A Letter to the Developers (Follow Up)

This is a follow up to the original thread “A Letter to the Developers”, from April 2016.

(I have learned my lesson from the first thread; this letter is addressed from me personally. I cannot speak for the community.)

Dear Developers,
I have supported and been a part of this community from day one. My opinion has varied from time to time about the state of the game. I was disappointed for a while, to say the least; I felt you had lost touch with the community and what the game was supposed to be. But I love where the game is right now.
The lobbies feel alive during peaks, with game world servers easy to find for most games. Just this year alone, we saw a complete lobby revamp, achievements, fishing, and an inventory/selling UI rehaul with two new maps and a visual overhaul of four. It’s hard for me not to see the spirit in the team to bring the best content possible. But there are some underlying issues that TU continues to have that hinder the game from actually reaching its potential even in its current state.

The Good
I love your current focus on the Plaza. One of the issues the game has always had was the plazas feeling empty, because what were you doing there other than gambling? Lobby 3 was a game changer. It pushed all the shops together, removed unneeded clutter, made the lobby look more in line with its theme, and added water slides. With the addition of achievements, it gave the plaza a reason to be played. You could explore and find hidden stuffed animals, you could play at the minigames pad, ride the lazy river, and now: fish. At release, I really didn’t enjoy fishing. It was difficult, and I thought it would be a plaza activity I wouldn’t visit. However, three hot fixes later, I really like it. Thank goodness for the added color difficulty indicators, and it’s now easier (if you were unfortunate enough to get a very difficult fish, you knew you were getting man handled). Which is another thing I really love about the team now: your ability to tweak your updates relatively quickly to adapt to how the community feels. This is also one of the very few dev teams to be so open with your development plans. I have checked the Trello at least once a week for four years now, just because I know I will see changes and new cards. We also get detailed (and updated) threads about future plans and updates. We are getting an arcade with over thirty detailed machines. If you check the Trello you can find an insane number of future updates (although disregard the time standards, all this will take years not quartiles) and this is only the content we are aware of. Who knows what the future holds, but it looks bright.

The Bad
During updates, we see a surge in players and the game feels alive and plentiful. During down season however, we get to average lows below four hundred. Now don’t get me wrong, GMT could only handle 124 players maximum, so TU beats that by a long shot. GMT was only two lobbies and a handful of game servers. TU is ten official lobbies and player hosted game servers, with individual condo servers. During peak, it would be suicide to attempt to join a plaza server with over forty players. The lobbies, as they currently stand, cannot handle their listed maximum. The experience is a jittery nightmare for nearly everyone. Even now in lobbies with twenty players, I constantly frame lag and skip frames. With the massive arcade update approaching, I think you should seriously consider looking into a resolution to this issue before the plazas get flooded. During down season, the plazas can feel quiet as the players are spread out on different servers or different areas of the large map. The chat issue was fixed with the introduction of global chat, but physical plaza loneliness exists. I believe you should strongly consider prioritizing bringing spirits back. If I recall, they were originally removed due to lag issues(?). They added a real presence to the server so that all the players across lobbies could see each other spiritually (although I wish the model was more creative). The plaza is looking great, but it needs to run well during peak and feel more alive during down season.

The Ugly
The worst crutch to the game, in my opinion, is the player hosted minigame servers and workshop models. With desync and lag, there have been several attempts to combat the issue of a player hosting the server, such as A Quality of Life update. These - updates - barely - help. Slashing in Virus is a coin flip. All Ballrace maps are a slip n’ slide for non hosts. The host can literally shut the server down at anytime. I understand the financial reasoning behind not paying for your own dedicated minigame servers, but I cannot enjoy anything outside of Minigolf or Zombie Massacre with how the current servers are handled. I don’t think there is a game mechanic that can be implemented to negate the immense upperhand the host will always have and the desync that the rest of the players encounter. Open a Patreon or donation panel on the TU website specifically for game servers. I don’t think the gameworlds will ever be an enjoyable experience for all players on the servers until a dedicated server is made; the host player has an advantage while the other players have mild to severe disadvantages. To touch on player models; make them resize to normal height in gameworlds. I am sick and tired of coin sized models on Virus.

If I didn’t love you, game, and community, I wouldn’t have stayed with the community this long or spent all this time writing a thread. Every update the team updates brings fixes and content that matters. I am so excited for the future, and I am glad you are adding a bunch to the Plaza to keep players occupied when focus gets shifted to gameworlds. I am glad you stay as transparent as you do, even adding a new community manager for community events. But you need to fix how the plazas handle players and how gameworld servers are managed.

Thank you for reading,
@profbonkars, corrected by @Arkive86

15 Likes

I definitely agree with almost all of this, though one negative I’d like to add is the lack of connectivity between the plazas, condos, and games. The current game ports are a tad redundant and locked behind a loading screen, and things don’t really feel super cohesive overall. Though changes to the game port system, main menu, and condo connectivity to the plaza are all coming down the pipeline, so it’s being addressed. This is similar to the plaza feeling empty thing (They have spoken on wanting to re-implement the player ghosts, they weren’t happy with how they looked, and weren’t obvious what they were to new players, so changes would be made).

Plaza servers definitely need some improvement, I stay away from populated servers because as soon as the player count rises above 30 or so ping is usually over 100, and not great to play. The one thing that could help with this is updating to a later Unreal version, which has better handling of dedicated servers from what I’ve heard (Because, y’know, Fortnite). Though that’s a very time consuming process with a whole host of issues, and isn’t realistically coming any time soon. Plus, I’m far from and expert, and maybe that won’t even help much at all. The alternative is getting better hosting/server system, but that’s a complicated and often expensive issue. In the end, there isn’t much they can do right now without a lot of extra time and/or money on their hands, though I really hope we see some changes.

The down season is kind of a sad inevitability, obviously there’s no way to directly counteract it, but hopefully a smooth Early Access release with marketing and whatnot will bring a more permanent population to the game (Technically all the current players are sort of beta testers). We can only hope.

On the opening a donation/Patreon thing, they just aren’t doing that. It’s come up many times, and they simply don’t want to accept donations of any kind. Perhaps it’s just a little iffy to accept donations for a paid game, and people might be expecting something, whatever the specific reasoning, they appear to be staying away from micro-transactions and anything similar indefinitely.

The only thing I’ll contest you on (At least from my personal experience), is the player hosted servers. I do agree Virus is quite the issue, though it’s hard to say exactly how much is latency, and what is just poorly handled collision, player position, etc. Though especially for non-PvP games like Ball Race, Minigolf, and Zombie Massacre, there are essentially 0 problems for me (I acknowledge you were good with Minigolf and ZM), if anything Minigolf is the most problematic of those, with Zombie Massacre in second. The delay on putting with higher ping is unbearable on maps like Sweet Tooth and Kingdom, and Zombie Massacre is downright unplayable if your host has a really awful connection, though with Ball Race the only noticeable effect ping has in my experience is the host seems to start slightly earlier, though with a lot of Ball Race being client-side, I don’t think that even gives them a better time/advantage (On the “slip n’ slide” note, that might just be TU’s physics, which are quite different from GMT’s, though maybe there’s a problem I haven’t noticed). If you can keep your ping around 30, it’s smooth sailing in my experience. The unmentioned elephant in the room being Little Crusaders, which I don’t blame you at all if it’s problematic for you, but although I can see where latency can feel like an obstacle (Mainly biting and pushing the button), as a frequent player it honestly doesn’t bother me, and you can make up for the lag with strategic gameplay that isn’t just rushing the dragon immediately. Though in my opinion, UCH/LC has been always been a bit gimmicky since GMT. My opinion on hosted servers doesn’t really matter if most people find it problematic anyways, I’m just voicing my perspective.

They have stated they’d like to have dedicated game world servers if possible, but it’s not very financially viable.

Also, I agree on workshop models, both in Virus and Chainsaw Deathmatch, I believe they might be addressing this, but I’m not really sure (It’s not easy to properly scale such diverse models).

Thanks for the read! It covered a lot of important things.

The donations was purely a suggestion as I know they don’t have the money to pay for dedicated servers for gameworlds. But is there any way to fix it?

It seems donations won’t be the solution, and they don’t have any other promising financial avenues (Considering they make so little from game sales). They’ve discussed a merch store around early access release, but I’m not sure that would be profitable enough to fund servers anyways.

So without funding for dedicated servers (Potentially, I’m just speculating, I don’t really know about their finances), it comes down to polishing the existing system.

Maybe the engine update I mentioned could help, or there’s some cleanup they can do somewhere, but considering I don’t have the issues you’re describing I can’t really say.

Personally, I think Virus might need a bit of a top-down overhaul though.

I definitely agree with a lot of this, namely the issues with gameworlds. For one, it’s really jarring if a host disconnects and everyone is immediately forced out of the game that they have been immersed in. This could be solved though by passing the host role onto another player on the server (maybe some form of calculation to see which would give everyone on average the best ping?) and would prevent the server from just immediately closing. Although, I’m almost 100% certain this has been suggested before and I can’t confidently say if this would be easy to implement or not (having to switch the connections of all the other clients to point to that other player and still have it be cohesive - maybe it’d restart the level players are currently on if that makes it any easier to manage).
Another thing I think might work out pretty well is a community condos-like system for gameworlds where a player hosts a dedicated server themselves. This server would operate as long as they have it running and would cycle through maps either randomly or via player voting each map. If no players are present the map is cancelled and the server will idle until someone joins again, allowing them to vote for a map or wait for more players.
As for the gameworld ports, the devs are very aware of the disconnect and have stated publicly they’re discussing some big changes to it.

People have thrown around a bunch of ideas so I don’t think I have much to say on this front but I’m interested to see what they decide on.
I completely agree in the case of server performance. Although I don’t think there’s much that could be done on this front as the devs have been well aware of the performance issues involving servers for a while now as far as I can tell, it would be well worth looking into before the Arcade is released. I wouldn’t be surprised if the player count passes the all-time high again, as updates 0.7 and 0.8 pulled in a ton of new players (as can be seen by all the new people on the forums and discord, etc). The more of these big updates that come out the bigger that spike in player count is going to be and in its current state I don’t think the servers would do well after a while.
Worth mentioning that my ideas here are purely ideas, I have no basis on how easy or hard these would be to implement, how practical they’d be, etc. Just thought it would be worth putting out there.

A lot of what’s here has been addressed in the past, but nonetheless I think it’s all worth mentioning in some form or another. Love all the incredible progress TU has made in the past 3 years and I look forward to what it has in store for the future.

3 Likes

I 100% agree with most of this post, I wouldn’t mind a lot of the empty plaza servers (see the US region) being used as gameworld servers instead.
I am confused by what you mean by “slip n slide” in Ball Race though? Ball Race is really one of the games where ping impacts you the least, both the host and clients have the same experience (except for a delay in picking up melons and entering the goal)

1 Like

Host Migration isnt possible with UE4, remember a dev bringing this up twice.

3 Likes

Maybe not a simple “call a method and be done with it” type host migration, but if someone crashes/disconnects surely there is a way to save the game state, determine a new host, then restart the server and restore the game state.

This would of course have to be implemented differently for every game world and would require a pixeltail-hosted server to store this information temporarily.

I am no game developer but I’ve seen a few Unreal games with a host migration system.