Unity Community


Page 6 of 36 FirstFirst ... 4567816 ... LastLast
Results 101 to 120 of 705

  1. Posts
    130
    this is looking good,

    got a few questions:

    1- since this is a non unity server you are building, i assume your server will not be aware of server side physics nor server side collision detection,
    like the other socket servers ie: photon, smartfox etc... is this correct?
    if so, are you guys planning to do something about it,
    otherwise, the server will be prone to lots of cheats, ie: walking thru buildings, speedhacking etc....

    2- is the backend an authoritative server?

    3 -how about AoI, will your setup have an Area of interest so the server will only synch traffic to nearby players instead of whole world or scene?
    otherwise the server might choke with more than 100 cc players in one scene/instance.

    4- how about the server backend you havent clarify what language the server is? java, c# c++ etc...
    is it windows only server? or is it multiplatform, ie: linux?

    5- one last question, are you the actual lead developer or are you the spokesman for the project, since by your posts you seem to know little about the server backend, is it someone else or a third party building the server?


    thanks.


  2. Posts
    177
    Quote Originally Posted by CaptainExtra300 View Post
    5- one last question, are you the actual lead developer or are you the spokesman for the project, since by your posts you seem to know little about the server backend, is it someone else or a third party building the server?
    I believe sooms (who's responded a couple times earlier in this thread) is the lead programmer. I'm not sure of cookimage's exact position, but he does seem to be more of a liaison/PM type role.

    I don't know about the other questions, so I'm eager to hear about them as well! We may need sooms to chime in to get those answers, however.


  3. Location
    Canada, Alberta, Calgary
    Posts
    415
    I am the CEO of the company Neojac Entertainment who is developing both the Atavism Online Server engine and also a MMO game which will showcase most of the things that the engine can do. Although I can answer allot of questions, when it comes down to the more technical stuff of the server Sooms is the best person to answer them. he is our Lead Programer. Everything is being developed in house so no 3rd parties. Server side physics and server side collision detection will definitely be implemented as you stated yourself the server and client needs to be on the same page with objects. The server is Linux based but also runs on a windows machine. Also I might be wrong but we have tried the other platforms available as servers for Unity and found that if we where to use them on a cloud server like Amazon we could not as they required a 3d card on the actual machine because they are Unity saved applications and Unity does require some form of 3d Processing to run where as ours is not a Unity direct application thus can run on any server type machine. But like I said I might be wrong about that and if I am please point it out as we would love some feedback on that topic. Ill see if I can get Sooms to elaborate a bit more on your questions but are currently extremely busy with a huge announcement that will be done around the 12 of August so hopefully I can get him to post something lol.


  4. Posts
    12
    Hey, I'm Sooms the lead programmer.

    Here's the answers to your questions:

    Quote Originally Posted by CaptainExtra300 View Post
    1- since this is a non unity server you are building, i assume your server will not be aware of server side physics nor server side collision detection,
    like the other socket servers ie: photon, smartfox etc... is this correct?
    if so, are you guys planning to do something about it,
    otherwise, the server will be prone to lots of cheats, ie: walking thru buildings, speedhacking etc....
    1) Currently there is no collision detection etc. on the server but we are looking at systems that will create a graph of the scene in unity and save that to a file which can be loaded in on the server. This will also allow objects to be added to it on the server on the fly for more dynamic detections when the world changes.

    Quote Originally Posted by CaptainExtra300 View Post
    2- is the backend an authoritative server?
    2) Yes the backend is authoritative.

    Quote Originally Posted by CaptainExtra300 View Post
    3 -how about AoI, will your setup have an Area of interest so the server will only synch traffic to nearby players instead of whole world or scene?
    otherwise the server might choke with more than 100 cc players in one scene/instance.
    3) Yes the server uses a quad-tree system to break the world up so you only get messages about objects close to you.

    Quote Originally Posted by CaptainExtra300 View Post
    4- how about the server backend you havent clarify what language the server is? java, c# c++ etc...
    is it windows only server? or is it multiplatform, ie: linux?
    4) The server is written in java and runs on windows and linux. It uses a mysql database and python scripts.


  5. Posts
    130
    Thanks for your reply sooms,

    implementing server side physics and server side collisions will be very tough to almost impossible to a non unity server,
    this has been attempted in a java server like smartfox but pretty much everyone fails to do so.

    so i hope and wish you have better luck to your server, hoping this could be done, since this is very important in an authoritative server. otherwise malicious players will take over your game and disrupt everyone eleses gameplay by simple wall hacking and speed hacks etc..
    and nothing the developer can do to avoid this or fix.


    now, one more quick question:
    - is this server you guys working on is this the multiverse.net open source mmo java server you guys trying to adapt to unity?

    just curios since i got the feeling that you may be doing so and sounds like it as per your responses.

    if it is, we too used it with its open source own c# client but reason we left it is because the lack of being able to adapt server side collisions and physics.


    thanks again.


