmed bug fixes for the next publish:</p> [*]Greater dragons are visibly named "Greater dragons" [*]Fixed a couple of worldbuilding trouble spots (places where folks would get stuck - one in Wrong, one in Magincia.) [*]Made several fixes that we ain't telling you about until they're live on all shards ;)

    Jacksonville Town Hall </p>

    The Jacksonville Town Hall is tomorrow! Looking forward to seeing all of you there!</p>

    For our mid-Atlantic players, the May Town Hall meeting is also posted - you can get details on the <a target="_blank" href=http://town.uo.com/luncheon.html> Town Hall Page </A> and RSVP for that one as well. If you're in the neighborhood, don't miss out!</p>

    Heading out to Jacksonville now - if I don't see you, have a great weekend!</p>

    Jeremy </p>
     
  2. <blockquote><hr>

    The Jacksonville Town Hall is tomorrow! Looking forward to seeing all of you there!

    [/ QUOTE ]

    Anyone going any other EA Reps / Stratics or players?




    PS: Larisa you getting faster... [​IMG]
     
  3. Larisa

    Larisa Reporting Manager
    Editor Reporter Moderator Professional Wiki Moderator Stratics Veteran Stratics Legend Editor Fallen Lords

    Joined:
    Sep 8, 2006
    Messages:
    4,345
    Likes Received:
    1,958
    *laughs* not fast enough...I knew it would be up early today because Jeremy is going to the TownHall but I forgot.....I'm slackin I know! *grins*
     
  4. *small bag of Quatloo* forrrrrr Larisa! ... [​IMG]

    HEY! whose gonna do "live feeds" for the townhall ... anyone? beuller?

    Hey! Florida? be sure to carry an uninteruptable power supply! [​IMG] just kidding ...

    Florida? Cooper? wasn't Florida Cooper?
    *cups hands*
    COOOOO O O O O PER !!!
     
  5. Frarc

    Frarc Stratics Legend
    Stratics Veteran Alumni Stratics Legend

    Joined:
    Mar 18, 2003
    Messages:
    7,770
    Likes Received:
    1,132
    Can someone gate me from Belgium? [​IMG]
     
  6. Larisa

    Larisa Reporting Manager
    Editor Reporter Moderator Professional Wiki Moderator Stratics Veteran Stratics Legend Editor Fallen Lords

    Joined:
    Sep 8, 2006
    Messages:
    4,345
    Likes Received:
    1,958
    You're in Belgium??? I have a friend there [​IMG] Wouldn't it be nice if we could do that???? *sighs*
     
  7. Guest

    Guest Guest

    Hopefully someone will be able to pin the team down and get
    some more info on stable slot increases. Maybe, that is why it
    wasn't referenced here. [​IMG]
     
  8. Guest

    Guest Guest

    <blockquote><hr>

    You're in Belgium??? I have a friend there [​IMG] Wouldn't it be nice if we could do that???? *sighs*

    [/ QUOTE ]

    We almost did but if insects happen to fly in the dna would get spliced into your own and turn ya into a freak.
     
  9. Deception

    Deception Sage
    Stratics Veteran Stratics Legend

    Joined:
    Oct 25, 2004
    Messages:
    643
    Likes Received:
    19
    <blockquote><hr>

    If the defender is not using a weapon, "weapon skill" will be EITHER Wrestling skill OR (Anatomy + Evaluating Intelligence + 20) / 2; whichever is greater (which means that if you have none of these skills and are barehanded, you'll have a functional weapon skill of 10

    [/ QUOTE ]

    This is capped at 120 right?

    120 anatomy and 120 eval would mean 130 equivalent weapon skill in that equation.

    Please tell me it's capped?
     
  10. <blockquote><hr>

    (Anatomy + Evaluating Intelligence + 20) / 2

    [/ QUOTE ]
    I think this is mistaken. "+20" should be dropped.
     
  11. Frarc

    Frarc Stratics Legend
    Stratics Veteran Alumni Stratics Legend

    Joined:
    Mar 18, 2003
    Messages:
    7,770
    Likes Received:
    1,132
    <blockquote><hr>

    You're in Belgium??? I have a friend there [​IMG] Wouldn't it be nice if we could do that???? *sighs*

    [/ QUOTE ]



    Yup, land of chocolate , beer and fries.......and sprouts! [​IMG]
     
  12. Jhym

    Jhym Guest

    Power troubles? In Florida? I have never heard of such a thing.


