Jump to content

  • Log In with Google      Sign In   
  • Create Account


Steps to manage a large project


Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

  • You cannot reply to this topic
6 replies to this topic

#1 l46kok   Members   -  Reputation: 107

Like
0Likes
Like

Posted 26 October 2012 - 05:17 AM

Software development is an area where parallel development to its fullest form is very difficult to achieve, although you could get reasonably close with the right design.

That being said, if you are designing a game from scratch from engine to front end, what steps should be taken in order? How would you efficiently manage your project and your team?

I'm asking this because me and several people are interested in working on a relatively large project for learning purposes. Initially we were going to use a proprietary engine like Unity but since we wanted to learn how engine works, we're going to start from bottom. I'd appreciate any suggestions you guys can provide me.

Sponsor:

#2 Tom Sloper   Moderators   -  Reputation: 9450

Like
3Likes
Like

Posted 26 October 2012 - 09:13 AM

0. Write a business plan. Know the target audience, the competition, the monetization method. Figure out how many copies of the game have to be sold in order to make back the cost of development and make profit.
1. Start with a GDD. Know that the game is likely to go through changes as the project progresses.
2. GDD in hand, make a TDD. Identify the challenges, make a plan for dealing with them. Define the major features of the engine, and the major tools the team will need to make. Determine how many people will be needed and what their roles will be and what qualifications they should have. Define the filenaming convention. Decide on a version control system.
3. Create the art and audio lists from the GDD. Determine how many people will be needed. Determine what assets need to be delivered to the programming team, and when, in order that the programming team not be kept waiting for assets.
4. Determine the target release date and make a schedule.
5. Determine the budget. Obtain money.
6. Pre-production is now coming to a close and production is about to begin. First step is to make a Proof-Of-Concept demo. Build a representative vertical slice of the game that demonstrates the heart of the gameplay. You want to be sure the game is going to be fun!
7. Production phase: build the entire game from start to finish.
8. Post-production phase: get game tested, and fix bugs.
9. Launch game.
-- Tom Sloper
Sloperama Productions
Making games fun and getting them done.
www.sloperama.com

Please do not PM me. My email address is easy to find, but note that I do not give private advice.

#3 Orymus3   Crossbones+   -  Reputation: 7538

Like
1Likes
Like

Posted 26 October 2012 - 10:07 AM

I'd like to propose an alternative to item number 7.
This would imply a waterfall model where production would simply stick to the original GDD. I'm assuming you may want to iterate as you go along.
You may seek to adopt a more agile methodology and make development cycles where you update your documentation based on changes in your requirements. I would advise to deliver builds to testers along the way and not wait until step number 8.

Other than that though, this is a fairly straightforward model employed in project management.

#4 Tom Sloper   Moderators   -  Reputation: 9450

Like
1Likes
Like

Posted 26 October 2012 - 11:21 AM

I'd like to propose an alternative to item number 7.
This would imply a waterfall model where production would simply stick to the original GDD. I'm assuming you may want to iterate as you go along.


I never implied waterfall. You inferred it. Step 7 can be iterative and agile.
-- Tom Sloper
Sloperama Productions
Making games fun and getting them done.
www.sloperama.com

Please do not PM me. My email address is easy to find, but note that I do not give private advice.

#5 BlueFalconLoyd   Members   -  Reputation: 128

Like
1Likes
Like

Posted 16 November 2012 - 12:43 PM

My personal experience derives from that of the Enterprise level, and less of the Independent developer level. Having said that any project that's looking to be taken seriously starts with good project management. It's just as important as the development itself as, if executed correctly, can save the team time, money, and resources. Based on this, I'll list my preferences when managing a project as well as some things that may help you down the road.

  • Legalities: This may seem a mundane task when first venturing into game development, but as your project moves on, it will be one of the most important processes - and you'll be glad you covered it. A Non-Disclosure Agreement(NDA) should be signed by each of the members of the project. This states that intellectual property remains that of the project itself, and that trade secrets aren't lifted from the project and taken to another to profit from. You can find several examples of NDAs online for guidance on how these should be treated.

    It may also be beneficial, depending on your budget for the project, to seek legal aid before finishing up production on the product itself. You'll also want to cover compensation clauses with your team members so that they're fully aware of what they will gain by participating in the work that is going on in the project.

  • Feasibility - Since you have a team in place already, a feasibility assessment of where your team currently stands within it's technical abilities and what it hopes to gain, knowledge-wise, throughout the duration of the project will help you in determining just how complex the scope of the project can be. Finding out team members strengths and weaknesses will go far in deciding what type of game you want to create, what platform it will be on, what features it will include, etc.

    This will also help in allocating the correct resources for each task on your project schedule.

  • Project Scope/Market Research - After determining what's realistically feasible to do with the team you currently have, you'll want to start brainstorming what type of project you'll want to start. The Project Scope is a 1,000 foot bird's eye view of the Project itself. Planning what type of game, what codebase, what engine, what platform, key features, etc. is what will go into this. You'll also want to do some market research based on your ideas. Is there a want for this type of game?, how large is the market segment for this type of game on this platform?, Is this genre trending currently?, Will this be available on a local, national, or international level?. All of these will shape the Project Scope and will need to be addressed prior to fleshing out your project.

  • Budget - Even if you plan on creating the project with the bare-bone essentials when it comes to monetary value, most likely you will run into some portion of the project which requires expenditures. Whether it's software licenses, requests for team training, trademarking, legal fees, contracted help, etc. You'll want to make quoted figures on any parts of the project that may cost you in some way. This will allow you to foresee any events that may arise in which you otherwise would not be aware. Once you get into the development phase of the project you want to make sure that nothing road-blocks you from keeping on task and finishing the project. Budgets are usually the one road-block that can place a project on hold indefinitely.

