72hr Contest Improvements

tyler

aa
Sep 11, 2013
5,100
4,622
Even when we're playing tons of maps, I know we aren't playing that many maps in a day/week. But I think what the rest of YM says makes sense.

The next 72hr has to be a lot different. I like the idea of a lighter contest with guaranteed testing and no voting, except possibly for a few maps hand-picked by judges. I think YM is totally right about the delays messing with development being bad for all the reasons he listed.

No matter what it has to change. It can still be a contest, but I think we all know there's only a few maps in the running for first/second/third. I don't think there's a reason to pretend every map deserves the same amount of attention when many of the entries are the author's first map or similar.

Might be worth looking at Ludum Dare. I'm not familiar with their setup but they do this all the time and probably have it figured out.
 

worMatty

Repacking Evangelist
aa
Jul 22, 2014
1,259
1,000
Hey, folks.

Glad to see a lot of really good ideas being floated for improvement of the 72hr comp. I don't really have much to add as some of my ideas have already been covered, so I'll just +1 them.

My concerns:
  • Voting does seem like it's going to take a long time.
  • I am concerned that being able to see what other people have voted for can influence your vote.
  • The voting system requires us to simply list the maps we like, and doesn't ask us to consider individual aspects of a map, like quality of layout in tailoring to all classes, clarity of game objectives and direction, innovation, detailing and lighting and so on. Perhaps any future voting system should ask voters to give each aspect a score.
  • Maps by well-known community members might receive more play testing than maps by authors that regular server users are unfamiliar with.
  • There are 24 slots on the servers. Each map is tested twice. So each map in the competition will be played by around 48 voters.

I feel...
  • That the format of a competition is exciting. It generates interest in the event from community members and also external organisations. Winners can get recognition that can help improve their profile and increase attention of their work.
  • Mappers that submit a map that has problematic logic, unpacked assets or an incorrectly-configured light_environment should be given a chance to fix it. We make allowances for logic prefabs to be used, so we could justify modifying logic later on. Such a small thing shouldn't stand in the way of a map that could provide some enjoyment.
  • That if you are having to work through one of 72hr periods, maybe you should book it off in future.

I agree that...
  • There should be multiple award tiers for things like best newcomer, most innovative game mode variation, most enjoyable layout, best use of theming, stuff like that. They could serve as runner up prizes for folks who don't stand a chance at getting a top three spot if those are dominated by experienced mappers.
  • If there are broken or poor maps that stand no chance of winning we should play test them so we can give the author feedback on how to improve them. I am concerned that new mappers will not get the encouragement they deserve.
  • Non-72 hour maps should not be forgotten. An imp list would really help because we would be able to see at a glance which maps need testing, without having to stay in the chat room a lot.
  • Server-based voting is a great idea. It's very convenient, and you can use data to decide which maps to put on the rotation so that more maps have a fair chance at being played. Such a system should be carefully considered, though, as people foreign with TF2maps' processes of feedback might vote heavily based on appearance.

Personally, ...
  • I am in it for the chance of getting a medal, as well as personal challenge, though I know I need a lot more experience and ability before I have a chance at getting the former. I do not prioritise any reason for entering over any other. They are a collection of positive factors that make up my eagerness to take part.
  • I am treating my spreadsheet like a bingo card. I want to play every submitted map before I make my votes. I'd happily play more than twelve maps in a week. Each map I see is interesting to me because I like to see what people have done, and try to understand why, and how it works, and see what plays nicely. I also want to help everyone improve their maps by providing the best feedback I can think of. I imagine people who have done this a few times before find the whole thing a bit tiring, which is perfectly normal.

Some things to consider:
  • Perhaps participants who have already won a previous 72hr competition should be disallowed from winning first place again, or their medal be given to the participant in the next place. Or, due to the growth in maps, perhaps the top three entrants could receive a medal.
  • Broken map logic can be fixed using the Stripper:Source Metamod plugin. It changes logic when the map is loaded in to the server, so a new BSP is not required.

Thanks, folks.
 

YM

LVL100 YM
aa
Dec 5, 2007
7,158
6,081
Might be worth looking at Ludum Dare. I'm not familiar with their setup but they do this all the time and probably have it figured out.

Ludum Dare's system wouldn't work for us because Ludum Dare's games can be played at your convenience, alone and require no group testing. Our 72hr maps require us to all get together in one place and time and play.

It's also a bit of a mismatch because nobody expects to continue those games to full products like we expect to continue our 72hr maps to completion. And if they DO want to continue working on them, there's no confusion over what's the Ludum Dare version and what's the continued version because the continued version isn't ever posted to Ludum Dare, it's posted to separate communities of people for their feedback. But here.... we can't do that. Everything is jumbled in together, the same people would play the contest version and the updated version and they'd play them on the same server in the same kind of events.


....