    Seriously though, in Florida it's required to have a UPS. Especially most afternoons during the summer, around 4.
     
  13. What are you talking about? It isn't "mistaken" at all; the +20 is correctly part of the formula.

    (120 eval + 100 anat + 20)/2 = 120 weapon skill. That is how it is commonly used. (Also 110 on either side in budget situations.) Anatomy over GM in this equation wouldn't be beneficial as it is capped at 120.

    If you removed the +20, the calculation would be worthless as you might as well just invest the points in wrestling every time. Right now as it stands the Eval + Anat formula provides a way to either save 20 skill points in tight templates, or give people a budget route to 120 evasion without buying a +20 scroll in any skill.
     
  14. <blockquote><hr>

    What are you talking about? It isn't "mistaken" at all; the +20 is correctly part of the formula.

    (120 eval + 100 anat + 20)/2 = 120 weapon skill. That is how it is commonly used. (Also 110 on either side in budget situations.) Anatomy over GM in this equation wouldn't be beneficial as it is capped at 120.

    If you removed the +20, the calculation would be worthless as you might as well just invest the points in wrestling every time. Right now as it stands the Eval + Anat formula provides a way to either save 20 skill points in tight templates, or give people a budget route to 120 evasion without buying a +20 scroll in any skill.

    [/ QUOTE ]
    On what source is your explanation based? My statement is based on KB:"How do I use Anatomy?":
    <blockquote><hr>

    Another aspect of the anatomy skill is that it assists in preventing spell casting disruption when used with the evaluating intelligence skill. Anatomy and evaluating intelligence are averaged to determine the value of your ability to not be disrupted in a similar fashion to the wrestling skill.

    [/ QUOTE ]
     
  15. My source is the fact it has been the way as described in the FoF since forever. It is common knowledge and was implemented as such in Publish 16.

    Here's a quote I've dug up from Evocare from way back when confirming the formula, originally posted on the uo.com message boards:

    <blockquote><hr>

    As a quick note, the actual formula for "dodging" a swing with anat/eval is: (anat+eval+20)/2, capped at 120 max. So, a character with 120 eval, 100 anat (for example) actually has a full 120 "hit evasion" ability

    [/ QUOTE ]
    Sourced from an old mage guide.

    You're welcome.
     
  16. Guest

    Guest Guest

    I hope they are fixing all the exploits I have been submitting. I just remembered another big one tonight.
     
  17. <blockquote><hr>

    My source is the fact it has been the way as described in the FoF since forever. It is common knowledge and was implemented as such in Publish 16.

    Here's a quote I've dug up from Evocare from way back when confirming the formula, originally posted on the uo.com message boards:

    <blockquote><hr>

    As a quick note, the actual formula for "dodging" a swing with anat/eval is: (anat+eval+20)/2, capped at 120 max. So, a character with 120 eval, 100 anat (for example) actually has a full 120 "hit evasion" ability

    [/ QUOTE ]
    Sourced from an old mage guide.

    You're welcome.

    [/ QUOTE ]

    Thank you very much for you that you kindly show me the truth.

    --

    This is the "official announcement" about this issue on uo.com: Publish 16 - Full list and Bugfixes"
    <blockquote><hr>

    Changes to Combat
    Unarmed "to-be-hit" changes (chance to get hit while unarmed is now the better of either Wrestling skill, or an average of Anatomy &amp; Evaluate Intelligence).

    [/ QUOTE ]
    Japanese official web site translated only this official announcement into Japanese, which results that we Japanese players are not correctly informed for 6 YEARS!

    From where could we know this truth? Why are we so long deceived by EA/OSI? Why could you devs do such an unfairness to Japanese players?
     
  18. [​IMG] not EA's fault


    [​IMG]
    Spiritless felt it WAS common knowlege ...

    thus and so! tis DONE !

    tadaaaaa!

    [​IMG]
    *toot!*


    Thx athos ...
    Say ... haven't got a copy of that Old Loot pack distribution chart ...

    by any chance?
     
  19. <blockquote><hr>

    Barehanded defense: If the defender is not using a weapon, "weapon skill" will be EITHER Wrestling skill OR (Anatomy + Evaluating Intelligence + 20) / 2; whichever is greater (which means that if you have none of these skills and are barehanded, you'll have a functional weapon skill of 10.)

    [/ QUOTE ]

    Does Jack of All Trades come into play for human chars?

