Preparing for Custom Mapping Events/Gamedays

[GrimGriz]

I thought I'd post what I plan on doing/how I best think to do setup on a server for special events. I'm hoping any server administrators that know how to do it better, or people that might have ideas on how to do it better post here also. Maybe we can end up with a helpful sticky in the end.

Control Conditions of Map Change

• Set conditions of map change in tf/cfg/server.cfg
  • mp_timelimit
  • mp_roundlimit 0 (thanks Apom)
• Set conditions of map change in tf/cfg/sourcemod/mapchooser.cfg
  • sm_mapvote_endvote "0" // Specifies if MapChooser should run an end of map vote
• Set conditions of map change in tf/cfg/sourcemod/plugin.playervotes.cfg
  • sm_votemap_minimum "-1" // minimum votes required for successful votemap. -1 to disable voting
    
• Disable rockthevote
  • Move rockthevote.smx from tf/addons/sourcemod/plugins to tf/addons/sourcemod/plugins/disabled.

Reserve Seating
How much would it suck if you couldn't play your own map for the event?
- Add the map authors as "admin" with the reserve seating flag.

• If you are using admins_simple.ini:
  • Get the steam ID of the author (they can find this by typing the word "status" without quotes in console. It should look something like this: STEAM_1:0:1386928
  • Give them the "a" flag for a reserved slot by adding a line similar to this in your admins_simple.ini file:
    

    "STEAM_1:0:1386928"      "a" 	// GrimGriz  - battleofthegods_a602

  • - set up reserved slots
• If you are using admins.cfg:
  • Get the steam ID of the author (they can find this by typing the word "status" without quotes in console. It should look something like this: STEAM_1:0:1386928
  • Give them an entry in your admins.cfg file with the "a" flag for a reserved slot, something like:
    

    Admins
    {
    	"GrimGriz" //battleofthegods_a602
    	{
    		"auth"			"steam"
    		"identity"		"STEAM_1:0:1386928"
    		"group"			"Full Admins"
    	}
    }

- Set up reserved slots.

• Make sure reservedslot.smx is in your tf/addons/sourcemod/plugins directory and not your tf/addons/sourcemod/plugins/disabled directory.
• Set reserved slot values in your sourcemod.cfg file, found in the /tf/cfg/sourcemod directory.
  • sm_reserve_type 1 (my understanding indicates type 2 is the best, but you have to have sourcemod 1.1 and I'm not sure how to find out if I do...
  • sm_reserved_slots 1 (this would have to be higher than 1 (one for each featured author) if using sm_reserve_type 9 (i think...)

Set the Mapcycle
Replace the mapcycle.txt file in the tf directory of the server with a mapcycle.txt file containing the list of maps in the order you want them played.

Get the show on the road
Restart the server and cross your fingers!

 

[Zeewier]

For the reserved slots, can they be selected by map? Because giving all gameday-mappers a reserved slot doesnt seem right :/

 

[Apom]

mp_roundlimit (at present, I'm operating under the belief that removing the roundlimit will cause timelimit to be the sole determination of map change, please correct me if I'm wrong).

Convars that are not specified by hand will use default values, I don't know what the default value for this one is (2? 3?). What you want is setting it to 0.

 

[tok_vaal]

For the reserved slots, can they be selected by map? Because giving all gameday-mappers a reserved slot doesnt seem right :/

I'm not sure about sourcemod, but I'm pretty sure that with beetlemod you can add the admin-permissions statement to a map-specific config file, and it would unload the setting when the next map loaded (giving you map-specific admins).

But, to tell you the truth, I think giving ALL the mappers a reserved slot is ideal... after all, who are these gamedays for?

It seems like it wouldn't really matter that much if some random public player can't get in (or was booted by res slot), but it would be very frustrating for the mapper, if they could not get in to play their own map. And wouldn't it suck if you (as the mapper) got kicked for not having a reserved slot when an admin joined a full server?

Besides who better to give you feedback on your map than your fellow TF2M members? Obviously everyone in this forum needs a reserved slot...

 

[VelvetFistIronGlove]

But, to tell you the truth, I think giving ALL the mappers a reserved slot is ideal... after all, who are these gamedays for?

Not that the map author needs to be around. In fact the playtest might well be better if the map author isn't present, as his preconceived ideas about how the map will play out will not affect the test. If the author is playing on the map as well, a lot of the effectiveness of playtesting (as opposed to feedback about mere appearance and detail) is lost.

 

[GrimGriz]

I agree...and if people can't talk (and what they say be forgotten and such) there just MIGHT be more WIP-thread feedback...

 

[StoneFrog]

Not that the map author needs to be around. In fact the playtest might well be better if the map author isn't present, as his preconceived ideas about how the map will play out will not affect the test. If the author is playing on the map as well, a lot of the effectiveness of playtesting (as opposed to feedback about mere appearance and detail) is lost.

I think the author should just spectate. Indeed, there's no point to playing your own map if your whole reason for uploading it was to see how others maneuver through it.

 

[MrAlBobo]

Not that the map author needs to be around. In fact the playtest might well be better if the map author isn't present, as his preconceived ideas about how the map will play out will not affect the test. If the author is playing on the map as well, a lot of the effectiveness of playtesting (as opposed to feedback about mere appearance and detail) is lost.

I disagree...in a game of 24 people 1 persons preconceived ideas will never effect the test unless he begins to direct the teams. If the author is not giving advice then all he is is a knowledgeable player, and that means little after a few rounds.

I happen to think the author should play, and form his own opinions on how it went, then use a sourcetv recording to see how it actually played out.

 

[MangyCarface]

I've always held we should just do the 2 round limit per map; 2 rounds is fair and is really all you need in individual playtest. Further iterations with the same sample player base will only stagnate as players become bored or ingrained in their roles.

Of course, players' knowledge of the map is a desired trait for testing, but really would occur over the map's iterations. Also, we used to do this thing where a couple moderators/players would hop on the server the night before and just run through the maps to check them out...

 

[kwagner]

Mappers being on the map during the test probaby taints the results. On the other hand, the live comments during play can provide information that latter comments can't. And seeing how people learn to play the map can be as important as the final results. So there IS value to being there.

Valves heat maps are a good tool, but I suspect it takes more than 30 minutes play to get telling results? It would be interesting to see a map that showed a line for the players paths, to see what areas get traffic. And that would probably show results sooner than a heat map.

Finally, is there some way to record the game play itself? I've read of SourceTV. Is that just for live spectating, or is it recorded? If it could be recorded, being able to replay the test would be great, as I'm sure much gets missed during a live test.

 

[ChronoTriggerFan]

Also, we used to do this thing where a couple moderators/players would hop on the server the night before and just run through the maps to check them out...

Yeah, we did each map for 5-10 minutes, just to make sure everything was working, and so that game day didn't explode when it was really happening.

 

[TheDarkerSideofYourShadow]

I've always held we should just do the 2 round limit per map; 2 rounds is fair and is really all you need in individual playtest. Further iterations with the same sample player base will only stagnate as players become bored or ingrained in their roles.

Well, that also depends on how things are going. For example, today we played 4 rounds on ghost town. The results were such: Loss - CP 1, Win, Win, Loss - CP 1. Thats interesting because a team that initially failed pretty badly ended up winning, while a team that adopted a rush strategy and won failed during their second attempt. As people came to know the map better, what worked and didn't worked changed, and it is something that I wish I had a demo of so I could see what changed, for both teams, during those 4 rounds.


As far as the map maker being present: If they are directing the action, for example telling people to jump out the window (guilty of that), because players are not naturally doing that on their own, well then they should realize something is wrong and that a change needs to be made.

 

[kwagner]

As far as the map maker being present: If they are directing the action, for example telling people to jump out the window (guilty of that), because players are not naturally doing that on their own, well then they should realize something is wrong and that a change needs to be made.

Quoted for truth. Part of my map wasn't nearly as intuitive as I thought. (Ok, it wasn't intuitive AT ALL. :blushing: )

Perhaps a practice could be put in place where each map gets 10 minutes/2 rounds, at which point the player/testers can vote whether to continue it for another 10/2, or to advance to the next map. For examples: 10 minutes would have been plenty to reveal the problems with my map. But for better, more developed maps (like Yukon which I enjoyed greatly) 10 minutes wouldn't be nearly enough.

 

[A Boojum Snark]

The problem I see with round based or dynamic timelimits is it throws the idea of a schedule out the window and make it so an author won't be able to ask for a certain time that he can be there. Granted some of you think the author doesn't need to or shouldn't be there, but it would suck if you wanted to come and didn't know what your map would run.

 

[Sgt Frag]

I agree that mapper being present and keeping to a schedule on gameday is important.

But I think this thread is drifting. I think Grim's server testing is a bit different. He's running complete maps for lengths of time's on a one-by-one basis. Not quite the same as tf2maps gameday.

I really don't see any reason NOT to have an author present if possible. Yeah sometimes we might get cranky after the same thing is brought up ten times
But still there is nothing like seeing how other people react/use the map.
I often get served abnormally high in my own maps because I am trying to keep an eye on everything. I'll usually go spectate for awhile to to get a birds eye view too.

 

[MrAlBobo]

yeah, what grim is doing is actually something I suggested the site do awhile back, and I happen to think both methods of testing are very useful. One gives you a quick look, one lets you find out smaller details of gameplay.
As for sourcetv...Ill probably end up throwing some cash at grims server so he can get an extra slot for sourcetv recordings to help the mapper even more with the long playtests.

 

[kwagner]

What kind of cash are we talking about? Since I'd like to see sourceTV of the game days it seems only fair that I help fund it (at least a little.)

 

[GrimGriz]

But I think this thread is drifting. I think Grim's server testing is a bit different. He's running complete maps for lengths of time's on a one-by-one basis. Not quite the same as tf2maps gameday.

I actually wrote this for one of my rate-off events (not what I'm doing now)...but I posted it here and didn't just keep it to myself because the little pre-event steps listed would have alleviated MANY of the problems that showed up at gamedays over the last few months.