Interaction: The Key to Depth?

You’ve talked a lot about depth and complexity in regards to game design, but what about interaction between mechanics? How does it relate to depth and complexity? Any examples done well/poorly?

Alright, if game quality was purely tied to the number of states possible then the scale would not be linear. It would be exponential or logarithmic, like decibels. A strictly linear addition of states does not create a big jump in quality.

For a mechanic to significantly improve the game it must interact with the other mechanics, multiplying or exponentiating the number of possible states.

Interaction between mechanics creates context across time, allows variables to occupy a wider range of possible values and combinations of values, and ultimately is what creates depth.

It’s not enough to have a ton of mechanics if none of them interact with each other, or even synergy, which I consider to be a step below direct interaction.

Here’s a simple example. You have walking, and you have jumping in a totally blank 2d environment. Just a character and the floor. Imagine for a sec that you’re only allowed to jump when you’re totally still. So you have all these states for the positions which you can walk to, and maybe the velocities you can be in all of those positions, then you have like 30 or so states for all the positions you end up during your jump, rising and falling (imagine it takes 30 frames), multiplied by all the positions you can be in when you jump.

Now imagine that you are allowed to jump while moving, so this creates new states, where you have the combination of your X axis ground velocity and Y axis air velocity. This is direct interaction between mechanics, which creates additional states which the character can occupy. This game, which allows walking and jumping to be combined, has a broader possibility space, and also allows events in the past (walking) to affect events in the future (jumping after walking).

This is part of why I love the kick glitch in mirror’s edge so much, because it’s so flexible. It inherits a lot from your prior state and can create a large range of new states based on your input. It’s affected by how long you’ve been running on the wall, what angle you face into the wall, how fast you were moving beforehand, where you face when you jump. Through all of these there is a wide range of kick glitches you can perform. It’s possible to do one dropping down and landing on a platform far below, possible to do one that is long and straight, possible to do one out to the side, possible to do ones that gain more speed or less.

Combos in Smash Bros are also a great example of the interaction or synergy that mechanics can have. Low tier characters have poor synergy, high tier ones have amazing synergy.

A game full of contextual actions that exclusively do one thing would be a game where no mechanics have interaction. Or a 3d zelda game. They have a lot of really segregated mechanics in those.

4 Criteria for Depth

Could you link to your four criteria for depth? You mention them often but I can’t find that specific blog post.

Don’t think I made a specific blog post about it.

My criteria are more a rule of thumb, but usually a helpful one. The criteria can be applied to any element or collection of elements within a game.

1. Every element has an individual niche (no perfect overlap or overshadowing of elements)
2. Any given element has a variety of uses (doesn’t just accomplish one thing)
3. Any given element can be manipulated or modulated to obtain better or at least varied outcomes based on the way it is used or context it is used in. (you can hold the button longer, press the button later, or relative to a certain position or velocity)
4. Elements have synergy or interaction between one another. (either directly creating a unique animation state or just allowing combinations of variables otherwise not possible.)

In Mario, jumping is a basic example. There’s no other means to get high except to jump, so it is the sole proprietor of its domain. Jumping can let you get on top of blocks, hit blocks, jump on enemies, get over pits. You can jump higher or shorter depending on how long you hold the button or how fast you’re moving when you decide to jump. Jumping + moving is what allows you to jump on top of things from below them, moving fast enough allows you to jump even higher, adding an extra level of interaction between the mechanics.

Something that lacks depth by contrast will usually lack in one of these areas.

If it doesn’t have its own niche, its own thing that its best at, then it will cease to be relevant to players. A lot of jumping attacks in street fighter games end up in this category, because there’s usually a couple go-to attacks that are preferable. Having slight overlap between mechanics is fine as long as none of them completely replace another. Like Mario 64 has a bunch of different jump mechanics, but they each go different heights and distances, so even though there is overlap, each has a distinct purpose and niche.

If something only has one use then it’s unlikely to be able to generate a large number of states.

If something cannot be modulated in some way to produce different outcomes, then it is less likely to have a variety of uses, and inherently produces less states. Smash bros attacks typically have multiple hitboxes that do different things based on where you hit. Street Fighter attacks have a collection of uniform hitboxes that do the same thing no matter where you hit.

If something has no interaction/synergy with anything else, then it cannot recombine to multiply the number of possible states. Street Fighter normals have interaction with specials, by canceling into them, creating depth, and also synergy in the form of blockstrings/links/confirms. Smash bros attacks only have synergy in that one can frequently set up for another.

But of course, this is still only a rule of thumb criteria system. You hit a bunch of weird examples in many games, like rolling in Mirror’s Edge that are related to depth but don’t cleanly fit criteria.

Could you expand on how rolling from Mirror’s Edge doesn’t fit your depth criteria? What other examples can you think of that are weird?

By itself it doesn’t do very much. It has one singular use, prevent hard landings, it can’t be modulated in any way, and it doesn’t inherit anything from any other state or feed into any other state, so it’s just kinda weird. However I wouldn’t say it’s not deep exactly, it ties into a larger system of depth regarding fall heights. The actually deep thing is trying to set up your jumps to land on spots that are of higher elevation than the spots around it, so you can roll instead of dying, or so you can avoid rolling at all ideally. So there’s a lot of spots where people will do crazy precise jumps to land on one little bit of geometry jutting up from a rooftop, which is cool as hell. Choosing to roll or not roll in the moment is a question of whether you’re high enough that a roll is necessary, whether you’ll actually hit that bit of geometry or miss it, and so on, so even if rolling doesn’t seem deep, it’s related to a system of depth.

Also I’m lying a little and rolling has another less conventional use. Rolling activates the fall reset glitch at the end of the animation, so if you jump to a lower elevation platform and roll exactly at the edge of it, you can fall during the roll animation, then get your fall height reset when the roll animation ends, allowing you to fall further than normal. This is used to skip the first half of chapter 4. You can even jump or sideboost off this fall reset. This isn’t really related to rolling’s primary use though, so overall rolling is kind of unconventional, not being deep in of itself, but being related to a deep system.

at 0:10

Another sort-of exception is the game of Go. It has one option for the player, placing a stone. It’s not really an exception because it recreates all of the 4 criteria through the combinations of stones possible. It sort of bypasses the criteria by not having many distinct elements that can exhibit the criteria, rather allowing you to build an endless number of elements that can fit those criteria.

This is why the 4 criteria are a rule of thumb and the definition of depth itself is the final judge.

Co-op Game Thoughts

what do you think about co-op games?

Okay, I honestly don’t play that many Co-op games. Lets see, I remember Tales of Symphonia, New Super Mario Bros Wii, Alien Swarm, Dark Souls, L4D, Final Fantasy Crystal Chronicles, and Smash Bros Brawl.

I think that a lot of the basic design applied to single player games overlaps into co-op ones easily. The big troubles are camera controls, griefing, overwhelming enemies, and giving players interesting simultaneous tasks.

In Tales of Symphonia, you have this trouble that a lot of time is spent with only one player controlling the main character, but during battles it generally runs great. Players can’t attack other ones, which keeps griefing to a minimum beyond pausing.

NSMBW was all about griefing. There was basically no reason to play it co-op, becuase you constantly got in each other’s way, except to grief one another. You could pick each other up, you could throw the other guy into pits, sometimes it worked. My brother and I trolled each other endlessly in this game. Camera controls are a bit of a problem here, because you can’t move too far apart and the limits of the camera are like walls.

Alien Swarm doesn’t have the camera problem, doesn’t have a griefing problem I know of, it was just fun.

Dark Souls is where enemy overwhelming is a big deal. No enemy seemed equipped to deal with multiple players. Also in games with melee combat, if combat is a rock paper scissors type of thing, imagine that you and your friend can both throw rock and scissors at once, or your friend throws rock then you immediately throw the thing that beats what the enemy threw right after. Or one player can stunlock an enemy, while the other lets their stamina refresh, trading places when one gets tired. It can get a bit dumb in that way. Players have a tough time dealing with gank squads when invading, enemies have much worse defenses.

Left 4 Dead cooperation works fine because it’s a shooting game and enemies don’t really get stunned. Sometimes you accidentally take up items that other players want though and can block other players’ way. The game has some features that sort of force players to rely on others though, like helping one another up, like hitting special infected who grab team mates, healing each other, rescuing from closets, etc. It’s pretty neat.

Crystal Chronicles has that chalice, which I always thought was a weak point in its design. Someone has to carry it, and it’s lame for that one dude. Also you gotta divvy up quest rewards, the menu interface sucks for multiplayer, you need a billion cables. It’s some shit.

Brawl’s co-op was alright, let player 2 warp to player 1 as a ways of mitigating platforming and camera bullshit. Had separate co-op events.

I like tag team climax’s concept in Bayo 2. I wish more crazy action games had Co-op. DMC3 lets you use a second controller for Doppelganger and Vergil, which is cool.

Input Movement Metaphors (QCF and More)

I absolutely adore Platinum’s stuff but absolutely cannot find anything to like in fighters. What differences between the two do you think causes this?

I can only guess, dude. That’s on you.

My guesses would be that fighting games have a very different control scheme from platinum games. I can only think of Ramlethal in GG Xrd who has chain combos of the same style as your average platinum game. Apart from that, the control schemes are extremely different, and fighting games are even different from normal 2D games in many ways.

And that fighting games aren’t really fun until you “get” them. It’s tricky to really understand what’s going on until it clicks for you, and before that it can seem like degenerate button mashing.

And that fighting games are a lot harder than platinum games. They have way more complex systems, more difficult commands to input moves, and tighter input windows for successful actions.

Perhaps it’s just because it’s multiplayer instead of single player? The single player modes in fighting games aren’t exactly great.

Do any of these sound correct? If not, you’ll have to elaborate on what you find lacking in fighting games, or on what you’d like to find in fighting games that you’re not finding.

The best I can come up with is that button combos in platinum games have associations. Like, in W101, if you want to do tombstone you press L2+R2. A similar move in a fighting game would be, like, up left left right down B. None of the movement inputs correlate to the move it’s just an arbitrary string of quick inputs. Like, in Smash, Down+B is not necessarily a downwards attack, just a different one to B with no direction. A launch in W101 is Jump+Attack – there’s that button association.

Alright, lemme cover Smash first. In Smash, you usually have Neutral B, Up B, Down B, and Side B. Up B and Side B very clearly correlate with directions, this is true on the ground as well as the air. So with Side B you get Fox/Falco Illusion, Raptor Boost (on falcon), Dancing Blade (Marth), Luigi Missile, and a bunch of other moves that move to the side. Up B is almost always a recovery move. Down B however can’t always go down. Sometimes you’re standing on the ground and you can’t really go down anymore. There’s exceptions of course like Falcon Kick, or Mario/Luigi Tornado which go down if you’re in the air. There’s a similar deal with down smash. Down smash can’t reasonably go down since you’re on the ground already, so it hits around you instead, or you get mewtwo’s which is a stronger attack hitting in front of him. Because Down Bs can’t really go down, most of them are designed to be utility moves like neutral B is. But apart from that, Smash mostly plays nice with the directional correlation.

In traditional fighters there’s a few very common input sequences that are common to most games. There’s Quarter Circle Forwards, there’s Quarter Circle Backs, there’s Dragon Punch motions, then there’s Charge moves. There are a few more complex motions than these, but they’re typically reserved for special moves or they’re not much different than these.

One example of a Quarter Circle Forward is Ryu’s Hadoken, a move that shoots forwards. Most projectiles are QCF motions. All of the ones in SFV are. One of the rare exceptions I can think of is Athena in KoF who uses QCB for projectiles. Chun Li’s lightning legs are another good example, now being a QCF motion, but they used to be a mash motion, which related to her kicking very fast, but now relates to the forward motion of the legs.

QCBs are usually less direct moves that serve some utility function rather than straight up aggression. For example, Ryu’s Tatsumaki isn’t as useful in most situations as his fireball. Nash’s sonic scythe has a backswing that kind of correlates with the QCB motion. Almost every character with a command throw has it on the Half Circle back motion, fitting this metaphor that you’re pulling the opponent in. Some characters have command grabs on 360s like Zangief, to strengthen the feeling of the pulling in.

Dragon punch inputs almost always do uppercut type of moves or something more special and less direct than standard moves. FANG can slide through opponents with his. Nash does this leaping hitgrab thing that drains meter and health.

Charge moves are a lot like dragon punches and QCFs, they all have this back to forward or down to up thing going on. Like Guile has back to forward sonic boom, down to up flash kick. FANG will lunge forwards on his back to forward charge and shoot a fireball up on his down to up charge.

The thing is, the movements are really not that complicated. Probably the two hardest movements in the list here are Zangief’s 720 and Guile’s back forward back forward.

These lists are comparable, except for the addition of Guile’s Delta inputs in SFIV (which are legitimately bullshit).

“similar move in a fighting game would be, like, up left left right down B.”
There’s basically no game with a move like that. You’ll run into a few games with weirder and harder inputs (usually just for the super moves), like King of Fighters and Guilty Gear, but for the most part, you can figure out most characters’ moves through trial and error once you know the common patterns.’98_Ultimate_Match/Moves_(normal_characters)

And beyond that, part of my issue with platinum games is that most of the moves are accessible through combo sequences, which have no directional cue at all. You just gotta remember that this arbitrary string of punches and kicks or light attacks and heavy attacks produces an attack with this effect at the end of it. It’s even more arbitrary than fighting game inputs.

Also, because platinum has shied away from using hard lock, they don’t have many back directional inputs or side directional inputs anymore. Bayonetta, MGR, and Transformers Devastation all seem to only use forward forward, back forward, and 360 motions for attacks (I think bayonetta had launchers on back + attack), so all of their attacks are generally in the forwards direction. DMC used to have back inputs, like for launchers, but also utility moves like windmilling the enemy to pummel them or freezing himself with cerberus, a rising uppercut or kick with Beowolf, drive or dance macabre with rebellion in swordmaster style, and so on.

Launchers in fighting games tend to be arbitrary moves, sure, but they’re arbitrary moves in platinum games and other crazy action games frequently as well, like prop shredder, charge shot, or a bunch of other combo sequences.

I mean, I think you’re just not following the directions very well if you think this is all arbitrary and come up with an example like, up left left right down B for your case. Try starting with a simple character like Ryu or Guile or Cammy. They have very intuitive motions and only 2-3 special moves.

On a related note, the fox costume in Bayo 2 got me thinking, what if there were a crazy action character who could shine like fox? Or maybe even multishine and wavedash?

Competitive Games Without Patches

Do you think the immutability of Melee and other classic vs. games is bad? What is your stance on the constant patching that esports games undergo?

I think there’s a tradeoff in patching. There’s a lot of tradeoffs actually. When you patch, obviously you can make the game better. You can bring it closer to perfect balance. The downside is, you upset the meta every single time you patch. If you patch really frequently, then performing well at your game is just about finding what everyone else is suffering against. Too many patches too fast will literally kill a game competitively, both because people can’t keep up, and because it’s impossible to observe how your changes perform if you’re constantly patching and never playtesting.

This happened with Brawl+. They updated nightly, and people couldn’t keep up, so everyone quit giving a fuck.

Other downsides are that characters that players like might become irrelevant, then it’s like they put all this time into the character, and now they’re trash. Sometimes you nerf characters and people don’t like the nerfs.

Beyond that, the advantage of never patching is it allows players to learn what’s there really thoroughly, should they decide to. Every time there’s a patch, players go back to surface level stuff, because that’s what’s most influential on victory, the big things. In Melee, we’re still having new tech discoveries and tier list shifts 15 years in. And 15 years from now, players will have the same game to come back home to.

