Novacanoo Crash Bandicoot Reply

Celia Wagar6 Comments

Critique these critiques?
https://www.youtube.com/watch?v=yZTD1b7ie9I

Holy crap, 11 reviews, jesus christ. Are 11 reviews really necessary?

Second, you’re really monotone. Like really really monotone. Sounds a little forced. Also are you actually british or just putting on an accent?

I don’t think the information on why you’re not covering the games prior to Crash is really relevant. I’m honestly not that interested in the history of Naughty Dog. Maybe someone else is, but I think it’s fluff when it comes to a game review.

Also I think your sentences are a bit long and drawn out, maybe using more words than necessary.

It’s weird to present the introduction to the game before the attract mode cutscene. It’s like you’re starting one idea then interrupting it with an earlier one.

Right here, you don’t mention what the character’s abilities are, how they work, or how intuitive they are.

I don’t think it’s fair to say that the X axis is barely used at all, it’s used about as much as the Y axis. Also don’t think it’s fair to say that obstacles that make use of the X axis don’t play into the strengths of the design. If they didn’t play into the strengths of the design in some way, then wouldn’t it effectively just be a different perspective version of a 2d platformer? A not very impressive 2d platformer?

Today we’d probably call it arcadey? Is that implying that arcadey is a bad thing? Today I think I’d call it uninspired and derivative same as they did back then.

Okay, so a game with high commitment jumps is alright as long as it’s, “designed around them.” This is a buzzphrase. What does “designed around them” mean here? It would appear to be, “the player is asked to jump any distance except the maximum length of the jump”, but castlevania, the example you just cited, requires you to jump the maximum length of the jump sometimes. Also doesn’t it seem like a good common sense rule of thumb to never ask a player to jump the maximum jump length so they don’t have to do ridiculously frameperfect jumps?

What does “poor physics” mean here? Are you sure you don’t mean the fact that you’re jumping into the Z axis, where distance is harder to judge?

“since most of the game’s jumps are the same length, there’s nothing stimulating about the act of jumping.” That’s a decent observation coupled with a swift judgment. I think it would have served your purposes better here to cite examples of how different jump lengths could have been used to construct more varied challenges, and how the system for controlling crash during jumps makes those infeasible, rather than jumping into the conclusion that this is compensation for an unruly physics system.

I don’t imagine that getting the physics right was a big hurdle at all. I think making a competent rendering engine was a much bigger hurdle. Crash’s system for rendering 3d graphics is legendary for its performance on the PS1 hardware. Physics calculations comparatively are very simple (at least, when you aren’t coding something like havok). Collision detection in 3d is harder.

Talking about what’s probably hard or the troubles the developers went through is very much copping MM’s style. I don’t think it’s helping your case, because like MM, I don’t think you have a background in computer programming or knowledge of game engines. Saying, “it must have taken a lot of effort” or “they put a lot of thought into this” or “I imagine that getting the physics right was one of the biggest hurdles” are kind of empty statements.

It doesn’t really matter what the intent of the designer was, or what was hard or what was easy. You can spend tons of effort doing something a hard way that doesn’t really pay off, and a reviewer should be impartial to that in my opinion. It shouldn’t matter to a reviewer how much effort something took. It matters to a development team. Saying “Man, they spent a lot of effort on this thing that turned out to be worthless” isn’t saying much about the final state of the game, it’s advice to make sure a hypothetical design team doesn’t fuck up another game. And maybe you want to include that type of thing as an aside, but it doesn’t properly indict the game itself for it’s own faults.

I’ll draw one exception when it comes to proposing alternative implementations that could have been better than what the game actually did. I usually limit myself to implementations of mechanics that existed at the time, to prevent hindsight bias. However this is still only loosely connected to evaluating the game as it is. If a game did something wrong and the correct interpretation was come up with later, you could just as easily say the game was made too soon to actually be good. It doesn’t matter if the developers tried hard, had the wrong tools, or incredible setbacks, either the game works out or it doesn’t.

I think you could have put in a better shot at explaining how that level doesn’t work so well with the dpad and physics. The dpad claim is especially weird here considering the dpad should make a level that looks like it’s just a straight line easier. It’s a digital input tool, it should be good at staying straight on a level that’s straight, and only moving you forward when pressed and not otherwise.

You’re reaching when you say that the developers tried to force intricate platforming after making compromises to their original design. I don’t think you have sufficient evidence to claim that. The developers are dead. You don’t know what their intent was. It doesn’t matter what their intent was.

Stating that players probably gave up once they reached this level is similar reaching. You don’t know that. Please avoid guessing at other players’ experiences.

The spin attack thing was probably intentional. It would have been easier to implement it the other way, where you stop when you release the dpad. They probably went to extra effort to make it move regardless of your dpad because they wanted you to feel committed to the motion of the spin attack, feeling like it has inertia of its own, like dash attacks in many games. Dying due to it sounds like it speaks more to incompetence on your part unfortunately.

I think the level analyses could have been better if you went over the way the individual levels attempted to challenge the player rather than the extreme cliffnotes versions. How was the level design actually utilized to make the player think? What sort of play does it bring out? How are the enemies used? For the water level, I think it would be appropriate to mention that aligning the camera with the primary plane the character is moving on helps players significantly with accurate movement across that plane. This is why we have 2d platformers with cameras from the sides instead of the back. In many Crash level, the camera is behind crash, which makes it difficult to judge how far things are ahead, but in water levels, they clearly felt that this wasn’t as important as simply showing the player exactly how far the character is moving relative to what’s below him.

I don’t think having a variety of stage types is necessarily a strength, especially if all of those stage types are bad or average. It’s better to do one thing well than 5 things poorly.

In what way do the hog ride levels rely on trial and error? You’re presenting a conclusion before your evidence. Same thing when you say the slippery climb level is extremely challenging but in a good way, because it “takes the limitations of the controls into account” which is incredibly vague. When you mention how it’s a pain to jump on moving platforms with these controls, that could do with some explanation as well. I’d presume it’s because you don’t inherit the platform’s movement when you jump and need to manually accelerate and deccelerate in the air to keep up. Also it would help if you had a better explanation of how bad the controls are in the first place. Also vague is saying that you were frustrated by the last level because again, the controls suck, especially versus the things introduced in that stage, which you don’t elaborate on.

I don’t really care how fancy the final boss’s scenario was for the time.

Game feel isn’t a concept that’s ethereal, though without knowing the exact game logic it can be hard to put a finger on. I think you could have broken down the mechanics of how crash moves and jumps much better than you did. I think this is particularly remiss given it’s a platformer game. This isn’t a factor of age, Mario 64 was released months before it. The game logic required to make a good feeling 3d platformer is not hardware intensive nor beyond the mathematical capabilities of developers of the time. Game logic is usually the least cpu intensive part of any game, barring AI. The issue is that they built a faulty implementation.

Also, it’s really bizarre how you can say that they basically screwed everything up, levels, camera perspective, physics, enemies, bosses, yet the game is somehow still worth playing. You haven’t really mentioned any of the merits of the game at all. Also, dark time when we played 3d games with Dpads, try playing the Ys games on PC. They have a very similar control system to Crash, and you can in fact play them with a dpad (or arrow keys). They control fairly well. The dpad limitation is something that can entirely be worked around. The Crash developers simply failed.

To look into the game a bit more, I downloaded a rom of it and had a try at it. Something I guessed from the video was that diagonal running was at the same speed as cardinal direction running, and that turned out to be correct. My guess is that they have an independent X and Z velocity value that both have physics calculations run independently, rather than a cental polar coordinate value that is later converted into X and Z velocity values using trig. This is a common error in games of the time, but I’d guess that it’s not an error here because of the boulder sections. If you ran with correct angular velocity, then you’d lose distance to the boulder across the boulder chases when moving left or right. Obviously having different speed ordinal directions is part of what makes movement feel so awkward. Another observation is that you didn’t really explain how the health powerups, the tiki masks work. More or less they’re like a mushroom from mario, giving you another hit before dying, but not providing any other function. Also you failed to mention that you can control your movement during a spin, but you have basically no friction and a smaller acceleration force than normal. And that spins can be canceled by jumps and vice versa, and canceling a jump with a spin seems to increase the speed at which you fall, so you don’t seem to get as much distance from jumps if you spin out of them (I can’t totally tell, all my testing seems to indicate the jump is unaffected, but it feels like you get less distance).

