|
Pauldelbrot
|
|
« Reply #1 on: June 01, 2010, 01:13:36 PM » |
|
There's something screwy about the ratings. Ones that had a four star rating a few hours ago, for example, have 3.02 or even 2.8 listed. Yet a four star rating obviously had to have been between 3.5 (inclusive) and 4.5 (exclusive). It's not that the numbers are simply lower by 1 either, because there is at least one 4.5 in there, and if the numbers were lower by 1, the highest possible would be 4.
As best as I can tell from memory and eyeballing the results, a sizable majority of the ratings are from half a point to two full points lower than they should be. None are higher. The highest-rated one in Fractal Fun for example had a pretty solid five star rating and now shows 4.5, which is barely qualifying for five stars; if the 4.5 rating is correct one single four-star or lower additional vote would have reduced it to four stars, and I don't think that's the case. If I were to guess at the "correct" rating for that one, it was probably in the 4.7 to 4.8 range, not 4.5.
So, unless several people dumped a rash of mostly one-star ratings on a fairly broad subset of the images during the final six or so hours of the voting period, which seems unlikely, something seems to have gone wrong with the calculations used to produce the result lists.
The ranked order is probably only moderately affected; in particular, the listed winners for Fractal Fun and Movies look like the same ones that looked to be winning just before the end of voting. This is because they're all lowballed and by roughly the same amount. But I am pretty sure the ranked order is incorrect for a fair number of the images other than the winners. One of the winners, unfortunately, also looks to be incorrect. David Makin's Forbidden Forest had been winning in Mandelbrot; as of a few hours ago it was the only one in that category sporting a five-star rating. It can't possibly now be in fourth place with a measly 3.65, yet supposedly it is. The "winner" listed for that category is a four-starrer with a nearly-unchanged 4.21 rating showing while Forbidden Forest seems to have been dinged a full point.
Any idea what happened?
|
|
|
Logged
|
|
|
|
Sockratease
|
|
« Reply #2 on: June 01, 2010, 01:21:50 PM » |
|
Any idea what happened? It's a Conspiracy!! The Gods are taking their Vengeance for our misdeeds. 2012 is coming!! Prepare thyself. The End is Nigh!! Other than that, nope. No idea. But a Computer calculated the results, and Computers are Never incorrect [/sarcasm] I'm sure Trifox will be Fair and Just, looking at the details and acting appropriately.
|
|
|
Logged
|
Life is complex - It has real and imaginary components. The All New Fractal Forums is now in Public Beta Testing! Visit FractalForums.org and check it out!
|
|
|
cKleinhuis
|
|
« Reply #3 on: June 01, 2010, 03:06:50 PM » |
|
those final results were indeed suspicious, but i have checked some of the votings, and it appears that many strategic votings took place right before end of the contest, i found many 1 star votes, far more than the 2 that were original there
but it is like it is, the final results !
|
|
|
Logged
|
---
divide and conquer - iterate and rule - chaos is No random!
|
|
|
bib
|
|
« Reply #4 on: June 01, 2010, 08:14:45 PM » |
|
I 100% agree with Pauldelbrot. I do not complain for my bad ratings, but I really think something strange happened, especially about David Makin's entry in the Mandelbrot section.
@Trifox : did the last minute 1 star votes come from newly registered users ?
|
|
« Last Edit: June 01, 2010, 08:16:51 PM by bib »
|
Logged
|
Between order and disorder reigns a delicious moment. (Paul Valéry)
|
|
|
cKleinhuis
|
|
« Reply #5 on: June 01, 2010, 09:01:18 PM » |
|
in fact there where 5 last minute 1 star votes, but only one user has registered on 5th may 2010 the rest has registered years ago
|
|
|
Logged
|
---
divide and conquer - iterate and rule - chaos is No random!
