Developer Story Time with Old Man Davis

[Anthony Davis]

I think it does. The example was really enlightening.

Vault Dweller how does this make you feel about the short dev time for W2 and PE?

My opinion is unchanged.

First, things like faction system or radiation systems are fairly minor (compared to more complex systems; and what Tim Cain did was the only logical course of action - take the poison system and tweak it). Second, the libraries and code examples help, but not that much. It took us a lot of work, for example, to change AoD character/combat systems from R3 to R4, even though it's the same system and the programmer is familiar with the engine and the system overall. It's the little things that take the most time.

So, as I said before, getting to the playable state fast is easy, if you know what to do and have the code. Getting it to the state where it's good and work as you want it to work isn't.

Third, Tim Cain said what he said after working on Fallout, Arcanum, ToEE, Bloodlines. It's safe to assume that he had the knowledge and the libraries and the code and the experience, and yet after all that he said that it takes at least 3 years to make a full-scale RPG. Now, maybe he lied, maybe he exaggerated, maybe he forgot that had the code and the libraries, but what he said didn't go against what we witnessed over the years. A decent RPG - 3 years, a decent sequel with the existing engine/systems/basic assets - 1.5 years.
Fargo's own track record doesn't disagree with this either.

So, maybe Fargo will surprise me and deliver a great RPG in exactly 18 months. Maybe Obsidian will do the same and we'll all praise Unity. Then again, maybe they won't...

There is nothing wrong with these statements, and I'm not sure what Tim Cain's original statement was. I'm not sure what anyone is trying to prove, can someone give me full illumination?

Without knowing all the variables, 3 years is a very accurate guess for building *any* complex game from scratch. An RPG certainly qualifies as a complex game.

Unity of course, is not starting from scratch. It allows for rapid development, the renderer is solid, the audio and music support is solid, the input and UI system is solid. Different from many engines, but still very solid.

Adding in systems, both major and minor, can be done quickly by smart and experienced engineers like Tim Cain, slower by smart and inexperienced engineers, and reeeeallly slow or wrong by dumb and inexperienced engineers.

Also, I would bet money that many of Obsidian's tools have been retasked to Unity which probably also saved time.

 

[tuluse]

I think VD and I just disagree with how much from scratch development InXile and Obsidian need to do.

I'm not really trying to prove anything. I'm genuinely very interested and just trying to foster informed discussion. Unfortunately, I don't think we can be informed enough without one of us be hired to work on on the games.

 

[Infinitron]

Anthony Davis This is a long-running argument on the Codex. VD believes that, unless a developer has an existing engine with existing assets, the only thing they can possibly create in 18 months is some kind of stripped down combat-focused RPG.

Hence, Wasteland 2 is really going to be Wasteland: Tactics and Project Eternity is going to be Icewind Dale 3, unless they are delayed.

 

[Grunker]

Seems to me that you misunderstood VD's point a little Anthony Davis. Correct me if I'm wrong, but it seems his points was:

- Designing systems and making them playable = fast

- Making the above good and reach finished state = slow

So no matter how good Unity is, that doesn't change that P:E is loads of completely new systems and content that you need to make work for the first time ever - no matter what resources you have/don't have.

Like Infinitron said, this is an argument that has been running since the start of Kickstarter. I don't think I agree with VD, but I certainly don't agree with Sawyer and Cain either who seem to think they have endless money and time for playtesting. I think the games will be delayed 6 or 12 months and be exactly as (un)balanced as any RPG, but my opinion is irrelevant, guesswork and (probably) stupid. I just don't think you really answered VD in your post, and since your arguments are more solid than mine, I'd like to see you do that

 

[Sensuki]

P:E is about 3 months behind already, if they don't finish the VS before August then that's 4 (not that they have to either really).

 

[Vault Dweller]

There is nothing wrong with these statements, and I'm not sure what Tim Cain's original statement was.

Here is the statement:

"With that said, the Vampire had been under development for three years. While that's not a long time for a role-playing game - Fallout had taken three and a half years to develop... The original schedule for Temple was 18 months, which was and is unthinkable for a full-featured role-playing game."

Didn't ToEE use a modified Arcanum engine? Wouldn't that give them more than an off the shelf engine like Unity?

I'm not sure what anyone is trying to prove, can someone give me full illumination?

We're done discussing what an RPG is and isn't, so now we're fighting over 'how long does it take to make an RPG?'

 

[Grunker]

We're done discussing what an RPG is and isn't, so now we're fighting over 'how long does it take to make an RPG?'

Though:

We're done discussing what an RPG is and isn't

Yeah right, in our dreams, lol.

 

[Anthony Davis]

There is nothing wrong with these statements, and I'm not sure what Tim Cain's original statement was.

Here is the statement:

"With that said, the Vampire had been under development for three years. While that's not a long time for a role-playing game - Fallout had taken three and a half years to develop... The original schedule for Temple was 18 months, which was and is unthinkable for a full-featured role-playing game."

Didn't ToEE use a modified Arcanum engine? Wouldn't that give them more than an off the shelf engine like Unity?

I'm not sure what anyone is trying to prove, can someone give me full illumination?

We're done discussing what an RPG is and isn't, so now we're fighting over 'how long does it take to make an RPG?'

I guess we'll see for PE and WL2 and Torment 2. I know it can be done in less than 3 years, and I know there are plenty of examples that have taken longer.

 

[Brother None]

"With that said, the Vampire had been under development for three years. While that's not a long time for a role-playing game - Fallout had taken three and a half years to develop... The original schedule for Temple was 18 months, which was and is unthinkable for a full-featured role-playing game."

Didn't ToEE use a modified Arcanum engine? Wouldn't that give them more than an off the shelf engine like Unity?

I do wonder how much freedom from publishers and advancement in tools changes this. Doesn't Tim Cain being involved in PE pretty much already counter his own statement?
Anyway, this doesn't seem like the kind of debate that'll be closed until WL2 is out (or it and more of these RPGs).

 

[Grunker]

Anyway, this doesn't seem like the kind of debate that'll be closed until WL2 is out (or it and more of these RPGs).

Closed, perhaps not. It is a pretty interesting discussion either way though, and all this talk of engines and resource libraries is interesting. I learned a thing or two about game development since the discussion started here, at least.

 

[Vault Dweller]

Doesn't Tim Cain being involved in PE pretty much already counter his own statement?

I don't read too much into such things as they are subject to change without notice.

If I recall correctly, they started a KS just to see what happens and developed the game's concepts after the launch. During the KS they added another big city (lotsa content) and a mega dungeon and a stronghold and a truckload of more stuff, without delaying the game by a day. There is no fucking day they can do it all in 18 months without rushing it horribly and cutting content left and right.

 

[Cursed Platypus]

Doesn't Tim Cain being involved in PE pretty much already counter his own statement?

I don't read too much into such things as they are subject to change without notice.

If I recall correctly, they started a KS just to see what happens and developed the game's concepts after the launch. During the KS they added another big city (lotsa content) and a mega dungeon and a stronghold and a truckload of more stuff, without delaying the game by a day. There is no fucking day they can do it all in 18 months without rushing it horribly and cutting content left and right.

It may seem like simplisitc logic, but they don't delay the game after these additions because they get the funds to hire more guys. Said guys will work at the same time as the others therefore no need to expand the timeframe?

Also they aren't using ground breaking technology in eternity thanks to their ''old school'' excuse. It does look fairly established and simple stuff for today's standards. Unlike TOEE or vampire which needed to look like they were at the forefront (all new and advanced) back then, to stay competitive.

 

[Cynic]

General rule of thumb in software development goes that if you have an existing framework to build from, things get done a lot faster. In the web development world, one example you could use is open source CMS solutions. These offer basic tools for getting what you need done - a database schema, the ability to add pages, fill them with content etc., Similarly a hosting environment can be rented as ready to use or as a blank slate which you have to set up and configure yourself.

If you had to do every single bit of development starting from nothing, the project would take around 10 times as long. I'm not a game developer so I can only imagine that using a pre existing engine or framework like Unity would definitely speed things up immensely. That is assuming of course that the engine either was easily extendable, or had the necessary tools for implementing the various complex methods needed for creating an RPG.

That being said, I think 18 months is an incredibly short period of time. If ToEE was made in this time, it explains a lot. It's taken years of patching to get it playable and while what was there underneath the surface of bugs is incredible, the game has to be seen as a perfect example of a product that was released far too early.

 

[Space Satan]

I guess most major Kickstarter projects will be delayed. And next ones wil adjust release dates according to previous miscalculations. I can wait a year if that'll make game better.

 

[Sensuki]

There is no fucking day they can do it all in 18 months without rushing it horribly and cutting content left and right.

It's gonna be at least 22 months man, they're already 3-4 months behind schedule and Adam Brennecke said in a recent interview the target launch is in Summer 2014 instead of April.

Pre-production has been going for 9 months (instead of the projected 6-7) and if it's not done by mid-August that'll be 10 months (or 11 if you count the fact that they started in September when the Kickstarter launched).

They have however been very good regarding the team size, the majority of this time was less than 15 full-time people.

 

[Infinitron]

There is no fucking day they can do it all in 18 months without rushing it horribly and cutting content left and right.

It's gonna be at least 22 months man, they're already 3-4 months behind schedule and Adam Brennecke said in a recent interview the target launch is in Summer 2014 instead of April.

Pre-production has been going for 9 months (instead of the projected 6-7) and if it's not done by mid-August that'll be 10 months (or 11 if you count the fact that they started in September when the Kickstarter launched).

They have however been very good regarding the team size, the majority of this time was less than 15 full-time people.

They're not "behind schedule" if they've modified the original schedule.

Anyway, 22 months still isn't enough for ol' VD. That's action-RPG development time!

 

[Minttunator]

 

[Space Satan]

Ironically, the best thing they can get to make people accept schedule change is some kickstarter project that will fail miserably or will be so raw and bug ridden that fear will drive peopel to accept any delay just to avoid similar fates.
Sadly, the closest we have is Shadowrun.

 

[Darth Roxor]

Ironically, the best thing they can get to make people accept schedule change is some kickstarter project that will fail miserably or will be so raw and bug ridden that fear will drive peopel to accept any delay just to avoid similar fates.
Sadly, the closest we have is Shadowrun.

You haven't heard of Broken Age, have you?

 

[Infinitron]

Ironically, the best thing they can get to make people accept schedule change is some kickstarter project that will fail miserably or will be so raw and bug ridden that fear will drive peopel to accept any delay just to avoid similar fates.
Sadly, the closest we have is Shadowrun.

You haven't heard of Broken Age, have you?

No, I think what he means is a game that ISN'T delayed and comes out bad because of that.

 

[Space Satan]

No, I think what he means is a game that ISN'T delayed and comes out bad because of that.

Precisely. Broken Age looks like "Sorry, people, hookers and cocke turned more expensive than we expected..."
There's whole article about kickstarted delays - Most Kickstarter Projects Fail to Deliver on Time

 

[Spectacle]

No, I think what he means is a game that ISN'T delayed and comes out bad because of that.

Precisely. Broken Age looks like "Sorry, people, hookers and cocke turned more expensive than we expected..."
There's whole article about kickstarted delays - Most Kickstarter Projects Fail to Deliver on Time

Sometimes a project just gets so far off track that no amount of delaying will have it turn out good, short of scrapping everything and restarting from square one. Broken age is starting to look like that kind of project, if it was published funded it would probably have been cancelled by now, maybe even before it was publicly announced.

 

[Roguey]

Ironically, the best thing they can get to make people accept schedule change is some kickstarter project that will fail miserably or will be so raw and bug ridden that fear will drive peopel to accept any delay just to avoid similar fates.
Sadly, the closest we have is Shadowrun.

SR's original release date was January.

 

[Anthony Davis]

I'm not gonna lie, scheduling is hard even under the best of circumstances.

Many times you are making a series of educated guesses based on data that is available and experience you have accumulated.

Think about this for an example:

I want to make a table. I need to figure out much this will cost me, in time and money.

How big is the table going to be? How tall? What kind of wood? More than one type of wood? Ornate or simple? Finished or unfinished? If finished, what kind and color for the finish? Do I have the tools to properly make this table? Do I have the knowledge to make this table? Do I need to get some training?

I need to estimate my time and money for this project. It can be very difficult for something as simple as a table. A video game is many times more complicated than a table and therefore many more times harder to schedule for.


This is why I get upset at DF, and other game companies, is that they know or SHOULD know that this is difficult and they should naturally adapt a more prudent stance rather than acting like crazy ass kids in a candy store. Any developer that doesn't take scheduling seriously is setting themselves up for failure. Yeah sure, they might get lucky once or twice, but luck is fickle and so are your fans.

Good producers are key to a successful project. They need to understand how to plan a project, what needs to be done and in what order things need to be done. They need to be savvy enough to not have the wool pulled over their eyes when developers try to give them numbers - they need to understand what those numbers mean. They also need to be experienced enough to look for alternatives for a feature that might take too many resources. They need to play the long game and understand the big picture. Sometimes this also means being a jerk and making hard decisions. It also means that sometimes you take a chance and play the odds, which is NOT the same thing as gambling btw.


Chris Parker at Obsidian was really good at this. Probably the best producer at Obsidian. Matt Singh is also very, very good at this.

TimeGate, which had problems at least had very, very good producers.

I'll never kick start anything for DF again, not that Tim and Co. would care, but I'm sure I'm not the only one. As much as I like and respect those guys, I can't and won't put my money behind someone this irresponsible. I feel like these guys get by on past accomplishments and how amazingly charming they are. That charm is wearing thin for me.

Like your buddy who is great fun to hang out with, but is always hitting you up for beer money.

I worked for a bit on WL2 before real life caught up with me and forced me to drop out. I thought that project was clicking very well.

Adam and Josh both are big proponents of a managing scope and scheduling. If they get within 10 or 20 percent of their target date, then their scheduling and scoping skills are already superior to just about every other game developer out there. I have high hopes for PE.

I have no idea for Torment and Shadowrun.

 

[Infinitron]

Anthony, have you ever worked with Unity?