I think the game feel issues can be chalked up entirely to the way acceleration works. The character moves at a slow speed across the screen, much lower than say the perceivable speed of mario in Super Mario Bros on NES, or Mario 64, yet the acceleration value is also fairly low. The friction value is the same as the acceleration value, so it creates this feeling like it’s difficult to get moving or to stop moving, where in mario 64 by contrast, if you hold a direction all the way, you get moving almost instantaneously. The dpad is a hinderance here because there is clearly extra acceleration and decceleration going on here compared to 2d games where acceleration is usually close to instantaneous, yet because of the digital controls, fine control over the acceleration is not possible, so the character always moves a bit further than you intend, or not as far as you intend when you press or tap the dpad directions. It’s like having ice physics, except the friction level is the same as the acceleration level, so it comes across as feeling weird rather than clearly like ice. Also friction is only applied when the dpad is released (I think), which is why it takes so long for crash to turn around (because if it’s applied all the time, then he’d lose speed from the direction he’s going at double the normal rate instead of the same rate as when the dpad is released).

This is the same in the air, making it so that if you’re directly over something at the peak of your jump and hold back directly when you’re over it, you’ll still overshoot it based on the time it takes to slow down. Unlike Castlevania, you’re required in some sections to accelerate in the air, which means you need to accelerate enough to get onto the target platform, but stop before you’ll overshoot it when you hold back. Because the acceleration value is so low, it’s easy to over/under shoot this. To accurately jump to another platform from no acceleration you need to accurately hold in your mind how long it will take to accelerate to the appropriate speed to get to that platform across the amount of time you’re in the air, and the point at which you need to hold back in order to not overshoot, plus the added constraint of how long to hold the jump button. This is so far from real-world physics and even commonly intuitive videogame physics that it’s a huge pain in the ass to jump accurately even in the 2d platformer levels. In Castlevania you at least have the luxury of always knowing the exact arc of your jump.

Speaking of the jump, something feels awful about it that I can’t quite put my finger on. I don’t have frame advance or a memory reader that can tell me Z position over time, so my best guess is that as long as X is held, you move up at a certain velocity until a timer runs out, then gravity handles it until you hit a terminal velocity equal to the rising velocity. The end result is that the jumps look consistent every time, but feel like you don’t have a lot of control over them. This is likely because the rising velocity and terminal velocity values, as well as the gravity value, are all fairly low. This produces a sensation that feels floaty and like you don’t have a lot of direct control (because you have no control over the jump height while gravity is doing its work, as opposed to the way 2d sonic and mario control their jumps where either your upwards velocity is instantly braked to a lower value when the button is released, or gravity is altered as long as the button is held, respective to the two games). On a more simple level, you get maybe 1.5 times as much height by holding the button down all the way as not. That’s a really shitty amount of variation.

That the running animation also is forced to play through a full cycle before coming to a stop doesn’t help matters. It seems like there’s an animation of him taking a step, and doing a full run, and it transitions between these when he hits a low enough speed threshold, but the full run can only transition to the step animation when his feet are together, at the passing position. So it looks like he has this exaggeratedly big level of motion when his velocity is very low, which is disconcerting. Animation blending didn’t exist back then, but they could have added extra animations to help buffer this transition out (like mario 64 did).

