So You Want To Make Video Games...
by Paul Dana · in General Discussion · 08/17/2005 (6:46 pm) · 16 replies
Flash Bios Marketing art by the very fine Nauris Krauze.There is a right way and wrong way to make video games
Flash Bios is a 3D Arcade Strategy game by Plastic Games. This game has been in development for two and a half years now and we are happy to announce that the Open Beta is now in progress!
If you want to beta test, just Sign Up and download the game. When you get done playing you will be directed to a questionnaire web page asking you 10 simple questions about the game. When you are done you can download more versions with more maps to test.
This forum thread is for discussing the practical side of making video games, using our experiences making Flash Bios as a real world exam[ple. If you want to talk about the game itself please post over in The Other Thread which is dedicated to that.
If you want to rant and rave about making video games, post here!
Start by reading my series of blogs, The Long Strange Trip. They are like a "making of" documentary. They give the account of the mistakes we made while learning how to make video games the hard way: by doing it wrong over and over until we learned to do it right.
Joe Maruschak encouraged me to write these blogs because so few projects ever make it as far as Flash Bios has. The most common reason is inexperienced teams fail is they do not know what they are getting into. Joe has seen so many game projects fail for the same reasons over and over. He wanted me to write about the specific hurdles we had to get over in order to get where we are today, and outline all the things we did wrong before we learned how to them right.
Use this forum thread to talk about these blogs, agree with us, disagree with us, and relate your own experiences learning the ropes of game development. We feel there is a definite right way and wrong to make games. Join the discussion. Tell us what you think!
#2
Currently in Air Ace, the physics model for the flight dynamics of the planes isnt complete.
I'm aware that this is perhaps 95% of what will make Air Ace a hit or a miss. But the physics IS underway, by a guy who knows physics code well. However its definitely slow going.
Now the developer/designer side of me KNOWS that this is a fundamental aspect of the game and that we shouldnt do anything else UNTIL this is sorted out.
But the producer/manager side of me says "ok, we know its work in progress, so we should just accept that it will come and continue with other stuff until its there".
But which side of me is right? To sit there waiting for the flight model to complete means no other progress can be made? Even though we know that without this being right, the game simply wont fly (pun intended).
I think its basically an act of faith that we have to put in our co-workers. That it WILL be done and that we CAN refine it into the model we want.
It helps that we have a very clear goal of how it is meant to feel, plus I personally have as Joe and Paul both mentioned, a very clear mental picture of the target audience.
Actually, while I'm on that subject, I should mention that its normally a good idea to have that mental picture of the target audience, but its also maybe a good idea to BE that target audience too. Having empathy for your own game audience by being a member of it.
My audience is very much like the thinktanks audience. Hardcore gamers who simply dont have time for hardcore games so much anymore. Or rather time for deep involved games with long play sessions. My intended game time is approaching 30 to 45 minutes for the online game. Considerably shorter for quick singleplayer missions and approaching an hour session for singleplayer "missions".
I want to make the players feel like theyre actually in the air. That getting shot is physical and that losing your plane is a BAD thing *imagine being shot at and having the realisation that if your plane dies, so do you*.
There are a number of ways of making this feeling work. Things like when you are hit, the plane should lurch or shake. Having damage decals appear across the cockpit glass. Having some audio (perhaps wingman or copilot) shouting to "Watch out!" etc. Reinforce that feeling of being in combat.
Anyway, rambling now.. great stuff guys! look forward to chewing the fat about all this at IGC!
In fact, maybe we all should set aside some quiet time for a real discussion about this, rather than just quick chats grabbed with different people during food?
08/18/2005 (2:57 am)
Ok, so one thing thats been thrown up by Paul and Joe's plans recently is this..Currently in Air Ace, the physics model for the flight dynamics of the planes isnt complete.
I'm aware that this is perhaps 95% of what will make Air Ace a hit or a miss. But the physics IS underway, by a guy who knows physics code well. However its definitely slow going.
Now the developer/designer side of me KNOWS that this is a fundamental aspect of the game and that we shouldnt do anything else UNTIL this is sorted out.
But the producer/manager side of me says "ok, we know its work in progress, so we should just accept that it will come and continue with other stuff until its there".
But which side of me is right? To sit there waiting for the flight model to complete means no other progress can be made? Even though we know that without this being right, the game simply wont fly (pun intended).
I think its basically an act of faith that we have to put in our co-workers. That it WILL be done and that we CAN refine it into the model we want.
It helps that we have a very clear goal of how it is meant to feel, plus I personally have as Joe and Paul both mentioned, a very clear mental picture of the target audience.
Actually, while I'm on that subject, I should mention that its normally a good idea to have that mental picture of the target audience, but its also maybe a good idea to BE that target audience too. Having empathy for your own game audience by being a member of it.
My audience is very much like the thinktanks audience. Hardcore gamers who simply dont have time for hardcore games so much anymore. Or rather time for deep involved games with long play sessions. My intended game time is approaching 30 to 45 minutes for the online game. Considerably shorter for quick singleplayer missions and approaching an hour session for singleplayer "missions".
I want to make the players feel like theyre actually in the air. That getting shot is physical and that losing your plane is a BAD thing *imagine being shot at and having the realisation that if your plane dies, so do you*.
There are a number of ways of making this feeling work. Things like when you are hit, the plane should lurch or shake. Having damage decals appear across the cockpit glass. Having some audio (perhaps wingman or copilot) shouting to "Watch out!" etc. Reinforce that feeling of being in combat.
Anyway, rambling now.. great stuff guys! look forward to chewing the fat about all this at IGC!
In fact, maybe we all should set aside some quiet time for a real discussion about this, rather than just quick chats grabbed with different people during food?
#4
I would just like to say that anyone who thinks they have to learn the engine before they can make a game will never complete a game..... or it wlil take them FOREVER to do so.
Smarten up people. learn what you need in order to complete your needed tasks and leave it at that.
Ok... I've had my monthly rant.
08/18/2005 (5:33 am)
Quote:
If you want to rant and rave about making video games, post here!
I would just like to say that anyone who thinks they have to learn the engine before they can make a game will never complete a game..... or it wlil take them FOREVER to do so.
Smarten up people. learn what you need in order to complete your needed tasks and leave it at that.
Ok... I've had my monthly rant.
#5
Yes, you can treat the engine as a "black box" which appears to be working quite well for you. And that's probably the way most of us have to learn the engine, by treating it as a black box until need to modify the way that it works by taking a peek inside.
@Phil: You're correct, it is an act of faith. Believe that they will pull through for you. There has got to be plenty of things to work on while the flight code isn't finished yet. Add shine to your menu's or GUIs, create some more missions, work on a communications system, sleep off that jet lag...
08/18/2005 (6:28 am)
@Chris: I'm glad you make games, and not Automobiles. Yes, you can treat the engine as a "black box" which appears to be working quite well for you. And that's probably the way most of us have to learn the engine, by treating it as a black box until need to modify the way that it works by taking a peek inside.
@Phil: You're correct, it is an act of faith. Believe that they will pull through for you. There has got to be plenty of things to work on while the flight code isn't finished yet. Add shine to your menu's or GUIs, create some more missions, work on a communications system, sleep off that jet lag...
#6
The funniest thing about your post is I am typing this post from work... I am an engineer in the automotive industry......
08/18/2005 (6:39 am)
Unsung Zero - I believe we've had this disucssion before ;) Didn't really result to a conclusion then either :) The funniest thing about your post is I am typing this post from work... I am an engineer in the automotive industry......
#7
it is a good fear to have that the flight model is not done. You have done this before, so I am confident that you won't paint yourself into a corner.
But, there are possible problems with the flight model being done. When it is done, you might play the game and realize it is not that fun, and if it is not, you might have to make massive alterations to the mission design in order to make what 'is' fun work. But you have the main problem solved.. you understand who you are making the game for, so you have a clear metric to test for when the game and flight model are right.. you have a sense of how it should feel, how difficult the controls can be, how 'core' the interface.
I still see danger, but there are things that can be parrallelized in development, and the shell and GUI can be started and a lot canbe done while the flight model is being finalized, provided that you don't settle for something that is not right for the audience and that you don't spend so much time managing all the supporting details that you forget to give enough attention to the main actor.
I think of it in film terms. The supporting actors support the lead role.. if the lead role is weak, the best performance of the support cast will not make it shine..
as for the quality time.. I really want to do this at this years IGC, so let's plan on it.
