Boghog

@boghog

Gamedev

he/him
boghog.itch.io/

Hi, I'm a game dev interested in all sorts of action games but primarily shmups and beat 'em ups right now.

Working on Armed Decobot, beat 'em up/shmup hybrid atm. Was the game designer on Gunvein & Mechanical Star Astra (on hold).

This is my blog, a low-stakes space where I can sort out messy thoughts without worrying too much about verifying anything. You shouldn't trust me about statistical claims or even specific examples, in fact don't trust me about anything, take it in and think for yourself 😎

Most posts are general but if I'm posting about something, it probably relates to my own gamedev in one way or another.

🕹️ My Games

boghog.itch.io/

🎙️ Game Design Vids & Streams

www.youtube.com/@boghogSTG

☠️ Small Updates + Dumb Takes

twitter.com/boghogooo

Boghog @boghog3/31/2024, 4:50 PM

Scoring Review : Guwange

FIRST DO THIS : if you're playing the game, go into service mode (hold start in retroarch/fbneo, use the service button in MAME). Go to Configuration, select C Button and enable Full-Auto. Some techniques depend on this.

A Quick Intro

GuwangeHUD
^ Guwange's HUD. The most important parts are the [Skull Meter], and the [Coin Chain] right below it. The ballz in the top left corner are your bombs. Also not visible here is your health meter - it's in the bottom left corner.

Guwange has 2 main goals - get as many coins as possible, and prevent the [Skull Meter] from running out in order to keep your [Coin Combo] going.

How you achieve this can be broken down into several rules of thumb, ordered by importance :

You wanna scoop up large clusters of bullets with your Shiki. Touch them, turn them pink, then kill any enemy with your Shiki to convert that pink mess into coins and enjoy the resulting golden shower. Bigger enemies tend to be the most obvious & lucrative Scoop Fodder.
You want to fill up your Skull Meter until it flashes gold & kill groups of small popcorn with your normal shot. Anything killed with normal shot will drop more gold coins. Sections with large swarms of bats, archers, etc. are usually good for this.
You want to get to 1000+ coins ASAP and stay there. Then use the Spirit Flux™ aka Coin Pumping™ technique - shooting enemies with your normal shot when your Skull Meter is flashing gold will give you extra coins. Bosses in particular are great for this type of milking. To do this effectively, hold down your full auto button and tap your normal shot button in a rhythm. If you see coins fall out and hear some dinging, then you're doing it right.

The game's chaining, despite being scary-looking, is actually very forgiving. If your goal is to just keep a chain going, your main obstacle will be a small handful of difficult links & bosses. The main challenge of the game isn't the chaining - it's maximizing coins while chaining.

Here's a more detailed guide on Guwange's mechanics, it might clarify some stuff in my review, but if you aren't already familiar with the game then it might be too much info being dumped at once.

With that out of the way, let me begin by saying :

Guwange Is The Best Cabal Shooter Ever (By Far)

The game's Shiki mechanic works like a much more robust & dynamic version of the crosshairs you see in Cabal Shooters like...well...Cabal (also Blood Bros, Wild Guns, NAM-1975, Sin & Punishment, etc).

Cabal shooters often let you delete enemy bullets by shooting them. Guwange takes this and builds on it. Instead of shooting them down directly, your crosshair slows them down. Actually deleting them requires you to kill an enemy with your crosshair. This takes a rather boring universal mechanic, and makes it highly contextual - enemy spawns & the surrounding environment dictate whether you can cancel bullets or not, and dictates how you do this.

Instead of the game being split into 2 layers, background and foreground, the game keeps most of its objects on the same layer. Bullets & enemies don't become a threat briefly when they enter your plane, they are always a threat to some extent. The game doesn't limit its interactions by locking them to a fake Z axis, everything that exists is a meaningful interactive element.

It also has a fantastic example of Double Edged Design - enemies are resources used to cancel bullets, similar to how you can use enemies as resources for iframes in 2D beat 'em ups. Because the cancel only happens within a short range however, the designers can simply use its level design to get around whatever small defensive barrier you create - you have clear limitations that they can exploit. And you will run out of enemies soon enough anyway.

