Announcement

Collapse
No announcement yet.

O3DE Game Engine Quickly Settling Its Linux Support

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #11
    Originally posted by rmfx View Post

    A full featured engine, that can handle AAA games, multiplatform and free, offering a great alternative to Unreal and Unity...
    Yeah who would want that ??? Let's complain with you.
    Lumberyard and CryEngine aren't exactly new, yet how many games use those?

    O3DE changes exactly nothing for game developers. They don't give a fuck about Linux and I can't blame them. The best O3DE can hope for is some adoption from developers targeting Stadia or SteamDeck, but Stadia is usually just one of the targets, and noone sane will develop a game for SteamDeck while ignoring Nintendo Switch.

    Comment


    • #12
      Originally posted by chocolate View Post
      Not so stellar C++ code quality at the moment. Time will tell.
      https://www.youtube.com/watch?v=IndGlm2uZCU
      It seems to require certain massive code cleaning and refactoring, IMHO.

      Comment


      • #13
        Originally posted by chocolate View Post
        Not so stellar C++ code quality at the moment. Time will tell.
        https://www.youtube.com/watch?v=IndGlm2uZCU
        Originally posted by timofonic View Post

        It seems to require certain massive code cleaning and refactoring, IMHO.
        Should read the pinned comment discussing this very issue. I'll quote parts of the comments from an O3DE dev on the matter:

        Originally posted by Tom Hulton-Harrop
        We still are carrying quite a lot of legacy code that isn't used anywhere these days (there's an ongoing effort to keep removing it). I think the first 12 minutes of the video were looking at legacy Cry code that is not used or actively maintained.

        I'd definitely recommend trying out a tool like Code Scene (https://codescene.io/) that shows you recently modified files to get an idea for what's under active development and changing now (Code Scene came out of a great book called 'Your Code as a Crime Scene' by Adam Tornhill which introduced a precursor to it called code-maat. A great book and very interesting tool to help identify high impact code.

        ---

        We're still wholesale removing whole files and directories that are no longer used. I believe the Content SIG will have more info on specifics but there's still lots to go 🙂 We have some metrics of how much code has been 'redcoded' (due to the color of the diffs) and it's hundreds of thousands of lines if I'm not mistaken.
        I wouldn't doubt that in addition to the legacy cleanup, the project is also looking at refactoring and overhauls where they are needed.

        Cheers,
        Mike

        Comment


        • #14
          Originally posted by mroche View Post
          I wouldn't doubt that in addition to the legacy cleanup, the project is also looking at refactoring and overhauls where they are needed.
          I guess the Star Citizen engine team could tell them a lot about that. Whether they will tell them a lot about that I don't know.

          Comment


          • #15
            kfriesth
            Junior Member
            kfriesth
            "Personally I am sticking with Godot and waiting for 4.0."

            Waiting for Godot? Ha ha ha. Couldn't resist.

            Comment


            • #16
              Originally posted by mroche View Post
              I wouldn't doubt that in addition to the legacy cleanup, the project is also looking at refactoring and overhauls where they are needed.

              Cheers,
              Mike
              Thank you kindly, I have updated my previous comment!

              Comment


              • #17
                Originally posted by uid313 View Post
                I wonder if anyone is going to do anything with this engine.
                I wonder if Amazon intends to continue the developement, or if they just to cut costs and have other people maintain it.
                AFAIK Linux Foundation only touchstuff when they are handsomely funded for it. (that i partly why they dont touch desktop linux)

                If Amazon wanted to dump this engine they could have without open source-ing it, or by making it an Apache Project.

                Handing it over to the Linux Foundation (even though it didnt support linux at the time) suggests there is a market and there is funding. Whether they can capitalise on this though is a different question

                Comment


                • #18
                  Originally posted by uid313 View Post
                  With id Tech 4 (the Doom 3 engine), nothing really happened with it. It was open sourced, but nobody really made any game with it
                  I think there were two big problems with Tech 4 from that perspective: mostly, that it was the first engine that required full DBS(I) texture sets, which were beyond most amateurs at the time and/or also required a lot more work than simple texturing; but it was also the first id engine predominantly written in C++, and I think that was a major stumbling block for a lot of hobbyists.
                  It also had significantly higher hardware demands than Tech 3, which put a huge number of people in a position of not really being able to run anything using it at acceptable framerates. You need the playerbase first, both to have a talent pool to draw devs/artists/etc from, and for that team to also have an audience for their work.

                  Fundamentally though, the biggest issue was simply that the console market at the time was exploding in popularity, while the PC gaming market was "dying" (yet again :P). With that added to the other issues, by the time the Tech 4 code was published there were other targets available for hobbyist teams to work with, D3 and Q4 were distant memories, and those newer engines were the new hotness.

                  Comment


                  • #19
                    Originally posted by schmidtbag View Post
                    Closed-source is often clean because usually, someone is getting paid to do it and any product worth buying was developed by someone who knows how to program properly. But, it can be messy because code behind closed doors only has to make sense to the creators and nobody else.
                    I would bet you will find more messy code in closed source than clean code.
                    Working with close deadlines, without code review, probably with nothing else than a few manual tests before pushing has been common in the few companies I worked with.
                    I would say clean code is more the exception behind the doors if you don't work on stuff than needs to be optimized or is security critical.

                    Comment


                    • #20
                      Originally posted by nabero View Post

                      I would bet you will find more messy code in closed source than clean code.
                      Working with close deadlines, without code review, probably with nothing else than a few manual tests before pushing has been common in the few companies I worked with.
                      I would say clean code is more the exception behind the doors if you don't work on stuff than needs to be optimized or is security critical.
                      This is the case even in famous engineering firms under contract with the government and the military somewhere in Europe, but you didn't hear this from me.
                      If you have OCD for clean code and best practices, you'll most likely leave in disgust. I'm totally convinced the same goes for most closed source out there.

                      Comment

                      Working...
                      X