Member
Incredible sleuthing there, Sherlock! Being a number junkie myself, I am always curious at the behind the scenes algorithms that put these numbers in place.
So with all the formulas locked down, have you conducted any 'audits' of players' final LB score? As in, gone through and grabbed their times for every single track, calculated their final LB score from the formulas, and checked to see if it matches what the actual LB score is?
I ask this question not because I doubt any of your conclusions, but because I wonder if there have been some discrepancies when the leaderboard was broken, where people got points for tracks that other people didn't.
Member
Also, what happens if your medal for a track goes from being a Gold to a Silver? I had this happen to me on Woodwork Conclave, where my 2 fault time gave me Gold, but I then got a 1 fault run with a worse time that gave me Silver. I don't think the system ever subtracts points... Does it?
Senior Member
Well, yeah I do have that; its what I was talking about when at the end of my postings I said: " Here is what I can share for now; its a small amount from what I originally hoped for, but its the best I can do at the moment".So with all the formulas locked down, have you conducted any 'audits' of players' final LB score? As in, gone through and grabbed their times for every single track, calculated their final LB score from the formulas, and checked to see if it matches what the actual LB score is?
I ask this question not because I doubt any of your conclusions, but because I wonder if there have been some discrepancies when the leaderboard was broken, where people got points for tracks that other people didn't.
I downloaded all times on all LB/platforms from Bonxy (I am afraid I hit his server pretty hard), and have it all ready showing rank changes, etc. I had every intention of going that last step and just saying; "here it is!".The scores are definitely "close" as they should be, but there are some clear cases where people are off by a couple of million.
(oh, and I think I removed almost all the Android hackers too!)
And I was going to post it when I poked around a little more at what I saw, and saw that carterclem7 dropped way down to rank 14... and it seemed odd so I looked into it and... Bonxys times seem wrong for him. (See this post):
http://forums.ubi.com/showthread.php...1#post11652508
Maybe when bonxy figures it out we'll know if its just carterclem7 or if this is happening to more people?
Plus my data for donkey/crazy LB is not as current as I would like... its from before Bonxy fixed his issue with Spinners Alley not posting on the Crazy LB so I am missing a tracks-worth of crazy-points...but I cant update it because I think I hit Bonxy's site so hard he blocked my IP address (quite understandable)... plus I found 4-5 Plat scores that were wrong over the 30 tracks I did, so it makes me wonder about the rest.... am I really showing good data, or just other bugs in Bonxys and the medals times we have that I dont know about?
Really not trying to diss Bonxys site... I know its going to be great one day, but what he is building is a big task and it just might not be ready yet. All that data is there and its not until you really poke at it that you find these odd things . . .
.. overall, I guess what I am saying is I dont know if what you see is really going to be... valid. It kinda bummed me out a little; I had high hopes.
Now that you now where I am coming from... do ya'll still want to see it?
Senior Member
In testing the cost of faults ... I KNOW I got points for have one less fault but with a slower score. I was doing this... pick a score (gold) and do 4 faults. Then try and get the same time with one fewer faults and see how many points I got. . . .sometimes I got slower scores but I still got points. They wernt slow by much... but they were slower.Originally Posted by Whitechapple Go to original post
But in my case it wasnt drastic enough to go from Gold to Silver like what you said.. still, my bet is that they know your previous "score" and just calculate your new one and compare. If its greater, then add the points.
One thing that donkey scoring seems to make pretty clear... how they score points doesnt have to have any real connection with the visible silver/gold/plat points they award medals for. Its like they are two different sets of logic entirely.
Senior Member
nice detective work xitooner, very impressive!
ps. it seems that I need to cut off that one fault from Woodwork Conclave...
and ofcourse we wanna see it.Originally Posted by xitooner Go to original post
Member
Sure let´s see it. Be gone cheaters!(almost)Originally Posted by xitooner Go to original post
Senior Member
Haha. I thought I seen my site drop out once or twice.