Another arguable advantage is it keeps the list of viable characters small, so players don’t need to deal with a billion matchups. They can focus really thoroughly on only a couple. It means less character counterpicking and more of people sticking to their favorite, especially if they know that pick won’t be going anywhere.

This won’t happen for every game. Not every game has potential like this. But it is something that can happen. That’s why modern competitive games are spacing out their balance patches into seasons. SFV is patching once a year from now on. Overwatch is going with every 3 months. If you patch at random times, then you make it difficult on players.

I just want to add that bad balance isn’t bad for competitive players. Competitive players are fine with bad balance. It makes it much easier to pick a character and stick to that character. Bad balance keeps things simpler for them. I’ve spoken with some people (who play multiple fighting games) who have remarked that having a smaller viable cast is actually a good thing for games like Melee.

The downside of having a large viable cast is it becomes matchup wars. I covered this briefly in the last post. People end up picking up a bunch of characters and winning just because they have the right counterpick. Changing characters might seem hype to a more new player in a game with a small viable cast, but in a well balanced game it can be irritating to a more experienced one who can no longer viably play just the character they specialized in.

Shouldn’t players that only play one character be punished for doing so? Why should the game cater to poor behavior? It takes less effort on their part and it makes the game far more boring for spectators.

Yeah, that’s what less experienced players tend to think. I don’t think specializing in a single character is a poor behavior.

It’s nearly impossible to totally iron out all the matchups to 5:5 status, so a tendency of more balanced games like SF4 and PM is that players pick up a few characters so they can prey on the bad matchups for other characters. This means you can pick up easy advantages at the character select screen and rather than working through hard matchups, people tend to just switch. This means people aren’t playing characters to the fullest potential, and also that the game is kind of reduced to blindpick RPS25.

Of course, I like PM, and I liked SF4, even if it was my least favorite street fighter. I love having a ton of viable characters. I like having a wider viable cast, but it ends up as just matchup wars sometimes and that’s not as interesting to see or play because you’re not playing at your best. This is a minus, but it’s not an overwhelming minus. There’s no real means of mediation here, it’s just a tradeoff. You get more characters, but need to contend with going into bad matchups more frequently.

That and realistically, given how much investment it takes to master a single character, players shouldn’t be expected to know multiple. Being able to choose a single character and stick with them should be a viable option more than playing practically rock paper scissors with character select.

Then, what’s the point of non-viable characters in a fighting game? Is it merely that roster size is a selling point?

From a practical or competitive point of view, there isn’t really a point to them, but people like large rosters, as you said. Developers don’t always know what’ll end up balanced in the development process, so sometimes you get characters that are just left over. If you get a viable cast of 8 characters, then that’s usually pretty good. Most old fighting games didn’t end up that balanced though.

Basically one thing that can happen is the devs make a bunch of characters, they don’t know where those characters will necessarily end up out in the wilds of tournament play. A lot of characters will end up discarded, but if they can manage to among those have about 8 that are viable, that’s pretty nice.

What do you think is the perfect size for a fighting game roster?

I don’t have an opinion on a perfect size, but if a game ends up with 12 viable characters, that’s great.

I think the upper bounds of what should be feasibly attempted are about 30-40 characters (only because Project M did a really good job). Once you get to about 50-60 characters, balance becomes nearly impossible to achieve. There’s too many parts that need to be aligned with one another, it’s like this gif:

I think a more reasonable number would be 23 like GG Xrd Revelator has. The Guilty Gear and Blazblue games have always had really good balance and really good character designs. The current version of Blazblue has 28 characters. AC+R has 25. I think this indicates that low-mid 20s are the sweetspot range.

People Aren’t Random

You’ve criticized randomness in games because it gives uninteresting variation, and can be unfair, making competition not about who is best, but about who lucked out. You’ve also criticized the “solution” to this as seen in poker, where many games even out the random variance to pick the best but even in deterministic sports (or as deterministic as real life can be assumed to be) there’s a significant randomness to determining who’s best. These made me think about this: Googled for the source but couldn’t find it. Thoughts?

The short of this seems to be that people are random therefore game’s outcomes are random. I’d personally rebut by saying that people are inconsistent, not necessarily random. We’ve gone over how people are bad RNGs before, but performance is a bit less in our control.

I’ve actually heard about streaks in sports just being clumps in random distributions, and that more or less baseballers and basketballers follow their average performance most of the time, even when you think they’re “hot” or performing well. Radiolab did an episode on this actually.

There’s two factors to the contrary here however. First is that we can improve our consistency over time. Even if we’re somewhat random, as in a worst case scenario, we become less random over time as we improve at the task. This is why good players consistently dominate over weaker players.

Second is basically what they talk about around 8:45 in this episode:
Basically, individual games or shots or successes are not statistically independent. When you’re on a hot streak, you’re being affected by your previous success, not merely odds happening to collude on that specific moment.

What I’d add beyond this is that humans are affected by a lot of factors determining if they’ll succeed or not. I wouldn’t call this strictly random. More that we change over time, and some people are more consistent and others are less consistent.

That and if things weren’t at least slightly inconsistent that games would be uninteresting in the first place. The whole point is to experience inconsistency.

Yeah, people don’t always perform the same every time. Yeah you need repeated trials to get accurate results (this is why tournament finals are best of 5), but you need a lot more repeated trials to derive accurate results if there are actual randomized in-game elements.

People won’t always perform the same, but we’re not swirling in a vortex of meaninglessness here.

Games Aren’t the Final Artform

Are video games the ultimate art form because they have the greatest possibility space?

What? They’re the only art form that has a possibility space, a state space, other than visual novels I guess.

If you’re saying because there’s more things you can do with games than anything else, then I don’t think it’s comparable. I can’t process it. This question seems like it’s trying to preach to the choir.

By possibility space, I mean, a novel, film, painting, etc. can exist inside a video game, so games have rendered all those art works obsolete. In those other mediums, at best, you can empathize with the characters, but in games, you ARE the character, so the experiences are better understood. Why read about a character’s experience of war or loneliness or whatever when YOU can experience those things? Games are the end of art.

You must be a new asker. I’d guess you’re probably either from LTC or Insomnia, but I wouldn’t be surprised if the ideas have spread elsewhere. There’s a lot to unpack here, lemme do my best.

Here are some things I’ve written on immersion previously as a primer.

Here’s the short, you’re not the character in a game. Games aren’t exclusively about controlling characters, many games have no characters at all. Characters are not a concept that is essential to games. The characters you do control are like a car you drive. They’re objects for manipulation. Unreal Engine’s code calls these pawns, a subclass of actors, and I find it to be an apt metaphor.

Next, sure, all those things can exist inside a piece of software, however those things are not games, the game is another thing that exists within the software separate from those things. The software can contain a lot of non-game things in addition to the game thing. Google Earth is a piece of functional software that happens to have a flight simulator in it. I can tell you a couple fun games you can play using photoshop.

Third, software hybrid media cannot render those original media obsolete. This is both because there are many traits of the original media that cannot be recreated in software (like for example, the texture and 3 dimensional attributes of paint on canvas, which some painters use to actual effect in their works), and because software containing a mishmash of a ton of media forms is not necessarily an efficient medium of delivery for those forms. Nobody is going to sit down to read an entire novel in say skyrim or another comparable game. Nobody wants to run an extremely system intensive process just to access that type of information when there are more efficient means of delivery. Same deal with movies/cinematography. You can’t realistically stick a movie in a game without people getting bored of it, even if you’re hideo kojima.

Games are a more specific class of object than software. Games are essentially systems that produce inconsistent outcomes that we leverage in contracts with ourselves to derive enjoyment from attempting to produce specific favored outcomes. The art of games is not a physical or otherwise tangible media. Games use objects like software as facilitators of the contracts we create. Game software is not a game by itself, it is an object designed in an extremely limited way to facilitate the specific functions of a specific game that we agree to play using the software, much like the physical game, Screwball Scamble. In Homo Ludens this contract is referred to as The Magic Circle. We can build software that contains all these mediums, but the game is a more specific thing than the whole software object.

Because you can’t really experience those things through games. You can’t experience the pain of a bullet wound. You can’t experience what it’s like to be isolated from everyone else in a war zone. You can experience systems of interaction that are labeled to be similar to those things, but they’re not the thing itself, and there’s really no substitute for that short of a Derren Brown special (his zombie apocalypse one was funny, as was the recent one where he got people to commit murder).

You’re at best like a tourist to these things, and more realistically you’re opening the cure menu in MGS3 and clicking a bunch of menu options until all the circles go away. You’re still vicariously empathizing through a pawn, or as some have said: a doll, that you are taking to be you. The systems of choice, especially dialogue, are so limited in games that they cannot approximate real conversation. There are always abstractions and this will not change with time.

Due to technological limitations, developmental limitations, and likely limitations in what is possible for a computer, software will probably never be able to simulate a universe to escape into. No developer can make enough content, and no computer will be able to simulate enough to calculate all the intricate physical interactions that govern our reality no matter what those “infinite detail” guys keep claiming but not delivering.

Hybrid Media Software is not the end of all other art forms. Games are a tiny non-encompassing part of that. Games are not the end point or zenith of art. They are what they are, and they’re fine being that.

As a parallel, consider Sleep No More, a site-specific work of theater that is essentially a building you walk through with actors performing parts of Macbeth in each room. The building was changed to look like a hotel. There are documents scattered that you can look through in each room. Haunted Houses are still popular around halloween time and I think some even run year-round. These are a hybrid media form present in real life, containing novels, films, paintings. They contain a type of media that cannot exist in software since it is a live theater production. They could easily shove a donkey kong cabinet in there somewhere and include games as well. Are they the end-point of all media? Don’t be ridiculous.

Concerning games, stories, and art: I see what you mean, but designers can attempt to simulate things like the pain of a bullet wound or the isolation of war. Take a bullet wound for example. Obviously you can’t feel pain, but the designers can attempt to simulate the physical motions of someone who got shot in the leg. He walks slower now. He stumbles. All the while, everything’s going to hell around you. Or loneliness for a more flexible and broader concept. The obvious example would be a “walking sim”-type game, but you could have a game where you’re so powerful, you end up feeling empty. You can’t feel pain of loneliness from reading a novel or watching a movie either, you can merely empathize with the characters portraying those characteristics. Wouldn’t you say that every great work of art was created out of a fear of death or, rather, the desire to use life meaningfully before dying. Videogames do that better than any medium because you can actually die in them or even transcend death. No other medium can even come close to the simulating this desire to live. And then you have games like Beyond Two Souls (a better game than any i-frame cheeser from From) or Everybody’s Gone to the Rapture (far more interesting environments than any fps), and these games portray so many interesting things about death and the human condition, but are ignored because there’s no shooting every 5 seconds.

At first I suspected you were trolling me, now with a statement like, “And then you have games like Beyond Two Souls (a better game than any i-frame cheeser from From)” I’m pretty sure you’re trolling me.

Designers can’t simulate pain or isolation. They can make mechanics that approximate and abstract them within the limits of what software can do. They can simulate a character that walks slower. They can simulate a character that screams in pain. However it’s still a matter of identification and empathization that you choose to interpret those things that way. You can’t actually die in a game. You can get sent back to a checkpoint. You can have a save file deleted.

Trying to connect the fear of death to a player’s experience of dying in a video game is absolutely ridiculous. On an existential level it might be valid to argue that all art is created to pass the time between living and dying, but trying to jump that to say that video games are the best art because they involve something that superficially resembles dying, that might be connected to a player’s, not the artist’s, experience? Are you high? In what world are those factors even remotely connected?

Your ability to feel loneliness from reading a novel or watching a movie is as great as that of playing a game. These are all very comparable experiences on a narrative level. Trying to prop up games as something special that puts you inside the work is straight up immersion talk and you should probably know how I feel about that.

Not all art is explicitly about death and the human condition, except in the most existential sense possible that all art is about death and the human condition, because we’re all going to die, and everything we do is a part of the human condition.

Games that are about shooting are a part of the human condition. Games that are about placing stones on a board, then removing surrounded ones are about the human condition. Or rolling things up into a giant ball. Or collecting coins. Or walking a tightrope. Or practicing a long sequence of precisely timed inputs for hours on end.

Not all music has lyrics. Not all music has meaning. Sometimes something just sounds nice. Same deal with games, sometimes it’s just an interesting system, and that’s frankly enough.

Bullet Sponges

What do you think of the phrase “bullet-sponge” enemies? Is it a valid complaint?

Yes actually. It’s a weird matter of pacing to have an enemy you just keep hitting that won’t die. This is also tied into poor feedback, about how much health enemies have left. I have some old notes and ruminations I wrote on this that were never really completed, but I’ll share it here:

I think this is why some enemies use multiple health bars, because feedback about how much damage you’re doing is more clear to the player when the bar moves more each hit.

It might also be because of how much enemies react after being hit, but I remember enemies in bioshock infinite’s hard mode getting totally fucked up and still feeling bulletspongey because seemingly I’d unload tons of shots into them any they wouldn’t die.

In NuDoom, the big charger enemies feel weirdly bulletspongey because sometimes I can kill them in 3 shots, and sometimes they take like 15 shots.

In Bloodborne, I love the way bosses will have their bodies reel like jigglebones when they’re hit, but continue playing the same animation. Great feedback there. Dark souls games in general have the right balance of health if you ask me. Gets a big boost on the first NG+ and much less for every one afterwards. Presumably because NG+1 is where it’s at a reasonable level to be hard, and after that is excessive.

In Nier’s hard mode, everything felt extremely spongey. Had way too much health relative to how much effort it took to fight them.

I think bullet sponginess comes from the perception that your shots just aren’t affecting the enemy. I dunno beyond that, just scattered thoughts.

You mentioned in that answer of bullet sponges that the base challenge of Dark Souls is so intrinsically interesting that it allows number buffs to be. What specific aspect do you find so interesting? Dark Souls is basically more about spacing, than it is about combat, combat is rather shallow by itself, and the spatial awareness and control challenge is incremented through the level design, and the great variation in enemy patterns. But beyond that, due to its relative slowness, it doesn’t seem that fantastically interesting.

Compared to your average FPS enemy, Dark Souls enemies do a lot to keep it interesting. Fighting them for a bit longer is more of a risk and more interesting than your average FPS enemy. I know I’ve told the story of how I had to learn the 4 kings boss fight very in depth on NG+ before, due to the added HP and damage in order to actually beat it; most critically, how to avoid having all the kings tag team me by staying close to one king, which makes the other kings more passive. More HP demands more consistency from you, demands better understanding. A simple HP boost works fine in a lot of action games across difficulties (which of course a lot of action games do across difficulties).

Also the slowness of Dark Souls combat is one of its relative advantages actually. Because both you and enemies are slow, it totally changes the whole dynamic of combat. It’s about this fuzzy evaluation, “is it safe to commit to an attack right now? Can I get away with this? How much can I get away with?” Then the enemies use their different patterns to play on this core dynamic in a bunch of different ways. When you know the game well enough it becomes a lot like this:

I wrote an ask a while back about all the things I love about dark souls combat which should help explain what’s interesting or deep about souls combat a bit better:

hm, but isn’t that advantage of its slowness a perceptual matter, rather than a real change in the way it’s played? Haven’t you said something similar, on twitter I think, about TF2’s heavy vs scout, with relation to the argument that the heavy promotes a more strategic play related to position whereas the scout is more twitch based, to which you argue that it’s just that slowness allows worse players to engage in strategic thinking, but it’s even more present in fast play, just that it’s behind a wall of a certain skill level to be able to appreciate it? Like, I’d argue that Vanquish has a similar spacing element to Souls, but it’s more interesting because the movement freedom gives a lot more options, and the speed and aggressivity of enemies is on par with that. Also bloodborne is often praised for adding speed and aggressivity (just a bit though) to the series.

I disagree that it’s perceptual, or at minimum I’d say that a perceptual change here is a very real change in how it’s played, because the timing of attacks is related to your reaction time, which is a big factor in how games are played. Most action games have very fast player character attacks, so that even if you see an enemy attack, you can always attack fast enough to interrupt them first. When your attacks are slower, you can’t always see an attack and still have time to attack them first. Beyond that, the basic idea is that your attacks move at a similar pace to enemy attacks, but enemy attacks are still well outside the reactionary blind spot to keep it fair. Bloodborne was faster, but it kept this principle up.

I don’t think I ever made such a statement about heavy and scout, but it sounds fairly on the nose.

What’s strategic or not depends on context. Slowness works when it works. I tend to prefer higher speed games overall, but I recognize that in some cases slower speed works. For example, Starcraft didn’t do so well in the transition to Starcraft 2 largely because the fastest speed was made even faster and that became the standard speed. A lot of players better than me say that it made the game much less micro oriented and more based on anticipating unit formations before they actually attack you, because once they attack it’s too late to micro your way out of it.

For the record, I think the Heavy (and the rest of the classes) should move faster. Like twice as fast or 1.5 times as fast. I think both heavy and scout are strategic, just in different ways. The heavy has superior damage output, but he needs to figure out where to plant himself and get revved up, which creates a unique tactical challenge compared to the scout.

The possibility of interrupting enemy attacks after seeing them comes from a relative difference in speed between player and enemy, not from speed of the game, and greater speed would increase possibility space. I’m interested in your thoughts of my comparison to Vanquish. I realize that I’m comparing a shooter to it, but despite the interactions with enemies being different, Vanquish requires tons of target prioritization and position choosing, so it seems comparable, and the dodge and boost (speed) give a greater access to available space than that seen in Souls games. Also there’s a comparable (sort of) dynamic to committing attacks in Souls, as shooting effectively requires a near detention of movement, thus making you vulnerable to attacks. Overall Vanquish just seems to have a more interesting dynamic, it just is inferior in level design.

Okay, I thought relative difference in player and enemy attack speeds was basically speed of the game, given that enemies can only feasibly be so fast, leaving you only the player to speed up.

I can see the comparison between Vanquish and Souls in a loose sense, but I don’t think they’re totally comparable, mostly because there’s no hitstun or attack startup time. I think the dynamic in Vanquish is generally very different, being concerned with position on a battlefield more than position relative to an enemy and immediate surroundings.

The other thing is, when you make a game faster, or more specifically the player character, the level design has less impact. It becomes easier to move around enemies and obstacles and they are less able to block your path and limit your options, even if they are fast too. This is why a lot of Crazy Action Games end up being about fighting in big empty arenas, because more significant level design generally just gets in the way.

Greater speed sometimes affects possibility space, sometimes it just feels nice. It depends on how the rest of the game is designed. I have a personal preference towards faster games, but I don’t think it’s better 100% of the time.

Matthew Matosis DMC Joint Commentary (Not Too Bad Edition)

7Okay, here’s another of these. I did this all stream of consciousness while watching, so sorry if it’s a bit hard to understand without watching the video as you read these.

Gotta say, doing an S rank, no damage, no items run is impressive. After beating Thi4f on similar terms, I know how hard that can be. My recent Mirror’s Edge run has mistakes because it’s goddamn near impossible to avoid fucking up in that game, even if you’re the best in the world, and I only spliced a little, hopefully so little that nobody noticed the more obvious cuts. So yeah, points for that, good way to establish that he’s not a shitter and can’t be dismissed for being bad.

It’s mildly weird that he doesn’t retell the actual history of resi 4 here, about how it was going to star a bioengineered superdude named tony redgrave, but whatever. I don’t think it’s a necessary tangent anyway.

I think there’s a benefit of the fixed camera angles for action games too, considering the issues many action games have with free rotating cameras. Wonderful 101 shows how a fixed isometric camera can help frame everything on the screen. Having a camera stuck in one place at a time and needing to cut to show everything however can be agony.

I would have personally mentioned literally any fighting game other than street fighter 2 for juggling. SF2 didn’t originally have juggling and it has a very limited implementation of juggling in its later editions. Also the entire fighting game renaissance had come and gone by the time DMC1 was out. You literally had GGXX and 3rd Strike by then. This is a nitpick though.

The remark that juggling is similar to grappling in beat em ups is kind of off point. It makes sense from an abstract perspective, but you could actually juggle in those beat em ups. It existed in Alien Versus Predator and Final Fight 3, to list 2 of the best known examples.

I wish he made a more detailed remark on how the style meter works here rather than just stating that you get a second of leniency.

I wish when he skipped the marionettes he made a remark comparing this to 2d action games, such as zelda and ninja gaiden. In 2d beat em ups it’s also easy to skip enemies, which is why they lock you on one screen to fight them, same as locking you in a room in DMC. I don’t think this is just a 3d space issue as he makes it out to be (because otherwise 2d beat em ups wouldn’t do the same thing), I think it’s because of the lack of contact damage and the use of individual attacks that are timed out to be fair to avoid or interrupt, making them trivial to run past before they can get started, where in a lot of 2d games there are ranged attacks they can use instantly or close to instantly at a distance where you can see the projectile coming and avoid that.

Fair remarks on the changes for the higher difficulties.

I’m guessing the gun switching is done through the menu because they didn’t think that people would want to switch guns instantly, and because there’s a lot more guns than melee weapons. We don’t know how this was done or why it was done. I don’t think it makes sense to guess at it. Just remark that it’s cumbersome. He says that originally they could only get 1 weapon working at a time, so maybe they encountered issues here, but my policy is to generally not accept technology compromises. If technological constraints made the game worse, then they made the game worse. It’s an excuse for the developers more than the game itself. They might have done the best they could, but a good game in the vision they imagined might have not been within their skills or the technological constraints of the time.

Also I think he read the book in this section by accident and tried to make a joke about it since he was stuck with the footage and didn’t want to redo the level.

The remark on the timer for the underwater skulls makes sense, but having health decrease also means that you have a drive to kill the enemies to earn back health/time, and that them attacking you is factored into the health/time you have remaining, keeping it one contiguous system rather than 2 independent ones.

Generally fair remark on health and devil trigger, especially that the necessity of devil trigger adds an extra element to make the boss fights more strategic. I think his guessing at what the intended level of health is is reaching, but it is fair to mention that it was balanced very deliberately across difficulties, in contrast to other games.

Fair remark on the scripted sequence with phantom and how you can fight him. Cool to show it off.

Nice jumping on the death scissors. Reasonable remarks on their placement. Sensible remarks on the audio cues and the fairness of the enemy.

The followup on audio cues being faster to react to is factual. I have a bit more information on it personally, like how visual processing is parallel, and audio processing is serial, and the average rate of difference, but whatever. Can’t think of any other older examples to contradict his claim that it’s impressive they thought of audio cues this far back, except that fighting games have used this for a long time, like differentiating zangief’s P and K lariats and Q’s dash punches.

I wish he described his tactics versus Nero Angelo a bit better here rather than just talking about the taunting mechanic. Also described the patterns Vergil throws at you here. He’s clearly doing something odd by going behind him after each hit to finish the combo, but he doesn’t talk about it.

Fair remark on keeping the player busy with the taunt button to fill in the gaps.

Muh ludonarrative dissonance. Muh reward. Muh eventual satisfaction.