08/18/2005 (7:46 am)
@Phil,it is a good fear to have that the flight model is not done. You have done this before, so I am confident that you won't paint yourself into a corner.
But, there are possible problems with the flight model being done. When it is done, you might play the game and realize it is not that fun, and if it is not, you might have to make massive alterations to the mission design in order to make what 'is' fun work. But you have the main problem solved.. you understand who you are making the game for, so you have a clear metric to test for when the game and flight model are right.. you have a sense of how it should feel, how difficult the controls can be, how 'core' the interface.
I still see danger, but there are things that can be parrallelized in development, and the shell and GUI can be started and a lot canbe done while the flight model is being finalized, provided that you don't settle for something that is not right for the audience and that you don't spend so much time managing all the supporting details that you forget to give enough attention to the main actor.
I think of it in film terms. The supporting actors support the lead role.. if the lead role is weak, the best performance of the support cast will not make it shine..
as for the quality time.. I really want to do this at this years IGC, so let's plan on it.
#8
I am in full agreement with you about getting that core mechanic. But it does reverb a lot with me about putting the cart before the horse and going on to things like making a nice gui whilst the core mechanic of flying around isnt nailed.
On the mitigation side of this, is that:
1) I already know how the mechanic should feel
2) I'm confidant that even if Kristen cant finish it in time, I can tackle it myself if we get really desperate
3) Technically there isnt any major problem with at least finishing up other functionality (for example AI) outside of that work, even though its secondary stuff.
I agree with you about the mission design stuff. But as we're still basically adding functionality rather than polish or finalised missions or artwork, I think its a risk worth taking to keep myself productive.
It really *IS* all about getting the core mechanic together though. As with thinktanks, Air Ace is basically at its heart an INCREDIBLY simple game. You fly a plane and you shoot other planes. Thats it, in a nutshell.
So obviously with something as simple as that, the majority of how well the game is recieved is in that aspect (flying and shooting). The rest is icing, its the flying and shooting that are the cake.
Anyway, Joe, you'll be able to see how well we've hit the mark at IGC. I'm fairly sure we will have a representative demo of the game available for the showOff to at least get feedback.
I'd love us to actually do some more in-depth stuff though. I wonder how we can plan a session like that? What we ideally want is to have an indepth session where we look at one product and perhaps discuss it and analyze it in some depth?
Less like a usability test kinda thing (which is what the showoff is) and more like a critique session.
One thing I would like to do this year, is maybe offer a talk about usability testing for indies. My colleagues and I have been looking at this area the last year or so and have some perhaps usable ideas for smaller developers to take a look at. Of course I'm eating my own pudding here too in that I've already had an uber-early version of Air Ace in our usability lab with my students.
08/18/2005 (8:31 am)
Thanks Joe,I am in full agreement with you about getting that core mechanic. But it does reverb a lot with me about putting the cart before the horse and going on to things like making a nice gui whilst the core mechanic of flying around isnt nailed.
On the mitigation side of this, is that:
1) I already know how the mechanic should feel
2) I'm confidant that even if Kristen cant finish it in time, I can tackle it myself if we get really desperate
3) Technically there isnt any major problem with at least finishing up other functionality (for example AI) outside of that work, even though its secondary stuff.
I agree with you about the mission design stuff. But as we're still basically adding functionality rather than polish or finalised missions or artwork, I think its a risk worth taking to keep myself productive.
It really *IS* all about getting the core mechanic together though. As with thinktanks, Air Ace is basically at its heart an INCREDIBLY simple game. You fly a plane and you shoot other planes. Thats it, in a nutshell.
So obviously with something as simple as that, the majority of how well the game is recieved is in that aspect (flying and shooting). The rest is icing, its the flying and shooting that are the cake.
Anyway, Joe, you'll be able to see how well we've hit the mark at IGC. I'm fairly sure we will have a representative demo of the game available for the showOff to at least get feedback.
I'd love us to actually do some more in-depth stuff though. I wonder how we can plan a session like that? What we ideally want is to have an indepth session where we look at one product and perhaps discuss it and analyze it in some depth?
Less like a usability test kinda thing (which is what the showoff is) and more like a critique session.
One thing I would like to do this year, is maybe offer a talk about usability testing for indies. My colleagues and I have been looking at this area the last year or so and have some perhaps usable ideas for smaller developers to take a look at. Of course I'm eating my own pudding here too in that I've already had an uber-early version of Air Ace in our usability lab with my students.
