uMMORPG Official Thread

Hey, thanks for your question.
Active development is a good thing. It doesn't mean that it's a beta or something like that, it already has the most important MMORPG features. And except for some UNET bugs, there are no (known) bugs in my code right now.

80$ isn't much money for a MMORPG if you consider the insane amount of time that it takes to do it right, or the amount of money that it would cost to let someone develop it for you. Also 3D models are not exactly cheap, just the monster model alone is worth 10x that much.

Furthermore, it's the first UNET MMORPG package on the store, and the most simple MMORPG solution in Unity.

 

Hello,

Thank to answer. I understood. it is true.

Before purchasing, could you confirm is rdy:
- broadcast messaging ?
- private channel ?
- trading player to player ?
- auction system ?
- world event system ?
- aoe system ?
- player transportation ?
- bind stone with ressurection ?
- sharing XP ?
- save position when i log off ?
- crafting ?
- harversting ?
- guild and player house ?

+++

 

I updated the uMMORPG version.(v1.3)
I try to remove monsters and NPCs from the Scene.
However...

 

• broadcast messaging:UNET automatically sends networking infos to all surrounding players automatically if that's what you mean.
• private channel: you can whisper other players if that's what you mean.
• trading player to player: will be in the next version, I already added it to the Roadmap in the first post a few days ago.
• auction system: it's not planned because I don't consider this to be part of the MMORPG core features.
• world event system: not planned because I don't consider this to be part of the core features either
• aoe system: good point, will add it to the Roadmap
• player transportation: what do you mean? Like mounts or like teleportation?
• bind stone with ressurection: not really part of the core features either
• sharing XP: will consider it if more people request it
• save position when i log off: already implemented
• crafting: will consider it if more people request it
• harversting: will consider it if more people request it
• guild and player house: not part of the core features

So generally speaking: uMMORPG is supposed to have the MMORPG core features that can be found in most MMORPGs. It's supposed to be easy to use and easy to understand and build on. Something like housing is cool, but it would overcomplicate the code for 100% of the people, why it would probably only be useful for 10% of the people. The same mantra applies to the other features that I considered to be not part of the MMO core.

The goal is not to build a complete WoW clone or anything like that. Think about it: if I'd sell you the WoW source code for 80$, you still wouldn't be able to do anything cool with it because it contains more than 1 million lines of code. All you would end up with is exactly WoW without any modifications. I'd rather have people understand all my 3500 lines of code and be able to build stuff with them.

Hope that helps,
Greetings

Hey, thanks for the screenshots. It looks like you run the server on your local computer, right? Can you please:
- make a copy of your project folder
- delete the Library folder inside it (its not needed)
- zip the project folder (should end up being only 9MB or so)
- email it to info[at]noobtuts.com
so that I can try it myself? I really want to find out what causes this UNET error to happen, the sooner the better.

 

Hi,vis2k.
I sent e-mail to you.

 

Hi there,

Bought the package today and I like what I am seeing. I am hoping to make a persistent multiplayer RPG to host 50-100 players so I think this should be doable with uNet and headless server. The simplicity makes it easy for us to drop in what we need so a +1 from me for that approach.

Some Thoughts:

• I see area of effect spells are on the list - +1 for that - can't wait!
• Another small idea is to have the player automatically target another aggressive NPC when combat ends. I don't think this need be complicated and is likely to be used by most?
• I'll be adding in MySQL support myself - I completely understand why it's not there but perhaps an option second package which plugs and plays? Just a thought.
• Personally, I think a proper uGUI implementation should be on the roadmap. I understand your reasons not to, but OnGUI is poor for mobile devices and just a legacy solution. I think in this instance the complexity would be worth it.

Just some feedback for you!

