No announcement yet.

Advice on a short but large project

  • Filter
  • Time
  • Show
Clear All
new posts

    Advice on a short but large project

    Hi, I’m new here so sorry if this is in the wrong section! Basically I’m looking for a bit of advice on a project me and two other people are running at the University of Bolton (UoB). It’s a student run project and we’re basically trying to make a game in a week, with around 40 people. These are preliminary numbers from a meeting we had to today, and the split goes around 16 artists, 16 designers (many can do art assets too), 6 programmers (limited unrealscript usage, amazing with C++) and a couple of sound designers. Now we’re all splitting into a hierarchy, groups of four working on each section with a lead overviewing a couple of groups at a time and then three guys at the top, so were organised from the human resources side. The game idea is being created by the designers as we speak and I’m the lead designer just trying to gauge what were capable of.

    So basically I have a few questions about what avid UDK users think we can achieve and for any advice on the project. So a couple of questions, If we were to say make a racing game, what sort of size of racing game do you think we could make if the programmers scripted it from scratch and all the assets used were only our own? Most of the level designers can use the UDK to a decent level and understand the standard things about the UDK, many also have specialist skills (but I won’t go into that)

    Another question is have you got any advice on things such as managing such a large team effectively; any techniques you would use or tools like file management tools? (We will be using Agile ‘scrum’) or just any computer programs you may think would help us.

    Other than that just anything else we should really consider before we start the week in June. We’re going to have a solid design document down by the start of the week and prototyped a couple of the high priority mechanics too. Any comments or advice would be amazing thanks!

    > Reniksuk

    Edit : Apologies if that's too vague, I can give more details about the project if you want

    I think a racing project could be done by a group of 40 in a week quite well.

    Just plan it out really well and figure in reasonable goals. And always imagine that things will take longer than you expect.


      Racing was just an example but thanks. I've been told that about the timings a lot too thanks, I just need to bring the everyone else's ambitions back down to earth. Any other comments would be great, we just want to make sure we consider all the bases before we start. My main concern is the programmers not knowing UnrealScript, are their C++ skills quite transferable? I know a few designers understand Javascript too and have done quite a bit of scripting in Unity, would this be of any help when game play programming?


        With good C++ skills you can work quite well with Unreal Script. Functions, Variables, loops and stuff like that are very similar to C++. I have learned that years ago and in couple of days you can get a well understanding of how the Engine works. If you want to create a racing game look at the gems:

        In that case you don't need any scripting, just some game logic with Kismet.

        If you want to create a game in one week, my advice are to have a playable version in the end of every day. Start with some basic stuff (simple Car-Movement/Physics), create some simple cars and one or two basic tracks to test. The second day create some basic waypoints, stuff for your tracks (ramps, side objects), and move on. In my opinion it's very important to sit down together at the end of the day, play your game and make some notes what can be done better or have to be changed.

        Management tools for one week is a little overhead for me. Daily meetings and a place to share your project will be enough. If you have a web server to share your project, install a SVN-System. Every change on your project are an extra version, so you can easily work at the same time on different tasks. But a dropbox or something like that are also okay.


          Thanks, the makes me feel a lot better about the programmers productivity. We would really want to avoid using something like that starter kit if possible, our main objective is to create a game, but our other objective is to get all the students involved some decent work for their portfolio. So programmers will most probably wish to design their own game type and character controllers etc. but that starter kit is something we can consider if we're having problems thanks. I've never used a SVN-system before, but we do have about 15gb of shared space on a network, I'll be backing it up ever hour or so and enforcing strict naming conventions, somebody mentioned 'assembla' to me so we are looking into any free versions of that to.

          Great advice about the builds at the end of the day, that something we will definitely look to do, somebody has put themselves forward to manage a QA team during the week too (anybody from the team who is doing little can be testing), a lecturer mentioned a good method of running meetings. Having a meeting during the day for 15mins where you say only positive things about the project, what's going well and just get positive ideas flowing, then having 15mins where everybody can moan and rant and let off some steam and bring some issues to up to the forefront.

          Any other ideas other things we should really consider? we really appreciate the advice! We will be uploading developer diaries at the end of each day too so if anybody is interested in how the project is turning out we can post it in the 'Works in progress' section.


            Well, your biggest concern is Unrealscript, so I would most likely do a game that extends UT and do some custom stuff with that


              Yeah that's seeming the most likely scenario at the moment, a FPS with boss fights now comes up more often than not, as it just extends on the standard UT stuff with some extra stuff of our own. But 5 very good programmers and about 4 designers that can do scripting should be able to put something good out. Has anybody heard of a project like this before? Such a large team in such a small time?


                The big amount of programmers might not necessary mean you can do it on time, if even one of them encounters a complex situation that requires enough time. Though with fixed limit, I assume you'd just change the design rather than spend additional time on such tasks.
                On art/level design/etc stage, it all comes down to whether you've got a team/project workflow established. Pre-production is important.
                A lot still can be done even within such short amount of time, but it would require tons of micro-management, correct planning and good skill to get the art done with proper style (otherwise it might look like a bunch of separate assets that came from different sources). Which is why (as you may know) teams usually create quality/style examples for everyone to aim for, before full production starts.

                So, plan ahead and get all the tasks and tech/style requirements ready. It'll save plenty of time for 16+ guys, which will be beneficial to project quality and amount of content delivered.


                  Yeah, the design can easily be shifted to avoid major problems. This is why were on here too, pre-production, planning, we want to make sure we cover all the bases before we start in a few weeks. Layout and design work will be covered before so were clear on what we have to do each day, clear on things like level streaming and know what each person in each team of 4 has to be doing. I think my plans of doing design work are slimming here, I'll probably end up micro-managing all the design teams, but that's the price of the lead role I guess. As for the art side we have some very good concept artists who will be concepting everything on our priority asset list before we start, and then doing quick stuff on the fly. I actually didn't know that about the quality example, but I'm not the art lead for a reason, I'll pass that bit of information on!

                  The art lead is making sure all the artists are up to scratch and will be priotising assets, before the week starts so that the fast/good modelling and UV mappers know exactly what they have to do next. I'm assessing all my designers skills so we know more about what we can do, scaleform is some daunting UI stuff! Still looking for some file management software if anybody knows of any? Thanks again btw!