My Info
Categories
- I suck at TF2 - like a lot. 14 hours ago
- Listening to college radio on Itunes... what happened to music... ugg this shit is terrible 2010/07/30
- Done 2010/07/21
- #ff33cc 2010/07/20
- New "The Guild" - yes, I actually watched it on bing because it wasn't on xbl yet 2010/07/20
Posts By Date
| M | T | W | T | F | S | S |
|---|---|---|---|---|---|---|
| « Aug | ||||||
| 1 | 2 | 3 | 4 | |||
| 5 | 6 | 7 | 8 | 9 | 10 | 11 |
| 12 | 13 | 14 | 15 | 16 | 17 | 18 |
| 19 | 20 | 21 | 22 | 23 | 24 | 25 |
| 26 | 27 | 28 | 29 | 30 | 31 | |
Cybercircuit main event scoring
The stated parameters are: We want 30% of the score to be based on a judging panel that will grade performance and costumes/hair.
I’m assuming that “30% of score” means a “maximum of 30% of the final score.” As in, if the band gets perfect scores from all the judges in all the categories, the boost to their final score will be equal to 30% of their final score.
There is only one way to make this mathematically valid:
We need: Final Score = Bandscore + Judges_Score
Thus each single point of a judges score should be equal to:
((Bandscore *30)/70)
——————————————————————————–
((#judges)*(#categories)*(max rating per judge per category))
So for example:
A band plays “White Wedding” and scores 1,400,200 points
Assuming we have 3 judges, and 2 categories (costume and performance) and a 10-point scale to judge each on, we plug the band’s score into the equation and figure out that each point the judge gives in each category is worth:
((1400200 *30)/70)/ 3*2*10
600085.714… / 60
= 10001.429 points.
If all 3 judges give a perfect 10 for both categories, the band racks up: 600085.714 points from the judges
Add this to the band score and you get 2000285.7143 total points.
Lets check our math. 30% of 2000285.7143 = 600 085.714 (which is exactly what the judges gave – woot! The math works!).
Whats wrong with this, though? It hugely depends on the actual in-game score (because we just want to add a “judges score” to the “band score”) A band with an amazing stage presence that scores very terribly on the song will get crap for points from the judges just to mathematically ensure that the score from the judges doesn’t exceed 30% of their final score. Then again, this ensures that the band that wins will both score very highly and have a relatively decent stage presence (two bands that score close to equal (within say, 10000k) will come down to the judges scores).
Lets break it down with an extreme example, and you decide if the “correct band” wins using this method.
We will use 3 judges, 2 categories, and a 10-point scale for each category.
Our 4 bands and their scores are:
Band A – Scores 1,500,000 points but doesn’t do much for costumes and performance. The judges give them a combined total of 20points (I guess we’re talkin’ that they just showed up in jeans and tshirts and maybe walked around a bit). Their total score would be: 1714285.714
Band B – Scores 1,200,000 points and does some for costumes and performances The judges give them a combined total of 30 points (so, they maybe have a couple of wigs on, generic rock t-shirts, and they moved around and interacted with each other a bit). Their total score would be: 1457142.857
Band C – Scores 900,000 points and does a fair amount of costume and performance work. The judges give them a combined total of 45 points (costume and moving really damaged the game score a bit). Their total would be: 1189285.714
Band D – Scores 500,000 and does a TON of costume and performance work (to the detriment of their score). Judges give them a combined total of 57 points. Their total would be: 703571.429
Band E – Scores 1,350,000 points (second place score overall) – but did a bunch of costume and performance work (not as much as D, but definitely more than any other band) – and the judges give them 50 points. Their total: 1832142.857
We see that what this (really) means, is that it ends up being a score competition with bonus points for costumes/performance that really don’t effect much unless the point totals between bands are fairly close (in this case, E wins by having the 2nd highest score and the 2nd highest performance/costume total). But that’s the way that anything is, especially when you have such a wide range in possible scores.
For reference, If a band can score over 2.5 million points odds are, they really don’t even need to bother dressing up – if they do, they can just stand there and get the score. Want to push it towards having the performance scores effect things more? We would have to do something like cutting the in-game score (to like ¾ of the in game counts towards the final score) and still basing the costume/performance off of the uncut in game score.
Notes about the Cybercircuit drum competition…
So in reference to: http://www.cybercircuit.net/2009/08/ion-drum-rocker-challenge-rob-prod1gyx.html
A discussion came up on twitter over what method will be used to score. Here’s my opinion on it… First, I took a look at the rundown of the known information about the way they want to score:
Playlist redux
My ipod playlist had way too many upbeat songs that wasn’t really reflecting my mood… so we cleaned it up a bit
jquery + canvas
Why hasn’t anyone written a good jquery plugin for interacting, manipulating, and animating canvas elements via jquery? I’ve found some very basic libraries, but nothing that approaches the point where they could be mainstream… project?
Rambling about books…
This is probably the most journal-like post I’ll ever publicly post… so err..
spoiler warning?
New playlist…because I *know* you care!
God knows I’m not stupid enough to *actually* blog. So I’ll let the new shuffle playlist do the talking…
Now this looks like fun…
Its nice that they changed the start date to be Saturday morning – it might give me time to do it. Looks like fun 
http://cssoff.com/
Holy Crap I need this on my ipod…
Found this on Digg earlier today…
Dude is a bit stuttery..but the info is right at least 
Tag your look and feel on a separate stylesheet, yo.
2 Liters of…what?
Heh, was in the deli over by work a few weeks ago and I’m not sure if this is too much or too little to pay…
not to mention I’m surprised as hell that my camera phone took an image big enough to read it
Invisible Myspace Trakzor
So, I’ll keep this brief – anyone who has trakzor probably knows I registered for the dumb thing and dropped it on my myspace just because I’m curious as to how it works (which is now obviously ip logging – I was testing with proxies yesterday, and they showed up today – though it may read your myspace cookie information which is just scary). Anyway, I don’t want to give them any free advertising and their graphics would look ugly on my myspace so all I had to do was modify their code a little bit and it works to not display the graphic, yet still track who looks @ your page…Heres the couple of steps to compile your code after the cut: