Jazz Jackrabbit 2 Online

"mini" tileset is probably the right term, not so much for the number of tiles but for their content. Looking at the titular grass shows the issue: it’s much closer in size to JJ1 Jungrock than JJ2 Carrotus, and JJ1 is displayed at a much lower resolution than JJ2 is usually played at. Jazz and other sprites look overly large against it, and the wall texture is sufficiently noticeable that it’s hard not to make the screen look repetitive—which, again, wouldn’t be an issue at a lower resolution. This would be a great tileset for a different game, because despite their size, the images really do look very nice. The treetops are my favorite things here, both the bushy-yet-plausibly-flat ones and the gnarly spiky foreground ones, though again the trunks are too small for this game. The background hill palette could use some more thought, it blends in too much with the foreground.

"Gorgeous" is a great word from minmay for this, but the other word I’m leaning toward is "natural." It’s easy for CTF levels in particular to feel very deliberately crafted, to maintain balanced gameplay—which is not itself a bad thing, as the crafting can produce very elegant results—and this is not that. Nor is this the opposite extreme where a level is completely devoted to being a picture with zero concern for playability. Instead Timberland somehow achieves the impossible of an asymmetrical, playable, utterly natural level to which nothing can be added and from which nothing can be taken away. Alcoves and slopes abound throughout the map. Powerups and ammo are sufficiently out in the open to seem as if they grew there naturally. The tileset—the truly incredible conversion work, blending Diamondus and Inferno to an impossible degree, with a perfect palette—is so organic that it is often difficult to believe the level is composed of tiles at all. The grid is invisible, the walls full of what looks like they must be nonce or bespoke connections. An absolute delight to play, an honor to experience.

There are a handful of things that don’t wow me in this set. The silver pots are the big offenders with their surface area to color depth ratio. The giant moth or butterfly with its incongruously black outlines. The destruct blocks that are too colorful to make out their specific markings. And…….. that’s it, honestly.

Again and again while scrolling or playing through Rainbow Runner, the impression I get is "these tiles could be a tileset all on their own." You could totally have a Waz-style set with the hamster tubes, or the "blueground" of course, or the clouds and rainbows. But instead they’re here all together in a giant kaleidoscope of everything that could possibly be related to the concept of rainbows. You can be most of the way through the level or set and suddenly there’s an entirely new set of tiles you’ve never seen or dreamt of before. It goes on and on, new trick after trick, cute animal after cute animal, and yet never sacrifices quality for so much quantity.

It’s hard to know what to write for this review that isn’t just listing things in the tileset and saying how cool they are, so maybe I’ll do that. It’s a damn shame that six months later nobody else has bothered to write a review for this, so I’ll have to. Let’s talk about the clouds. They’re based on the same flat color design that bothers me in other tilesets, but here they work beautifully. They’re soft and bubbly. The option of unmasked rows of clouds against the flat background keeps things fresh and interesting, and so do the curvy and bubbly edges at every possible angle. The masked cloud rows look especially delicious when set against the unmasked ones, but even without them they’re still great, reminders of how many tilesets we’ve seen that failed to have clouds of any quality at all. And there are giant clouds too for hanging around in the foreground.

The blueground is lovely in a Skulg sort of way, and works well with the floating background trees, but what really makes it stand out is the giant staircases which I’ve never seen in any other set. Castlevania’s tiny background stairs can sit right back down, these are the real deal and they work perfectly and look phenomenal. Besides that, all the tiny little roots coming out of the bottoms of the ground reinforce the idea that these are floating platforms ripped up into the air—as they should be, up in the clouds—not just regular earth that happens to be cut off at the bottom because video games are weird. Coming up with the exact three colors of grass to look great together must have taken some time but it paid off, even though one’s green it’s not a default grass green by any means, it’s something much more interesting and fantastic.

The animals! I’m sure many of them are space fillers, like the flowerpots and other such tiny little objects taking up a tile or two in a larger section, but they’re emblematic of what may honestly be this tileset’s most distinctive feature: there are so many individual drawings. It would be impossible to catalogue them all were it not for the fact that Blade actually drew them all. It’s the exact opposite of a Raging Inferno with its single trident to distract from the tileset’s two or three textures. It’s luxury, vast, unending. The theme is rainbows and yet there is more of everything than in any dedicated tileset. And then somehow on top of all these other tiles there are animations too, scores and scores of them, territory basically untouched save for RainV and that was probably originally drawn by professionals. Not only does everything look incredible, but there are so many things that nobody else would ever even have thought to draw and include.