    EDIT:
    By watching the demo video you guys have in the first page of this thread, i think i can confirm that you guys using the multiverse.net open source java server, at least i am pretty sure since saw the mob spawner model used in the original multiverse c# client and java server.
    Am I correct?
    Last edited by CaptainExtra300; 08-03-2013 at 05:21 AM.


  6. Location
    Canada, Alberta, Calgary
    Posts
    415
    We started of in 2008 using the then Multiverse Server before Multiverse Closed its doors and moved it to open source. We where the leading team using Multiverse and was in constant contact with Bill from Multiverse. As we where developing our stuff faster than what Multiverse could keep up with, with their updates we moved away from it and started building our own server side. Because we had so many systems done that was integrated with their server we had to follow the same structure as Multiverse to not loose all the work we had done. The server we have today is our own build so I am pretty confident that we will get it working with collision and Physics. You must know we have been working on this for some time and Sooms has been the lead programer all that time so he can create miracles with this product lol. We are already breaking ground with the integration into Unity that makes it super easy to set up a server using our in game tools and much faster than any other server platform available currently.


  7. Location
    Canada, Alberta, Calgary
    Posts
    415
    Quote Originally Posted by CaptainExtra300 View Post

    EDIT:
    By watching the demo video you guys have in the first page of this thread, i think i can confirm that you guys using the multiverse.net open source java server, at least i am pretty sure since saw the mob spawner model used in the original multiverse c# client and java server.
    Am I correct?
    As far as I know multiverse never had the capabilities to spawn mobs in real time in game or be able to edit the world in real time in game, they had a world editor outside the game where you had to do all your editing and mob spawning, lol we might have used their models at that time to safe us time in making them.


  8. Posts
    130
    ahh, yes, no wonder
    cuz i saw the mob spawner model and i remembered that that was the same as multiverse mob spawner, and the server infrastructure is just identical to multiverse as per your lead programmer comments, since multiverse is java server and scriptable in c# and ironpython, and the spawns and items are managed thru templates too as multiverse.
    and now that is open source and MIT license, i thought you guys were the braves to do a port to unity and take advantage of the nice multiverse server backend infrasctucture and scalability it offers out of the box and all the mmorpg game logic already included from its demo game.like, combat, inventory classes, factions, pets , advancement system etc....




    well, the reason i say this is because i wanted to confirm that it was multiverse, I wish it was, since i know for fact that multiverse server is quite powerful, and it has been a proven technology for a while, i know that it works great, since it was the main attraction and reason we used multiverse due of its server mmo capability.
    but only drawback and reason we stepped out from it was due of the c# client, since was really outdated, i believe the Axiom3d rendering version is using as of now is a build that was integrated into the MV client since 2005, so it is already very dated, 8 years old.

    we used multiverse and were very happy for its server capabilities, high scalable server, in fact i always wanted to get rid of the C# client and use the server with Unity, but since there are already some unity servers outhere, like:
    smartfox, Java server, photon, ulink, etc..

    so we just went and tried all of them and at the end we ended up with ulink, so far the best of all, for only 3 major reasons:

    1- both client and server are scriptable in C# , easier than using or learning 2 different languages.client and server

    2-client and server both are aware of unity physics and server side collisions so makes it very very hard for hackers to ruin the gameplay, unless one gives a reason on the poor coding for hackers to take advantage.

    3-high scalable server backend, one can have a multiple zoned game and use a hardware node per zone so no bottlenecks and have your whole game spread in a cluster of machines or datacentes all over, since all zones/ hardware nodes can talk to eachother internally and can handover the player seamlessly even thru datacenters.


  9. Posts
    177
    I've been following uLink off-and-on for quite a while now. The fact that it leverages Unity for server-side physics and collision detection was a huge draw in my eyes, since I wouldn't have a clue on how to implement those features on the server. (Good to hear that Neojac's will incorporate physics as well.) I believe you can run a uLink server in headless mode so your server machine doesn't have to have a 3D graphics card, but you have to have Unity Pro for that feature, correct?

    I'll be interested to see how server scalability is handled by Neojac's product. uZone's load balancing looks to be quite robust.


