Public Discussion: Feedback Quality and Quality of tests

I have not really been keeping up with this topic but here's my 2 cents I guess

It's quite simple why the feedback is not what it used to be. The feedback plugin is cool and all but it fails in more than one way.

it's apparent to me that providing feedback WHILE playing is a detriment to both gameplay and feedback.

back in the days, it used to be that people would just play the map and leave feedback later on in map threads. playtime and feedback were 2 distinct things. While less immediate, this was a good way for people to take a step back and think about what they experienced during tests which ultimately led to more elaborate/thoughtful feedback. The way it's done now, people can just leave a feedback annotations and never have to worry again about their half baked ideas.

the other thing about feedback is that it's text based and unidirectional in nature. Actual discussion during playtime will be much more beneficial than having individual mappers !fb-ing in scattered areas.



Now for the good part!

what to do when feedback is shit:
don't wait on feedback to make your map great. if you want to make a great map, just make the map you want to play, make the map that will foster the gameplay scenarios that you find interesting, create new and exciting opportunities for players.

wild tangent: notice how no amount of feedback can give you any of this? feedback is grossly overrated. Now is a good time to question whether your craving for feedback is misplaced or not but I'm telling you, feedback does not develop great maps, individuals with goals do.

If you have any idea of what it is you would like your map to do, let that guide you. "how do I want this to play", "how do I want this to feel", "does this element of my map contribute to what it is I am looking to do" these are very easy questions that you can answer by yourself (and more importantly, that others can't answer for you, unless you are a terrible spineless mapper). strong goals, strong maps.

and that's that!

I just want to share this with all of you. I had to leave the server a few days ago because testing turned into a huge conga/"get into the skybox" party. Even though it wasn't my map I felt insulted. Someone wanted their map tested and instead we just degraded into "friendlies" and partying. Its a huge insult to the person who made the map and was expecting feedback, and the people who were playing it and intending to give feedback.

On a side note, it is not WRONG to have fun when playing, but you really should not be abusing a bug or exploit. If you find one, report it and move on.

And if you cannot think of anything good to comment on the map, then just play it casually for the entire duration of the test and reflect on it later. There is a map thread for a reason.

@Fr0z3n I agree certain taunts should be banned. No one likes it when the ENTIRE OPPOSITE TEAM IS CONGAING IN THE SPAWN ON YOUR MAP. Also, I don't own any of those taunts.

After seeing test yesterday with six rocket jumper soldiers just jumping around, contributing nothing, i would like to see Rocket Jumper and Sticky Jumper blacklisted off the server, instead of taunts.

Wilson said:

After seeing test yesterday with six rocket jumper soldiers just jumping around, contributing nothing, i would like to see Rocket Jumper and Sticky Jumper blacklisted off the server, instead of taunts.

Click to expand...

I agree with this...

Fr0Z3n said:

Since it was brought up, just spitballing this idea: What do people think about banning the Conga, maybe even some of the other perma-taunts (Rancho Relaxo, etc) from our servers during tests? (Don't shoot me for just mentioning it).

Click to expand...

Personally, if I saw someone congaing on my map in the middle of a test, I don't think I would be upset about it. I might ask myself, why are they doing this? The answer is probably that they're not having fun on my map, and I need to figure out why.

This isn't exclusive to congaing, of course - whenever people get bored of a map, they tend to go off and do something silly and probably unhelpful. They might equip a silly loadout, farm their stranges, go hit a rock with a pan, uber a sniper... No matter what you do, people are gonna find a way to fuck around if your map isn't fun, and that's frankly the mapper's responsibility to deal with.

Kicking people who are being a nuisance for an extended period of time is of course not something I oppose, but I do think it's unfair and ineffectual to put out a server wide ban on just one of the dozens of things people do when bored.

I think mappers always need to remember that they're making a map for a video game, and thus it shouldn't be your primary concern that all your testing data is perfectly stable and everything is meticulously laid out in accordance with this; your job is first and foremost to make an area in which people come to have fun, and fun comes in many flavors.

Alright, time to bring in my horribly wrong opinion I guess.

To me, the saddest thing I've seen is how much feedback the valve beta maps are getting. I'm not going to lie, it's made me loose a lot of motivation in mapping.

Clearly, the community is able to give high quality feedback already. They've been doing it in the pageloads in the valve beta map threads. So the question remains how do we get people to give this feedback to maps that actually need it (more).

If people already have the ability to give feedback, why aren't they doing it on community maps (much). I thought about the possible reasons, and what I thought of was things like, uncertainty as to if the map would be finished. If the maps not going to be worked on, what's the point in feedbacking it, right? This is a bit of a vicious self-fulfilling prophecy, but I think it's one of the reasons.

The other reason I could think of was simply apathy. People genuinely don't really seem to want to see maps go forward. Perhaps there's just too many junk maps that never make it past one or two alphas that people are just tired of bothering trying to promote maps that don't stand out on their own.

And of course, there's always the time that it takes to do a truly good feedback post. But this one seems null looking at the feedback posts in the valve beta threads...

I'm not sure on the solution for either, honestly. Whatever we try, it requires a major attitude change or a hell of an incentive, neither of which are easy if at all possible.

Half the problem is, as you say, people use tests as a casual play session. I think that's unavoidable. A lot of the maps we play are, not trying to offend anyone, frankly...bad. And of course they are, they're early alphas. That's why we're getting feedback. Except at the same time, it has the effect of making the map not fun, at least for me. And when a map isn't fun, people either leave...or they entertain themselves some other way, like taunting, playing stupid classes, etc.

I don't think getting rid of the feedback plugin will solve anything. The feedback plugin didn't replace forum feedback, I think it created a new kind of feedback. Things like clipping errors, visual bugs, stuff like that is where the feedback system is really useful because of it's coordinate tagging. I don't think it's caused people to actually forget about giving proper feedback, rather I think that proper feedback has just been a thing always slipping away as much of what I'd consider the "core" of TF2M leaves as it's members who are truly great mappers leave for bigger and better things.

I don't think we can restore that, TF2's too old now. It's community is larger than it ever has been, but it's a different type of community. It just hasn't got what drew lots of people to level design for it originally, and that's what time does.

People just don't use the tests anymore as tests. They seem to use it more as a way to satisfy their hunger for new custom maps and a place to play with friends. And it's certainly a great reason, and really the only place for that...but still.

On that note, there's another thing I'd like to discuss. I don't like how this community seems to be so focussed on alpha as the only development stage. It sounds weird, but I truly think this is a problem. People seem to stop caring about maps, and giving feedback (what little they do give) once a map reaches beta. I tried to test a rather large change to escarpment a few years ago, and people just voted past it because "this map is RC, surely it can't ever need testing for feedback again". Consequently I never released it and moved on. Even beta maps suffer from this...feedback shouldn't stop when the dev textures disappear.



In the end, I don't really know if there's any solutions to the problems that will actually work that aren't simple band aid solutions. We can talk all we want about solving community problems but at the end of the day I've never really seen any of that work effectively. The TF2 community, and our community has changed, and that's no one's fault but time.

It sounds really critical and useless, I think I lost myself in this post a bit. Sorry.

Then I guess something that we have to keep in mind is to treat maps equally. Whether in beta or in alpha or even in release candidate stage. Something I am guilty of just joining test to play custom maps because I am bored of the stock ones.

Which brings up a good question: why does the server die as soon as the tests are over? Why can't we just have custom maps running all day?

(I don't want this thread to get too off topic, I just feel as though people might give better feedback in actual testing if they have the opportunity to play maps all day long. Then the testing period becomes something special instead of just another custom maps server.)

