Using Pinball Construction Set
Jun. 2nd, 2019 09:55 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
mmcirvinSo, having reawakened my long-dormant memories of how to make a usable Atari disk image for saving files, I spent some time actually messing with Pinball Construction Set, Bill Budge's 1983 game that introduced most of the world to the concept of a game editor. Is it possible to make a decent pinball with this thing, one that will be fun to play in 2019?
Amazingly, it is! The game's limitations become immediately apparent when you mess with it, and there are a few things I wish you could do that you can't, but I think the program's limits were actually a big part of what made it enjoyable for the general public. You can whip together a table that can do most of what Pinball Construction Set could manage in a couple of hours, and the rest is just subtly tweaking and decorating it. There are modern PC pinball editors designed to let you put together accurate simulations of complex modern tables, but to use them well you need to master half a dozen technical disciplines. There's not much of that here. It sticks close to the goal of being a fun toy. Everything in the game editor works through a simple joystick-and-fire-button GUI, maybe the first time many people ever encountered something like that. The beautifully written manual (a time capsule from the absolute high point of software marketing and packaging) is 12 pages of text and three cheat sheets, and it covers everything comprehensively, even giving design hints and subtly acknowledging some of the game's anomalies and quirks.
There's some graphical crud here that is probably from bugs in the application, which I haven't bothered to clean up. Also made no effort at all to decorate the table or the "backglass", which you can do; these are just the visual elements the game gives you. (Adding artwork to your pinball is one of the more cumbersome and unpleasant elements of the editor, unfortunately, and I didn't mess with it much. Editing the polygonal table geometry can also be a bit frustrating in this pre-mouse era.)
The first thing I did was to save a file with just my rendition of an "Italian bottom", that is, the more or less standard arrangement of slingshots, flippers and wire guides down at the bottom of the table. I figured I'd be going back to some variant of that a lot.
What I was going for here in general was nothing particularly original, but a homage to Ed Krynski's 1970s Gottlieb drop-target-fests like Big Shot or El Dorado. It seems like the first tables people make with PCS often clutter up the middle with a lot of bumpers and stuff, but I went for an open layout where the main goal is supposed to be bashing away at that bank of eight drop targets on the right (actually two adjacent banks of four). The tightly grouped pop bumpers in the upper left actually have a lovely Pat Lawloresque straight-through-the-bumpers shot off the right flipper, that also nails the spinner below them. That was an accident, and I ought to put some target in the path to make it worth more, because it's great. I put in a little diagonal barrier to keep the ball bouncing around in the pop bumper area once it gets in there, and then I put a couple of stand-up targets on the bottom to make it more than just a passive barrier, and because designers like Lawlor do that a lot.
The gizmo behind the small third flipper is a simple multiball lock. You can't do anything fancy like require some action to activate the lock; it's always active, it holds three balls and when it's full it dumps them all out for multiball. Simple to understand.
Lots of reviews of PCS complain about the weird and unbalanced scoring options: for some reason, the game emphasizes gigantic end-of-ball bonuses that dwarf the amount of points you can directly award for a target. Maybe the reason the bonuses get so big is that they are also the only really nontrivial table rules you can establish. There's a single type of logical element, an AND gate, and all it can do is AND up to three items together to do one of two things: add points to the bonus, or bump up the bonus multiplier. I did the common thing of making completion of those three rollover lanes at the top increase the bonus multiplier, and I have a bonus for getting the two standup targets (maybe too cheap a victory, I don't know), a bonus for getting multiball, and a relatively big bonus for completing all eight drop targets on the right. That last goal is very hard, and maybe it should be worth even more, but I didn't want to make it too much of a Golden Snitch. You can't award specials or extra balls or anything like that, as far as I can tell, or I'd definitely put one there.
But for purposes of a program like this, I'm not sure you really want much more complex table rules. I mean, I get it, I once had great fun wiring up a baroque state machine in Disney Infinity to play the old Merlin Magic Square game (with added killer robots) just to see if I could, but keeping it very simple probably feeds the player's sense of accomplishment and avoids getting people into really frustrating situations where they break their table by wiring up unexpected behavior they can't figure out.
What did I miss? The main thing, overwhelmingly, is that I wanted more control over the geometry of the flippers. You have big and small left flippers, big and small right flippers, and all of them have a resting state where they're canted at a weirdly steep angle, much steeper than a typical bottom flipper but not as steep as a sideways flipper. This makes it really hard to aim a shot with a trapped ball (though I appreciate that you can trap balls) or a ball coming down the inlane: the ball slides down the steep flipper in an instant, much faster than on a typical real pinball. It makes my target-shooter table design particularly challenging. I'd like to either have some control over the angle of the flipper, or at least have a better default.
Still, this pinball I made using this ancient toy editor from the dawn of personal computing isn't the greatest, but it is actually fun to play! I got it to the point where I wanted to play it more than tinker with it some more. You do get just a little taste of the pinball design process, where the designers make a plain prototype of the physical design called a "whitewood" and then spend some time "flipping the whitewood", that is, playing the naked table so that they can tell what works and what doesn't, and what shots are fun to make so that it's fun to reward the player for making them.
Amazingly, it is! The game's limitations become immediately apparent when you mess with it, and there are a few things I wish you could do that you can't, but I think the program's limits were actually a big part of what made it enjoyable for the general public. You can whip together a table that can do most of what Pinball Construction Set could manage in a couple of hours, and the rest is just subtly tweaking and decorating it. There are modern PC pinball editors designed to let you put together accurate simulations of complex modern tables, but to use them well you need to master half a dozen technical disciplines. There's not much of that here. It sticks close to the goal of being a fun toy. Everything in the game editor works through a simple joystick-and-fire-button GUI, maybe the first time many people ever encountered something like that. The beautifully written manual (a time capsule from the absolute high point of software marketing and packaging) is 12 pages of text and three cheat sheets, and it covers everything comprehensively, even giving design hints and subtly acknowledging some of the game's anomalies and quirks.
There's some graphical crud here that is probably from bugs in the application, which I haven't bothered to clean up. Also made no effort at all to decorate the table or the "backglass", which you can do; these are just the visual elements the game gives you. (Adding artwork to your pinball is one of the more cumbersome and unpleasant elements of the editor, unfortunately, and I didn't mess with it much. Editing the polygonal table geometry can also be a bit frustrating in this pre-mouse era.)
The first thing I did was to save a file with just my rendition of an "Italian bottom", that is, the more or less standard arrangement of slingshots, flippers and wire guides down at the bottom of the table. I figured I'd be going back to some variant of that a lot.
What I was going for here in general was nothing particularly original, but a homage to Ed Krynski's 1970s Gottlieb drop-target-fests like Big Shot or El Dorado. It seems like the first tables people make with PCS often clutter up the middle with a lot of bumpers and stuff, but I went for an open layout where the main goal is supposed to be bashing away at that bank of eight drop targets on the right (actually two adjacent banks of four). The tightly grouped pop bumpers in the upper left actually have a lovely Pat Lawloresque straight-through-the-bumpers shot off the right flipper, that also nails the spinner below them. That was an accident, and I ought to put some target in the path to make it worth more, because it's great. I put in a little diagonal barrier to keep the ball bouncing around in the pop bumper area once it gets in there, and then I put a couple of stand-up targets on the bottom to make it more than just a passive barrier, and because designers like Lawlor do that a lot.
The gizmo behind the small third flipper is a simple multiball lock. You can't do anything fancy like require some action to activate the lock; it's always active, it holds three balls and when it's full it dumps them all out for multiball. Simple to understand.
Lots of reviews of PCS complain about the weird and unbalanced scoring options: for some reason, the game emphasizes gigantic end-of-ball bonuses that dwarf the amount of points you can directly award for a target. Maybe the reason the bonuses get so big is that they are also the only really nontrivial table rules you can establish. There's a single type of logical element, an AND gate, and all it can do is AND up to three items together to do one of two things: add points to the bonus, or bump up the bonus multiplier. I did the common thing of making completion of those three rollover lanes at the top increase the bonus multiplier, and I have a bonus for getting the two standup targets (maybe too cheap a victory, I don't know), a bonus for getting multiball, and a relatively big bonus for completing all eight drop targets on the right. That last goal is very hard, and maybe it should be worth even more, but I didn't want to make it too much of a Golden Snitch. You can't award specials or extra balls or anything like that, as far as I can tell, or I'd definitely put one there.
But for purposes of a program like this, I'm not sure you really want much more complex table rules. I mean, I get it, I once had great fun wiring up a baroque state machine in Disney Infinity to play the old Merlin Magic Square game (with added killer robots) just to see if I could, but keeping it very simple probably feeds the player's sense of accomplishment and avoids getting people into really frustrating situations where they break their table by wiring up unexpected behavior they can't figure out.
What did I miss? The main thing, overwhelmingly, is that I wanted more control over the geometry of the flippers. You have big and small left flippers, big and small right flippers, and all of them have a resting state where they're canted at a weirdly steep angle, much steeper than a typical bottom flipper but not as steep as a sideways flipper. This makes it really hard to aim a shot with a trapped ball (though I appreciate that you can trap balls) or a ball coming down the inlane: the ball slides down the steep flipper in an instant, much faster than on a typical real pinball. It makes my target-shooter table design particularly challenging. I'd like to either have some control over the angle of the flipper, or at least have a better default.
Still, this pinball I made using this ancient toy editor from the dawn of personal computing isn't the greatest, but it is actually fun to play! I got it to the point where I wanted to play it more than tinker with it some more. You do get just a little taste of the pinball design process, where the designers make a plain prototype of the physical design called a "whitewood" and then spend some time "flipping the whitewood", that is, playing the naked table so that they can tell what works and what doesn't, and what shots are fun to make so that it's fun to reward the player for making them.