  • Project Planning- My absolute favorite part of any project. Project Planning is where you make your money. Simply put, it's the process of managing your resources, time, and the overall health of the project by splitting the project up into "tasks". In doing so, everyone on the team understands what's going on currently with the project, what their particular job function is at the time, what everyone else's job function is, what's next on their list, what resources are currently tied up, and what will be the next steps after their task is finished.

    Remember in Programming 101 where your professor/teacher asked you to name the steps it takes, in detail, to create a PB&J sandwich. That's essentially what you're doing - but for a much more complicated tasks. You'll want to take your Project Scope that you created previously and start listing all the bits and bytes of what will need to be done from start to finish, then assign resources(aka your team) to those tasks. It's important to note that a good project plan will never have a team member sitting and waiting for something to do. Each team member will always be engaged with a task, and know what their next task will be in case they finish early.

    A good investment during this portion would be a Project Management application. Microsoft Project 2010 is what I currently use, though there are other options available if this is not in your budget. Some notable collaboration applications that can be found for online use are BaseCamp, Wunderkit, and TeamLab.

    I've included a screenshot of an example project I'm currently working on for work, to help showcase what I'm speaking of in the above paragraphs.
    http://s17.postimage...development.png

  • Development- I'm sure many will disagree with me here, but I do not believe in creating a Game Design Document prior to the Development Phase. This should be incorporated with the development phase as brain-storming, writing of the storyline from your writers, specific hardware specifications from your infrastructure team, etc. The Game Design Documents that I've viewed in the past are specific step-by-step of what every item is, what every spell is, etc prior to any development being done.

    I state this due to Game Development and Design being dynamic. Things WILL change in that design document and it's much better to parse this out at certain periods during the task/resource allocation of the Project Planning than to mess with it prior to initiation of development. Your Project Plan should be just as agile as your design of the game. In doing so, making sure that you've not only allocated the correct amount of time for each task, but that you've extended that by a certain % to allow for changes within the plan itself is imperative to this type of design.

    Before you know it, you'll be managing your project like a pro - and when something unexpected comes up you'll be able to calculate to a "T" what it will take to get the project back on track and being able to positively mark what, if any, resources you've lost.

    That being said, the development process should fall in line with what's currently being tasked to your team members, and therefore, with the technical knowledge shouldn't be too hard to discern what needs to be done.

  • Testing/Pre-Production- This should be handled as a "running" list throughout the development phase. During development make sure you keep a list of all known bugs and process improvements that you've found. It's best, in my opinion, to leave bugs that do not directly effect the functions of the core code till this phase to be dealt with. This will better keep the project on schedule and allocated time and resources can be managed during pre-production to take care of these oddities.

    It's also best to reassess your current product at this point and see how in-line with your original Project Scope/idea that it is. Quality-wise, is this the product you intended to create and does it feel as though, based on your market research, that it's something that's appealing to others?. A complete re-evaluation with the whole team will allow for a better understanding of your product after the end-rushes of the development cycle.

  • Production - Releasing a game product for production has so many variables that a whole book could probably be written about it. By what means are you going to release the game? How will advertising be done? How will continued support and updating of the game be done? Are investors/publishers involved at this point? Will you keep all of your staff, or are their resources that are unneeded at this point?
-------------

Most of the ideas listed above probably have conflicting interests with the game development community as a whole, and I'll restate that this is all from my own experience. Some suggestions that aren't included, to keep up with your team, during the project but not necessarily pertaining to the product itself I've listed below.
  • Continue to reassess your project plan on a daily basis. A good project coordinator will know when he should reallocate resources on an over-allocated task. It will also keep your plan up-to-date.
  • Plan training events or extra-circular activities for your team. There are many great sites out there for learning, and learning together as a team will not only keep them motivated but help to define your resources as a whole. Such websites come to mind as www.coursera.org, www.udemy.com, www.learni.st, www.scribd.com, www.safaribooksonline.com. - All can be valuable resources. If someone on your team learns how to do something new that may be beneficial to the team as a whole, ask them to create a presentation to teach the rest of the team.
  • Don't panic! If your project turns a complete 360 for varying reasons, remember that you've planned well and with your management of the team and your team's drive, you can accomplish any obstacle.
Anyways, just my two cents.

Edited by BlueFalconLoyd, 16 November 2012 - 12:54 PM.

~Medical Interface Developer and Project Coordinator~


#6 Tom Sloper   Moderators   -  Reputation: 9450

Like
4Likes
Like

Posted 16 November 2012 - 03:38 PM

I do not believe in creating a Game Design Document prior to the Development Phase.


But before you design the game, you plan the project?

Project Planning- ...Simply put, it's the process of managing your resources, time, and the overall health of the project by splitting the project up into "tasks".


How do you know what tasks are needed if the game hasn't been brainstormed and designed yet?
-- Tom Sloper
Sloperama Productions
Making games fun and getting them done.
www.sloperama.com

Please do not PM me. My email address is easy to find, but note that I do not give private advice.

#7 heavycat   Members   -  Reputation: 383

Like
1Likes
Like

Posted 07 December 2012 - 02:20 AM

Keep it simple. Every feature adds exponentially to the implementation effort required, then also adds to the support costs. If in doubt, drop it. Always ALWAYS better to have simple working than complex on a whiteboard.




Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.



PARTNERS