#9
Given that we have this plane that you fly around. How would you go about giving players an introduction to how to fly the thing? Assuming that its a fairly easy flight model already (we have 3, the default is very stable and easy to fly in my opinion).
Is a tutorial even worth the effort, knowing most players? Keeping in mind the market is ex-hardcore players, most of them would be at home with keys, keys + mouse, or joystick usage.
Would it even be worth the effort to go through a series of scripted tutorials on flight, combat and the like?
I guess maybe some kind of "helper" mode which you can toggle, which would prompt for various actions during flight?
For instance.. in the normal game sequence of events, the player starts off on the side of the runway of thier home team, pointing at it at a 90 degree angle. In order to take off, they have to taxi onto the runway, turn 90 degrees onto it and then throttle up until they have takeoff speed.
Is that something we'd need to make a tutorial about?
I guess I can do a usability test for that at work next month and answer it.. but intuitively, I'd say that most of my players would be able to handle that.
These are the questions that are hard to answer.. how to make it easy for someone who is maybe less of a gamer than yourself.
08/18/2005 (8:37 am)
Sorry for following my own post, but I just remembered a question I would like to ask fellow designer/developers..Given that we have this plane that you fly around. How would you go about giving players an introduction to how to fly the thing? Assuming that its a fairly easy flight model already (we have 3, the default is very stable and easy to fly in my opinion).
Is a tutorial even worth the effort, knowing most players? Keeping in mind the market is ex-hardcore players, most of them would be at home with keys, keys + mouse, or joystick usage.
Would it even be worth the effort to go through a series of scripted tutorials on flight, combat and the like?
I guess maybe some kind of "helper" mode which you can toggle, which would prompt for various actions during flight?
For instance.. in the normal game sequence of events, the player starts off on the side of the runway of thier home team, pointing at it at a 90 degree angle. In order to take off, they have to taxi onto the runway, turn 90 degrees onto it and then throttle up until they have takeoff speed.
Is that something we'd need to make a tutorial about?
I guess I can do a usability test for that at work next month and answer it.. but intuitively, I'd say that most of my players would be able to handle that.
These are the questions that are hard to answer.. how to make it easy for someone who is maybe less of a gamer than yourself.
#10
Don't limit your market by excluding those not in the directed target.
08/18/2005 (8:47 am)
Quote:
Would it even be worth the effort to go through a series of scripted tutorials on flight, combat and the like?
Don't limit your market by excluding those not in the directed target.
#11
'just got this one off the assembly line with this new |33t engine.. lets put it through it's paces.. .. good, good, everything looking stable.. fire off a few rounds to make sure those bozos back at the hanger didn't forget to re-aling the machine guns.. good, good... alright, looks like this one checks out.... oh crap.. looks like we got some enemy fighters coming in.. sorry to cut this short, but we've got to go take these guys out before they get to the airfield...."
I do think the tutorial is important.. at the very least, I would need a playground to try the plane out in a way that I feel safe and I am not getting my butt kicked by others and crashing all the time before I even had a chance to get a feel for the physics and input.
on the session.. Phil, email Jay.. we talked about this breifly, and a 'designers roundtable' would be a great thing.
08/18/2005 (8:58 am)
@ Phil, if it were my game, I would make the 'tutorial' the first mission.. something where you are flying around blowing up target blimps or something.. where there is a direction for you to 'adjust your firing range' etc... I really like the way the original half life did it.. where it showed you the suit as part of the 'testing' of the suit, and a similar thing could be done with the planes.. 'just got this one off the assembly line with this new |33t engine.. lets put it through it's paces.. .. good, good, everything looking stable.. fire off a few rounds to make sure those bozos back at the hanger didn't forget to re-aling the machine guns.. good, good... alright, looks like this one checks out.... oh crap.. looks like we got some enemy fighters coming in.. sorry to cut this short, but we've got to go take these guys out before they get to the airfield...."
I do think the tutorial is important.. at the very least, I would need a playground to try the plane out in a way that I feel safe and I am not getting my butt kicked by others and crashing all the time before I even had a chance to get a feel for the physics and input.
on the session.. Phil, email Jay.. we talked about this breifly, and a 'designers roundtable' would be a great thing.
