Jump to content

Thai_tong

Members
  • Posts

    1175
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by Thai_tong

  1. I realised that dagganoth under waterbirth had different charm drops to the lighthouse ones so I tested and fixed their drop rates. Also players who don't get enough charms from slayer will now kill dagganoth to get the remaining charms.

  2. [Arbitrary update on how much xp suomi has left because it is super important]

     

    I'm actually enjoying watching further down the list, the competition of sorts (loosely using that terminology) between Alkan, Dragonseance and Kingduffy. I know Alkan is smashing through serious Attack xp at Icefiends, but it is still rather close. Looking like he'll be next to 3B though. I am, however, also looking forward to seeing DrSeance smash out some serious xp in the near future.

     

    Arbitrary remark supporting Suomi which he obviously needs]

     

    Oh come on... don't be so negative about it. This kind of thing doesn't happen every day, so just let people enjoy it.

    This kind of thing has been happening every day for a year. "This" being supporting suomi.

     

    Do you cheer for your favourite football team every game during the playoffs, but cease to do so when they reach the Super Bowl? He doesn't need motivation, and I don't think anyone here is trying to give him any - people are simply commenting because they're excited.

    I'm explaining why Aussie posted that because you went on a side track. People are bored of seeing the same posts for the past year, does that explain it?
  3. [Arbitrary update on how much xp suomi has left because it is super important]

     

    I'm actually enjoying watching further down the list, the competition of sorts (loosely using that terminology) between Alkan, Dragonseance and Kingduffy. I know Alkan is smashing through serious Attack xp at Icefiends, but it is still rather close. Looking like he'll be next to 3B though. I am, however, also looking forward to seeing DrSeance smash out some serious xp in the near future.

     

    Arbitrary remark supporting Suomi which he obviously needs]

     

    Oh come on... don't be so negative about it. This kind of thing doesn't happen every day, so just let people enjoy it.

    This kind of thing has been happening every day for a year. "This" being supporting suomi.
  4. cant believe this topic was technically started in 2006.... and now over 6 years later someone is utterly close to achieving this goal.

     

    glad ill get to see this happen :)

     

    anyone come up with some kinda ETA ?

    Why do people use the buzz word "technically" when it adds nothing to the meaning.

    • Like 2
  5. This is great, might be more suited to Data and Research though?

    I keep it up to date so that people can use it as a tool for their own slaying not a place to go to find various averages like herbicide xp per kill (not that anyone would want to know anything but effigy drops and ganodermic drop rates).

     

    I'm not sure how this data holds up in credibility, but you might want to cruise around it a little bit, in case you haven't already.

    Thanks for that info, I'd use the clue scroll data if I was ever to add an option to complete hard or elite clues.

    The rest of the data is not accurate enough though. The old formula for effigy drop rates which they say was reduced by a factor of 2.8 pre-dates March 2011. As they say in relation to frost dragon vissage rates, their whole calculation could be skewed by people picking up effigies and not opening them immediately- I have several measures to help prevent this from happening to my data. However, even though the actual rates are inaccurate, the relative rates of items dropped should be very accurate with their sample size, I will be able to use this to find a better approximation to special drops such as dragon plateskirts/legs dropped by iron and steel dragons :)

  6. I should have a better formula for the drop rate of effigies within the next 2 months. I am following the adventurers logs of 40 bots to see how often they get effigies and from those grotworm/frost dragon/green dragon kills I will be able to find a formula for drop rate of a monster of any combat level.

  7. Also, I don't think they will enable xp lamps or anything similar for the new skills when they get released (like they did with dungeoneering), so I don't think people will just buy their way to top ranks :)

    Its a golden opportunity for jagex.

  8. Enaid kindly helped me with testing accuracy.

     

    Firstly this is to test how attack and defence change when using the wrong style

    I used a rapier with 1521 attack bonus against enaid's 279 melee defence bonus.

    I got 550 hits and 51 misses which is an accuracy of 0.915±0.022

     

    If defence/attack level gets adjusted for using the wrong style too then

    attack rating= 1521+(99*.5)= 1570

    defence rating= 279+(99/3.2)= 309

    This would give an accuracy of (2*1570- 309-1)/(2*1570+2)= 0.901 which is inside the confidence itnerval.

     

    If the defence/attack level doesnt get adjusted

    attack rating= 1521+99= 1620

    defence rating= 279+99= 378

    Giving an accuracy of (2*1620- 378- 1)/(2*1620+2)= 0.882 which is outside the confidence interval. So we can be 95% sure that defence

    In fact the accuracy 0.882 is only within the 99.6% confidence interval so we can be 99.6% sure that defence level does get adjusted.

     

     

    So the maths I did with the previous data was wrong.

    In the first set where the chance of hitting is 0.5418±0.0616

    Attack rating= (99/2+81)a= 130a

    Defence rating= (99/1.6+85)d= 146d

    0.5418±0.0616= (260a- 146d-1)/(260a+2)

    Let the chance of hitting= P

    P= (260a- 146d-1)/(260a+2)

    -146d= (P-1)*260a+ 2P+ 1

     

    The the 2nd set accuracy= 0.5100±0.0618

    Attack rating= (99/2+25)=74a

    Defence rating= (99/1.6+85)d= 146d

     

    0.5100±0.0618= (148a- 146d-1)/(146a+2)

    Let the chance of hitting= R

    R= (148a- 146d-1)/(148a+2)

    -146d= (R-1)*148a+ 2R+ 1

     

    (P-1)*260a+ 2P+ 1= (R-1)*148a+ 2R+ 1

    a= (R-P)/(130P- 74R- 56)

     

    Putting in upper and lower limits for P and R still shows that a is less than 1 :mad:

  9. More combat formulae research

     

    Turmoil boosts attack and drains defence. I tested a bit to try and identify what the multipliers are.

