Search Unity

So, UNET is basically useless and abandoned?

Discussion in 'Connected Games' started by Artaani, Jul 25, 2017.

?

What do you think, UNET is ready for serious projects?

  1. Yes, UNET is much better than alternatives and very stable

    7 vote(s)
    30.4%
  2. No, UNET have many serious problems and not ready for production. Alternative solutions should be us

    16 vote(s)
    69.6%
  1. Artaani

    Artaani

    Joined:
    Aug 5, 2012
    Posts:
    403
    I'm tried to work with UNET many times during its development.
    Developers of UNET forces us to use UNET, but it is still have a lot of critical bugs and flaws. Very bad tutorials, documentation and overall usability.

    And seems like developers does not read feedback from the forum during long time.

    It is sad.

    Someone even use UNET for serious project?
    I know UNET used in '7 days to die", but I don't have an idea how they managed to do it.
    Working with UNET it is pain, even with the most simpler tasks which can be implemented easily using old Unity network.

    I want to prepare a full list of thing which should be fixed in UNET with maximum priority, but I' looking on this forum and afraid that no one even will not read this.
     
    Romenics likes this.
  2. Artaani

    Artaani

    Joined:
    Aug 5, 2012
    Posts:
    403
  3. TwoTen

    TwoTen

    Joined:
    May 25, 2016
    Posts:
    1,134
    https://blogs.unity3d.com/2017/03/17/update-on-unity-multiplayer-current-and-future/
    This blog post was released a bit ago.
    But I know what you mean, it really feels slow. But the bitbucket repositories is getting updates now and again. I guess they just have alot of internal priorities. Can't really see UNET being the biggest one.

    And as for feedback. There is a guy that's really cool and listening to feedback in the Networking section.
    @aabramychev
     
    Artaani likes this.
  4. pKallv

    pKallv

    Joined:
    Mar 2, 2014
    Posts:
    772
    This is, from my point of view, the biggest problem ...to get help when needed.
     
  5. TwoTen

    TwoTen

    Joined:
    May 25, 2016
    Posts:
    1,134
    I see the strong issue. But really, Unity probably has a lot bigger priorities.
     
  6. robochase

    robochase

    Joined:
    Mar 1, 2014
    Posts:
    213
    what's a 'serious' project?
     
  7. Artaani

    Artaani

    Joined:
    Aug 5, 2012
    Posts:
    403
    I mean, commercial, ready projects (not a test projects).

    Thanks for replies and links.
     
  8. aabramychev

    aabramychev

    Unity Technologies

    Joined:
    Jul 17, 2012
    Posts:
    574
    I am not sure what i need to comment. OK, i'm trying
    1. When we started we understood that we cannot satisfy all our customers with one library, first of all because requirements from different customer groups are different. Newcomers want to make network game immediately and transparently (by pressing couple of buttons and left magic to unity), experts want access to raw sockets. The decision was made, to implement foundation, (without significant care how difficult api will be) + different libraries on top of them. HLAPI was implemented (and I repeated that here on during all unites where I presented) as EXAMPLE how it can be implemented. As HLAPI was develop in the same time when LLAPI implemented, some features are duplicated (if same feature was not ready yet in llapi, it was implemented in hlapi and do not block development). But again it is just example, we open source code so everybody can use it change it and so on.
    2. We did hope that community will create different library on top of foundation and publish them. We was wrong :( Nobody want to do this.
    3. We did hope that community will help us to find the bugs during pre-alpha alpha and beta testing. For new transport we published and keep in-sync special Unity build almost for a year. Very limited number of developers tried it and helped find and fix the bugs (thanks guys, i appreciate your help). But again - it was very limited amount.
    So? we are not wizards (and I'm sorry about), we just try to do our best.

    Sincerely yours
    Alex
     
    PhilSA, Artaani, Romenics and 5 others like this.
  9. marcV2g

    marcV2g

    Joined:
    Jan 11, 2016
    Posts:
    115
    Poor docs and reliance on post compile magic make this very difficult.
     
  10. robochase

    robochase

    Joined:
    Mar 1, 2014
    Posts:
    213
    AFAIK, The code weaver stuff is only used by the HLAPI and wouldn't interfere with building a custom library on top of the LLAPI.

    I think the main reason people aren't building their own interpretations of the HLAPI is because it's a pretty hard thing to do well. The vast majority of people using unity want to make games, not build networking libraries.

    Personally I ended up getting my hands super dirty digging around in the HLAPI code. While i would call it a worthwhile experience, at the end of the day it was a major setback for my project. I would have rather spent that time making my game more awesome instead of working on plumbing issues the player will never see.
     
    recon0303 likes this.
  11. aabramychev

    aabramychev

    Unity Technologies

    Joined:
    Jul 17, 2012
    Posts:
    574
    @robochase, @marcV2g understood... Cannot promise that we will create that fast, but we will try... Anyway, according llapi you can always ask me, and as I know that or documentation is hmm, "not good" I'm ready to answer for any question related this.
     
  12. vis2k

    vis2k

    Joined:
    Sep 4, 2015
    Posts:
    3,220