    If so wouldn't the lowest functional weapon skill be 30 for humans and 10 for elfs?
     
  20. Guest

    Guest Guest

    I can't give an explanation one way or another for pub 16-era communications, as I'm sure you understand, but I saw the formula in the code with my own eyes, and the +20 is in fact correct.
     
  21. Would It be possible to give the new "greater dragons" a new spawning ground in destard?

    I know a lot of people who have hunting troubles there now, especially when I go. I got insta killed walking in the dungeon entrance. They are a bit.. strong, but I don't believe its great encouragement to goto a dungeon to be killed on the spot upon entering it.

    They should be moved to another area of teh dungeon (further from the entrance). Perhaps on the hill back where the shrine resides.
     
  22. <blockquote><hr>

    Would It be possible to give the new "greater dragons" a new spawning ground in destard?

    I know a lot of people who have hunting troubles there now, especially when I go. I got insta killed walking in the dungeon entrance. They are a bit.. strong, but I don't believe its great encouragement to goto a dungeon to be killed on the spot upon entering it.

    They should be moved to another area of teh dungeon (further from the entrance). Perhaps on the hill back where the shrine resides.

    [/ QUOTE ]

    It would be helpful also if they looked a bit different. Maybe a different color if not a new graphic.
     
  23. <blockquote><hr>

    <blockquote><hr>

    Would It be possible to give the new "greater dragons" a new spawning ground in destard?

    I know a lot of people who have hunting troubles there now, especially when I go. I got insta killed walking in the dungeon entrance. They are a bit.. strong, but I don't believe its great encouragement to goto a dungeon to be killed on the spot upon entering it.

    They should be moved to another area of teh dungeon (further from the entrance). Perhaps on the hill back where the shrine resides.

    [/ QUOTE ]

    It would be helpful also if they looked a bit different. Maybe a different color if not a new graphic.

    [/ QUOTE ]OHH !!!! yes a new color would be nice. Like the color purple [​IMG]. I hear most of you saying purple? but I have to at least try for my purple dragon, I need it to match my outfit, Its a girl thing ya know.[​IMG]
     
  24. <blockquote><hr>

    "Why do I seem to miss so often when using Archery?"

    I've seen several variations on this question, so Leurocian scoured the accuracy code for me, and we did not find anything that would specifically affect Archery accuracy (with one exception, covered below.) Accuracy in general is a fairly complex topic with many variables, so it's a tricky thing to test at the best of times. Here's an explanation of the different factors - if anyone sees any results that aren't consistent with this, please let us know!

    Attacker's stats: Weapon skill and Hit Chance Increase are the stats the formula uses on the attacker's side. HCI is increased by HCI items and decreased by the defender's Hit Lower Attack, if any.

    Defender's stats: Weapon skill and Defense Chance Increase are the stats the formula uses on the defender's side. DCI is increased by DCI items and decreased by the attacker's Hit Lower Defense, if any. (Exception below.)

    Barehanded defense: If the defender is not using a weapon, "weapon skill" will be EITHER Wrestling skill OR (Anatomy + Evaluating Intelligence + 20) / 2; whichever is greater (which means that if you have none of these skills and are barehanded, you'll have a functional weapon skill of 10.)

    Mage Weapons and Use Best Weapon Skill: UBWS uses whichever weapon skill is highest, and Mage Weapons use Magery (with whatever penalties listed on the weapon) assuming it's higher than the appropriate weapon skill for that weapon.

    The minimum chance to hit is always at least 2%. If both attacker and defender have the same weapon skill and HCI/DCI, the chance to hit is 50%.

    But wait, there's more!

    Attacks can be parried or evaded - those are completely different calculations that happen after the hit chance is calculated. If you're trying to track your accuracy, make sure you're taking parried or evaded attacks into account.

    Archery has one more caveat - if you move, it resets the swing cycle. There's only one place where this might break the swing - between the "swing" and the "hit" calculations - and we couldn't make it break in a few minutes of testing, but it's possible that a move could break the swing and it would look exactly like a miss, when in fact the attack was simply aborted. Archers, hold still!

    The actual formula is conveniently listed in the question below!

