Plane Havoc - my game about plane combat on Roblox struggling to get players

Help and Feedback Creations Feedback
, , , ,
#1

Plane Havoc is a game about semirealistic plane combat. Its main selling points are flightsim-style controls, a part-by-part plane damage system and flight tools. There are three eras that players vote between before rounds - ww1, interwar and ww2. There are 15 planes in total spread out over the eras.

The game includes terrain generation, creating 961 km^2 of triangle terrain for each round, sometimes giving the teams a slight advantage, or both teams an additional challenge, such as fighting around a mountain.

The game includes a tutorial to improve the onboarding experience and allow players to learn how to use the planes.

Recently, I have rolled out an update, remaking much of the in-game UI, adding a currency system and skins that players can purchase to improve retention by creating an additional long-term loop.

However, ever since its creation the game has been struggling to get players. I have tried adding the tutorial, improving the game images and thumbnail, and rolling out the aforementioned update, however the results are horrid so far. In fact, placing a better thumbnail and game images has pulled the game out of search, which used to be the main way it was found by new players, which has killed off the miserable 1-6 new players I had a day.

I have an account on youtube and tiktok where I post videos of the game in hopes of attracting new players. I suspect that they would look up ‘Plane Havoc’ on roblox after watching the videos, which is why about 99% of my players were from search.

I would appreciate constructive feedback and improvement suggestions.

1 Like
#3

My 2-cents:

  • The logo is non-existent. If you want players to join you must lure them in with good-looking icon and thumbnail(s).
  • The description is unreadable. You need to add some paragraphs and/or shorten it. You might want to add emojis.
  • Thumbnails: the letter type of the thumbnails looks bad, try something more colorful or outstanding.
    And last but not least:
  • Add mobile support: it will add a lot of players to your game.

Your game has potential, don’t waste it. Goodluck!

2 Likes
#4

Almost a week has passed now since I released the UI, skins and currency update. I have also tried to increase my output of social media videos. However, the game’s visit counts are not increasing at all - held back at the 1-6 visits a day position. It feels like growth is limited entirely to shovelware on this platform.

#5

dude you cant rely on social media you’re gonna have to save up some money and just start advertising the game, social media can only get you so far depending on how viral your videos have gone

not to mention its pretty much a numbers game and luck when it comes to posting on these platforms, if you’re looking for immediate players you’re going to have to invest into advertising my friend thats the only advice I have for you

#6

the game is actually pretty good, plane physics feel nice. but if I may recommend, you NEEEEEEEEEEED to advertise, unfortunately THE CORPORATION has removed user ads and gave us a bad advertising setup to use. social media does work so try tiktok and youtube shorts to see which one works better.

get a discord too, really good way to gather a player base.

1 Like
#7

It’s really good, but I think the landscape can be improved. Since it’s a semi-realistic game, having more realistic land would be good. Maybe making the thumbnail more attractive would work?

#8

getting a discord isnt gonna help him gather a player base if he has pretty much no funnel of players to actually join it…

and yes youtube and tiktok can work but the chances his videos get millions of views is just low, its pretty much just luck and having a somewhat decent video alot of people watch almost the whole way through

#9

to the bank it is then, he’s gonna have to pull out a 1,000 dollar loan. im kidding, atleast spend 100 bucks on ads

#10

we do have a discord, and i meant tiktok, youtube and youtube shorts by social media; thank you for your feedback

#11

The problem with ads is that they are not guaranteed to work - the ad credit system could just get me about 10 visits at ridiculously terrible player per credit rates. If we could run ads with 100, or even with a single thousand robux then I would be able to do something, but currently I just don’t have anywhere near the required almost 3000 robux for the 10 ad credits.

#12

Have you ever considered using deep reinforcement learning AIs as a way to increase average playing time for players? Like making them as fake players?

More longer playing time means more likely people would play with each other. And this would create a stronger feedback loop on other players will join.

Before doing anything rash like spending money, why not do that first? I could give you resources related to deep reinforcement learning AIs if you want.

#13

I looked into it and it sounds like an interesting idea, however the problem with using something like deep learning to create bots is that it would need huge amounts of training, especially with something like piloting a plane, potentially exponentially more than tasks like walking and sword fighting which by themselves are much simpler, and still need a large amount of training hours to be good. Additionally, with all that training, the data built up would likely break the new limits on datastore usage, as while the current player data doesn’t take up a lot of space in the datastore, we don’t have that many players and therefore probably below a gigabyte of total datastore space, which would not be enough to store the training data for deep reinforcement algorithms.

#14

This is just against ToS, you cannot inflate your player count. You will receive backlash from the entire community and then get banned from ROBLOX.

#15

I don’t think that’s what he meant, if I understood him right he suggested adding bots that use a deep learning algorithm to be capable opponents

#16

Then that’s fine, just the way he worded it seemed like he was advocating for botting. In that way, it would require connecting to an AI software, or making your own, training the AI and telling it how to play the game and make sure it doesn’t cause bugs or break anything. If this were done it would be awesome but really complex.

1 Like
#17

How the hell did you arrive to that conclusion… If I were to really want to encourage botting, I’d say something along like “disguising AIs as real players”.

And yes, I meant what the other person said.

1 Like
#18

Everything what you said is mostly correct. Though, I’d disagree on the limitations because that would be based on outdated knowledge.

I’ll address this one by one.

huge amounts of training, especially with something like piloting a plane, potentially exponentially more than tasks like walking and sword fighting

This is correct. However, recent research in deep reinforcement learning kind of already solved quite a number of issues related to long training times.

For example, instead of going full deep reinforcement learning approach where it would make training very long, you could initially train your AI through deep learning approach first with small amount of collected data. This would give a baseline performance, potentially cutting down 10% - 50% for training time. Then you refine this model through deep reinforcement learning, so that it can really learn on how to fly the plane with baseline performance.

Not to mention, researchers still kept increasing deep reinforcement learning training speed through creating new algorithms. I was having a hard time trying to catch up with these researchers and put them into my DataPredict library so that people can train their AIs faster without needing to know the underlying maths. The most recent ones have maths are a headache for me to follow, and so my current list of algorithms that I have are pretty recent, but not too recent.

Additionally, with all that training, the data built up would likely break the new limits on datastore usage

In deep reinforcement learning, we don’t store data to an online storage, but instead gets stored to an AI’s “memories”. What you need is the AI’s neural network’s weights. Those can be stored more easily and doesn’t even need a lot of storage. Believe me, I have done this as many times as before in my own projects.

To be honest, I don’t think it would hurt to just try. After all, deep reinforcement learning AIs does make a game stand out because the other games doesn’t have it and might bring in some curious players.

I can also recommend other libraries if you’re not comfortable with my library.