  10. Posts
    130
    yeah,
    well in ulink you dont have to do much to manage the server side physics and collisions,
    pretty much do the same as single player mode scene with rigid bodies and clone the scene and use one for the client and the other as server kinda of deal without too much effort, look at their rigid body demo, quite simple.

    and yes the uZone is an awesome deal, we just love it , it auto spins up instances on the fly of the zone when new client connects and after reaches players limit , then spins up another instance automatically, and the other way around when the last player leaves the zone/instance uzone automatically destroys the instance to save resources. then have the zone to intercommunicate with other zone in other datacenter or hardware node to handover player sleeplessly, i just love this deal.

    yes indeed you need unity pro to build a headless server but not really necessary tho, just get a good hardware node, have a zone and instances of that Zone per hardware node and spread the game all over the datacenter and youll be growing without any lag or bottlenecks.

    i mean i really wanted to port multiverse to use unity client but then looking at ulink, then i found no reason to invest time on multiverse server since there is already something working great.

    but i think if NeoJack comes with something simular to how multiverse used to escalate the server thru nodes and have the world seamless communicating to other hardware nodes, and the server side physics and server side collision then ill bet it will be a winner.
    if not it may just fall in to another smartfox type plugin, but have to remember that smartfox is very well polished and powerfull jave server since its has been in development for so many years already.
    so Neojack needs to come up with something new as incentive like server side physics/collisions
    and server scalability if they want to survive because of the existing competitors.

    i think they need to forget about the complex housing deal they working on and focus on the main server infrastructure to make it stand out from existing polished competitors.

    just my .02
    Last edited by CaptainExtra300; 08-03-2013 at 02:25 PM.


  11. Posts
    177
    I think Neojac is presenting a different product to the Unity community than the other server options out there. As powerful as uLink is, and despite all the features it has, it is still something that requires a lot of programming and development to act as a full-blown MMO server. Certainly, a small indie team can take the bones of uLink and the rest of UnityPark Suite and more easily build out an MMO solution than, say, SmartFox or Photon. But you'll still need someone who is skilled at programming to make that happen.

    Neojac's product stands to provide an MMO solution that requires no programming at all. The tradeoff, obviously, is in the flexibility of the MMO game one can make if he/she cannot or chooses not to customize the game via coding/scripting. You'll be "limited" to the gaming systems that Neojac offers. I would venture that such a solution is not what you (CaptainExtra300) would find as useful, since it sounds like you have a team that can build an MMO without that much handholding. For someone like me, however, the prospect of an out-of-the-box, no-programming-required MMO solution is quite compelling. I'm not a great programmer (changed my major after struggling with computer science for two years lol).

    Also, with respect to the housing part, remember that Neojac is building their own MMO to launch alongside the kit, and the modular housing system was a feature that was asked for by their player audience. So Neojac is bundling that functionality in with the kit, since it is something they have already established for their own game (which hypothetically uses the kit or some version of it, I would hope ).


  12. Posts
    130
    yes, you right,
    for non programmers this will be ok, but like you said they will be limited to the features NeoJack offers only, so no future for them, since they "NeoJack" are building their own, so none programmers wont be able to compete with them since they are the actual programmers. and your game will be much the same cookie cutter as their.

    so still needs someone to learn or at least a little coding to change or adjust the gameplay to be a little bit unique.

    but like i mentioned before, NeoJack needs a little bit more than what they offer since there is one mmokit (Zerano multiplayer kit) already and unfortunately did not do that good, hence the developer stopped the developing for a reason no future and was using the same principle, unity and smartfox java server.

    but ill, guess they have a plan, one is just stating some facts so they can at least learn from other people's experience/loses and wont fall into the same path.

    hopefully they 'll succeed and come out with something unique that people really need.

    P.S
    Hell, ill even drop ulink on the spot and get a copy of this kit if they were to manage to add server side collision/physics, encrypted network and find a way to make the java server as scalable as multiverse since that was my main intention initially but did not have the time to start such undertaking job port process. but if they can come up with a solution , ill get it with out a hesitation, i would not even care for the actual gameplay, ill code my own, but the actual port and server side physics is the most important thing atm for us.
    Last edited by CaptainExtra300; 08-03-2013 at 03:13 PM.


  13. Posts
    177
    Not all of us are looking to make a revolutionary, massively popular MMO. I know that such an undertaking is outside of any single hobbyist's capabilities like my own, and probably even the vast majority of smaller indie teams. I do think the fact that Python scripting is supported seems promising. Python seems easier for me to pick up than C#.

    To be fair, I don't think we know everything that Neojac has already developed for the kit, nor do we know everything that they have planned for it before launch either (which is slated for sometime next year). I think you've pointed out some good things to lookout for, but at the same time, I don't think you can definitively say their product is lacking just yet, since none of us have seen more than the handful of videos, screenshots, and posts on this thread. Let's wait until we see more of the product before rendering judgment on whether it will be successful.


  14. Posts
    130
    I see you point.

    i have been there before so i understand you, in my early years of wannabe multiplayer programming, i just wanted a startup/push something already made multiplayer kit /no programming required to play with my friends, but at the end you will see that that is not enough, you will feel very frustrated not being able to do your own stuff and waiting for the actual kit developer to come up with something for you to show off to your friends.

    believe me been there and done that loooong time ago. it never works that way you think you look for.

