Thank you again to all the sponsors, judges, and participants who make this an aswesome competition. The winners are:

1st Ferdinand "cityy" List with 'Fluorescent'
2nd Tom "Phantazm11" Perryman with 'Windsong Keep'
3rd Russell "bst" Vint with 'Terminatria'
4th Patrick "kaffeewunder" horn with 'knc_dm3'
5th Justin "StormShadow" Ingels with 'Satanic'
6th David "Noruen" Adámek with 'Penumbra'
7th Pawel "ShadoW" Chrapka with 'Sanctum'
8th Peter "fKd" Ward with 'Server Overload'
9th Ricardo "EmeraldTiger" Stevenson with 'Tactical Assault'
10th DaEngineer and Bliccer with 'Roman Duel'
11th Fubsy with 'Fbtdm1'
12th Anthony "Kaustic" Tullo with 'Damed Be Thy Name'
13th Andy "dONKEY" Dunn with 'dk_ng3'
14th Hendrik "InsaneKid" Berger with 'Ebirah'

Again, thank you all for participating. This was great fun.

Prize distribution:
-The top 10 won a copy of Steel Storm from Kot-In-Action (http://www.kot-in-action.com/blog/?p=677)
-1st, 2nd, and 3rd place get a web and email hosting package from Lunar Module (http://www.lunarmodule.co.uk/)
-id Software has given 4th and 5th place Premium QuakeLive subscriptions and the top 3 Pro subscriptions (http://www.quakelive.com)
-3rd place gets a year long subscription to http://www.cgtextures.com
-1st place get to pick two items from the following list with 2nd place getting the third item:
Modo 501 Rendering Software from Luxology (http://www.luxology.com)
Sapphire HD 6870 FleX graphics card with Eyefinity (http://www.sapphiretech.com)
Intuos4 Medium pen table from Wacom (http://wacom.com/)

For those who won a prize I'll be sending you an email with the info as soon as I can. Note if you have changed your email and you won a prize please notify me as soon as possible at campagnolo90@hotmail.com.

Oh, and don't forget to leave feedback (even if you weren't in the competition):
viewtopic.php?f=14&t=16950

Good job, everyone.

Congrats winners
Lol, seconded most crap map...guess that what lack of time does....ho hum, guess I better get on with the map I should have made ))

The maps seem to get better every competition. Good job to the participants and Maverick, and well done to the top 3!

i was checkin the scores, again:
City: 524
Phantazm: 526,5!

True - I do not fully understand why you used this kind of calculation Maverick. I think it is not applicable here.

Counting of partial scores is more relevant here, because each judging of judge is de facto different measure method and no judge is absolutely objective. They are judging on their feelings and preferences and thus SD is not applicable in this case.

EDIT: Also I see in my Excel, that in "average" and "standard deviation" cells there are not last two scores of maps "Q3shw27" and "fbdm1_a" included in computation.

I don't really understand why SD is used either. I suppose its a way of standardising the judges scores though? But I don't know why the final scores added up isn't good enough. I don't know enough to agree or disagree tbh, but I would be interested to learn why its scored the way it is.

There were also some others btw:

Storm3dm6.................465.5...5th
nor3dm3_final..............466.5...6th
aty3dm8MSC2011v.......367...12th
dk_ng3.......................373...13th

Calculating by raw score alone means that each point from each judge is equal, when in fact it is not. Because each judge is different and their average score is not the same and because there are no grading standards setup a normalization of points is needed. In the Olympics, for instance, every judge knows the exact deduction each competitor gets when a flaw is detected in their form or performance so their resultant scores are very standard and relatively predictable. In this competition, however, average scores range from 85 to 69 which is a huge amount, and every judge values things differently. Each judges bell curve is wildly different than the next (standard deviations ranged from 6 to 16).

Using standard deviations to judge equalizes the point value from each judge. Also, because each judge has a different average score and judges based on different standards using standard deviations avoids the situation where a judge with a low average can be completely overruled by a judge with a high average and gives lots of points (imagine a situation where a judge gives a maximum of 10 points to each map and combine their scores with another judge who gives a very high average. The judge giving fewer points will essentially be overridden).

I apparently uploaded the wrong version of the judging excel sheet. I'll go ahead and update the original attachment which should have the formulas right.

But it is not this case, you can see that even after normalization places changed only in some cases. The fact you want to normalize their judging brings an artificiality into results (as we can see). SD from 6 to 16 means range of around 20 points between all judges from total 100 points per judge and it is normal range. It is a question why to try bring objectiveness into personal feelings, but Occam's razor is a good law.

Thanks mav I think I understand it now. Its a way of normalising the value of points, to unify the judges different style of grading.

You could look at it in an anology between a homeless person and a rich person judging houses. The homeless person would always rate higher than the rich person because any house is better than none. So when you come to the end of judging 14 houses, the homeless persons average score would be higher than the rich persons, but only because of a bias that shouldn't have anything to do with the judging in the first place. To get to the real nitty-gritty of the true ranking, to make it fair, you need to eliminate those kinds of things.

The amounts/differences may be small, but in the end I think its a fairer way to do it. I think its really nice that the effort has been put in

more feedback would have been helpful. the only reason i ended up entering was to get some feedback. but there really was none, (numbers mean nothing) kinda disappointing. at least a paragraph per map would have been nice. so we could hear their opinions and what they were thinking with their scores. it really helps us grow.

still, lessons learned. comps are not for me. im too slow a mapper

also mave, can we get a solid date on when the maps need to be ready for the final mappack? im working as hard as possible to get something a bit more playable. so a solid date would be helpful. cheers and thank you

Since the final judging was a bit late let's just aim for this coming Saturday, September 3rd. I'll go ahead and post a sticky about this as well.

i think, that some judges
judge (or not)
how they feel ...
and without long-sight!