Horizontal panning

[Sumez]

It seems that ever since the first Tatsujin game, and probably in certain instances before then as well, being able to scroll slightly left and right has been quite common practice in vertically scrolling games, probably a lot more often than not (or at least int he games I'm into).

Snooping into a couple of Cave titles I've come to the conclusion that this means the playfield is typically ~24.1% wider than it appears, or going the other way around, that around 19.4% of the total available playfield is actually hidden outside the borders at all times.

In effect, this means a lot of (the majority of) vertical games are in fact almost quadratic in size (providing they are viewed on a 4:3 monitor), placing them somewhere between vertical games without panning and horizontically oriented vertical games (like Giga Wing).

Of course, this is common knowledge among shmuppers (at least without the exact numbers ), but doesn't seem to be something most people think much about. Personally I think it's barely noticeable that games do this, and I actually had to replay a couple of games to see whether they did it. For example I was surprised panning isn't present in Psikyo's vertical games, and I was absolutely certain it was a part of Grind Stormer (arcade), but it isn't, so apparently it's something developers are able to implement into the game without most people really noticing.
What's confusing me is that when developers port their games to home platforms and reformat the screen to a 4:3 horizontal format, they usually retain the panning with some added letterboxing, when they could easily use the available playfield to fill out most of the screen. (for example Donpachi is a lot wider on Saturn than on arcade, but still pans left and right)
Why do you think they do this? The obvious answer could be to keep things fair, comparing scores with people playing the game in true tate mode, but I honestly don't think most developers think like that (considering a bunch of other things they tend to do), maybe the panning is considered an important part of the gameplay, or is it simply done for the visual effect (making the game look more smooth and less confined/old school)?


Since I'm currently developing a vertical shoot'em up I'd really like to hear people's thoughts on horizontal panning in vertical games, both from the perspective of a player and a potential developer, anything that crosses your mind.
How far do you think you can get away with panning the playfield? Would it become a nuisance if I exceeded the ~24%? Would you rather prefer to not have it at all, for a complete overview of the situation?
Does it even have any real effect on gameplay or is it merely cosmetic?
Would it be wrong to give people the option, when playing on a non-rotated monitor, to view the entire playfield without the need to pan, to get the most out of the monitor space? (while still panning in tate mode)

It may sound like I'm putting way too much thought into this, but when you're on the developer side of things, these things turn out to have more of an impact than you'd think.

[drei :3]

Whatever you do, if you have view modes that affect when a player sees stuff, have separate scoring tables. Don't use bullet warbling or whatever it's called, that is when otherwise non-homing enemy bullets move a bit with you when the screen moves sideways.

If you don't want that offscreen enemies shoot at the player, panning makes development considerably more complicated. Do you want that the player basically controls when enemies on the sides shoot at him? And what about games with 2-player mode? Do you want that a player has the power to let the other player crash sideways into an enemy ship?

Call me a perv, but I like vertizontals. I think that vertical scrolling can work very well with a non-flipped monitor if you make the most of it, making the most of it means putting score info at the sides instead of the top or bottom and making the most of it of course means not hiding some part of the playfield at the sides. You can also have translucent parts at the sides, so while the player can't move into these parts and enemies can't shoot in these parts, the player can see some enemies come in from the sides or leave sideways and also hit them there with a bomb.

[BarfHappy]

In cave x68000 games, usually the field is a tad smaller in total width compared to height, but not by much. When pixel format is taken into account (not talking about raw resolution), total width is 96% of height. (the visible portion is 75% of height).

I like when there is horizontal scrolling as well for a 1player game, it gives you some surprises sometimes, some bonuses you didn t see previously (like DDP series bees).
For 2 players games, however, it just melts my brains... i cannot play a cave game with a friend.

[Sumez]

Thanks for some very intelligent feedback so far.

I like drei's idea of having translucent sides, though it does seem like something that could have a rather big (almost game defining) impact on the game experience. I still sort of have the idea that I want tate mode to be the game's "true" setting, but I wonder if I should just get that idea out of my head and realize that only really makes sense in actual arcade games.
By "warbling", do you mean a potential bug where enemy bullets won't scroll along with the background? So far all my "panning" code is done entirely in the drawing process, so it's not affecting any game logic.


@Barfhappy: That's basically the calculations I made too
Personally I think I like the scrolling too, but considering I hardly ever notice it, I still wonder how big a difference it makes. Thanks both for bringing up 2 player modes - I hadn't even thought about those complications. However I'm currently considering a sort of unconventional 2 player mode using a split screen interface and some unique gameplay ideas to go along with it. (probably closer to Kingdom GP than Twinkle Star Sprites).

[Sumez]

Some additional thoughts on the gameplay affects of panning:

The obvious route here would be to look at how the well known commercial titles do it, but as already stated I hardly ever notice the panning, but I'm sure some of the vert-fanatics in here know all about this.

As already stated the panning in my game is done entirely in the drawing phase - in fact I'm basically drawing everything off-screen using the full width of the level, and then displaying that in a slightly thinner frame with the scrolling applied. But are there other things that would be "good practice" to consider? Eg. not allowing enemies to fire unless they're inside the visible area of the playfield - obviously you wouldn't want to have enemies off the edge of the playfield fire at you, but should the player-controlled scrolling be considered too?