Thanks,

Rob

 

Here is what I would like to see implemented as features of this package:
looks like only melee is only implemented atm.
so here my suggestions that all mmos should have as core feature.:

-range attacks(arrows, bullets or spells ie: fireballs etc...)
-AoE area of effect spells
-DoT damage over time spells
-Buffs
-mounts(Ground and Fly mounts)
-healing spells
-harvesting and crafting
-factions, races and classes
-player teleportation
-abilities
-grouping and share xp and quests to accomplish elite mobs or enter a pvp area for group vs group fights etc..
-guilds
-Player vs Player combat
-mobs ability to loot items(helmet, weapons etc..) besides gold and xp
-mobs ability to cast spells
-player housing can be a very nice and attractive feature.
-uNet has a lobby system, why not add one with global sql or xml account login/authentication db, and can list all current servers/realms online so player can pick available one, this way players wont be limited to a single sever/realm which can be filled up quickly or over populated and create performance or crashing issues.
-ohh and please convert to uGUI, I know you trying to do simplicity, but old legacy gui is just way out of the picture due of too many reasons that we all know specially performance issues and most of all, it is already deprecated, it will be gone soon in future unity 5 releases.

some may think all the above are not mmo core features, but if one pays close attention to most if not all mmos out there all have all of them features, so therefore should'n it be a core feature from the getgo?.

anyways, there should be more features that are not listed here or above in other posts, but to start having all of there will make the asset way more attractive to buyers.

cheers

 

For everyone: I am currently testing the server with huge amounts of entities and found some interesting things that I can change to make it faster. One of the next versions will get some performance improvements, I also added that to the Roadmap as usual.

Got it, status can be seen here: http://forum.unity3d.com/threads/unet-readbytes-out-of-range-and-readstring-too-long-errors.379585/
I am currently trying to track the bug down in order to create a Demo Project that I can submit to that thread.

Hey,
I am glad that you like it! I will give this auto targeting a thought, but my first feeling is that it makes game Bots far too easy. About MYSQL: I already have some examples in the NetworkManager's IsValidAccount function, this might help you. When I get the time then I might setup mysql myself and have a more complete example that can just be uncommented. About uGUI: see below.

Hey,

• Yes only melee is implemented right now. The next player model that I want to add (see first post) will be an Archer, so range support will be added soon.
• The game already has a healing spell.
• PvP is a good point, will add that to the Roadmap.
• Items can already be looted from mobs.
• Lobby: I only took a quick look at the lobby demos, but my impression was that the whole server has to be in the lobby state until the game starts, then the whole server is in the 'world' state. This doesn't apply to MMOs, where only a few players are in the lobby while logging in. But I could be wrong, perhaps I missed something in the Documentation.
• About OnGUI: see below
• About the features: I will keep them in mind. One step at a time

OnGUI vs. UI: I modified the part in the FAQ a bit more to give some more insight to my point of view:

• Why OnGUI and not UI? Because it's a immediate mode GUI system. Items, Skills etc. live in SyncListStructs and OnGUI allows us to draw them to the Screen without any additional state. UI is retained mode and would require one additional UI GameObject for each Item, Skill etc., hence twice as much state, more complexity, more code. When it comes to software development, state is bad. In other words: without a doubt, the new UI system looks better, it's faster and it has more features than the OnGUI system, but OnGUI is immediate mode.

 

Ｈｉ.
Thank you for your reply on e-mail.
I was given a hint to solve to you.
and uMMORPG is working!
However, I have errors.

 

Progress Info: V1.4 has HUGE performance improvements, see first post (headless can run 1000 Monsters instead of 50, even on my mediocre i5 CPU). The performance boost comes mostly from modifying the Layer Collision Matrix, and I am really excited about the result. Unity is so fast, it's incredible!

Hey, I get those warnings too, they don't seem to do anything so far. Also I managed to track down the ReadBytes error further by creating a test scene with 500 monsters, but then all of the sudden it didn't happen anymore. Also seanr suggested that we take a look at the server logs (http://forum.unity3d.com/threads/un...adstring-too-long-errors.379585/#post-2463769), can you send the log file to me right after it happens? You can find out where it stores the log file here: http://docs.unity3d.com/Manual/LogFiles.html so if you use Linux then it's probably in "~/.config/unity3d/unity3d/noobtuts.com/uMMORPG by noobtuts.com/Player.log".

 

Hey guys, I reported two more UNET bugs that affect all of us here:

• http://forum.unity3d.com/threads/bu...ect-function-is-never-called-for-host.379756/
  
• http://forum.unity3d.com/threads/bug-player-ondestroy-is-never-called-on-server.379752/

Those bugs are the reason why the player is sometimes not saved after quitting the server when using "Server & Play" mode.They are also relevant if we want to save all players before quitting a dedicated server. If you want to get them fixed, then please go to those threads, try the project and write a reply where you confirm that it happens to you too.
EDIT: received some more info on those bugs, all is good.

And another bug/missing feature where Debug Log/Warning/Error output isn't shown in the Terminal for headless mode executables (which would be incredibly useful when running game servers):

• http://forum.unity3d.com/threads/re...dless-mode-when-running-from-terminal.379758/

 

Would it be possible for you to integrate UMA 2 into this? Allowing for easy equipment setup of armors etc.

Also, how would you currently add armor? Does it support Unity's particle system?

What is the possibility of adding or switching over to uGUI? Since it is after all uMMORPG and uses uNET you might as well go with uGUI as well, otherwise old GUI is yuck and won't ever be pretty

 

Hey,
UMA sounds interesting. I didn't plan character modifications in detail yet, so I will keep this in mind for when I get to it.
Weapons, Armor etc. are Items, which have a ScriptableObject representation in the project's ScriptableObjects folder. Each item can also have a 3D model (for equipment). The 3D models are just prefabs, so you can easily add particle effects to them if needed.

About the GUI:

• Why OnGUI and not UI? Because it's a immediate mode GUI system. Items, Skills etc. live in SyncListStructs and OnGUI allows us to draw them to the Screen without any additional state. UI is retained mode and would require one additional UI GameObject for each Item, Skill etc., hence twice as much state, more complexity, more code. When it comes to software development, state is bad. In other words: without a doubt, the new UI system looks better, it's faster and it has more features than the OnGUI system, but OnGUI is immediate mode.

Greetings

 

@vis2k
It would be interesting you add support for UMA.
Guild/Party are part of the core of a MMORPG, it's hard for you to find an MMORPG without Guild and Party.

I think KBEngine more interesting than the UNET. But I know that your goal is to let code to be easy interpreted.

Congratulations, you are doing a great job.

 

Progress Info: UNET Server Hosting Tutorial added!

Hey,
I will look into UMA when I get to character customization later. Also I like UNET a lot because it's Unity-native and I believe that the UNET developers will make it really fast and reliable in the future.

Thanks!

 

I didn't see this asked so sorry if it already has been answered. Does this support dynamic server side collision? So if someone plants a tree everyone's own client would collide with the tree.

 

Progress Info:

• fixed a NetworkNavMeshAgent bug (the one with the Vector3 Magnitude workaround)
• reported http://forum.unity3d.com/threads/bu...creentoguipoint-both-dont-do-anything.380193/
• reported http://forum.unity3d.com/threads/bu...ayers-not-cleared-after-disconnecting.380190/
• reported http://forum.unity3d.com/threads/bug-ground-texture-very-gray-in-webgl-build-if-fog-enabled.380199/
• reported http://forum.unity3d.com/threads/bi...ercontrollerid-higher-than-expected-0.380205/

UNET supports that by default. Just call NetworkServer.Spawn for a GameObject with a collider and then all the clients get it too.

 

This is a great project and love where it's going.

To second some features that people are requesting, I totally am looking forward to the ranged attacks and area of effect implementations. I also second anyone that said they'd like to see a harvesting/crafting system, and if possible the world event system.

But this project is awesome, diffidently worth the money in my opinion. The stuff I'm learning from this source is priceless. Keep it up!

 

Late night progress update: V1.4 will use the 'intended' way to do character selection via ClientScene.AddPlayer+OnServerAddPlayer+extraMessage, which also gets rid of the "controllerId higher than expected: 0" warning and makes the code shorter.
Also big thanks to @seanr for posting the UNET Bitbucket link on another thread. Seeing the source code helps a lot.

Thanks, I am glad that you like it! Every feedback is welcome. V1.4 will contain important UNET related fixes/improvements. V1.5 might see some new features then.

Greetings

 

I been playing MMO's since the early 90's when most where made... I also worked on some, these are standard features today and some even back then... I have played nearly every AAA and most indie MMO's in my life.. So I would agree with this statement.. But 80 bucks is dirt cheap for what he has for this asset.. If he where to add some of this I would raise the price.. Just my opinion.

 

Progress Update: V1.4 was submitted for review. Changes can be seen in the first post.
I also had an idea regarding SyncLists and the new UI system, perhaps that's the solution.

Hey,
I added some of those features to the roadmap already. One step after another and we will get there!

Greetings

 

Great stuff, looking forward to it!

 

Progress Info: found a way to use UI with SyncLists, looks promising so far:

Good to hear!

 

vis2k - The new uGUI is looking amazing!! Woot!

Once thats complete and as soon as you get UMA 2 integrated, I will be buying this and donating

Also how does this work with animations? Like Mecanim/Legacy? Can you add additional combat animations? Like for casting spells or 1-handed swords vs. 2-handed etc.

 

hello,

UI will be dragable ?

++

 

Progress Update: more UI; got rid of 500 lines of OnGUI code.

Hey,
thanks, I like it too.
Don't expect UMA 2 too soon (if at all). Character customization is still a bit further away and integrating a 3rd party asset may not be the best idea either, will see about that.

Hey,
of course, each window is dragable and each item/skill/quest as well. I actually developed those features about half a year ago because I started this project with UI. Which means that most UI elements are already very stable and responsive.

 

I'm looking forward to next version.

 

Hello,

I'm waiting for better minimap integration because it seems second live camera, i think, and the performance and client experience are decreased by this.
It will be better if was snapshot, maybe.

++

 

Progress Update: UI Skills are done, those were the biggest challenge. What remains are NPC Buy/Sell, Respawn, Loot, Quests, Character Selection, Character Creation, Messagebox and then some testing.

Great to hear!

Did you try to disable the minimap camera and compare the FPS? It shouldn't be that much of a big impact actually, Unity is fast that way.

Greetings

 

Update: V1.4 was released to the store, changes can be seen in the first post!
Progress: I finished the last UI window for V1.5 today. A few more adjustments and fixes remaining..

 

Hi vis2k,

I've been following this for a while and loving the responsiveness and updates. Is the demo that you have hosted running the current release?

Is it possible to test with multiple users in one webgl on the demo? I have tried creating a dedicated server/server & play, changing login users but they never show up on the same instance.

I haven't found this to be the case it just displays connecting.. and it vanishes with doing nothing else.

Thanks

 

Hey,
I didn't update the demo to the current version yet, because V1.5 is almost done and I will update it to 1.5 soon.

About WebGL: I don't think that you can host a UNET server via WebGL that allows other WebGL clients to connect to it. I am pretty sure that has something to do with browser security / sandboxing.

Greetings

 

when v1.5 to be release?

 

Hey,
very soon, definitely this week. The code is finished, now I just have to update the documentation, screenshots, WebGL demo etc.

 

V1.5 was submitted to the Asset Store for review and the WebGL Demo was updated to V1.5 already! I will post here again when it was accepted to the Asset Store.


Changes:

• Note: make sure to delete the old Database folder because CharacterData has changed
• Used UI instead of OnGUI and removed 500 Lines of OnGUI Code
• Added [TextArea] attribute to Items, Skills and Quests for easier editing in the Inspector
• Renamed PlayerSkill.cs file to Skill.cs
• Moved IsHeadless detection from LoginMask to NetworkManager
• Fixed a bug where Skillbar Hotkeys were checked for all players and not just for the local Player
• Added string.ToInt() extension method
• Player Skills are now all in one list instead of using a Learned and Learnable list
• Workaround for a UNET bug where skills/quests wouldn't be synchronized to clients if we loaded them before the AddPlayerForConnection call
• Moved all Player logic into the Player class and all Player UI updates into another class for simplicity
• Simplified NPC Quest/Trade window logic by using two NPCs and getting rid of the initial NPC dialog
• Removed NetworkManager prefab because it's in the Hierarchy at all times anyway
• String IsNullOrWhitespace is now a static method instead of an extension, because it wouldn't be able to detect 'null' on a string that is 'null' if it's an extension
• Removed 2 unnecessary client variables in NetworkManager message sending
• Added Player quest limit to avoid database size exploits etc.
• Set 'Trees' Layer to 'default'
• CmdNavigateTo now checks if the Player is alive before navigating somewhere (dead players now can't walk anymore if they were to send a navigation command to the server)
• Loading a Player with 0 HP now also sets the initial state to "DEAD"
• Fixed "Command function CmdCancelAction called on server." by creating a CancelAction function that can be called on the server without any errors.
• Simplified Skill logic by replacing the 'lastCasted' variable with 'cooldownEnd' and 'casttimeEnd'. This makes cooldown/casttime remaining calculations and database saving/loading much easier and allows for skills to be canceled without starting the cooldown.
• Simplified Monster skill cast logic by replacing the 'lastCasted' variable with 'castTimeEnd'.
• CancelAction improved: it now stops movement or cast first and only clears the target when pressed again
• NetworkManager save interval decreased from 5 minutes to 1 minute
• NetworkManager OnDestroy now doesn't try to save the host character anymore because it only worked sometimes anyway; it also fixes a skill saving bug where Time.time was '0' already it in OnDestroy, causing skills to be saved with high cooldowns.
• Increased WebGL memory size to 512MB

 

Hi, that are great news! I learned a lot, but until now I could not figure out how to handle multiple scenes. The simple use case that one player walks to a portal(trigger) to get from "Town" scene to "Dungeon" scene. Did I miss some documentation or is it not possible at the moment with Unet?

 

Hey,
there is no easy way to do that with UNET right now, because the server always runs on one scene. So you would need different servers that communicate with each other, which also isn't easily doable with UNET right now.

Instead you could put your dungeon on the same scene as your main level, just far away from it. And then you could teleport people from point A to point B on the same scene.

 

Hey vis2k, thank you for your fast answer. I made the same conclusion with UNET HLAPI. Putting all into one scene would be a valid solution, but it isn't suitable for my use case, because the client runs on mobile devices with limited resources.

 

For those of us who would like to try out the multiplayer capabilities of the demo, would you be willing to private message us two logins to a running server so we can test latency and multiplayer behavior? Or perhaps make some generic logins like Test01...20 for everyone to use?

 

Progress info: started to work on V1.6, changes can be seen in first post. Everyone please support this request: http://forum.unity3d.com/threads/suggestion-globalrpc-observerrpc-targetrpc.382486/ . We desperately need this to make the Chat, Party invites etc. much easier.

You could try Unity's Occlusion Culling: http://docs.unity3d.com/Manual/OcclusionCulling.html

Hey,
the WebGL demo works exactly like when connecting to a server, just that you would have a small latency depending on your location. My VPS is located in the US, so if you are from there then you probably wouldn't notice any difference. If you are from the other end of the world, then you would notice some latency. Uhm I guess I could setup an extra demo that connects to the VPS, but only if you really need it

 

I don't really need it because this kit isn't at the maturity level yet where I'm ready to make a purchase, although I am impressed and encouraged by the progress I see.

You're telling people they can host their own mini-MMO with this, but you're hesitant to host one yourself? That is a red flag. You may have trouble getting people to do more than just watch this thread if they can't test the multiplayer before purchase.

I want this to succeed. I think it's in your interest to demonstrate what this kit can do.

 

Hey,
sorry about that, I didn't mean to be rude. I just use my VPS for testing the latest changes all the time, hence my hesitation. But no problem, I'll put a online demo up tomorrow!

Cheers

 

I didn't think you were being rude, just acting like it was a lot of trouble. Glad to hear it is around the corner and not a big deal.

I'm in California, so if the VPS is on the west coast, I'm sure the lag will be negligible. But I'd like to make sure things like animations stay in sync. Other multiplayer kits have had issues with that.

 

Okay so I just spent my whole day trying to get a WebGL demo connecting to the VPS. It only kinda works, hence why I won't update the WebGL demo just yet. Here is why:

• There is a WebGL bug where 75% of the time, my build comes out with black textures for some of the entities. If someone else encounters this problem: reimporting all textures and then immediately building WebGL sometimes fixes this. Also setting the texture compression mode to Truecolor seems to fix this sometimes.
• WebGL has to use WebSockets to connect to the server (which also has to use WebSockets). This can be enabled in the NetworkManager with one click, and it generally works. However, there are random disconnects without any error messages on the client or on the server. I assume I could track them down further and try to report them to get them fixed. My last three UNET bug reports went completely ignored, so I don't feel like spending my time on that right now to be honest.
• The alternative was to build for Windows/Mac/Linux and run the server without WebSockets, which works perfectly fine. The downside is that the Assembly-CSharp.dll file in the Managed folder contains all the source code practically in plain text, so uploading that is not an option. Unity doesn't seem to use IL2CPP for Windows/Mac/Linux builds yet, so there is no good way to protect the source code / project files.

So long story short: I'll focus on the project features/improvements and try a WebGL<->VPS build again in the upcoming Unity versions, perhaps those bugs will be fixed then.

EDIT: I found out why WebGL clients crash: http://forum.unity3d.com/threads/un...structs-before-addplayerforconnection.381606/

 

uMMORPG V1.5 was just accepted and can now be downloaded from the Asset Store!
This is a huge update, changes can be seen in first post. Make sure to delete your Database folder once.

Enjoy the new UI!

Edit: V1.6 was submitted to the Asset Store for review.

 

Hi Vis,

I've been following this project since it was first released and the progress is going pretty fast! Great job on keeping the project going!

What's planned for V1.7?

 

That's nice to hear, thanks!

For 1.7: fixes/improvements have highest priority, so if people find issues with the new UI system then 1.7 will fix those. Otherwise my list of open issues is pretty much empty, so I'll probably add player to player trading.

 

I know this has been asked before, but are you planning to add male/female option within the next 5 versions?

 

'Player type selection' is on the roadmap already. I want to add a whole new player type (archer). People can then easily replace that with other genders/classes. It could happen during the next 5 versions.

 

Alright, what happened here?

First let me say, I got the product when it was version 1.3 or 1.4 who knows. I started creating playable world. All the basics were there and I loved how you implemented everything. There was definitely still a lot could be done and I looked forward to seeing it develop.

Now here we are on 1.5

So first I have tried for hours to get my content to work with the new version. I get nothing but massive errors, I did not change a single line of code, everything I did was build with in your design functionality.

Next, what did you do to NPC's? making a Shopkeeper and a Quest Giver separate like they can never be the same person? Totally eliminating the greeting text from both of them. The greeting text was a good idea why take it out? What up with Quest Givers now only giving a single quest instead of a list like before?

Let's get a little positivity in here. I do like what you have done with the player and am excited to hear you are close on the class implementation. I can already see how the equipment types could tie into that. Have you given any thought into allowing us to increase damage and defence on level instead of just HP and MP? I saw in one of your posts you were working on WASD keys. Id really like to see that functionality

Now back to ranting. Having all quests repeatable indefinitely is a huge problem. Quests really need a way to be 1 time only and repeatable. However I would rather have quests that can only be completed once then every quest be repeatable. That being said, are there any plans to expand on quest dynamics. For example, collect X items, go to X location? with the ability to turn in the quest to another NPC?

Now let's talk about skills, they are off to a good start. Do you have any plans to implement buffs or debuffs, increasing or decreasing Health, Mana, Damage, or Defence for a duration? Do you have any plans to implement skill ranks, allowing a higher level of a skill to replace a lower version of the same skill so the skill list does not get out of control? Have you given any thought into making skills percentage based rather than flat damage, allowing the skill to scale with weapon or player.

Like I am OCD, I am very happy to see the chat system working.

In the general area, do you have any intention on adding an avoidance state or a resist state, I guess that would go hand in hand with the skills being classified at melee or spell so it would require an upgrade to skill mechanics.

Anyways, being a game designer I could go all night. I look forward to your feed back.


PS
If you could get your giant GUI out of my world, I would appreciate that.