I also feel it’s weird you didn’t mention the life system and how collectibles work, granting an extra life when you gain 100. Compared to most games in the genre, there are a lot more collectibles dispensed, and the game chastises you for missing collectibles at the end of each level. Also in your review I felt like the remark about how it was unconventional that you could destroy collectibles, even really important ones, was weird. Playing the game, I feel like the reason it’s there is so there’s a hazard as you open boxes, that in your rush to destroy more boxes, you might destroy collectibles too. I just destroyed a save point collectible like 4 times in a row while playing. A more appropriate criticism of the mechanic would probably be to mention that there’s no benefit or alternative use to destroying powerups, or objects in general, they’re just gone.

Also you brushed over way too many of the level challenges. The core of the game is moving forward in the direction the level goes, and around obstacles as they crop up, jumping and spinning as appropriate. You could have explained how the level designs, the enemy arrangements, and so on, combine these 3 key aspects to make the player need to think (or not need to think too hard) about, planar movement, jumping, spinning.

I think overall you need to focus more on describing how the core mechanics work, or don’t work. You’re going to need to develop a better eye for that sort of thing. I think you brushed over a lot of the central interactions between the core mechanics and the enemies/levels. You frequently remarked that they were trivial or not worth mentioning, sometimes that they were hard or frustrating, but rarely got into detail about what goes into any particular section that makes it trivial or frustrating.

Your review ends on an upnote for a reason that is not at all made clear by the contents of the review. I mean, I really have no reason to not think this game is a walking talking abomination all things considered.

Well, thanks for your effort. There’s certainly plenty there to consider. I will say though that the last few paragraphs on controls are where we differ – people like you and me who can understand all that jargon are few and far between, and there are even fewer who find it interesting.

What do you mean we differ? Like, you don’t agree with that part, or you didn’t quite have the knowledge to pull that type of thing out?

Anyway, I find it super interesting. I think it’s one of the most interesting things about games. I’ve programmed a number of simple physics simulations, collision detection systems, and other stuff, and the vast number of different ways these things can be implemented is fascinating to me.

“What do you mean we differ?” I just meant that, while we can both understand everything you meant when talking about acceleration/friction/gravity/velocity values, you find it super interesting where I don’t. And, while I usually hate the phrase, to each their own.

Hahaha, damn! That’s unfortunate.

I like math, and physics, and animation. Games are like the intersection between those things.

With that said, you’ve got a follower in me if you ever get around to making videos of your own. You say you’re lazy, but take out the references to me and paragraphs 51 through 60 could fill out a focused video on Crash 1’s controls just fine. I just dunno what kind of audience you’d find. To show that that 3000 words doesn’t go unappreciated: http://pastebin.com/r12Jjafy I’ve taken the parts I unreservedly agree with and written a little something to keep myself in check. In particular, I really needed to hear that part about making claims about a game’s development as justification for a fault being a totally worthless endeavor. Sorry for quadruple post – I appreciate how constructive your criticism is, and this is the way to act on it that makes the most sense to me. Also it’s a New Zealand accent, believe me you aren’t the first one to wonder.

Thank you and you’re welcome.

You’re right, my remarks on the game from trying it is practically an analysis in of itself. However writing is easy for me, video editing is harder. Why do you think I answer so many asks?

Thanks for your pastebin summing up some of my remarks. That fits into this ask from earlier: http://ask.fm/Evilagram/answers/138183828245 At the time I couldn’t remember some of the more specific procedure I had, but your pastebin sums it up nicely. I might append a rewrite of some of that when I post the ask to my blog.

Good luck, have fun.

Thief 3: Deadly Shadows Review

Celia Wagar2 Comments

Did you ever try Thief: deadly shadows? What do you think?

Yes. I think it’s good, but obviously not as good as the first two. It has % loot requirements in the levels rather than an arbitrarily chosen number, so playing on the hardest difficulty can frequently mean combing the level multiple times for that one little thing you’re missing, which is unfortunate.

I got stuck by a weird glitch where I’d be able to move in the falling state, sort of hovering along the floor, frequently.

There’s chairs that guards can sit in that totally block your ability to backstab them. This sucks.

Guards don’t perform psychic searches when they lose track of your position while investigating, so it’s a bit too easy to shake guards in investigation mode.