No no no no no. I completely disagree with what puxorb was saying and pretty much every aspect of the conversation after that up until the very intellingent posts made by Aly and umb.

The imp puxorb was talking about was in relation to koth_amazing_a5a, which featured an "easter egg" that allowed players to get into the skybox by sticky jumping onto the UFO and waiting there.

This is the crux of the matter, and to my mind, it means that it's not the players at fault, but the mapper. Seeing people hanging out in the skybox is fun and all, but it completely wipes any chance of serious gameplay. It turns it into an April Fools' map, and a bad one at that. Now I'm sure abp just did this as a joke and he fully accepted that the consequence of it was a completely useless test. Maybe he's fine with that. Maybe we should even complain at HIM for "ruining" an occasion when you wanted to play "sensibly". That's up to you.

Too long; didn't read? Gameplay can only ever be as serious as the map. If your map encourages silly behaviour, expect silly behaviour.

This is reflected, to a lesser extent, by Conga parties and rocket jumper action. Consider a map like Hightower. People moan that there's always a few rocket jumper soldiers there, and that play devolves to deathmatch. But it'd be insane to blame that on the players: the map is designed in such a way to foster that playstyle, by using an abundance of height variation and healthkits. Don't want to play with rocket jumper soldiers? THEN DON'T PLAY HIGHTOWER.

When we're testing maps here, a similar effect occurs, with the only difference being that most players are unable to predict how the map will play until it's already too late - after it's dissolved into a deathmatchfest. Testers should not complain about other testers in this scenario; it'd be like whining that no one's taking cp_orange seriously.

The real kernel of this complaint though, the reason it turns up again and again, is a failure by mappers to appreciate what you have created. If people are not playing on your map seriously, then some aspect of your map is allowing them to do that. Lemme give you a brief rundown of this kind of cause and effect:

• Excessive number of healthpacks -> Players rely less on their team to provide backup when on low health -> players deathmatch instead of focusing the objective.
• Skybox is extremely high up, with little clipping above buildings to impede aerial movement -> It's fun and effective to sticky jump across the entire map -> map devolves into stickyjumping.
• Map is extremely vertical. with the majority of player interaction occuring with 512 vertical units between them -> rocket jumping becomes the only efficient movement scheme -> Best rocket launcher for rocket jumping is the Rocket Jumper, which players use in order to avoid running up stairs all the time.
• Map contains easter eggs that have massive visual impact on the game -> players find an alternative objective to the one specified by the gamemode -> players ignore the original objective.
• Players lose interest in the playspace, either because it is dull and poorly designed, or because they have explored all interesting gameplay choices within it -> players find an activity other than gameplay that is more interesting -> CONGA

In my opinion, Crash's April Fools' maps - koth_trainsawlaser and koth_wubwubwub - are perfectly designed for a balance of fun and serious competition. Notice in wubwubwub, the gameplay is completely unaffected by what makes the map fun - in other words, there are no ancillary activities, no easter eggs to activate the dubstep - so the result is that people play as per normal. It's a koth map made into a joke, but the joke doesn't stop the koth.

Why does the server die? It is partly because the map list management is sloppy. But the main reason is because it is expected. People don't expect more playing, so they do something else. And don't forget about the schedule. People have other things to do. And then there is the exhaustion factor. People get exhausted by doing anything for a long time uninterrupted.

But we can all blame the quickplay system. Everything is the fault of quickplay.

idk ive sometimes congaed just for the hell of it, interesting playspace or not

my bad


on topic of complimenting the map, thats just how it is. unless you've fixed a horrible problem, are popular on tf2m or have done something really outstanding, you aint gonna get compliments for making an alright map cuz that is what you're supposed to be doing. take the lack of hate as a silent praise and get used to it.

on topic of feedback plugin being bad, it aint. it lets me leave my feedback then and there instead of having to remember it until the test ends and post it properly in the thread which most of the time i just cbf doing.