Over the years many fun and memorable levels have been made from limited tilesets, using flipped tiles and transparency and other funny tricks to make much out of little. Rainbow Runner presents the opposite challenge, providing so much bounty that we must learn to reduce, to find ways not to use every single element on screen—or even, perhaps, in the same level—at the same time, lest the player be utterly overwhelmed. I hope it is a challenge that somebody shall soon take up.

The background layers of Aztec 2 have taken on a life of their own since the tileset’s release, to the point that going into this review, I was crafting a narrative in my head about how they completely outweighed the rest of the tileset and people mostly struggled with how to use its vast variety of themes. But that was wrong! There are at least half a dozen levels that use Aztec 2 as a tileset primarily or exclusively. I just forgot about them somehow.

Maybe it’s just me that finds it overwhelming. Sure, there are a few sections that don’t need to be there because they hail from a time when layer 5 couldn’t be trusted as a reliable source of 1/1 layer speeds, but even putting them aside, there’s still a whole lot of stuff in here. But it’s not confusing, hard-to-use stuff by any means. It’s not all blocky (except the first section), but it’s all close enough to blocky that it’s easy to tell what goes with what. Tile order is pretty easy to follow, I can’t spot more than a handful of tiles that had to be shuffled into the next section to make the grid work. And besides being easy to use, the different tileset components (blocks, pillars, branches, foliage, etc.) all have lots of options for tiling in elaborate ways. There are a dozen or so different treetops. Rocks have short and long slopes at every angle. Tree branches can be diagonal. Water offers more options than any other tileset ever.

So why am I still kind of lukewarm about Aztec 2? Ultimately I think it’s the flat colors. They work fine for the water, which needs to be no more than a tint so it doesn’t obscure the actual gameplay behind it. Fine for the background mountains, which similarly shouldn’t be too distracting. The foreground foliage I can live with because you can have a lot of borders to break up the flat insides. Nothing is as flat as in Islands. But it’s the main blocks and the cave sections that don’t appeal to me, and they do happen to be the main building blocks of the whole tileset. They don’t look very solid and they don’t look very interesting. You have to put in a lot of time to make shapes that aren’t just rectangles, to keep things attractive, and granted you should be doing that all the time anyway, but it’s also easy to think that structures made of lots of stone blocks should be rectangular. It’s easy to get sucked into making sections of Aztec 2 levels that are nothing more than seas of undifferentiated brown.

Is it fair to criticize Aztec 2 for making it possible for level designers to not put in enough work? I’m not sure. If you stick to the more decorative stuff, the trees and the water and the gems and things, I’m happy. But it’s hard to make that carry an entire level. So do you combine with something else, and in so doing, sort of perpetuate the Aztec 2 Background Mountains issue? I dunno.

Aztec 2 is very cool. It can do a lot of things very well. It’s absolutely packed with cool little features and single-tile drawings, the antithesis of a tileset that’s nothing but textures stretched across shapes. It would be great to love it. But the flat browns are a turnoff for me personally.

This leans heavily on some elements that generally turn me off—trigger crates with unclear targets, dark lighting, general difficulty discerning where to go next—but by wrapping everything in a haunted house it ends up working really well! It’s haunted like disconcerting haunted, like you never know quite what’s going to happen next. Floors collapse under you, crates fall from above, doors open mysteriously. It’s really good use of atmosphere (in 1999!) and near-flawless use of a fairly complicated tileset. And the moment-to-moment gameplay is good too, with cool uses of gems and enemies that pose a good challenge (even if the darkness helps them a lot). Not the longest level in the world, but one with a lot of craft to it. A disconcerting pleasure to play at 1 AM.

oh, the end of the level seems to be messed up though, there are activate boss events but no boss, weird.

Over the years JJ2 has seen an explosion of multiplayer gamemodes. Sch wizards as BlurredD and EvilMike wove together arcane contraptions of sucker tubes and trigger scenery to create levels that, even when played in vanilla JJ2, still followed the scoring rules of Team Battle or Last Rabbit Standing or the like. JJ2+ canonized these, and others with more complicated rulesets, from Pestilence to Headhunters. Mutators expanded the playing field still further. But somewhat lost in all this was the idea of single player gamemodes.

In general there have only ever been two goals in single player levels: either get to the exit (which may involve a boss battle), or collect all items (usually coins). Within collecting levels there are two subtypes, one where collecting everything makes you win immediately and one where you have to go to a specific place afterwards, but they’re largely the same thing.