Coked Up Pacing

The other thing that really elevates Guwange in my eyes not just above Cabal Shooters but even most other shmups is its insane pacing. The game has no chill - enemies spawn non stop, waves overlap if you don't quick kill them, bullets quickly fill the screen unless you're proactive in canceling them.

Even stage transitions don't let you rest. You get maybe 5 seconds of relaxing, before being immediately assaulted by enemies & having to keep your chain going. It's a game which doesn't let you disengage and immediately overwhelms you if you're not ready, and that's truly beautiful.

High Pressure Tug of War

Guwange's scoring is a game about constantly juggling different things that help you achieve the same end - all of which are quickly running out at different rates! It's a game where nothing is free, all you can do is minimize the negative cost of choices.

You want coins, but they drop in different locations and constantly require you to reposition because otherwise they'll leave the screen. However, you can't reposition willy nilly because you always need to keep your offense going - trapping bullets with your Shiki to cancel them, precisely timing kills, shooting enemies before they also leave the screen.

At the same time your Skull Meter is constantly depleting and DEMANDS BLOOD. But feeding it means killing enemies, which are your main Bullet Producers and, as a consequence, source of coins. You cannot hover over enemies with your shiki without damaging them, and even isolating single stray bullets isn't exactly free - sure you might not be damaging enemies with your Shiki, but your character's still shooting their normal shot, potentially killing enemies right above you, ones which you wanted to leave alive.

Even the extent to which your Skull Meter's filled matters - you can't get coins by normally shooting enemies unless you have 5+ skulls, and maxing out the skulls gets you more coins.

Everything has a cost, every system & mechanics is very tightly interlinked, they all feed back into the spacing game. And you will exist in a constant state of stress, trying to juggle all of this shit while dodging nasty patterns.

Variety & Context Sensitivity

Like Psyvariar Revision, Guwange ties its scoring mechanics directly to its most varied & engaging gameplay elements - the enemy bullets and enemy spawns themselves.

By detaching its "grazing" equivalent (Shiki rubbing bullets to slow them down) from your hitbox & player object, it captures a lot of the dynamics I talked about in my Psyvariar Review and even adds some unique ones by forcing you to kill enemies to cash out on all your "grazing".

You will trace bullets with your Shiki in line, curve & circular patterns
You will keep as many bullets as possible "trapped" before you cancel them all
You will "wipe" bullets like a windshield wiper
You will try to line up your Shiki with an enemy's center, so the entirety of a circular pattern gets canceled
You will time your Shiki's movement so you can cancel as much bullets as you can before killing an enemy - move too quick and not enough will come out, move too slow and they'll spread out too much to be "trapped"
You will strategically pick which enemies to kill, which to leave alive so you can optimize your tracing "paths", because without enemies you can't "cash out"

Each of the game's screens genuinely feels very different in terms of the strategies you have to use, and even tiny differences in popcorn placement around the same big enemy can make for a massive change in how you approach the encounter. For example, if Stage 1 didn't have little spear dudes come out of the big tower/turret, players wouldn't trace its path several times and would instead just trap as many bullets as they can and quick kill it.

The same cluster of popcorn can require very different strats just based on how many bullets they shoot in total, where they shoot them, and how frequently. Everything's connected.

C A T S P I D E R

Like with Psyvariar Revision, Guwange's boss milking is also a highlight of the game - each one feels very different and intense.

Bosses have some universal rules of thumb - you want to use the [Spirit Flux/Coin Pumping] technique to constaintly gain coins by shooting bosses. You also want to keep your [Skull Chain] up without damaging the boss - either by killing objects that the boss releases, or by keeping some bullets trapped.

The dreaded Cat Spider is the boss that best highlights how crazy Guwange's boss milking is. Its second phase is absolute hell of tracking randomly bouncing ballz that you have to destroy, picking up coins that killing those balls generates, keeping an eye on your chain making sure it doesn't fully drop, dodging fast bullets at full movement speed and reading the nightmarish, heavily randomized arrowhead patterns in between. It's all heavily randomized. And it gets harder on each repetition because the bullets get faster & denser.

RNG & Counterstop

Unfortunately, as good as Guwange is, it's also a very messy game. The game is full of RNG. While most of the time, this averages out over the course of a run, some instances of it are particularly nasty. It starts near the beginning of stage 1 and doesn't truly let up - which direction the big tower/turret rotates is random which can make you miss out on a lot of coins. Stage 2 has the whales with seemingly randomized shot order. Bosses are full of overlapping RNG across the board.