Aly said:

"this map is RC, surely it can't ever need testing for feedback again".

Click to expand...

that surely is a retarded mindset. even pushing the rc to the end of test is.

henke37 said:

Why does the server die? It is partly because the map list management is sloppy. But the main reason is because it is expected. People don't expect more playing, so they do something else. And don't forget about the schedule. People have other things to do. And then there is the exhaustion factor. People get exhausted by doing anything for a long time uninterrupted.

But we can all blame the quickplay system. Everything is the fault of quickplay.

Click to expand...

This is getting on to a different topic, but I think that the best method of combating this would be to have more admins for the steam group. Why? Because admins announce imps. We already have enough people who can change maps, we just need people to tell all those who have nothing better to do. Admins are the bottleneck, so more admins would make more imps which would make more people on the server. Eventually, we could reach a player density where there's always people to play and always something to play.

What else do steam group admins do?

People behaving stupidly in tests be it continuing to use exploits, being a 'friendly' player and not attcking anyone or congaing around is a problem with that singular playtester. I just think we as staff need to be more vigilant with dealing with these people and behaviors.

We could disable those congas but those people would find some other way to be useless in map tests, they did it with Hi5s before the new taunts even released for instance.

Aly said:

Clearly, the community is able to give high quality feedback already. They've been doing it in the pageloads in the valve beta map threads. So the question remains how do we get people to give this feedback to maps that actually need it (more).

If people already have the ability to give feedback, why aren't they doing it on community maps (much).

Click to expand...

Because they play the community maps once. rd_asteroid and pl_cactuscanyon is getting played by thousands of people every day. If you want to try it out, all you have to do is refresh the server browser and there's 20+ servers available.

For community maps? You get a 30 minute window once a week on one server.

This is really only something Valve can fix and goes back to the topic of them helping mappers out. tf2maps simply doesn't have enough people to build any critical mass.. we need some mechanism to tap into the beta program and get more opportunity for playtime.

One thing I've noticed is that people seem to care about the exact mapping process far too much. The whole reason we have the alpha-beta-rc format is so that people who aren't familiar with the process have some guidelines on how to do iterative testing. I can think of a particular test on a map of mine where one person did nothing but complain about how I made my map: "Why did you use vertex lighting on an alpha?" (to get rid of ugly track shadows) "Why did you do a final compile on an alpha map?" (i didn't) "You could've easily textured this map." (????) Other people left comments about "You shouldn't make your roofs this complex in an alpha stage!" Why?

Feedback like this is completely rubbish and useless. During a map test, how the mapper made the map should be the least of your concerns. Unless the map is functionally broken, the only thing that matters is how the map plays. I can almost guarantee every 1-2+ year mapper here does things differently than The True Way To Map, because that's just how they're comfortable mapping.

More or longer tests is definitely good, but unless the active population of tf2maps explodes by an order of magnitude you're going to suffer from low turnout or burnout if it's not aggressively curated.. which places a large workload on the admins.

I think everyone here is pretty good about spotting a promising map, maybe some kind of nomination system like the map showcase or the heatmap marathon test could work. Pick one map every two weeks to get a couple hours of playtime, then schedule it. Make some qualifications, maybe like "map thread must have measurable activity, cannot be eligible more than once every three months, and must have passed some kind of community stamp of approval from a normal gameday."

It'd even better to get that one map onto the Valve beta servers for a week but I don't think we're there quite yet.

You know, there's a source plugin that lets you vote on maps. That might be handy as another data point. One simple question: "Is this map fun to play?"

If you're getting generally positive feedback from those bothering to use !fb and post in the thread, but most of the players who have played your map are voting it as 2 out of 5 stars, it could help let the mapper know to look for something else that's off.

Combine voting results on the feedback page, maybe with a voting breakdown (no names, but something like "X negative votes were on the losing team, Y negative votes were on the winning team" etc) and it'd be a great way to determine the overall impression your map is giving.