I run another unrelated site on that server which is quite active and I thought it was that causing me to crash.
Xitooner, private message me your email address and we can chat. I'll also unblock your IP tonight and try and implement your logic!
Good work though.
Junior Member
Hey!
You did a really great job!!!
But are you sure about 3M points?
I drove at Cave on Donkey just now and got exactly 1,750,000 points for the first attempt (but it was improvement from 10.999 (you said ultimate time is 11.000), not the run from scratch), and no points for the seconds one (maybe just too bad improvement for noticeable difference in points?).
Spoiler: Show
Senior Member
Thanks (to you and everyone earlier) for liking my work.You did a really great job!!!
But are you sure about 3M points?
I drove at Cave on Donkey just now and got exactly 1,750,000 points for the first attempt (but it was improvement from 10.999 (you said ultimate time is 11.000), not the run from scratch), and no points for the seconds one (maybe just too bad improvement for noticeable difference in points?).
Besides those runs recorded by EHVdeG and I a year ago, this run I made this last weekend is the only other time I've actually measured a 3-million run (I started up a new ID for testing and it took 2++ weeks to earn the donkey to do this test)
So there havent been many measurements of this...and its always possible I got something wrong, or there is more going on here than we know.
My intepretation of your results is that you already had 10.999 before you ran, so you must have already had your full 3 million awarded before. Once you get 3 million, any futher time improvement wont give you any more points (you maxed your points on the track). That would match what I have seen.
As for the points you did get.... I am guessing maybe those are points awarded to you from one/more previous runs? I have seen something like that before; points that arent awarded right away (you were out of RL server communication, RL servers were slow to process your results right away, etc)
And so the points get awarded on your next run.
You know the oddest thing of all to me? How you got such an even number of points.
Here is my procedure to do my data gathering; its how I've been trying to avoid some of these pitfalls (my apologies if this gets really detailed)
1) At the start of a data-collection session against a LB, pick a different track, and improve your global score. (I designated a single track for this and just made slight changes to the score each time). Record the points; until you see LB points awared, you arent ready for testing. That will avoid "extra" points from before.
2) That LB score is your "base" number. In the table below, its the number in red for my test.
3) Pick your time target before running, and always do 0-faults unless thats what you are testing (its just easier); then record the time you actually got. DONT accept the run unless you like it; just reset back and rerun again (and again..)
(down below you can sorta see I was aiming for 40 sec, 30.7 secs, 24, 22, etc)
4) When you are ready to record, accept the run results, and record the results as fast as you can (write it down, make a snapshot, etc I often ran on my ipad and took pictures with my iphone). If you miss recording the points... I would stop/restart the game and go get it from the LB (that always seemed to work for me).
5) Subtract that new LB number from your "base" number. (I just found that easier than using the previous LB number; very easy to do this in excel and set up a forumula)
This was my Donkey run results; my "base" number was in red.
Cave Times LB Points LB-Base 831925 39.910 918659 86734 30.575 1086589 254664 just past silver 23.891 1345550 513625 just past gold 22.199 1557050 725125 just under plat 18.116 1670198 838273 14.110 1761243 929318 12.016 1808834 976909 11.842 1812789 980864 11.037 1831084 999159 creeped up to Ultimate 10.948 3831925 3000000 crossed Ultimate 10.641 3831925 3000000 I had maxed points
At this point all I can hope for is that these findings do all keep making sense if we do this the same way. If it doesnt, we'll try and figure it out together...
I think I am going to go add this to my original post....
Senior Member
this is an amazing job xitooner, big thumbs up for this, i have few questions tho:
1)A)about the night life on donkey\crazy
im not sure if i got it right, RL based all scoring system for donkey and crazy on night life medal times?(in contrast normal LB has an individual scoring system for each track, is that right?)
1)B)so short tracks are the most efficient way to gain points?
2)so basically we get the biggest amount of points on tracks where the achieveable time is far from plat time?
3)forgot what was my last question so im just gonna ask: how are you so smart?