Also, could some enemy waves behave differently depending on the pan/scroll of the playfield? Eg appear further to the right if the player scrolled all the way to the right of the playfield? Or any type of AI that follows the player only activating when they're scrolled into view? I'm fairly sure I've never seen any game behaving like this, but you never know..

Finally, should I still calculate collisions outside of the frame? Normally you wouldn't be able to shoot enemies outside of the playfield, but as far as I remember from games like Espgaluda, etc. you need to actually scroll to the side to have any chance of killing popcorn enemies that aren't currently in view, even though your main shot spreads out a lot, thus meaning those little guys that are all the way to the right or left can turn out to be quite persistent. Or maybe it just seems like that to me because I can't see them?

[BarfHappy]

I personally let enemies fire when they are off screen but still on playfield but yes, maybe that would be best not to...
What i will do i think is that enemies will fire once they arrive on the part that is always shown. Enemies that will be always in a zone that can be masked will be just popcorn stationary foes or hidden bonuses.

I think one of the key things to keep in mind anyway is to try to keep as much as possible the bosses in the area that is always visible. Othewise... just try to see what works best

edit: i just modified my shifting way. Before i converted the actual position of the player into a pan value... it was getting really annoying when dying in the right corner, the screen would repan abruptly... instead now, i just scroll when there is a left/right input by (speed of the player -that s pixel-equivalent value-)*(fieldTotalWidth-fieldDisplayWidth)/(playerXMax-playerXMin) until it reaches the borders. that feels much better overall, when the player dies the screen does not pan.

[louisg]

imo enemies should never be firing from way off screen. This always bugs me in games like Revenge of Shinobi. I know that's not a shmup, but still, I think it tends to make games seem unfinished. You might be able to get away with it, though, in a manic "bullet maze" style shmup.

Also be sure to avoid bullet wobble That is, when the screen scrolls, the sprites should scroll with it. If bullets don't move with the screen, it makes it seem as if the bullets *are* moving when the screen scrolls horizontally.

My two cents about panning overall is that it adds a lot to the feel of a game, even if it's very slight. It makes the player feel as if they are actually moving around a map and not just on a guided tour (even though they are).

[S20-TBL]

My two cents about panning overall is that it adds a lot to the feel of a game, even if it's very slight. It makes the player feel as if they are actually moving around a map and not just on a guided tour (even though they are).

I know they're horis, but this is what I love Xexex and Mystic Riders' wraparound maps for. I have freedom to move around and explore my surroundings, fight varied enemy types/patterns and maybe discover some hidden goodies.

The real danger is in suddenly exposing an enemy you didn't know was hanging out on the edge of the screen all the time and you bump into it, or it fires at you point blank. Adjusting the wobbling rate should technically take care of this.

[Blackbird]

Panning serves a very important visual feedback function in a scrolling shooter. In a scrolling shooter, when you reach the edge of the screen, you cannot move any further - in other words, you run out of space to run from bullets. If the screen begins panning as you approach the edge, that creates a sense of dramatic tension as the player realizes he's running out of room, unlike a still frame, where you just stop cold.

Furthermore, panning introduces a strategic element to the game, where you can reduce or eliminate the number of bullets fired at you by panning the originating enemies off the screen.

[Sumez]

My two cents about panning overall is that it adds a lot to the feel of a game, even if it's very slight. It makes the player feel as if they are actually moving around a map and not just on a guided tour (even though they are).

I completely agree, and even though I originally wanted to make it possible to see the entire "scroll area" when playing horizontally, I'm probably going to decide against that anyway after playing around a bit with the concept. It simply adds a much better feel, and I can't help thinking the games that don't do this feel a bit old and deprecated.


As for enemies not firing depending on your scroll, I played around a bit with Espglauda and confirmed that it definitely doesn't allow it. And not because the far edge is a "no fire zone", but simply because the enemies aren't in view. I also noticed that a lot of enemy patterns changed depending on where on the screen the player is located!

The real danger is in suddenly exposing an enemy you didn't know was hanging out on the edge of the screen all the time and you bump into it, or it fires at you point blank. Adjusting the wobbling rate should technically take care of this.

Though it might be obvious to scroll like a typical sidescroller and scroll when the player crosses the center of the display area, it seems to me that most, if not all, vert shooters scroll "constantly", slowly decreasing the space between the player and the visible edge of the screen, so that you can only scroll all the way to the side by moving as far to that side as you can get.
I think this does create a better effect as the scrolling seems more smooth that way, and there's probably some more ideas behind it, like keeping more of the center visible at all times. But it also makes me worry about bumping into unexpected enemies, though I don't think you'd ever have any static enemies in the far edges except from those ground type that you can fly right over anyway.

Panning serves a very important visual feedback function in a scrolling shooter. In a scrolling shooter, when you reach the edge of the screen, you cannot move any further - in other words, you run out of space to run from bullets. If the screen begins panning as you approach the edge, that creates a sense of dramatic tension as the player realizes he's running out of room, unlike a still frame, where you just stop cold.

Interesting point of view!