#12
Several reasons for this - I was sure we could get them done, we already knew golf was fun (so no need to test for that 'fun' quality), commercial game with deadlines for demos and such (so 100% accurate physics wasn't the most important thing). So we have a ways to go on ball physics even though quite a bit of the rest of the game has been done.
So, I definitely see your point about letting the physics come along with the rest of the game. Can't wait to see Air Ace at IGC...
08/18/2005 (9:11 am)
Hey, Phil, I just wanted to comment that we are approaching Golden Fairway the same way - golf ball physics haven't come as far as they could have before we started to develop the rest of the game. Several reasons for this - I was sure we could get them done, we already knew golf was fun (so no need to test for that 'fun' quality), commercial game with deadlines for demos and such (so 100% accurate physics wasn't the most important thing). So we have a ways to go on ball physics even though quite a bit of the rest of the game has been done.
So, I definitely see your point about letting the physics come along with the rest of the game. Can't wait to see Air Ace at IGC...
#13
I'm sure that the planes will be fun to fly, because I've got models of such in other games.
Joe: I like that tutorial idea.. I see where youre heading with that.. Nice little "intro" kind of thing.
08/18/2005 (9:55 am)
Joseph, you make a good point. Its ok to skip ahead with some other elements as long as you KNOW that the element thats really important is both going to be fun and is technically do-able.I'm sure that the planes will be fun to fly, because I've got models of such in other games.
Joe: I like that tutorial idea.. I see where youre heading with that.. Nice little "intro" kind of thing.
#14
From a gamers pov.
I think i been flying all simulators out there from amigans bomber to IL2 , im have been a flying freak for so long i cant rember.
When it comes to multiplayer aircombat it most be easy to handle the plane , not like the heavy simulator games Fighter Squadron , European Airwar and more .
Grimson skies is a lovely example of a fast game , easy to fly for noobs and they still have a change against skilled players.
Fighter Dual was another lovely online game with lots of nice airfights , but alittle bit to hard for people that not are flying freeks , so if Air Ace comes close to the games above then it gonna kick ass.
Make it fun fast with some humor and i think people will love it.
But it looks really good so far !!
Only some thoughts .
-Billy
08/18/2005 (10:21 am)
@PhilFrom a gamers pov.
I think i been flying all simulators out there from amigans bomber to IL2 , im have been a flying freak for so long i cant rember.
When it comes to multiplayer aircombat it most be easy to handle the plane , not like the heavy simulator games Fighter Squadron , European Airwar and more .
Grimson skies is a lovely example of a fast game , easy to fly for noobs and they still have a change against skilled players.
Fighter Dual was another lovely online game with lots of nice airfights , but alittle bit to hard for people that not are flying freeks , so if Air Ace comes close to the games above then it gonna kick ass.
Make it fun fast with some humor and i think people will love it.
But it looks really good so far !!
Only some thoughts .
-Billy
#15
08/18/2005 (5:11 pm)
As a student I feel a little under qualified to be posting here but, my take on usability is that if ever you question whether or not someone will know what to do there will be someone out there who won't. And keep in mind when designing interfaces (or interactive objects) that there are color blind people out there. I'm color blind (and an artist *sigh*) and it's a big pet pieve of mine when designers don't think of it. It's a really really simple thing to avoid too. For example if you have a light that indicates when to go and you use red and green colors for the lights, make the lights different physical entities. Using color as a guide is fine, just be sure to have some redundent indicator in there like position. If you can look at your interface in black and white and still know what is what your golden.
#16
A lot of people had CRASH bugs trying to play the Open Beta...I am happy to announce that those issues have been fixed. If you had problems before please sign up again and re-download the first version and see if the crash bugs are fixed. They probably will be!
Also there are FOUR versions of the game posted right now...each letting you test a new map. Now would be a great time to sign up and check out the game!
09/25/2005 (10:07 am)
You all make good points. Thanks so much for your posts.A lot of people had CRASH bugs trying to play the Open Beta...I am happy to announce that those issues have been fixed. If you had problems before please sign up again and re-download the first version and see if the crash bugs are fixed. They probably will be!
Also there are FOUR versions of the game posted right now...each letting you test a new map. Now would be a great time to sign up and check out the game!
Torque Owner AllynMcelrath
Good luck to you =)