At this point fighting the shadows, it occurs to me that it would be nice if he described the patterns of enemies a bit better than he does. He kind of glosses over all of them instead of really getting into their design. Fair remarks on the lacking audio cues. “you end up constantly dodging these massive attacks, which is very satisfying” C’mon, you can do better than that.

Don’t care about the backstory on Shadows, I’d say it’s really lame that they’re only vulnerable to guns initially since guns aren’t very interesting to fight with by themselves. I don’t think the backstory matters this much and if it’s a gameplay contrivance, they could have made the shadows more sensitive to gunfire at certain times or from certain angles to stress skill in another way.

The Plasma strategy is interesting. I don’t think the remark on how economical they were to produce really has a place in this review, especially given they have a unique model, and a unique effect. I think the speculation on bat form is reaching.

Nice that he mentioned pause combos. I’d say he could have been more explicit on how the controls feel right. Specifically, pause combos feel good because you are hitting a tight timing window at a point where there’s a natural feeling of recoil from the move’s animation as it recovers, which translates into the idea he mentions that we’re “building up power as well,” but worded a bit less awkwardly. Hitting tight timing windows with clearly distinguishable visual/auditory feedback is something that intrinsically feels good, like how PM made L cancels feel better by having the white flash, providing more clear visual feedback. Nero in DMC4 would later take this principle to an extreme with his pause combos that could be extended far past their normal length if you keep hitting the attack button at the end of each slash in a faster and faster tempo. The yuganon fight in Zelda ALBW similarly shows off a raise in tempo in one of the magic orb volleys, and it actually will force your sword to miss if you mash off tempo.

Funny, I’d think that timing for the last attack off a pause combo would be hardest, but it really depends on how big the windows are for each possible attack. I’d also add onto his remark that the last hit of a combo doing the most damage is not just a reward, but something that creates risk as well, because you’re stuck in the combo for all that time, and might get interrupted if you stick all the way to the end. Small enemies appear in groups so you cannot attack without another enemy potentially threatening you, and bosses have super armor so they can attack indiscriminately of you hitting them, so you’re always weighing risk and reward.

I also appreciate the remark that the attack button can do so many things by itself. That’s one of the things I like about Devil May Cry versus other beat em up games, it gets a lot of mileage out of individual buttons instead of relying on multiple buttons.

His tactic versus phantom is nice and well explained. His remarks afterwards about satisfaction, and fighting smarter, not harder, are all fluff.

I don’t really care that the crazy key item pickups make more sense in dante’s world than resident evil’s. I think it would make more sense to remark here on their affect on the game’s progression rather than their believability. Not to mention comment on the way the whole world is designed to steadily gate progression this way as opposed to other ways they could have done it. It might be thematically satisfying, but they could have cut down on backtracking and avoided people wandering off into useless areas with other solutions, for whatever reason it’s possible to walk from whatever area of the game you’re on back to the beginning in every DMC game, even though there’s nothing to do there. If you don’t know the way forward, then this can seriously waste your time. I know from doing that by accident in DMC3, and finding a hidden battle at the start of the game, which just confused me even more.

Remarking on the combo staying alive while charging is a decent touch. I don’t think combo scoring systems matter as much in other games as in DMC, and I don’t think they matter much in DMC to begin with. Ifrit attacks being held indefinitely to extend combos forever probably wouldn’t break anything, I don’t think it’s a serious abuse considering you’re motionless during them.

I think remarking on all the polish effects for E&I is a little extraneous, but not completely so. Feedback is important, it can help guide expectations about function and verify that input is registered and having an effect. I think the 2 room comparison is dumb considering one version has more than the other with no tradeoff between them.

I don’t think E&I are actually hitscan, even normally. I’ve taken pictures of the hitboxes for them in DMC4 and they’re definitely hitboxes there, just very quickly moving ones. That MM notices there’s a delay across larger distance lends credance to the idea that they’re not hitscan. It’s nice that he sums up the downside of devil trigger bullets and how they introduce some element of strategy into shooting. I’d like it if he mentioned angles of movement a little to be just slightly more explicit here, like it’s harder to catch Griffin when he moves across, but easier when he moves towards you or stands still.

It’s reasonable to say that having more tension after killing a boss is something that can be exploited. Not much to say about this.

I don’t think the last hit is a tension thing so much as a fairness thing. Bayonetta does this too. I think DMC3 and 4 don’t do it, but I’m not sure offhand. There’s something to be said for having this and not, like it wouldn’t really fit in Dark Souls if you ask me.

I don’t think the remarks on possible influences here is very interesting. I sympathize with thinking that games can be a valuable source of inspiration as opposed to books, but I think this line of commentary is pointless.

I think the remark here on how because enemies can only be 1 per room (or only happen to be 1 type per room) and therefore were designed with both close and far ranging attacks is a great observation, probably the most valuable thing I take away from this video as a whole.

The remark on standing still being fitting for dante is silly.

The explanation of how the devil trigger contextually links encounters is a reasonable observation. I’d have gone into slightly more detail for how it affects the game though.

Commenting on the music of nero angelo is outside my scope.

Makes sense to comment on how the game guides the player through the levels. Especially good observation is the red orb placed in water.

I don’t think the remarks on first person allowing them to avoid animating dante in water are relevant. I wish he commented on the swimming itself instead.

His explanation of smears kind of muddles what smears are versus squash and stretch. A Smear is meant to emulate the way an object in motion blurs as it moves by avoiding changing the proportions and retaining the same overall volume of the object or limb as in traditional squash and stretch, but rather dragging speedlines or an extended version of the object behind the object’s current position. It’s not really possible to do these in 3d games using skeleton animation, you’d need to deform the mesh of the object, which isn’t easily possible. It hasn’t been possible to use smears in 3d animated films until recently (I remember one of the first examples was in Kung Fu Panda) And in games at best they can resize individual bones, as used in smash bros occasionally, or they need to make completely new models to smear properly (see guilty gear Xrd, the first game to really do this type of thing correctly.) In the end though what he was referring to are the fire and electrical effects on ifrit and alastor, which do technically fulfill a similar role to smears, so I’m just confused as to why he’d suggest other 3d games have them. You can see some smears on this blog or in this classic cartoon.

Yup, Dante’s coat is a secondary action.

I think meteors you knock back are kind of uninteresting, considering how often it’s been done and it’s nothing but a timing challenge It’s alright though. I like that the plasmas have friendly fire though. The shadow observation is great too. It’s cool that he mentioned how there are so many ways to use moves back on enemies.

He’s definitely overthinking his commentary on the freebie mission.

Decent reasoning for why they split up the missions, so the backtrack is more obvious.

Slash canceling probably works based on IASA frames (Interruptable As Soon As). Typically followthrough animations (where the sword is sheathed for example) are cancelable by any action including movement in most action games, because having followthrough animations makes the action feel complete when they actually play, but allowing it to be interrupted prevents that awkward feeling when the animation hasn’t totally completed but you’re still not allowed to move. Most action games make use of this to some degree. Oddly fighting games don’t for the most part, except for Smash Bros, which is where the name comes from. In most games there is a separate cancel window for followup moves that comes before the IASA frames, so leading into the next hit of the combo is faster than ending it. This can be seen with every character’s jab combos in smash, and marth’s side B. It’s understandable that this principle wasn’t developed or popularized by the time DMC1 came out.

I totally disagree with MM’s statement that he prefers when glitches are “recognized and canonized as proper gameplay mechanics.” I especially hate the statement that some gameplay mechanics are “proper” while others aren’t, ignoring all the processes in the code that make these things work. Mechanics aren’t something specifically designated as such, they’re the emergent result of whatever tools were used to build the game, be it code or plastic. What’s there is there.

However in single player games specifically, the ontology is a little weird. Playing a singleplayer game is a contract between yourself and the game to use the game software in the way that you see fit to play. To that end, you can deliberately decide to play a certain way, and it’s as legitimate as anyone else’s. Theoretically there could be a tool that completely breaks the game and trivializes the challenge, and you could agree for the sake of your own experience to abstain from using that tool. For example, the Death ring used with the light and dark glyph fusion in Order of Ecclesia. The thing is, this is an arbitrary imposition on the game, as much as choosing to not use a glitch. I don’t think it’s fair to attempt to separate this based on presumed intention or non-intention. I think if you’re going to make an arbitrary ruleset for yourself, you should base it on what you know about the game and the elements you’re setting as off limits. For example, I recommend anyone playing dark messiah bind kick to the opposite end of the keyboard, so the option is there but not incredibly easy to access.

Of course, this attitude that only some functions of the game can be “recognized and canonized as proper gameplay mechanics” is an actively dangerous attitude to take about competitive games. Competitive games are subject to higher standards than single player games, because it’s not a contract subject to the interpretation of an individual, it’s a joint contract between people to compete. To this end, the developers, or the game itself, are to be held at least somewhat liable for anything that is possible in the game, because in a fair competitive environment there is absolutely no restriction on what you’re allowed to do, especially online. If something is in the game, you should use it to pursue victory no matter what it is. The idea that only some of the mechanics are “proper” leads to scrubby behavior, to legitimizing those who would move to ban or patch out glitch elements that define or enrich multiplayer games. There should be no assumption by developers that there will be a central organizing committee of players that can empirically judge the effect of a mechanic over a long period of time to determine whether it should be banned, they should take all appropriate measures to prevent glitches which would lead to tactical degeneracy, because those are always canon in a multiplayer context. This is part of why I fear MM ever reviewing competitive multiplayer games in depth, especially ones like smash, street fighter, or quake. Even CS:GO might be liable.

The point about the boulders is reasonable. Random spawn boulders that fall too fast to react to are kinda sucky. I can imagine that killing a few attempts at this chapter damage-less. I think his statement that unfairness is more tolerable in less challenging games makes sense in a general light, but he could be more specific here as to what type of challenge, because obviously there’s different components of challenge, different levels of punishment for a mistake. Something being unfairly challenging, but only giving you a slap on the wrist is tolerable.

Most games already use windups of 20 frames or so (333ms) for attacks at the minimum, to account not just for input lag and display lag, but also people just having bad/inconsistent reaction times, and most critically because the reaction time test is for a single stimuli with a prepared response. Almost no fighting game has an overhead with a startup of 13 frames (Hi GGXX Sol), because in the middle of battle there’s so many things to look out for that reaction time increases.

Uh, no. If the framerate drops lower it will not reduce the amount of time the player has to react, it will either not affect it, or increase it. All 3d games past the N64 (and a few on the N64) are programmed using Delta Time, a variable that records how long the transition is between one frame and the next, to scale how much things are supposed to move for a given frame, so if the framerate drops in realtime, game time is unaffected. In the older style of game before delta time, if the framerate drops, in-game time will literally become slower, making it easier to react. It’s harder to act during lower framerates because the delay between action and reaction increases, especially for very fine levels of interaction, like carefully positioning a character in a certain spot, or aiming a mouse.

I really wish he used frames here instead of milliseconds. It’s a 60fps game, frames are the standard unit of time measurement for games. If he really wanted to he could even explain the conversion rate. 16.666 milliseonds per frame. 300 milliseconds is about 18 frames

His standard of fairness here is kinda narrow, there’s more to fairness than just startup times of windups. Also he says that blades, “don’t telegraph their attacks well” when he should say that’s the fault of the animation. The enemy’s animation for attacking is not visually distinct.

Then he mixes fairness with good gameplay, seemingly addressing an unspoken statement that fair gameplay is de facto good. This is a weird muddling of two topics, depth and fairness, that don’t have any overlap from a design standpoint. His example is that if there was an enemy that could only be beaten with guns, then it would be fair, but it wouldn’t necessarily make for good gameplay. It’s just the strangest conflation of fairness and good gameplay I’ve ever seen.

The near instant grab doesn’t break the 300 millisecond rule, because the player needs to be close to the marionette before it can use the grab. It will technically be unfair the very first time it happens, before the player knows it exists, since there’s no possible way to avoid it the first time you encounter it, but every time after that it’s fair because you know if you’re close you’ll get grabbed, and you can see yourself getting closer to the marionette outside the 300ms window. I don’t think he considered that the key is having a tell of ANY KIND outside 300ms. Like flashing yellow, then a second later, sucker punching you. Or sucker punching you on regular intervals of 15 seconds.

Dude, you could have just spliced the footage after griffin. The cutscene would make the splice unnoticeable. Also commenting on how you practiced this is not exactly cool.

Shrug, the jump commentary makes sense. So does mentioning that green orbs add exra red orbs if you’re at full health. I like how God Hand does the same thing with fruit.

I don’t really care about the “whoa, below absolute zero!” thing.

Yeah, I’ve written about rewards and so on before. Sensible rant. Not a fan of the unlocking moves as a soft tutorial approach personally, but I don’t see a better alternative. I think in games like this you should be able to do everything from the start if you really want to, but I also know that I was overwhelmed my first time playing on a friend’s finished save file. Actually, one mistake I think DMC makes (among other games) is spacing out the upgrades further and further as you spend more red orbs. If the upgrades are to serve as a soft tutorial, then it would make more sense to space upgrades evenly, or to have them get spaced closer together as you get more advanced at the game and can understand what new upgrades do for you more easily.

Fair description of the nightmare strategy. Funny that he remarks that something might be just a bug, like bugs are totally uncaused. Good notes on the cores.

I wish he explained how jump canceling worked a little better. Also what? He says that jump canceling is almost certainly not intentional. “It only works because a few different mechanics happen to collide with each other” Are you serious? This is like saying that L canceling is almost certainly not intentional. They made a specific animation that jumps off of enemies, they allowed this animation to interrupt the shotgun blast, they made this animation RESTORE YOUR AIR HIKE. They then had this mechanic return in every single sequel, and Kamiya brought it into Bayonetta as well! You might as well argue that dodge offset is unintentional. You can’t really know this stuff.

Yeah, having an enemy attack you just after a transition is some bullshit. I didn’t like it in Bayonetta either.

Fair remark on the T-Rex meteor reflection teaching you about the phantom meteors.

Taunting vergil first or taunting him back “feeding into the story” is silly.

Pointing out that the jump has iframes so it can complement the trickier roll input is a good observation.

Predicting a creature is easier than predicting a floating sword because creatures have a lot more limbs, they can be more expressive and explicit in how they wind up.

Honestly activating and deactivating devil trigger every time you attack would probably be quibbling over so small a resource drain that it’s crazy to bother. Having a definite cost associated with deactivating devil trigger means that you need to get mileage out of it every time you use it and incentivizes draining it to the bottom or recognizing when you won’t be able to get further hits in and cutting your losses. Also later games in the series have DT distortion, activating DT right when you hit boosts damage by a lot, so literally the thing he talked about being bad happened and to greater effect.

I don’t really buy his emotion in game design talk. I don’t buy most emotional theories regarding game design. I think that’s something outside the purview of analysis most of the time.

Oh man, a technical framerate talk. Eh, good enough.

As for the widescreen thing, I don’t think it affects 3d games much. It affects fighting games or other 2d games a lot more, especially those that use the borders of the screen to some effect. Guilty Gear Xrd actually made all the characters wider to compensate.

No, emulation is a perfectly viable alternative. It’s arguably the best alternative, assuming you can emulate accurately. The argument in the GDC talk below is that ports are not the original game, they contain their own quirks, and emulation, barring faults of the emulator, is literally the original game. If the emulation technology is smoothed out, then there’s generally no problem. First party companies, or those with a good relationship with the console manufacturer can usually get good emulators for ports of their games. We saw that happen on the PS4 recently with PS2 titles such as Dark Cloud. Emulators are the only road to flawless preservation. Source ports have as many issues as the originals. Matthew Matosis probably doesn’t know this, but the grapple move on Kalina Ann in the DMC3 port on the HD rerelease he’s playing for this video literally doesn’t work.

Oh god, please not another statement of this, “If it’s always the best option, it should just be automatic instead of manual, the controls shouldn’t be needlessly complex, decisionmaking is what’s most important” argument. I’m tired of hearing this. It’s totally amateur as I’ve been saying for literally years now. Decisions are important. Not everything needs to be a decision. Please never review DMC4’s ACT or MGR’s Blade Mode Canceling, or Smash Bros or any other fighting game for that matter. Please also never review an FPS, since literally aiming accurately at a single target would fall under “needlessly complex controls that lack decisionmaking” If you have a problem with dodge canceling by itself, criticize that. Writing off this entire category of possibilities is bad form, since there are valid implementations and it leads to thinking execution skill is bad in general.

Ninja Gaiden had an online leaderboard, or at least NG2 did.

I don’t think online leaderboards generally turn out the way MM describes. Didn’t Devil Daggers base itself on a leaderboard recently? Also don’t a lot of games with online leaderboards have regional leaderboards or ones among friends so you can get a smaller pool of generally worse people to compete with?

That and does DMC really seem to you like a scoring game? Especially given those points are then spent on items and upgrades? It seems pretty superfluous to me across the franchise.

I’d make my usual argument about linear versus branched stories but I don’t really want to hash that out again. Story choices are shallow.

Yeah, non-deterministic dodging sucks. Especially when there’s really no way to counter it, it just happens. It’s like the enemy just randomly has a lot more health or something. There’s a random chance of enemies blocking in Curse of Issyos and sometimes you need to hit enemies like ten times before they actually die.

Wish he out and out said that if there’s only one enemy, you can stunlock them rather easily, and with two enemies you gotta watch out for at least one.

That’s the simplest argument against Length versus Price? That Price changes over time? How about that Length is not Quality? There are short games that suck and long games that suck. Also judging games by whether they accomplish what they set out to do is a shitty shitty metric. This is exactly what I blasted him for in my original MM article. The developer intention might be garbage, and the game might be great despite that. The developer intention might be garbage, but they accomplish it perfectly, but the game is still garbage. Games shouldn’t be judged relative to intent except in perhaps the most general strokes for non-competitive games.

I don’t know why MM’s expectation would be to go to the menu given the preference for inverted versus noninverted controls wasn’t present there from the start. The commentary on the shmup fight is kinda bland, lacks detail. Doesn’t really explain why it’s good. I think the speculation that challenging people by proving themselves at more than one game type is total grasping at straws on Matt’s part. No founding. Adapting to a shallow set of mechanics for one section doesn’t necessarily make the whole playthrough more impressive. It can split development resources to program an additional segment like that which could be reinvested into the core gameplay, resulting in a more deep game rather than a more “broad” game. Here he praises this segment basically just because it adds to the “experience” in his eyes rather than because it has unique merits.

Good commentary on mundus part 2 and all the abilities he has. Also in noting the layered style of design. I think most of the commentary on remastering is mundane/out of place. It’s worth mentioning faults with the HD release, but the rest of the commentary is not only obvious, but off topic if you ask me.

It’s fair to say that mundus feels the least like a standard enemy encounter, however I wish he attributed the strengths of mundus’s fight less to the spectacle of it. I don’t think mentioning spectacle has a place in a game review.

I’d ignore the “fill your dark soul with light” cutscene personally.

I don’t care about his commentary on the use of the battle themes.

Second encounter with mundus is a way to beat him on the first try after dying to him a lot? Shrug. Whatever. More experiential review. I don’t care for it.

I think a lot of the details he harped on are extraneous personally, except for the gameplay ones of course, which are the only ones he lists here.

I think it’s rather interesting that the game randomly chooses between 2 different enemy placements. That’s not a bad idea honestly.

Shrug, ending comments are alright.

Okay, so overall, lets have a look. A lot of his remarks are just fair commentary. A few are things I didn’t know, like all enemies being designed with close and far ranging attacks because there’s a limitation of only 1 enemy type per room. A few I nitpick for the wrong terminology. I picked on him a lot for covering things that I feel were non-essential to the base game review/commentary. I wish he’d comment on things from a more practical perspective in general. He does do a bunch of practical commentary across this, and that’s pretty alright. I guess it represents a step up for him. He still has stuff to work on clearly, because I still had to make a large number of corrections, but it’s still much more gameplay commentary than his videos have previously contained and in more detail.

So not bad, 7/10 overall. (I’m a harsh grader)

Air Gear Game Design Ideas

Which mango series do you think would have some great material to be used in a videogame?

Air Gear, no doubt.

I’ve been rereading it recently and just going, holy crap this would make such an amazing game, but it would be so impossible.

It just has this amazingly rich system it built for skating. It was inspired by Jet Set Radio, but took the concept a lot further and made it more into its own thing, focusing less on graffiti and more on turf wars + the powered skates themselves.

It has this core concept of powered inline skates, these can be customized with different parts and there are different types, riders that operate in teams and compete over turf in different competitions that are varied within each category, progression from simpler tricks to harder ones, different styles of tricks that different riders specialize in, the 8 roads that categorize the different styles of tricks and 8 kings who are the masters of each road, and even tuners who help repair and “tune” the different skates to perform their best.

Here’s a few wiki entries explaining things, there are unmarked spoilers for the biggest twist in the series, so be warned.

And imagining a game of this is awesome, because it has this incredible hero’s journey thing going on with it. The main character starts out barely being able to walk in air trek skates, to riding up flagpoles, to being able to go up walls, to eventually riding on pure air. Other characters learn to shoot fire from their skates, freeze time, and shoot layers of wind at their opponents. There’s a lot of assorted powers throughout the series, but the key is there’s this sense that characters learn them through effort and experimentation. Naturally, having played fighting games and speedrunning mirror’s edge, I have a good idea what this could feel like. When I first learned to kick glitch in mirror’s edge, it was a world opened up to me. It was magical. One of the core themes of Air Gear is this magic of pulling off a neat trick.

The trouble, the part where it’s impossible is, how do you allow players to do all these moves? How do you allow players to learn these moves? How do you pack all these tricks and moves into a system which is naturally designed to have this progression from normal skating to grinding to wallriding to flying across buildings and eventually flying through the air? How do you make players steadily go through these stages?

Part of the reason there can’t ever be a game of Air Gear that lives up to the source is any normal game designer would solve this effortlessly by just making a stat system. Okay, you’ve done 30 dashes, your speed levels up. 30 wallrides, now you can wallride a bit higher. Your overall trick level is 20, now you have access to elemental tricks, all of which are one-button that you can bind to whatever face button or trigger you want.

What I’d really like to see is a control system that somehow emulates the way characters learn tricks. You can take some cues from fighting games, speedrun tricks, mirror’s edge and of course Skate for this.

But how do you feasibly build that, introduce new concepts to the player over time, and avoid players accidentally riding on sheer air by buttonmashing? I don’t honestly know. Is it even possible on current controllers?

Maybe you could have leveling up for the styles and tuning just to keep the insane number of possible tricks you could do reasonable?

The perfect inspiration for air gear tricks is super metroid tricks. The walljump, shine spark, space jump. They’re all really nice design inspirations for low affordance tech. There could be RPG systems that actively change how easy the inputs are to do, like lengthening the windows or even changing the read algorithms to add extra shortcuts.

It’s not a style of game that really conforms to the types of game design that is common nowadays. You’d have to work out how the environments would work, all the things players could trick off of, how to combine tricks, a lot of things.

But damn, it could be awesome. I might write up a design doc if I get any good ideas for it.