But there’s a third mode that’s lurked around the edges, only making an appearance every once in a while: horde mode. The level design stays constant, but enemies keep appearing, and either they die or you do. A major example is TDI_07.j2l from Moonblaze’s “The Demon Invasion,” where a regenerating coin and an invisible coin warp serve as a countdown timer. hgfDiamondusColosseum.j2l from happygreenfrog’s “Operation Cleanup: Turtle Terror Revisited” uses scripting to implement an enemy quota, so the level will not end until you’ve killed enough (regenerating) enemies. Both make the assumption there must be some sort of measurable goal that the game itself keeps track of for you, or else why would you bother?

The Lapidarian Chaparral says screw that noise.

The Lapidarian Chaparral decides that you, the player, are in charge of keeping track of how well you’ve done and whether you’ve accomplished whatever goal you might choose to set for yourself. It demands you make up your own emergent gameplay. It puts itself in your hands as clay to be molded into whatever experience you prefer. It is single player by way of sandbox. It is, in a way, a metaphor for JJ2 and JCS as a whole.

Sure, with some work Lark could probably piece together a complicated system of a generating coin, an electro-blaster SCE, and various belts, animated tiles, bridges, and so on, that periodically dropped a coin into a pseudo-random spot in the level, so that you have to keep moving around to find the next coin. It would be pretty cool. But there’s a boldness to giving up on that altogether as irrelevant.

Sure, the concept still has some limitations as implemented. In the current version of JJ2+, regenerating enemies and pickups don’t give points, but pickups spawned from regenerating barrels do, so if score is your goal, you should focus there. Sure, certain areas are purely safe, so if your goal is actually to set a stopwatch and see how long you can last, you’re just gonna hang out next to a carrot and walk away from your computer. But does Lark really want you to do that? More importantly, do you really want to do that? Or do you want to blast turtles?

Anyway. Cool to see Jungrock. Level looks nice. Didn’t notice any bugs.

Gosh. Lark. Long time, kid. Thanks for dropping by. Happy new year.

Ah, I missed this one when I was doing my Psych3-edits tour the other day! This has some good moments with warps but doesn’t always do a great job of trying to impose layout edits. I miss the fly carrot but reluctantly accept the float sucker stomp sequence is also neat. Definitely on the more interesting side of the genre, though imperfect.

A decently lengthy pack with highs and lows. The highs show up more toward the end, so you leave with a good memory, but it takes some work to get there. I like the original, memorable sequences like the airboard level, the frog level, the fun and colorful rainforest level—I’m less into the trigger crate chasing, and there’s a lot of it.

I think Turtemple sounds good with these instruments, very ominous. The others fall into the standard remix trap: a bunch of random noises but every once in a while you get to hear the melody of the original song and think “oh yeah I like that original song!”

A bunch of duel-sized, high-concept CTF levels of varying quality. The pack starts off strong with krcage, which has some interesting gameplay around the carrots, with vines and electroblasters. The bottom area is a bit too simple, though, and the double vine rows could probably be reduced to single vines to make the level play a little faster. krdeep also has a focus on two +1 carrots instead of one full energy, this time paired with copters, but the rest of the level is too difficult to parse visually, and Spaz can get up the central castle much faster than the other characters. krlz I’m not a fan of, only Spaz can get up the central column in any reasonable timeframe and the rest of the level is blocky and hard to see in. There are other CTF maps that do this diamond layout more gracefully. krnj is a bold attempt to make a vertical CTF level, but I’m not convinced it’s any more successful than most others, since there’s nothing to keep you from falling down from one base to the other. Also I asked my brother, who lives in New Jersey, about its accuracy, and he says “Seems a little desolate for the most densely populated state!” krsand seems fine to me, I could imagine giving it a little more polish in certain spots but the basic layout is totally functional and enjoyable. krvolc is a letdown though, the bases are so incredibly close together I don’t see how this ends up playing well.

In conclusion, some of these levels are better than others, but it’s not a bad effort, and I like getting more bite-sized maps out there for quicker games.

Ah, 2015! A heady time full of promise. Let’s look at the angelscript API and see what it can do for a single player level. Different palettes at different stages of the level? Absolutely! Enemies with more health, other sprites used as enemies, other sprites used as (unremarkable) weapons. Sure! Totally sick fire-breathing platforms moving in all directions! Anything seemed possible. People were building on JJ2 gameplay to add exciting new elements to dazzle and amaze and entertain.

Good times.