    I will strongly recommend you to start with basic stuff, single player game programming, then when you get acquainted with programming then start to go multiplayer , you will feel free and can do whatever you want.

    if they are taking the same approach like mulltiverse server had which they mentioned, which so far seems kinda identical, then is no walk in the park, since you need to learn not just 1 but 3 coding languages, c# for the unity default client, IronPython and not just regular (python) which is to me a little bit more complex than normal python because is a mixture of java and python and this is for client to server game logic , plus you need to learn java for the server side authoritative game logic which is the most important in an mmo.

    Ill give you an example demo of how i think this mmo kit be like incase you really want to know how is going to be when they release it.

    well at least youll have the feel of the server side and its backend infrastructure.

    go to http://www.multiversemmo.com or to he wiki http://www.multiversemmo.com/wiki/ and download the open source c# client and java server and install it , there are lots of great game logic examples in their wiki and forums, to add to it since they already have a demo world with most basic mmo fantasy rpg game logic scripted, but you will get the feeling on how this kit be like when they release, offcourse with the unity client instead of the C# client multiverse come with.

    i mean dont get me wrong here, if this kit was close to multiverse server , ill strongly recommend it but not for noobs, you will bang your head to the wall, so if you really looking forward to this kit, go download multiverse and install the server and play with it, get acquainted with it and try to learn the 3 languages, c#, ironpython and java so youll get the best of it, and ill assure you youll love it, but you still need coding skills requirement.

    I wish NeoJack could tell me is plain Multiverse server integrated with unity port and with server side physics/collissions and scalable server with encrypted networking and ill be the first one to jump in and pay the first copy.
    that should tell you how good multiverse network is, but if no server side collsions/physics and no server scallability nor encrypted network communication, then might aswell just go for SmartFox server which youll get more polished server backend.
    Last edited by CaptainExtra300; 08-03-2013 at 03:57 PM.


  15. Posts
    177
    I poked around with Multiverse a few months ago. The server side was easier to follow, although again, I'm not a good programmer, so it was still far too complex for me. The client, however, did not seem great. I have a feeling Neojac's server has some parallels to Multiverse, since that is what they started out using, but it sounds like it's their own engine now--even if it's written in Java still.

    I understand and appreciate your thoughts, CaptainExtra300, but it's clear that you and I are different people with different goals and interests when it comes to game development. What you've said largely seems applicable to your tastes and requirements; I don't think they apply to mine. I personally am not interested in trying to do any heavy coding on my own. Coding is not my strength by any means, and it would take me a lot longer and cause me a lot more frustration to code on my own. I would much rather pay someone with the knowledge and expertise to customize an MMO engine to my requirements. I will very likely still look into that option once Neojac's product is available, as I'm sure there are things I will want to do with the kit that will require coding beyond my meager capabilities.

    I would recommend you wait until August 12th, which is when there is supposed to be an announcement made about the MMO kit.


  16. Location
    Canada, Alberta, Calgary
    Posts
    415
    Good to see some talk about the MMO kit, and yes we don't announce everything we have as then there would be nothing to announce after that hehe. We will be catering for both parties so if you can code you will be able to code your own things in and if you don't then just use our pre-set. With Other people coding systems they will be able to add in their code in forms of plug-ins and maybe even sell them....... Hmm wait now I'm saying to much!!


  17. Posts
    130
    yup, now you saying too much, lol joking

    but you get me a multiverse like server with server side physics/collisions plus server scallable infrastructure + network encryption,
    and i think this kit not only will be the attention for noobs or kids that want to make the next WoW, but to some of us already coders that do not want to start from scratch and like the multiverse coding style or had the multiverse background and with the unity as client will be a good startup for all of us.

    then ill be the first to knock in your online shop's door to get me a copy.
    i dont need to extra code/game logic, just the above

    i just love java/ironpython and c# combination since am very acquainted with multiverse engine

    cheers


  18. Location
    Istanbul, Turkey
    Posts
    85
    Hello, first of all nice going there.

    May i ask where did you get Multiverse models? I'm developing a character system and that needs customizing and this models will be awesome for prototyping?

    Thanks


  19. Location
    Canada, Alberta, Calgary
    Posts
    415
    Only thing we had that was a multiverse model was a spawning marker and that was 4 years ago lol. What models are you talking about, if its the one's currently in the game those are from unity assets store we used for prototyping.


  20. Location
    Canada, Alberta, Calgary
    Posts
    415
    We will be doing a Kickstarter for our game and one of the rewards will be alpha and beta access to the MMO construction Kit for Unity. Feel free to check for updates on the Kickstarter and a link to it next week or two.
    Last edited by cookimage; 08-18-2013 at 05:42 PM.

Page 6 of 36 FirstFirst ... 4567816 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •