Hey folks,
So this is something that Erik and the rest of us internally need to discuss, and he'll likely speak to some other indies about their experiences, so nothing is being decided right now. No reason to panic if you feel strongly about one way or another.
But I wanted to solicit some feedback from this crowd as well, since a diversity of opinions and experience tends to be very very useful in decision-making of this sort.
The Core ProblemOverall there has been a LOT more discussion here on the forums than I expected, which is both a blessing and a curse. Things are moving incredibly fast and I think that there has been a huge amount of hashing-out of things that will be very exciting to talk about as part of the kickstarter campaign itself, and that will also result in a smoother post-kickstarter experience with better initial designs leading to fewer rewrites as we go.
A great example of this is the ship quadrangle discussion, which is easy for me to dismiss as "just a data problem we can work out later." But ultimately it does play a lot into clarity of the game, and that's one of the selling points of the sequel, and so a lot of discussion and revisions have happened that make a really positive difference.
Another good example is having multiple playable races. This is something that solves a whole lot of design problems and inconsistencies, and it's actually a really super-cool marquee feature to focus on in the kickstarter pitch in general. But this does come at a certain cost in terms of us needing to actually design that (Cinth is pretty well single-handedly taking that on, though, which is a really big help to me). But then when we jerk around the core quadrangle, then that sets HIM back, and so on.
Overall there are simply more features and options and whatnot than I remembered (hahahaha -- and there were a bajillion that I had remembered), and getting all those things written up, organized, and evaluated for inclusion or revision is a big task. To some extent "can't those little things just wait until later?" is a valid question. However, in order to do a kickstarter with confidence, Keith and I and Blue all need to be able to properly estimate how long things will take us, with appropriate buffers. That then translates into a cost figure, again with buffer.
Not going through the proper budgeting and cost analysis process is a great way to wind up in a pickle 10 months from now, and that's something that I want to avoid. It also wouldn't inspire confidence in the kickstarter itself for prospective backers. So that stuff is important, and being able to really break things down is very important to me both from a professional presentation sense to a "good thing to do in general" sense.
Visuals And Cinematics And VideoThis is something that I feel like is really important, too. Being able to show people that the game will look awesome, but also demonstrating that they can run it is important to me. I've been working hard on that, but to be honest for every "day" that I've been working on that, it seems like I wind up actually getting like 1-2 hours done on it at most because of the aforementioned design discussions and so forth, which I don't want to lose momentum on. It's very much a matter of juggling a lot of things at once.
Given enough time, there's all sorts of things that I could do, ranging from scripted battle examples with a cinematic camera view going through them, to various special effects that I plan for in-game in general like an awesome warp-in effect when ships arrive from another planet versus the boring old fade-in.
Do I have to show those things? Absolutely not. Would they help sell the idea of the kickstarter? Arguably so. "Sex sells," so to speak. There's a middle ground in there between showing a lot of polished things that normally I'd leave until later, versus being too boring for people to get that compelling feeling from the visuals. A presentation like kickstarter is about the entire package: visual, audio, text, narration, etc. I have to record video of myself talking to the camera, which I always hate to do because I get flustered when there's a camera and not someone actually asking me questions. We'll see how that goes.
Time PressureRight now the time pressure is, well, money. We spent all of 2015 making a game that didn't release (Stars Beyond Reach). We spent half of this year working on two games that didn't lead to money (Keith on SBR again, and then Blue and I on Raptor). We had an awesome game that came out early this year called Starward Rogue, and it was made very quickly, but that's one that has still not hit the break-even point yet and likely won't until next year (so, in other words, we lost money in the process of making it).
This puts pressure on me, certainly, and Arcen in general. It also puts pressure on some of our other staff, at least one key member of which who was considering needing to look for part-time work. That's not good for us in terms of our ability to get you the game you want in a timely fashion. It's also not good for the life-balance health of the staff member involved (it's not me, but my life-balance is already terrible).
I Had Considered Patreon
One thing that I'd considered was setting up a patreon account so that folks could help us make it through an extended pre-kickstarter design period. However, I feel like that would then lead to an increased risk to the kickstarter itself, since kickstarters are all-or-nothing: either you hit your funding goal, or you get none of what people pledged. That's a very good system, because getting part of your budget is a recipe for failure.
However, if you're then asking people to help fund your earlier design period prior to the kickstarter, that gets dicey fast. So ultimately I've decided not to take that approach.
Overall Budget
The overall budget will be somewhere like $200k to $250k, probably, for the kickstarter. It depends on exactly what the scope of the central design document is, versus what we decide to pull out into stretch goals to make the core goal more attainable. Plenty of projects get funded at this level, and overall we need about 10 thousand backers to reach the $200k mark. We already have 71 (and growing every day) people who have emailed us to say "let me know as soon as the kickstarter is live!" That's a start, at least. AI War Classic has sold 300 thousand copies of the base game, and about 1 million copies of the various expansions, so hopefully 10 thousand is an attainable number.
But I want to be SURE that really is the number. I also want to find out what the average "failed pledge collection" rate is for games like this, because that's something I don't know and it's not accounted for in the number above. If anyone has some source stats on that from the recent past, that would be awesome.
Oh, and I'd like to have time for Blue to mock up more of the GUIs than had previously been planned. There are enough new ones that to fully communicate what is going on with them I think we should have mockups to demonstrate them.
The Original ScheduleOriginally my schedule plan was to start the kickstarter prior to the end of September so that things like part-time work wouldn't need to be contemplated by anyone, amongst avoiding other financial unpleasantries. THAT said, to do that would mean launching the kickstarter next week.
While that would be possible, there's still so freaking much to the design document that needs to be added (that thing is already 46 thousand words long, but it's going to practically double I think). And then there's the whole video and presentation and so forth, and the question of how much time I can spend on really making that amazing versus having to rush something less compelling out. That seems like a rush would be a Really Bad Idea.
My Current ThinkingThis is still pushing it, but my thought at the moment is that one extra week ought to do it (so launch two weeks from now rather than one). In an ideal world I'd take even a bit more time, but I think that I can get things to the state that they need to be in for a compelling kickstarter and a proper budget in that extra week. I think it would be foolish to take less than that extra week.
If this sort of schedule slippage sounds familiar, it's basically what happens when you start without a design document and try to do a lot of experimental stuff. It's what happened to Raptor and SBR, and it's precisely what the giant design document for AI War II is intended to prevent. Originally I was going to be slightly more conservative with the AI War II design, but too many compelling things came up either from myself or others, and I think that those elements ultimately make a kickstarter more likely to succeed rather than less.
So as a calculated risk, I let the "design the kickstarter" phase of this increase in scope somewhat, and now that's biting me in the butt a bit. I'm more concerned about it biting certain staff than myself, though, to be honest. Ultimately if this kickstarter gains traction at all (which it already seems to have momentum to say the least), then I think that will have been a very good call.
We all complain about the "suits" at big publishers giving arbitrary deadlines that cramp the creativity and quality of AAA games that are coming out, and in this case
I'm the suit. And I
hate that! I feel like one extra week is a pretty good compromise, but it can't go slip-sliding on from there.
What I need From YouAnything that folks can do to help focus the many (many) conversations on the design such that we can resolve them "sufficiently for now" would be helpful. There are a lot of discussions where there are things that are primarily data-driven that can be experimented with during alpha and early access, and it's easier for us all to make judgments and decisions at that point when we can all play with it and test it ourselves.
There's a line there somewhere between "this needs to get resolved to make things make sense and for the central design architecture to be correct" and "I think that this would be a bit better than that, for these detailed reasons." If the latter category is something data-driven anyway (like should MLRS be demolition class or siege), then for goodness sake please let's table that one for now.
I can make a special subforum specifically for discussing post-design-document-locking stuff, if you prefer. I don't want to stymie those sorts of discussions, since ultimately we do have to make those sorts of decisions. And we can discuss that DURING the kickstarter, no problem.
Anyway, I'm very open to new lines of inquiry into the design as well, but ultimately we need to think if a given thing is relevant for pre-kickstarter-design-document (aka, fundamentally how a big part of the game works) or if it's a refinement-after-the-fact question (aka a number of the solutions for refleeting might fall under that category... but then again it might be a more fundamental thing depending on what the solution is). This is something that I have to contemplate with each thread that someone opens, but I'd also like for you folks to start helping out with that as well.
The Kickstarter ItselfAt Arcen we're used to thinking of things in terms of calendar months, so going an extra week and passing the magic threshold into October is seen as Very Bad since usually we're paid on a net-30 basis and that would be the difference in getting paid at the end of October versus the end of November. But with kickstarter it's not actually tied to calendar months, so I need to set that mentality aside. I actually haven't read up on when funds are released to the dev -- I assume it's right as pledges are collected, which from my experience as a backer is immediately after a successful funding.
In terms of the staff member that is worried about needing another part-time job, I can potentially just pay them for an extra week out of pocket to solve that issue. That probably would be all the same difference to them, and it's not much difference to me.
Regarding the actual kickstarter dates, though, I understand that when kickstarters start and end is a big deal. You want to end on certain days of the week to maximize the final push for funds. You want to start when press are able to cover it. Etc.
I'm not too worried about press not covering this one, because we've already gotten some coverage from the press on it, and we have direct enough relationships with the press that they either will or won't cover this and that's that. We won't have trouble getting their attention with this, unlike someone new to the market, so it's completely a matter of if they decide to write/talk about it or not.
From what I've read, a shorter kickstarter (say, 21 days) has a lot of benefits in terms of creating more urgency for people to back and tell others to back. The most common downside is the inability to reach enough press in time, but again I really don't think that one is a factor here when we're talking 21 versus 30 days. The other big downside I've read about is that if someone sees that a KS has 20 days left and is only 12% funded, they might think that it's destined for obvious failure; but actually it's just the first day of the KS, so it's doing fine. This is the part that worries me the most, honestly.
Frankly outside of actually talking to people and whatnot during the kickstarter and doing all those Q&A and marketing-type things, we're going to be working on the actual game during that period in the first place. I'm planning on pretty much all my working time being sucked up by the KS, and Erik is as well, but Keith and Blue will be working on the actual project during that period. Given my druthers, I'd rather be able to get back to actually working on the project 9 days sooner, too.
But not at the expense of some important perception relating to backers. Though I'm also told that a shorter kickstarter shows confidence to backers in a lot of ways, which can be positive; I don't know which one weighs more in this case. I also have the difficult decision of whether to include the soundtrack from Pablo (aka him creating one) in the base budget, or whether to make it the first stretch goal. That's a different topic, though.
Halloween and the sales and promotions that presumably go along with that would be near the end of our kickstarter if we did 21 or 22 days starting on the 6th or similar, but we'd be in a few days ahead of it at least, which is good. A number of AAA games would come out during that period, but we wouldn't be into the super-heavy period of November.
Clear as mud, eh? Anyway, this is one of the things that's been on my mind. I've tried working overtime and so forth to speed this up, but ultimately my brain is exhausted at the end of each week and there's only so much extra time I can squeeze out and still remain productive. I've cut my hours back from the 70-80 that I was working prior to Raptor coming out to being more like 55-65 hours a week the last few weeks, which feels like really slacking off, but it's freaking mentally exhausting thinking through so many things, and my brain needs a rest every so often. My wife and son are quite happy seeing more of me, to boot.
Like I said, these are the things I'm trying to make decisions on right now. The TLDR:
1. When do you think we should launch and how long should it run?
2. What data do you have from the market that could help me make a decision to agree or disagree with that?
3. Does anyone have good data on the uncollected-pledge amounts?
4. Anything you can do to help in terms of resolving the design document's many open ends would be appreciated.
Thank you so much!
Chris
