Building A Game In XNA Part 3

   16/02/2009 at 07:36       Nick H       4 COMMENTS.
The goals for my third week on the game were to put in the last of my planned obstacles, to start producing levels, and get a graphic style finalised.

A problem I've often heard mentioned in game developer blogs and interviews, particularly developers of projects that experienced delays, is that of feature creep. I found myself experiencing this first hand this week as I began putting in what I thought would be the last obstacle into the game. While I've been building the game and thinking about it away from the PC I've been trying to visualise levels that will be made with the obstacles I'm putting in. Each new obstacle offers a new set of opportunities which often give me ideas for new obstacles to increase the options available when creating levels. I've been struggling with the conflict between the wish to make the game fun and varied whilst at the same time being aware I need to finish it. I could keep adding new objects to the game probably indefinitely, and never finish the game, so around Wednesday decided to draw a line under adding new content with one additional object having turned into five. Even as I write this there's a new object type I'm fighting myself not to include.

My second goal was to get started on creating levels for the game. I've built the level system along the lines of what I used in the n-gage demo that the bear game is based on. It reads in a string of letters with each letter representing a type of tile e.g.

B B B B B B B B E E B X E B B E E U E E B B E E H E E B B E E R E E B B E E B H H B B B B B B B B ,1,1,7,7

Each of the "B"s is a brick wall, "E" being an empty square etc. The numbers separated by commas are the start square and the level dimensions which I intend to change as the game progresses. The above code gives a level that looks like this.
AllAboutTheGames screenshots

It's a little hard to visualise the level in text form, particularly now my objects have become so numerous that their letters don't really correspond with their names, "M" being a pressure pad switch for example. To make it easier I thought it best to build a level editor that could spit out the level strings from a more WYSIWYG input method. I've found that building a few tools right early on can really speed up content creation when making flash games, so put my time in the second half of the week into getting an editor built in flash (where I'm most comfortable).
AllAboutTheGames screenshots

Editor built I started making a few levels starting with a few basic tutorial like levels. It didn't take long to knock out 5 very simple tutorial levels showing all the different tile types I'd come up with. Having built them though I started thinking about the best way to introduce the player to the different tiles available in the game. Is it in the best interest of the player to show everything up front before the game proper starts or gradually introduce them as the game progresses letting the player get comfortable with tiles and their rules before introducing the new tiles that freshen things up. I've not made up my mind completely on this, though I'm leaning more towards gradually introducing new toys for the player as they progress acting as rewards for success as well as keeping things interesting. I've also got the demo version in the back of my mind, which will need to explain the rules and give a demonstration of the different stages of the game.

The final goal for the week, deciding on a graphic style, is where the least progress was made. We made a few broad decisions about the style, a more arty impressionistic look than the current sprites I'm using. We've been collecting up examples of styles we found interesting and talking about which bits could succeed in a character design. It's a challenge thinking what kind of look to target, the hope being you can achieve half the success of something like Little Big Planet's graphic style which treads the line between cute and cool fantastically well appealing to a large number of people. One gameplay related design decision we did make is that the character will have something that will show which way the character will turn when he walks into a wall, he always turns right if he can but this isn't always easy to guess when the character is walking in a direction other than directly up.

Next week will hopefully see the graphic style come together, and I'll hopefully get a good chunk of levels built.

Helpful Links:

XNA web site -

XNA getting started -

Microsoft Visual C# -

XNA game studio 3.0 (Installs into MS Visual C #) - Here

XNA Beginners' Guides -

C# Tutorials I found handy -

Part 1 of the XNA diary - Part 1

Part 2 of the XNA diary - Part 2

Associated screenshots.

User Comments:

You must sign up for an AATG account and login in order to post comments

nekotcha - on 16/02/2009 at 10:31 wrote:
Keep up the good work chap, I'm really enjoying reading these articles. Fair play for keeping a grasp on feature creep too - that's probably one of the hardest things to do when making a game. Personally I'm a big advocate of a small number of devices and features intelligently implemented than a broad swathe of stuff that's only half-finished. Things like Flower, for instance.

Oh, and I definitely agree that new gameplay features are best gradually introduced to the player rather than thrown at them all at the start. My approach on stuff like that is always the same - do what Nintendo do. :)

Tiger_Walts - on 16/02/2009 at 11:33 wrote:
I agree with keeping the specialist tiles few in number. Keep the features that you can later combine to create depth rather than giving the player something alien to contemplate.

Combining two, or more, things that the player is comfortable with smooths out the learning curve.

Give this a quick read. It'll get you thinking about how you present elements and puzzles, in both pacing and representation.

Trip SkyWay - on 16/02/2009 at 12:38 wrote:
Thanks for the ideas and the interesting link fellas. Good to hear other peoples opinions on things when your trying to see the best way forwards.

eviltobz - on 17/02/2009 at 19:30 wrote:
the thing about controlling the feature creep is that it gives you a starting point for expanding the game at a later date, either as an add-on or sequel. if you give a little consideration to such things now, you can try to make sure that your code will be able to cope with the sort of changes you have in mind further down the road, even if you don't implement them from the start.

4 comment(s) in total.
Call of Cthulhu Review (0) (08/11)
Fishy Goings On
Red Dead Redemption 2 Review (0) (07/11)
Wicky Wicky Wild Wild West
WWE 2K19 Review (0) (01/11)
Push Tyler Breeze
Vroom Kaboom Review (0) (19/10)
Old Banger
Call of Cthulhu Review (0) (08/11)
Fishy Goings On
Red Dead Redemption 2 Review (0) (07/11)
Wicky Wicky Wild Wild West
WWE 2K19 Review (0) (01/11)
Push Tyler Breeze
Vroom Kaboom Review (0) (19/10)
Old Banger
Hearthstone Beta Review - Gather Round (1) (24/02)
Gather Round Blizzard's Latest Cash-Cow
Dream: The Hypersloth Interview - Part 3 (0) (18/11)
Fears before bedtime
Dream: The Hypersloth Interview - Part 2 (2) (17/11)
I can't get no sleep...
Angband Retrospective (4) (25/10)
Learn to fear the letter D.
Why TitanFall is no Call of Duty Killer (5) (18/02)
But it's still bloody marvellous
Gioteck's PS3 Controller Range - Preview (0) (01/08)
Striving for excellence
Havin' a Go: Mass Effect 3 Multiplayer (4) (27/10)
Don't cry wolf just yet...
Shank 2 Set to Kick Your Ass in Early 2012 (0) (26/10)
Not actually a protein. Go on, google it.

Better late than never, eh Ror?
Khanivor - In response to: Battle Chasers: Nightwar Review - 217day(s) ago.
Enjoyed this, cheers!
evilashchris - In response to: Reflecting on the Life of a Tomb Raider - 302day(s) ago.
Looks who's back. Shady's back.
GOD - In response to: Sniper Elite 4 Review - Xbox One - 632day(s) ago.
Micro Machines was my favourite!
ClaytonNotClive - In response to: Mantis Burn Racing Review - 674day(s) ago.
i agree chris, the Aliens table makes the others look bad.. because its so goood!! but they arent that bad.. haha! ...
neosalad - In response to: Aliens Vs. Pinball Review - 935day(s) ago.