     

    Turmoil drains defence over time, in the first test I waited 3 mins for it to drain fully before starting. There was no break between the first and second tests so before the second test started it had been draining for 13 minutes.

     

    First test was steel dagger at 99 attack vs iron helm,plate,legs,shield at 99 defence.

    In melee vs melee the attack bonus of the weapon is halved and the armour bonus is divided by 1.6. The results from this very test indicate that attack and defence levels are not affected by using the wrong style. I assume that attack rating= (weapon attack/2 + attack level)*prayer

    and defence rating= (armour/1.6 +defence level)*prayer

     

    Call the attack bonus of turmoil a and the defence drain d

    So in this test attack rating= (99+81)a= 180a

    Defence rating= (99+85)d= 184d

    I hit 136 times and missed 115 times

    I'm 95% sure the accuracy is in the region 0.5418±0.0616

     

    Where A is attack rating and D is defence rating the formula for accuracy where A>D is

    accuracy= (2A- D- 1)/(2A+2)

    0.5418± 0.0616= (360a- 184d- 1)/(360a+ 2)

     

     

    The second test was bronze dagger vs iron helm,plate,legs,shield

    Defence rating is still 184d

    Attack rating= (25+99)a= 124a

    I hit 128 times and missed 123 times.

    I'm 95% sure the accuracy is in the region 0.5100±0.0618

     

    0.5100± 0.0618= (248a- 184d- 1)/(248a+ 2)

     

     

     

    Let 0.5418± 0.0616= P± p

     

    P± p= (360a- 184d- 1)/(360a+ 2)

    Solving for 184d

    184d= 360a- 1- 360Pa- 2P± (360ap+ 2p)

     

     

    Let 0.5100± 0.0618= R± r

     

    R± r= (248a- 184d- 1)/(248a+ 2)

    Solving for 184d

    184d= 248a- 1- 248Ra- 2R± (248ar+ 2r)

     

    So we have that

    360a- 1- 360Pa- 2P± (360ap+ 2p) = 248a- 1- 248Ra- 2R± (248ar+ 2r)

    a(112- 360P+ 248R)± a(360p+ 248r)= 2P- 2R± (2r+2p)

    I've done it in terms of P,p,R,r so that its easy to re-calculate if someone wanted more accurate values.

     

    Using this equation to find a gives results around a=0.002 when it should be something like a=1.1

     

     

    What are your thoughts?

  10. The amount of facepalms I just had... :wall:

     

    On a topic worthy note, what does the numbers inside the curly brackets mean? Ex. (55). Does that mean increased hours?

    It means he now needs 55 more hours than he did last week. Its because gemeos lowered the xp rate on magic.

    • Like 1
  11. I'd advise against getting the rudder first. I got it and regretted it. Only gives +200 speed compared with the cherrywood one. For reference, the best attainable Bowl voyage time is 6:57 if I'm not mistaken.

    Friend said his times went from avg 9 hours to avg 7 hours on bowl with the rudder upgrade, I'm definitely going for it first.

    This could help on weekends but if you are in full time education/job then the 2 hours might not allow you to make a 2nd voyage.

     

    Does a palmist's "good fortune" make your voyage more likely to be successful?

  12. Basically one can calculate how likely it is (given the known expected probability function) that you will achieve the results you did.

     

    From what I gather, you are noting what the game says your chance of success is and if you fail or succeed.

     

    If from N trials the game says you have chances p1, p2, p3, ..., pN with average p

    and from N trials your ship succeeds T times so that your overall success rate h is equal to T/N

     

    The difference between your success rate and what the game says is |h-p|

    You can find the chance of this difference occurring using confidence intervals.

     

    If the theory that the game give the correct chance is true then the variance of the distribution is p-p2

    If it is a coincidence that your data differs from the game then |h-p|= z*((p-p2 )/N)1/2

    z= |h-p|*(N/(p-p2))1/2

    From your data you know the values of p, h and N so you can find the value of z

    Next use this tool http://www.mathsisfun.com/data/standard-normal-distribution-table.html

    to find the % for your value of z

    Make sure "0 to Z" is selected, not "Up to Z" or "Z onwards"

    When you have found the percentage that corresponds to the value of z multiply it by 2 and subtract it from 1. The final figure is the likelihood that the actual chance of success is the same as what the game says.

    For example if z= 2.2 then the percentage for that is 48.6% or 0.486

    0.486*2= 0.972

    1- 0.972= 0.028

    There is a 2.8% chance that the actual chance is the same as what the game says

    • Like 3
  13. Well master rc only saves you 45 hours of runecrafting getting from 99 to 200m so do you think that getting master rc and getting runes for transportation + 170k essence would take less than 45 hours?

    Why is that even relevant. In all my runespanning I've never had shortages when I return after leaving. Seems like common sense says trade 45 hours for something that has zero value.

    I have never played runespan before, his post makes it sound relevant and I'm telling you what my calculation showed.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.