I didn't realise ido only scheduled 3 days per week for 72hr testing during phase 1. It feels like that's a million miles away from what we did last time and a step in the wrong direction. I remember big sessions last time where we did two rounds then moved on and did 15+ maps in a single night multiple times. But now we seem to be running no maps, ever. I haven't even avoided the tests and I've only noticed two so far. The contest ended over a week ago, by this time last contest I'd played dozens of maps.
 

UKCS-Alias

Mann vs Machine... or... Mapper vs Meta?
aa
Sep 8, 2008
1,264
817
  • Perhaps participants who have already won a previous 72hr competition should be disallowed from winning first place again, or their medal be given to the participant in the next place. Or, due to the growth in maps, perhaps the top three entrants could receive a medal.
.
There is still a problem. Its very hard to judge how many medals should be given out. As its not just down to the top maps, but also to the overal quality. With 10 or 50 maps you still can get only 3 decent maps in total. It might just be that the other 40 maps are just from beginners that in the 10 map situation would not take part.
If however there would be 10 quality maps that are nearly getting equal to major contest quality then its definitely worth to consider.

Passing on the medal to the first one that doesnt have one will also only work when there is a huge list. With only 10 maps it will happen too fast that someone who barely managed to finish his map suddenly would get the medal. And to me that would actualy be counter productive to the status of the medal.

The contest part to me is for a part a good one because there are not enough major contests to allow everyone that is capable of getting a medal to realy achieve one. But with too many maps it just gets out of control.
But i personaly dont see a problem to restricting maps that get a chance. A contest allways will show someone who is better, thats something a contest cant exclude.

But maybe some reward system for taking part into the contest regardless of result could be interesting.

What if for example you would reward people that manage to get a fully working map in the contest? It could be as simple as a badge on the forum stating that you took part in X 72h contests. No rank on it at all. It just says a number. So even if you were last position for 10x, you would see that you attempted one 10x. And someone who got first twice but only did 3 of them would only show 3x attempted.
For beginners it will still show them they did take part which especialy for a beginner is good as it will make his name show up better than the one who never did take part into one (note that i never did a 72h contest and only did 1 major one and that means i wont get that badge). It gives a status to showing the effort you do to take part into them.

Alternatively you mark the maps in the workshop with an icon stating it was a contender in a contest. It could be a major contest, artpass contest, 72h contest. or whatever. Even if they were bad maps early on and only became a good map 6 months later, that badge can make the map get just that extra bit of attention.
 

YM

LVL100 YM
aa
Dec 5, 2007
7,158
6,081
Playing the maps today.

Even one of the best ones we played today has massive problems. I'd vote yes for phase 1, but the massive problems could be seen and addressed after just one playtest, but it can't be updated until sept 6th at the very earliest and if it gets through to stage 2 (because it is one of the better maps) we'll be playing the same massively flawed version until early 2017.

THIS IS SO FUCKING DUMB

Why are we putting such flawed maps that haven't had any testing through a rigorous voting and assessment regime? No one is benefitting from it. We should be testing them once, or twice max, then letting the authors get on with completing their maps
 

Fruity Snacks

Creator of blackholes & memes. Destroyer of forums
aa
Sep 5, 2010
6,394
5,573
Why are we putting such flawed maps that haven't had any testing through a rigorous voting and assessment regime? No one is benefitting from it. We should be testing them once, or twice max, then letting the authors get on with completing their maps

You've made your point time and time again about how you don't like the current way voting is going, saying it's dumb (even with previous reasoning) doesn't really help anyone. It's annoying. Honestly, did you or anyone expect 60+ entries into the 72hr contest? If you did, you should've spoken up so we could come up with something better.

Why are we putting the maps through the testing? Because thats what we decided to do on the crunch time that was there. We get it, you want people to be able to work on their 72hr map and you don't like that they can't. You would rather have changed the voting. This thread about finding a solution to the voting system for future contests, not this current one. Posting that you don't like it every time the through crosses your mind is energy wasted. Channel that energy into DOING something about it, not just talking about it.

This is currently how this contest is being ran right now. (this is to ANYONE) Don't like it because you want to work on your map? You're welcome to drop out of voting and openly develop your map.
 

killohurtz

Distinction in Applied Carving
aa
Feb 22, 2014
1,016
1,277
Playing the maps today.

Even one of the best ones we played today has massive problems. I'd vote yes for phase 1, but the massive problems could be seen and addressed after just one playtest, but it can't be updated until sept 6th at the very earliest and if it gets through to stage 2 (because it is one of the better maps) we'll be playing the same massively flawed version until early 2017.

THIS IS SO FUCKING DUMB

Why are we putting such flawed maps that haven't had any testing through a rigorous voting and assessment regime? No one is benefitting from it. We should be testing them once, or twice max, then letting the authors get on with completing their maps

I think you're forgetting the point of the contest now. This is to see what you can make in 72 hours, and allowing entries that are still in the running to be updated defeats the whole purpose of the challenge. Yes, the length of the voting period sucks and I wish I could get on with development as well, but we should sooner streamline voting than allow for updates.
 