Tweedle Wheedle is not without bugs, I suppose, and the eponymous boss battle is frustrating as all Tweedle fights seem to be, and the layout occasionally gets confusing, but it’s unrelentingly cool. And even without all the scripting stuff the layout is still really good, doing interesting things with the psych3 base, then adding lots of really distinctive design elements to the rest of the level. Just a fun time throughout.

This is a good glow-up! It’s striking how much of a difference the fade color and the foreground layers make, and the intro area from psych3 has a lot of cool edits with pickups off to the side and stuff. Using a Cheshire2 is very smart—surprisingly few psych3 edits seem to do this—though I don’t understand why the fly carrot was removed here, as it was the original’s most distinctive feature. The spring section that replaces it is pretty cool, though. The rest of the layout is also generally good and fun with some fun takes on Psych wall shapes, albeit with the occasional dud moment (everything relating to TNT, some alcoves it’s easy to get stuck in). This isn’t a transformational take on JJ2 gameplay but there’s a lot of aww yeah neat. A great take on psych3.

Pretty much the most bare bones Psych3 edit you can imagine. A handful of enemies, almost no pickups, no layout changes. Functional but uninteresting.

I’m not thrilled with the secret level entrances, particularly the Tubelectric one, but I guess they’re not horribly out of line with existing JJ2 secrets… if this were really part of the game, eventually players would have found them over a period of months or whatever. Still, more visual cues would have been nice I think. But whatever that’s not the focus here.

Far Out is actually pretty good in terms of new event placement while sticking to the existing psych3 level. There’s a decent variety of pickups and enemies, and the little pipe organs on the platforms after the warp target are a good touch that make that area much more interesting to navigate. The new parts of the layout kind of lose me, though. The idea of warping through a bunch of little rooms (kind of like in Medivo1) is nice but eventually I got tired of hunting for trigger crates, and there are too many invisible warps and weird blocks that need to be destroyed or removed in unpredictable ways.

The other, all-new level here, though, Pinball Land, I actually really liked. Toward the end it gets annoying in the part when you track down four trigger crates behind lots of layer3 and stuff, but the rest of the level is I think a really good take on a Tubelectric secret level. There are clever pinball tables, and crucially, they’re not all at once, they appear from time to time within the level so they don’t get stale. There are spike platforms, one of Tubelectric’s other main unique features. Goodies feel fairly plentiful and the graphics are nice and varied. This is a nice one.

The way to access Far Out is a neat idea… if you find a secret earlier in the level, you get an alternate ending at the end of the level. It’s an understandable way to attempt to get the mid-level secret level experience while not knowing how to fix vanilla JJ2’s treatment of secret level area (it’s possible but takes some setup). The implementation has issues, though: the tiling at the end is pretty dubious, and more importantly, nobody’s going to find the trigger crate without knowing to look for it. If you find an area totally covered in layer 3 with no goodies in it, you’re not going to think “oh, I should stomp everywhere in case there’s a trigger crate.” And even if you do stomp it, there are no nearby effects, so there’s no way for the player to know what happened or why the secret level entrance is/isn’t open. Finally, this is a level with a bouncer powerup, and it’s trivial to shoot those bouncers through the trigger scenery events and hit the secret level signpost that way.

I couldn’t actually figure out how to get to the Tubelectric and Labrat secret levels in these edited versions. :?

Anyway, the main attractions here are the secret levels themselves, and they’re not great. “Shocking Experience” is just an extra (non-thematic) boss fight—is that really a good reward for finding a secret? And while this version of Far Out makes a good decision by using a unique palette, like Return of Birdland does, there’s not much to recommend it past that. It’s ‘finished’ to a pretty minimum extent, with strings of pickups and no other edits, and the foreground layers are a mess that you can barely see through.

The basic idea here is good, but I don’t actually want to play these secret levels, so I don’t want to apply the patch to my official labrat2 and tube1.

There’s not a lot of level design flow here. You kind of wander around, maybe collecting a bunch of goodies, maybe jumping over some enemies, not sure if you’re going the right direction, and suddenly there’s an exit. But what’s interesting about this pack is how much Talec is (already, in 1999) pushing up against the limits of what the game is supposed to support. Frogs, invisible enemies, secret levels, next level settings. (You’re asked repeatedly to use the JJNXT cheat code in order not to break the secret level bool, though in 2022, we know that using other events could have avoided this issue.) And there’s something thematically fitting about the pack’s focus on secret levels in light of how some of those features are kind of secrets within JJ2. The first level is an edit of psych3, an unfinished secret level, if not the most interesting one. There are more pickups, the warp target doesn’t go where I’d expect it go, but nothing truly remarkable. In general the disappointing factor in this pack is that secret levels are never really treated as a reward. Talec learns from Labrat3 that secret levels are supposed to involve turning into a bird, but there aren’t the profusion of goodies you’d expect from a JJ1 secret level. And you need a secret level to have some kind of desirable experience or else the player is actually working to their advantage to skip it, because they save time that way.

