The State of (My) GURPS Vehicles Addendum: My Process

After my previous post, someone asked me how long it took me to build a spaceship currently, to which I lied and said "about two hours." Well, "lie" might be a strong word, but I simplified a lot.  This spurred some discussion about how things get made, and there was no quick and easy way for me to explain that, so I thought I'd turn it into a blog post.  It would certainly help anyone who is building their own designs.

If you're a patron, you can follow along with these supplementary documents.

First, start from scratch, then realize you need to go back

The first step most people might imagine is loading up a spreadsheet and working everything out in excruciating detail, but that's not actually where I start.  Or rather, it is, but I quickly realize I run into problems with this approach.

• how much armor do I need?
• how much firepower should I pack?
• how big are my carriers?
• what sort of electronics do I need?

This brings me back to my endless discussions of an Ultra-Tech framework.  Everything in your setting will interact with everything else, and in principle, you run into a chicken-and-egg problem, because to balance your ships, you need your ships to balance, and to make your ships, you need to know your balanced stats.

This is why I started this whole thing, over a year ago, talking about frameworks.  You need a sense of how everything comes together.  There are a lot of ways to do this, but this is actually where the spaceships from the GURPS Spaceships series are really useful.  Sure, you might not be using that design system, but that design system is a simplification of the vehicles system, so the vehicles that pop out of both are more or less compatible.  You can grab some of those to get a rough idea of your basic values. Or you can borrow an existing model, such as WW2, and see how that plays out, or mix and match.  For Psi-Wars, I took the rough concept of WW1/WW2 ships crossed with some of the concepts from GURPS Spaceships and centered a lot of damage around the concept of an isomeric torpedo, so I created weapons that could deal as much damage as said torpedoes, and then gave a battleship three turrets with those cannons and saw what popped out, and began to balance everything around that.

So, we can start to get a rough sketch of what everything might look like as we build this framework.  For this, I built my Generic Fleet, which did double duty as a "Look, if you don't know what your ship looks like, just borrow one  of these," and a prototype for the rest of my fleets.  They tended to be rough sketches, rather than completely finished vehicles with exhaustive detail.

I always build these with spreadsheets.  So many spreadsheets.  What I find is there's a lot of ping-ponging around, like you decide you actually want a fourth turret, which means you need more space for that, and you need more reactor to power your fourth turret, and you need more framework to account for all of this space, and if you want to keep it as fast as it was before, you need more engine, which needs more space and more reactor, which needs more framework, which needs more engine, and so on.  Doing it all by hand is tedious at best and utterly unworkable at worst, so you don't do it by hand: you get a spreadsheet to handle the mess for you.

Second, alright, we can go back to building something from scratch

Once we have our framework, we can go back to building from scratch, because at this point, we can begin to accurately describe our vehicle.  For example, if we know conceptually that our vehicle has "a lot" of firepower, we know what that might look like in actual numbers, or if it's "not particularly tough," we know what that means too.  We've bounded our context, so we know what we're looking for, so now we can settle down and really build our new vehicles in enormous detail.

Only, in practice, we don't really need to do that.  We can take one of our existing generic vehicles and elaborate on it.  Perhaps your Imperial Cruiser, is really just a slightly larger Generic Destroyer with a much better hyperdrive and some typical ideas you might like to have in imperial capital ships (such as a strategy office and some hangars for close-range fighters).  So you take that old spreadsheed, clone it, and start working on the new vehicle.  You might end up with something very different, but what matters is that you can start somewhere.

This is the part that I meant when I said "it takes me about 2 hours," but it's probably closer to one.  I'm often surprised how easy it is to make a vehicle at this stage, because most of it is just borrowing existing weapons, armor, and other structures and pasting them in place, or perhaps looking up one new concept and seeing what it might look like, and then fiddling with it until the numbers look good.

This tends to be a self-reinforcing cycle too, because each new vehicle you build can serve as inspiration for another (the Redjack Drifter, which you'll see soon, started as a Tempest-Class Imperial Interceptor), each new weapon or structure can serve as inspiration for the next, so what you're building is not just a vehicle, but a series of concepts that you can use anywhere else.

Finally, a Usable Document

A GURPS Vehicle is not a spreadsheet, it's a statline and some notes, though those notes tend to be pretty heavy, like what sort of weapons does it have, what sort of DR and, especially what sort accommodations and electronics does it carry.  I'm still trying to figure out the best way to get all that information to the player quickly.

This part takes a lot longer, because as I write, I realize that one important bit was actually more of a sketch than I realized and it needs more detail, or that something doesn't work the way I intended to, so I need to go back and revise the vehicle so that it matches my concept better, which can result in some pretty dramatic changes.  

Most of it amounts to quibbling, though, and as a result, I often lie. It's easier to round up or down than to give you exact prices, and you don't need exact prices anyway.  Similarly, handling and SR are more a guess than a hard number, so those get fudged.  Those who sift closely through my spreadsheets will quickly realize not everything lines up perfectly. Instead, you can see the spreadsheet as a fuzzy cloud of precise values that can change or might be wrong, and the document is an attempt to describe what that precise cloud should look like in gameable terms.  The purpose of a design system is not to set hard rules, in my opinion, but to give you guidelines on creating a reasonably internally consistent set of vehicles; what comes out of it should not be the quirky, excessively precise designs that vehicles encourages, but a gameable something.

This part takes much longer than the actual design. So, at least another hour, sometimes longer.

Actually, I'm not done yet

As I work on new vehicles from old vehicles, I find errors or rethink designs.  I've found something that makes my vehicles "too expensive," a small mistake easily fixed, but now I have to back and pore over my old designs to make sure they're correct.  And as I worked on the Redjack vehicles, I came up with an advanced hyperdrive that seems more appropriate than the advanced Valiant hyperdrive, so I had to go back and update the Valiant, and then I decided the Tempest should really have one, which resulted in some serious updates to the Tempest.

None of this is strictly necessary.  We're already "lying" about the stats, so what are a few more lies?  Your designs, especially if you have so many, will always have problems, with the most centered on the first set of vehicles you worked on.  Revising them takes time and adds to your work-load, so you have to decide what flaws you're willing to accept and which not.  Sometimes, the flaws don't amount to much (the Valiant's changes were below the rounding threshold, so you won't notice them) while others are pretty big (the Tempest definitely changed some numbers with its added hyperdrive).  At what point you're willing to accept your mistakes are up to you.

I do find that it helps to realize that you will make mistakes and that this is okay.  I think that's one flaw of the Vehicles system: people want their designs to be perfect, and the complexity of the system makes that very difficult.  If you instead seek to gain a sort of "educated guess" as to what your vehicle would reasonably look like and then narrow it down with your final document, I personally think that's likely to be fine.  This is also why I don't think that even with its flaws that the SS system is actually too bad.  Sometimes it's so off that it's really noticable, but often, that's not the case.  The main reason to use the vehicles system is that you can get extra precision if you want it, and that certain aspects really ring untrue to me (electronics, weapons and ground performance, primarily).