|
|
|
Jesse
Download Section
Fractal Schemer
Posts: 1013
|
|
« Reply #6 on: June 01, 2010, 09:09:36 PM » |
|
This was a really chaotic contest with some strange results, but maybe typically for a fractal forum that is founded on chaotic systems? With respect to everyone who feels disappointed, i would for example spend my mug for Daniels forbidden forest. But i guess he has already some trophy? Or can u put Daniels image on it, that would be a compromise However, many thanks to all
|
|
|
Logged
|
|
|
|
cKleinhuis
|
|
« Reply #7 on: June 01, 2010, 09:20:46 PM » |
|
a final result message will come up later, but from me also, thank you all for participating this "chaotic" contest, one reason for the dissusion might be the grown attention, in the previous year we had around 50 entries, and far fewer page impressions during the period, but the admin team is aware of the problems concerning a fair voting in an online forum, we try to improve this! also we (I) are looking for sponsors to get a little more attractive prizes, so, everyone else is also invited to get a sponsor, especially those who could provide special things e.g. like a 3d printing of the winner item, a lenticular pseudo 3d print or holographic items, my dream would be to produce nice, neat things as competition prizes, and sell them over the year as merchandise for the forum ... but these are just thoughts/dreams and congratulations to all of you cheers!
|
|
|
Logged
|
---
divide and conquer - iterate and rule - chaos is No random!
|
|
|
David Makin
|
|
« Reply #8 on: June 02, 2010, 12:04:05 AM » |
|
@Jesse: Did you mean "Dave's Forbidden Forest" ? Thanks to all for pointing out that something "iffy" happened Being a moderator I restrained myself from mentioning it
|
|
|
Logged
|
|
|
|
Jesse
Download Section
Fractal Schemer
Posts: 1013
|
|
« Reply #9 on: June 02, 2010, 12:15:47 AM » |
|
@Jesse: Did you mean "Dave's Forbidden Forest" ? Argh, too much matrix, quaternion and vector interpolation in my poor little brain, cant remind me of certain member names
|
|
|
Logged
|
|
|
|
Sockratease
|
|
« Reply #10 on: June 02, 2010, 12:33:06 AM » |
|
...Thanks to all for pointing out that something "iffy" happened Being a moderator I restrained myself from mentioning it And I was so busy being Sarcastic about it - I forgot to congratulate the winners! Well done, Everybody
|
|
|
Logged
|
Life is complex - It has real and imaginary components. The All New Fractal Forums is now in Public Beta Testing! Visit FractalForums.org and check it out!
|
|
|
Power 8
|
|
« Reply #11 on: June 02, 2010, 01:12:13 AM » |
|
Yes congratulations to winners. My personal fave was the Fractal Fun category with Jesse's Mountain Monastery and Bib's Atomic Ruins so results are mixed for me... But win or not this selection is certainly one of the most interesting fractal image collections on the web today. Videos and Mbrots showed lots of great stuff too. So bravo to all participants!
|
|
|
Logged
|
|
|
|
Tglad
Fractal Molossus
Posts: 703
|
|
« Reply #12 on: June 02, 2010, 01:12:23 AM » |
|
Congratulations to Jesse, LhoghoNurbs and JAM they are all fantastic entries. Remember, tactical voting just before the end is just as valid as tactical voting at any time in the contest. Everyone has 1 vote so the day that they choose to use it is up to them, the above winners are fully deserved winners. Trifox, some of those ideas sound awesome for next year, particularly a 3d model of the winning 3d entry, good luck with getting sponsorship! As far as next year goes, I would say that the voting system to use depends on who people think should win, Bill, Bob or Britney? Bill's image has 3 five stars, 4 one stars Bob's image has 2 five stars, 2 four stars Britney's image has 1 five star ( http://www.fractalforums.com/fractalforums-com-spring-2010-fractal-art-competition/fractalforums-com-spring-2010-competition-submission-period-has-ended/msg17377/#msg17377)
|
|
« Last Edit: June 02, 2010, 01:59:19 AM by Tglad »
|
Logged
|
|
|
|
Buddhi
|
|
« Reply #13 on: June 02, 2010, 09:16:16 AM » |
|
Hi,
I tried to analyse how is calculated final result score. In result table everything is without sense. Some arts have higher number of votes, more comments, higher totalrating but have lower result. Example (video competition):
3. JackOfTraDeZ Universe #6X 8 13 50 3.85 4. subblue The Formula 11 23 85 3.70
Is there somebody able to describe algorithm for calculating result?
|
|
|
Logged
|
|
|
|
Pauldelbrot
|
|
« Reply #14 on: June 02, 2010, 09:21:32 AM » |
|
Eh, simple, the third number divided by the second number is the fourth number.
|
|
|
Logged
|
|
|
|
|