I don’t have much to add to Stijn’s solid review. These are good levels for sitting back, tuning out, and playing some JJ2 with nothing else to worry about. I think the Carrotus levels are the weakest, feeling too rectangular and not very organic, but it’s nice to see the float lizard copters getting some use, even if not to the same extent as in The Big Rescue. It’s great that the layouts give you a lot of choices of which directions to go in and have lots of extra goodies just off the main road. There are a lot of uses of buttstomp blocks to mark out goodies just for Jazz or just for Spaz(/Lori), which mostly works well until the level with TNT, at which point they all start to be accessible to either character, and it’s not clear if the pack realizes this. Likewise, powerups and shields frequently appear behind thin walls that can be shot through with electroblasters or bouncers. But the variety of goodies and powerups is a welcome one (even birds and invincibility carrots make appearances), and they help against the enemies, which are fairly numerous but never feel particularly difficult when faced with so much ammo. It’s a good time to play.

abgrenv: this tileset is expected for use in JJ2+, which lets you put traditional skyboxes in any level, so there is no need to include them in the tileset:

	jjPIXELMAP skybox(TEXTURE::NORMAL);
	for (uint x = 0; x < 256; ++x)
		for (uint y = 0; y < 256; ++y)
			skybox[x,y] = (skybox[x,y] - 176) / 2 + 192;
	skybox.makeTexture();
	jjTexturedBGUsed = true;
	jjSetFadeColors();

What I particularly enjoy about this level is the degree to which it has distinct areas. Instead of being constants, elements like giant carrots or snaking vines or wooden platforms will appear in clusters, making those areas memorable and giving the player a feeling of progression. More such focuses keep turning up throughout the experience, including progressively larger cave areas, bigger uses of wooden platforms, and a lategame forest and underwater section. Besides that, there’s a good healthy amount of pickups all through the level, on the main path and in alcoves and in secrets, and enemies are varied even if never particularly dangerous. The layout and graphics do a lot to avoid becoming boring: the floor is constantly rising and falling as you move along, you change directions a lot, there’s some variety in wall thickness (although it would feel more Carrotus-y with some thinner platforms), various bridges and wooden things appear to keep things fresh.

Other than the rock pushing section, at the bottom, though, nothing in the course of the level ever feels unique… which, of course, is totally fine, it just means that the level isn’t transcendent. But I’d like to see more attention paid to how the level can feel distinctly Carrotus/Easter, particularly when using events instead of just stuff the tileset provides. Bird cages would feel at home here, for example. The wooden platforms and floors didn’t always feel great, due to a combination of how thin they are drawn and how inconsistently solidly they were placed (layer 3 vs 4). One area is covered in thorns that don’t hurt you, followed immediately by a small pit of thorns that does hurt you. Speed blocks are used in place of lock blocks. Diagonal carrots only ever sprout from the ceiling. Both(?) powerups can be obtained by shooting bouncers at them instead of using the intended warps.

It’s a good level with good, if never particularly challenging, gameplay, but it lacks a certain identity.

The aurora in layer 8 is quite lovely, but the rest of the level doesn’t charm me as much. The layers of mountains and trees in front of it are dull gray, neither silhouettes in front of the green glow nor illuminated by it. Frosted Peaks does a better job of blending these background colors. The darkest color in the blue grass gradient is still too bright, distractingly so. Electroblaster pickups are common but the layout does not seem designed for them. Barrels and masked trees appear because the tileset has them but mostly don’t seem to contribute to gameplay, instead getting in the way of easy movement. Multiple top right corners of walls can’t be jumped up against because of improper masking (and no One Way events). Most egregiously, innumerable platforms are simply too high up to reach unless playing with a double jump. Several in particular are placed ever so slightly higher than the highest point reached by a spring that appears to be intended to take you there, but, only if you are Spaz. Other springs bump you into ceilings or are hard to find, set not quite flush against the nearest walls. The design process seems to have prioritized graphics (though with some questionable palette choices) above gameplay flow, but both are important. The areas with bridges or vines are generally nice and interesting, but it’s frustrating to get around the rest of the level.