sweet idea. if you guys wanna see that, click on any game on the results page, and you’ll notice the judge’s names are now hyperlinks 
Wow. I never expected to end up that high - thanks! 
My personal favorites were Miners, Fuzetzu, Goomba, Roll, Xero and Ares (in no particular order). I do think it’s a pity that Ares got lowered that way; a trimmed median seems to be a good idea. Also, I suggest that next year, we encourage the judges to use the full scale (one of the judges gave 80 for games he didn’t like, while Mahlokan used the scale better, giving 70 for games he liked but not having as his favorites. A couple of the judges hardly gave any game less than 50).
Great work, everyone. And big thanks to the judges!
[quote=“Myself,post:12,topic:26524”]
No, there’s no need to change the scoring. I’m very happy with the results of the judging this year, especially in comparison to last year. While I really hate losing on a technicality, it’s simply par for the course in any competition. In this case I took a risk (full screen) that I felt was worth taking in order to make a better game. I ended up getting burned by it. Using sports games as a comparison, referees often swing the tide of the game by showing bias in making penalty calls. Such is life, and that’s just one of the hurdles that competitors must overcome to win. 
Granted, I would have loved to see how Miners4K and Xero squared off in the end. Sadly, the world may never know. Them’s the breaks.
A much better solution for next year is to give the judges guidance on how to handle situations like the game not running. For example, the JSquares entry never should have been rated without the judge having the opportunity to run it. The game should have been put aside until the judge was done with other games, then discussed with the moderator (Woogley in this case) to see if anything can be done to resolve the issue. What happened may have been completely secondary to the game (extra components?), and it was unfair to the competitor not to take a second look.
[quote=“Markus_Persson,post:19,topic:26524”]
From what I can tell, the median score already is the final score. The votes from each judge are tallied to a total, then averaged based on the number of judges. So the scoring is overall pretty fair. Extremes happen, and they show how much opinions differ.
BTW, Markus. Congratulations on winning!!! I really enjoyed the competition this year (including our friendly arguments ;)) , and Miners4K was more than a worthy competitor. Excellent job! I’m really looking forward to next year.
From what I can tell, the median score already is the final score.
Median is taking the score in the middle and not the average. Say there are scores: 20, 30 and 47795. Then the median would be 30. In the case of an even number of scores you take the two in the middle and average em. Eg 10, 50, 60 and 100 would be 55.
edit: “Eg 10, 50, 60 and 100 would be 55.”
Holy crap example, batman! :o
Ok… better one… 5, 10, 20 and 100 would be 15.
[quote=“jbanes,post:23,topic:26524”]
Thanks, and agreed.
The j4k competition is much more fun than it should be. I’m sure the kick-ass competition and good spirits on this board is a big part of that.
For some reason, I suck very very bad at all games, though.
Including miners4k… Back when the timelimits in miners were a lot harder, people were emailing me screenshots of the bonus level… I have no idea how they got there, heh.
[quote=“oNyx,post:24,topic:26524”]
Ah, my bad. Thanks for the correction.
EDIT:
[quote=“Markus_Persson,post:25,topic:26524”]
Indeed. I’ve noticed that hard-core gamers almost always surprise the programmers on how well they can do at a game. This was something I actually counted on in Xero. Just looking at the high scores in Xero is enough to show how amazing the players can be. I was expecting high scores to be around 75,000 or so. But the scoreboard shows scores well over 100,000 points! :o
Thanks for organizing this great little compo !
I had lotsa fun (both playing and participating
I’ll try hard to do better next year !!
[quote=“ctrl-alt-dev,post:27,topic:26524”]
Hi ctrl-alt-dev! Welcome! I’m glad to hear that you enjoyed the competition! Were you the author of Daleks4K?
Noooh. I ended up at #36 with Jyruss4k, which is kinda nice for something i decided ‘just to have a go at’
Just on a lark, I did a little statistical analysis.
I want to preface this by saying that the three winners definitely deserved to win, and I’m not recommending any changes in the results at all!
But this might be useful info for deciding on a judging system in the future.
If you remove the highest and lowest scores (the “trimmed mean”), the top three scores are: Miners4K, Roll 4K, and Xero.
Interestingly, the trimmed mean results (on average) in a 1.21 higher score for entries. Although there was a wide range. The best gain was +12.89 for Xero, and the biggest loss was -2.71 for Daleks4K. It’s interesting that the change between mean and trimmed mean was usually positive: that means that there were more extreme low scores than there were extreme high scores.
If you use the median instead of the average (or mean), the top three scores are: Miners4K, Xero, and Goomba4K.
The three biggest standard deviations (on untrimmed scores) belonged to: Xero, Kana Invaders, and Daleks4K. These were games whose scores had some wide variations. The three smallest standard deviations (again, on untrimmed scores) belonged to: Goomba4K, Roll 4K, and Gandalf 4K. These were games with very consistent scoring.