    "The formula for Hit Chance on Stratics is: Hit Chance% = ( ( [Attacker's Combat Ability + 20] * [100% + Attacker's Hit Chance Increase] ) divided by ( [Defender's Combat Ability + 20] * [100% + Defender's Defense Chance Increase] * 2 ) ) * 100

    Does HLD lower the defender's combat ability by 25%? Does HLA lower the attacker's combat ability by 25%? Or are both applied to the DCI and HCI portion of the calculation?"

    HLA/HLD are directly applied to the HCI/DCI portion of the equation - when the HLA/HLD effect goes off, it reduces HCI or DCI by 25 (and yes, it can make that value negative.)

    [/ QUOTE ]

    Thanks for this great information! :&gt;

    I think I see what you mean about a movement-interrupted shot maybe masquerading as a miss client-side. Like the results you describe from your testing on this, I've never noticed anything like that; even in a laggy connection the client seems to accurately show hits vs. misses vs. shots that don't get off due to the archer moving too soon. Even if you don't get to hear the final sound of the hit or miss until the lag spike is finished, it always seems to show the right thing.

    Regarding odd client displays with archery, I can make something happen that is the opposite side of that possibility. You can be on an archer on a laggy connection and walk off at what is the correct time for the weapon's swing speed, and a few steps after you've moved, you hear and see the shot get off (be it a hit or miss). It's a bit like a correctly-executed, standing shot masquerading as a moving shot in the client. I have no idea what it looks like to other clients, I've never asked anyone (thats easy to find out, though).

    I have a sense as to why this thing happens, the client and server are really making an effort to not send the wrong story to the player as to what's going on with the archery shot. But in contrast, the client is going to represent the archer as taking those steps when the player asks for it (and bounce him back if that representation was in error) because otherwise the game would feel laggy and horrible in any sort of lag. It doesn't handle the shots that way, though, it's set to be really sure on those.

    Applicable point is: the results of your brief test seems right to me: the client appears to be good at not mixing up misses, hits, and shots that didn't make it out of the bow. Even when client-side it lets the guy appear to move before server verification, it doesn't seem to play those games with the shot details.

    Before this, I've thought about really, seriously testing the funky hit chance stuff that seems to some of us to happen with archery. And yeah, the advice to be darn sure to understand the parry, weapon skill, and HCI/DCI interaction is hella important for anyone who wants to try. I should maybe actually do that now that this point's been specifically looked at. It would also make sense to check to see if other skills that aren't supposed to affect it actually, somehow, do, either on the target or the archer.

    So yeah - - thanks for taking a look into it! I do feel like checking now, and seeing what happens.
     
  25. <blockquote><hr>

    I can't give an explanation one way or another for pub 16-era communications, as I'm sure you understand, but I saw the formula in the code with my own eyes, and the +20 is in fact correct.

    [/ QUOTE ]
    I have tested this.

    I have made a character of elf wrestler with wrestling 0.0, anatomy 0.0, EI 0.0. She should have (anatomy 0 + EI 0 + 20 )/2 = 10 weapon skill. She fought with a human wrestler with wrestling 20.0. Human wrester had a base damage of 2-8, so she could give at least 2 damage on each hit, and elf wresler has only 45 physical resistence, so she could reduce the least damage at most to 1, this meaan every hit would do her a damage at least.

    If her weapon skill will be substituted by (anatomy 0 + EI 0 + 20 )/2 = 10, attacker's hit chance will be:

    [(20+20)/(10+20)]*50(%) = 67%

    therefore defender's defence chance is 33%. If her weapon skill is not substituted by the equatation with +20%, but by (anatomy 0 + EI 0 )/2, her weapon skill for defence will be 0%. This time attacker's hit chance will be 100%([(20+20)/(0+20)]*50(%) = 100%).

    Result is as follows:

    attempts:hits:dodged
    100 : 95 : 5
    100 : 93 : 7
    100 : 96 : 4
    100 : 95 : 5
    100 : 97 : 3

    I made one more elf attacker of the same skills:
    attempts:hits:dodged
    100 : 96 : 4

    (*)Note: I can't understand why this is not 100%. Does the parry work? But I set the parry skill of Elf wrestler 0.0, besides if the parry worked, an effect of round would appear, but I didn't see any effects.

    To compare, I made an elf wresler of Wresling 10.0, anatomy 0.0, EI 0.0. Hit chance of attacker is theoretically 67%([(20+20)/(10+20)]*50(%) = 67%).

    Results:
    attempts:hits:dodged
    100 : 65 : 35

    I could get the expected results at this time.

    --

    How do you think devs on the result of this experiments?