YM

LVL100 YM
aa
Dec 5, 2007
7,158
6,081
I think you're forgetting the point of the contest now. This is to see what you can make in 72 hours, and allowing entries that are still in the running to be updated defeats the whole purpose of the challenge. Yes, the length of the voting period sucks and I wish I could get on with development as well, but we should sooner streamline voting than allow for updates.

Nah, we had the contest initially to fill a small gap, not to make it a two month long event. It was a quick fun thing and now it's a quick fun thing with two months of unhelpfulness following it.

@frozen:
I said all these points last time after the ~40 entries, everyone shrugged it off.
Plus energy put in that might convince people we're doing things wrong this time, is not wasted energy. Geit's suggested method is what we should do next time, hell I think we should switch to it for this contest if we can. We could prevent another 6 weeks of this slog and reduce it just 2-3 more weeks!
 

Turbo Lover

Fight me under Glasgow Central Station
aa
Feb 15, 2011
333
344
Geit's suggested method is what we should do next time, hell I think we should switch to it for this contest if we can. We could prevent another 6 weeks of this slog and reduce it just 2-3 more weeks!

I really don't think it's fair on participants to change the rules of the contest this late in the running.

I'm not sure Geit's method for testing maps will ensure that each map gets played equally or even with a full server of players, but we should only be considering it as a system to improve future 72hr events.
 

YM

LVL100 YM
aa
Dec 5, 2007
7,158
6,081
I really don't think it's fair on participants to change the rules of the contest this late in the running.
Has the process for phase 2 been announced yet?

I'm not sure Geit's method for testing maps will ensure that each map gets played equally or even with a full server of players, but we should only be considering it as a system to improve future 72hr events.
Why should they all be played equally? If we're doing guaranteed testing, then yeah, sure, but if we're doing a big old public voting thing, then the maps that are best should receive most time so that we can put our effort into choosing the top X maps with most accuracy.

Buuuuttt Geit's system involved a plugin that removed maps from rotation when they're being played more than others. So the very nature of his system prevented maps from receiving too little testing time/feedback.
 

Idolon

they/them
aa
Feb 7, 2008
2,128
6,144
Buuuuttt Geit's system involved a plugin that removed maps from rotation when they're being played more than others. So the very nature of his system prevented maps from receiving too little testing time/feedback.

That system was introduced because there were a number of maps that just never got played at all. Nobody knew if they were good or bad, because they were too busy playing the maps that they already knew were good.
 

Turbo Lover

Fight me under Glasgow Central Station
aa
Feb 15, 2011
333
344
You know the Winter 72hr contest isn't that far away, and we're going to have at least one other contest happening in close proximity to it. Perhaps we should cancel or postpone the Winter 72hr until we have executed a plan to address the issues that have been raised in this thread.
 
Last edited:

UKCS-Alias

Mann vs Machine... or... Mapper vs Meta?
aa
Sep 8, 2008
1,264
817
Rather than having a full map, just have a 2k-2k-2k detail contest instead. They are much easier to judge. Sometimes people cant design complete maps but are still very good at detailing. We shouldnt forget these people.
 

tyler

aa
Sep 11, 2013
5,100
4,622
I'm not sure there needs to be a contest between the next one. Let people have some time to work on their stuff imo. If there is a contest it should be as tiny as possible.
 

Vincent

&#128296 Grandmaster Lizard Wizard Jedi &#128296
aa
Sep 5, 2009
912
684
I really enjoy starting and finishing an entire map. If it was a detail contest I wouldnt enter.

EDIT: To clarify, if the 72hr contest was replaced with a detail contest I wouldn't enter. I'm all for having a miniature detail contest for others, as long as it doesn't impede anything major.
 
Last edited:

YM

LVL100 YM
aa
Dec 5, 2007
7,158
6,081
We absolutely cannot replace the 72hr with a detail contest. But perhaps 72hr should be rebranded as a major contest so we can start doing smaller detail contests in addition. The 1024^3 unit area is enough that it can be done quickly and the most important thing is that voting can be done quickly and offline. We can do it when it suits us.

Introducing smaller contests doesn't reduce the scale issues of the 72hr though.

@ido I played every single map on our servers with at least 12 people on. Your claim that there were unplayed maps isn't entirely accurate. It would be more fair to say the good maps were circlejerked to getting more playtime. If the server configuration is good, we can avoid that entirely!

I really think the current system could be compressed. 3 days a week isn't enough, it's taking too long. Can you shuffle up the schedule a bit so we're doing 4 per week? I think we could manage 5 tbh
 

Vincent

&#128296 Grandmaster Lizard Wizard Jedi &#128296
aa
Sep 5, 2009
912
684
I really think the current system could be compressed. 3 days a week isn't enough, it's taking too long. Can you shuffle up the schedule a bit so we're doing 4 per week? I think we could manage 5 tbh

I agree that 3 days feels a bit slow. We just need people to run them. Or is that also not an issue?