Are you kidding me? That game rocks! You did some really great work there, even if it was a smidge jerky on a few people’s machines. I hope we see you enter again next year!
Well, my game honestly ended up pretty much where I expected it to; I’d hoped to put together another few entries but didn’t quite have the time.
Regardless, congratulations to the winners!
- HC
Oh well can’t win them all. I really thought SpyHunter4k would have rated higher than it did. Of course, Dave and I approached it from the wrong viewpoint–how much can we squeeze into 4k. We were trying to be as close to the arcade version as we could, same screen size, same sprites and didn’t focus on playability at all. It wasn’t until after we sumbitted that we realized that if we had quatered the graphics we could have added more of the original game play.
In creating the game we came up with three new techniques of byte saving that none of the other games use. For that alone we are proud of what we submitted.
I just wish more people would have recognized the technical achievements for what they were.
Hopefully next year we can break into the top 10.
It was lots of fun guys! Congrats to the winners!
chris
[quote=“cajurabi,post:33,topic:26524”]
I have to say, if there’s one take-away lesson I’ve gotten from the 4K competition over the years, it’s this: It’s not about the technology.
The technology is merely a means to an end. If it helps you make a better game, then it’s worth investing in. However, if your game becomes nothing more than the technology, it will fail.
The reason why I find this to be such an important lesson, is that it also holds true in the commercial game market. There are tons of games that use technologically impressive engines. Yet, unless they back it up with a complete game that jumps out and grabs you for hours at a time, they will fail to make inroads into the market. Just look at some of the all-time classic games for an example of this. Street Fighter can’t hold a candle to Soul Caliber’s technology, but which game do you think players remember more fondly? If you place Kameo: Elements of Power and Super Mario Bros. in front of someone, how many will chose to play Super Mario Bros.?
When it comes down to it, it’s just so amazingly important to realize how little the technology actually matters to how people perceive a game. It may be frustrating to us technologists, but it’s just the nature of the beast. Not all is lost, though. Feel free to discuss your improvements that made SpyHunter4K possible. I think you’ll find that accolades from your fellow programmers are some of the highest honors you can get for your technological solutions.
EDIT: BTW, aren’t there a few games missing from the final lineup? I could have sworn that there was a 3D Poker Game in the competition.
Yup, Poker4k was mine, and was Java 1.5, so I didn’t even submit it as it would be disqualified anyway.
Without System.nanoTime() the animation was very jerky so I’ll probably wait for next year, and stuff a whole lot of gameplay in
By the way- Woogley, you said in the Judging thread that there were a number of ‘techie’ judges and a number of ‘pure gamer’ judges. Out of curiosity, who was who?
- HC
peggy - gamer
shelton - gamer
tim - gamer
borkert - gamer/tech
malokhan - tech
kingaschi - tech
nonnus29 - tech
wow! Congratulations to Markus, Kev and Woogley, i was sure that miners4k would be the winner this time congrats
This edition has been great, both imagination and tech used has been awesome and i’m sure that next year’s competition will be even better ;D
Thanks to all the judges, a 7th place is far more than i have expected for my game, thanks ;D
Interesting results!
I noticed that the bulk of the games ended up very near a score of 80… with very little difference in the scores.
I placed about were I figured by rank, better than I expected by score out of 100 though.
The judges comments are very interesting, really shows the differences in taste. I am honestly quite surprized that my Kung Fu game score lower than my Bungie Bill game. I thought the kung fu game was way more playable and having single and two player modes might have given it an extra point or two. Just wish I had the time to fill up those extra bytes. Blocking and high/low hits were planned but I just didn’t have the time to deal with it. And yes, my sctick figure graphics were even bad for stick figures
Overall I think my games got a fair rating, but next time I see a game review with a score of 8 out of 10 I know that the game is ‘average’ at best
I really enjoyed this competition. I agree with jbanes, the gameplay is more important than the tech (I spent too many bytes on tech and not enough on gameplay methinks).
I noticed that different judges gave different spreads in their scoring. Some folks seldom (if ever) gave a score below 60, while others gave scores over the full range available. This does tend to increase the score weighting of judges who use the whole range available (as was seen in the cases with a close to zero score). Perhaps next year, we could have a scoring guide for judges, to try to make the scoring spread more consistant. Or maybe just rate the games in order (a bit like the Eurovision song contest). Or perhaps, continue as now, but normalise each judges scores (takes some maths) before summing them up. What do people think? Maybe we could work out a generic scoring system in advance of the next competition.
Finally, I’d like to thank all the judges. I really appreciated some of the detailed comments on my entries. Hopefully these will help me make better entries next year
Alan