Another big turnoff is that the game has a counterstop. Amazing players will reach 99,999,999 puntos and basically cap the scoring system. This isn't as bad as it seems though - it's extremely difficult to do, and because the game essentially has 2 representations of score in the form of your main points & your coins, you can simply start optimizing coins themselves instead of score.

One benefit of the counterstop is that it slightly helps alleviate some of the nastier RNG - as long as you get to the counterstop, you're good.

Conclusion

Guwange is an incredible game that has been a massive influence on me personally, and one that I appreciate more & more as I get better at it & learn more about it. It's very unique, it's very intense and it's very well designed.

If you've ever enjoyed timer/cooldown/resource juggling in Overcooked or RTS (maybe MMO cooldown management?), Guwange may tap into a similar kind of appeal - give it a shot, you might fall in love!

Extra Notes

The 1000+ coin mechanic is really fun and I kinda wish it was an element during more of the game. Not being able to milk coins by shooting enemies makes the first stage feel very distinct, and getting to 1000+ coins is not only a great tangible way to see your progress as a player, but also gives a really exciting "cash out" moment. A single stage mode which starts you at 0 coins would be amazing for this type of thing.

The game has a lot of nuances to how the Shiki works. For example, the bombs it drops are delayed, so you can do all kinds of crazy shit by for example hovering over an enemy, then moving away before the bomb drops, and then timing a big cancel just as the bomb hits its target. You should check out some high level replays, they're full of all kinds of really cool subtle tricks.

You must log in to comment.

in reply to @boghog's post:

@noldea4/11/2024, 7:33 PM

played a round of guwange and managed to get to cat spider on my first credit... on super easy dip switch

Something that I like about guwange (it was the reason i grinded it a bit a few months ago when all i had was a barely functional copy of fbneo) is the life gauge. Most verticals shmups work with the "3 lives 1 hit per lives 2 bomb" model, which means that if something look like it have a 33% chance of killing the player, bombing it is probably the correct choice. It also means that scaling the amount of hit the player can take up will automatically scale the amount of bomb the player have, and both combined result in players frustrated of restarting stage 1 over and over for small progress may switch for trying to clear the game in a minimum of credits instead, only to just bomb everything that look remotely scary (which is everything in later stages), survive 5 seconds, die, and go back to bombing everything. But in guwange, since there is a dip switch that allow you to survive more or less hit per life segment, means that the player have a bit of a buffer before trying to solve the problem with a bomb; it might be tempting to try to tackle a situation not fully mastered with only the normal shot and the shiki if the player believe it can get more mileage out of the bomb later on.

It's not a game for absolute beginners, and the strong amount of snowballin and difficulty after the mid game means that a mediocre credit feeder play still look like a bomb button mashing competition, but I would recommend it to intermediate players who bounced off others cave games like dodonpachi that give you even more bombs the more you die.

(circling the subject back to score, I think it's kind of a shame that scoreplay and survival are completely disconnected, in the sense that cancelling bullets is obviously good but getting coins is worthless. Reaching the 1000 coins threesold in a zone is much harder than surviving it, but even in a zone that the player master fairly well he's not gonna be tempted to do that because it's just taking risk with no reward for his goal. Idk what would be a good option in a game with somewhat exponential scoring and very finite ressources. Two things that crossed my minds were giving a bomb every 1000 coins or having the player take less damage by hit the more score they have and they both have issues, especially the second one that require reblancing a whole bunch of things)

Boghog @boghog4/13/2024, 11:56 AM

I do love health bars myself but I gotta say the fact that you have to make such a hard decision wrt bombing in 1 hit death shmups is a big part of the appeal cause its so holistic. To use bombs well you need to :

Understand the game enough to know when you're losing controls of the patterns
Know your own limits as a player and being able to recognize when you're about to hit them on the fly
Think through the tradeoffs of bombing vs not bombing in your head - do you need the bomb for a harder pattern later on? What are your chances of success? Will you just exchange a high risk dodge now for an even higher risk dodge later?
So you have a whole skill going on that you build over time and is the result not of you training 1 thing but rather everything put together. A 33% of being hit is def not grounds for bombing, hell thats the default state during a bunch of bosses

Guwanges health bar is crazy though, the fact that you take less damage while shiki-ing legit makes for some fun decision making cuz some patterns are much easier to dodge in normal speed but mistakes will fuck you over big time. The fact that it lets you do a 100% resource reset in the final stage is also INSANE and makes 1cc attempts fun as hell.

Yea I dont really like scoring being integrated into survival on a moment to moment basis, I think itd be cool if the game was outright based around scoring on an extra difficulty/in an arrange mode/with a specific character. Where dropping the chain would kill you but you could give yourself a buffer with every 1000 coins or something where dropping the chain would reset it back to 100 and full meter but you wouldnt instantly die. That sorta stuff is a lot better cuz it doesn't interfere with survival unless you opt into it, no weird counterintuitive feedback loops or anything

@noldea4/13/2024, 9:04 PM

I'm not saying that 1 hit death is necessary a always bad decision but that it's a decision that come with tradeoff, something that a large part of the shmup community seems to not think about. It basically turns the spectrum of skill at which people can enjoy a difficulty in almost a laser (oh no if you're slightly too bad not much is worth actually engaging with), hence why zun have to bruteforce that problem with three easy modes. I'm less advocating for more shmup with healthbar as the default and more than using buffer between hit and death should be considered in the list of tools to create easy option especially compared to some variation of more lives and thus more bombs (I'm including continues in that definition). It may denature the game to some extent but all easy options does that and adding a big piles of bomb definitely damage the experience, and while you may argue that bombing is a skill it's one highly depend of the player performance in others area (basically the definition of a situation where it's advantageous for the player to bomb better change between the moment the player start the game and the one where he can think he can give a shot at hard mode). The alternative solution is either auto bomb and I kinda wish they had completely different utility than normal bomb since they still trigger my loss aversion and as a result I continue to pre-emptively bomb even when it's arguably not mathematically a good idea, or keeping the bomb on death like caladrius blaze which is basically over correcting since replacing the overdose of loss aversion by a complete lack of it cause most players to not care much about the incentive for bombing (the scoring giving bombs and the multiplayer resetting on death) and end up at the final boss with something like 9 bombs on stock.
(I think that others interesting things could be done with healthbars, like you could have little to no invulnerability on hit to give the player a small penality if he get it by a stray shot but a bigger one if he let himself get cornered by the enemies, in order to put emphasis on the "control the enemies" aspect. I think that guwange kinda try that but it's pretty easy to use the post hit invulnerability to reset the character none the less)
(the 33% was estimated by running with the idea that the player have the default bomb stock because that idiot keep dying (of course if the player keep getting that much past some point you need to kick him out !). When dying and not bombing can mean less bombs for future though pattern, in top of the intrinsic advantage of bombing, the question of whether or not the player shall bomb at a <50% situation become imo debatable)

On survival and scoring my thought are a bit messy at the moment, but basically imo it give something to do in early levels when they well mastered rather than avoid taking the slightest risk. Games that used other values to do that kinda disappointed me since the scoring usually encompase more things and thus would give more to think about, unless it's exponential or you get a bonus for each lives at the end of each stage.

Boghog @boghog4/13/2024, 9:33 PM

Little invincibility on hit sounds like a good mechanic on paper but is dogshit in practice - both Guwange and euroshmups suffer from this. It just feels awful. At that point it's better to simply slightly lower damage from stray bullets like how Deathsmiles lowers collision damage but still keep things clear & discrete. Otherwise you just get euro/kusoge where you get this disgusting gradient of mistakes that doesn't sufficiently punish sloppy play and lets you get by with mistakes.

If anything shmups should keep their harsh punishment of mistakes and difficulty overall but simply have modes that offer persistent progress - the more sloppily you allow players to play, the shittier the genre gets cause now instead of engaging with all of its elements you're just kinda stumbling your way through it.

Many games show that players can accept difficulty if they're constantly encouraged via structure/other mechanics. The #1 goal isn't to diminish what shmups are, it's to encourage players to engage with what shmups are in the most effective way. So instead of diminishing bomb decision making, devs should force extreme bomb decision making but in smaller bursts - Celeste style segments. If a bad choice doesn't doom your run but only dooms the boss attempt, it's a lot easier to engage with the choice-making.