The Cradle level is super freaky. One of the best horror moments I’ve experienced in games, though I did play it when I was younger, in college.

No rope arrows. 😦

Way too many sleep and noisemaker arrows. Makes the levels really easy.

The combination 1st/3rd person control system functions really awkwardly in first person, because you’re still bound to your third person hitboxes and movement. You frequently need to switch to really understand what’s going on.

The locks always use the same few patterns, which you could say is a good or bad thing. On the one hand, observant players can notice this and pick locks faster, but on the other it becomes a memorization test instead of a skill test. Also the lock picking process is poorly explained.

The loading zones that cut up levels suck. Guards can’t chase you across them, so the contiguous areas that guards patrol are much smaller than the first two games, limiting both your movement and theirs.

I personally liked the hub world, but it was really limited in scope.

The inability to swim sucked.

Breaking out of prison when you get caught in the hub world was a cute idea, but it gets old after the first time.

Having gold and items stack across multiple missions maybe wasn’t the best idea, removes incentive to perform well on the next mission and lets you steamroll with a big inventory.

I can’t believe Thi4f seriously chose to take inspiration from this game over the other 2. It commits all the same mistakes but worse. I also can’t believe they decided to reboot it when Deadly Shadows had such a great sequel hook.

What Should be in a Review?

Celia WagarLeave a comment

How would you sum up what you want from a game review?

Ostensibly the purpose of a game review is to tell the reader whether the game is good or not, with or without existing knowledge of the game. (usually without, but it becomes more of a tastemaking thing over time instead of a consumer awareness thing). The goal is either to give a consumer enough information to make a purchasing decision, or to justify/influence a player’s opinion of a game they’re familiar with.

Assuming this is the job of game reviews, then the question is, what are all the things necessary to accomplish this? Assuming the reader has no knowledge of the game beforehand, the review must make clear how the game plays. Thanks to streaming video, we can do this much more easily than we used to be able to, but everything about a game’s operation is not clear from video alone.

From there, I think it’s about evaluating the way the game is played, and the content that is played through in a way that is made clear through predefined criteria. You need to establish what traits make a game good and how this game lives up to that. If this game violates norms, you need to argue for how it redefines what makes a game good, and not in a localized way that explains this one game without offering a broader, more general, explanation for games categorically.

Beyond that, there’s probably a bunch of really specific examples of what I’d like to see out of game reviews that are hard to sum up generally. I’d like to hear less of, “This mechanic feels loose and difficult to control” and more of, “This mechanic functions exactly this way, which may produce a loose feeling.” I want to see more of the scientific process going into game reviews, more evidentialism. Cite examples and use them to build conclusions. Do not mix opinions and observations. Do not attempt to tell us how you feel about something in the process of describing it. Describe it first, then explain why it does or doesn’t work.

Beyond that, try to get at the heart of the game. What’s the central enjoyable thing the game is about?

Have you talked about what you think makes a good review? Is it anything besides a precise description of mechanics and an evaluation of depth?

Yes, I answered that in this older ask:
http://ask.fm/Evilagram/answers/138183828245

It’s about user experience, depth, challenge, game feel, and maybe a couple other things.

Novacanoo wrote a summary of what guidelines he inferred for reviews from my critique of his review:
http://pastebin.com/r12Jjafy

Oh yeah, I’ve updated that. http://pastebin.com/CjuxfCbn The first and second points are now somewhat contradictory, but that’s the eternal struggle.

That’s cool, thanks for the update. Don’t totally agree about the judging intention thing. I’ve gone over that in other asks.

Frame Trainer Tool & How Long Are Frames?

Celia WagarLeave a comment

A lot of people ask me how long a frame is. I reflexively measure timings in games using frames (assuming 60fps), and I have a rather good sense of it. This comes both from being an animator, and experience with games in general.

So I had the idea, why not make a tool that helps show how long a frame can be by giving people an interactive example? So I got into flash, putz’d around for a bit, and ended up with this.

https://www.dropbox.com/s/nt7dbdu6hg56tcu/Frame%20Trainer.swf?dl=1

https://www.adobe.com/support/flashplayer/debug_downloads.html (download the flash player projector and drag the swf from the zip file above to use the frame trainer, since flash is deprecated on every platform ever now.)

To operate it, click the Go button to start the arrow moving from left to right. Press any key, or click the stop button to stop the arrow. The goal is to stop the arrow when it’s yellow, just before it hits the end.

You can configure how long it takes to get from the beginning to the end by changing the Frame Total text box on the left. And you can change how long it turns yellow with the Frame Window text box on the right.

By changing frame total, you can give yourself more or less waiting time before it turns yellow. By changing the frame window, you make it turn yellow for longer.

If you stop it while it’s green, you did it too early. If you stop it while it’s red, you did it too late. Use the arrow getting close to the end as your visual cue.

By default I set it to 40 frames total (to give you a decent amount of reaction time before you gotta press the button) and a window of 7 frames, which is the window for parries in 3rd strike and L Cancels in Melee. Try setting it to all different periods and trying it out. Try setting it for different periods of time just to see how long they are, like 60 frames is 1 second, 30 is half a second, 20 is a third of a second, 10 is a sixth, 5 is a twentieth.

For reference, here’s some other frame windows from various games and my description of how easy/hard they are.

1 frame: Reversal window in Super Turbo and Guilty Gear before Xrd. Kick Glitch Window in Mirror’s Edge. 1 frame link timing. This is the hardest possible single input in a 60FPS game. Obviously combination inputs can be harder. Coincidentally, I had to jerry rig the setup to allow this window (it would otherwise show red on the last frame), and I managed to test it was working successfully on the 3rd try.

2 frames: Reversal window in 3rd strike, power shield window in Melee, throw tech window in guilty gear. Boost Smash/DACUS window in PM. This input is almost perfect, allowing just enough leniency that people can feasibly get consistent at it.

3 frames: Smallest possible window for a link in SFV, reversal window in SFV and GG Xrd, common window for links across fighting games. This one has a tight timing, you’ll feel that it’s really tight. It’s practically the exact moment that the thing hits, except significantly more lenient than 1 or 2 frames. I can do these consistently in SFV. Any mediocre fighting game player can do these in their sleep.

4 frames: Perfect Shield window in Brawl. Slightly less tight, but still enough to be difficult.

5 frames: Reversal Window in SFIV, parry window in DMC3/4. This is where the window becomes wide enough to let you get the input even if you mashed it (unless of course there’s a lockout period to dissuade mashers, like the DMC parry has)

7 Frames: 3s Parry Timing, L Cancel Timing in Melee. There’s a bit of wiggle room here. You’re no longer pressing the button just as you reach the end, just as the fireball is about to hit you, or you’re about to hit the ground. If you do it a bit early, you are forgiven.

15 frames: Average human reaction time. Throw Tech Window in Blazblue.

20 Frames: The Tech Roll window in Melee. This window is so wide, there should be no reason to miss it if you see it coming, it’s completely outside average human reaction time.

30 frames: Half a second. Blazblue has a 27 frame throw break window for throws during hitstun/blockstun. The Parry Window in Metal Gear Rising is this long. The parry window in Rivals of Aether is this long.

50 frames: Seth Killian once said that the counter window in Batman Arkham Whatever is like 40-50 frames. This is so long that it’s practically impossible to miss.

By the way, if anyone wants me to make a 20 fps or 30 fps version of this tool, then I can do so easily. I tried to add another box that let you change the framerate manually, but it didn’t work.

The Cooldown Effect

Celia WagarLeave a comment

What do you think of some games that have special moves with cooldowns?

I don’t really like cooldowns. Cooldowns tend to promote play based on rotations. Cooldowns tend to promote abilities that aren’t designed very interestingly. I’ve spoken on this before, how I dislike the way RTS games and the genres influenced by them tend to have “abilities” instead of “moves”. Like a move or action in a traditional game has different periods of timing, it moves through space in an interesting way. There’s startup, active, recovery. “Abilities” tend to function instantly and ignore all the physical elements of how the action could function. They don’t feel grounded.

Compare a Moba to a 2d Zelda game and how different the items and attacks feel.

The other thing is, when abilities are designed with cooldowns in mind, generally the design ignores things like drawbacks to those abilities. Think of how many cooldown abilities in various games would be broken if you could spam them over and over again. What stops you from doing the same in say bloodborne, a fighting game, a first person shooter? Recovery time, ammo limits, reloading. There’s risks associated with these actions that make you not want to perform them all the time. In “ability oriented” games with cooldowns, if you could you would mash all your abilities at once usually.

So what’s the drawback to not using an ability? If you use it now, you can’t use it later. The real cost is opportunity cost, which is sensible, but kinda lame if you ask me. This leads to rotation based play, such as seen in MMOs, because when one is cooling down, you use the next one, then the next one, then loop around to the first one as it comes off. In some cases, there’s a good order to do these in that each boost one another for maximum efficiency, so you follow the rotation. Leads to very efficiency race styled play, because there’s no drawbacks to the moves that would make them counterable.

Rising Thunder made abilities work a bit better, but also those abilities were designed like Moves, they had the solid grounding in mechanic design that abilities don’t tend to have. Notice how all the Overwatch abilities tend to function instantly, where something like say, Mirror’s Edge (to use another first person example) has a bit more kinaesthetic sensation in its moves.

Thiefer’s Edge

Celia WagarLeave a comment

How would you design a game with stealth elements from the Thief series, mixed with the parkour system with all of it’s advanced techniques found in Mirror’s Edge?

Interesting idea. I wasn’t sure how this could work at first, but I think I might have an idea based on an older concept I thought up, of a stealth game based on speed. The idea is that as you go faster, you’re less visible/audible, so stealth is about trying to keep up speed and not let it drop. Keep on moving, don’t get tripped up.

Mirror’s edge has a system that supports this, a lot of the game is about avoiding getting tripped up. Many things like climbing up ledges or over fences have varying standards of success. Do it higher and more smoothly and make less noise, and go faster. This could also be applied to fall damage. Landing without a heavy fall is optimal, rolling is less optimal, then two levels of fall damage that make less and more noise. More perfect sideboosting could create less noise too.

A basic thought is, how do you inform the player that they’re being loud or quiet? One idea is that loud sounds can be lower pitched and more bassy where the quiet sounds can be higher pitched, softer in tone, and less bassy, so players can clearly distinguish them, yet still receive auditory feedback.

It would make sense to have footstep sounds change radius at different speed thresholds, so that things like wallboosts and the like can temporarily push you over the normal speed threshold, making you quieter until you slow down. It would also make sense to add a sound for hitting a wall at high speeds.

There should be a more mundane stealth system based on going slow on top of this, because it’s hard to go fast without knowing the level layouts. So you have the high level, gottagofast stealth, then the low level stuff. The low level could function more like thief, the high level more like mirror’s edge. Throw leaning on Q and E, move use onto F. Implement the fancy lighting system. Add a blackjack that can disable guards who don’t detect you.

The hard part is the level designs, the enemy AI. To make something suited for this would be difficult. You need to design around the fact that the player doesn’t know where stuff is in advance. How do you plan guard patrol paths that players zip by without getting a chance to study them for flaws? You could give players wallhack vision like everything does these days, show guard vision cones too. Whatever.

The levels probably make more sense being linear than open if speed is the focus, but you gotta provide a lot of paths to get around guards, otherwise the game doesn’t really make sense. I dunno what the final product would look like, would require a lot of thought to put together, maybe research into similar games, if any exist.

Stunning Detail: Hitstun in Depth

Celia Wagar1 Comment

Not sure if there’s much to talk about on this subject, but care to do a writeup on hitstun in videogames?

Hmmmm, yeah, there’s actually a bunch that could be talked about I think.

The big factors are interrupting attacks, advantage time, and pushback. Other things are knockdown, juggles, counter hits, meaties & multihits, special hitstun states, dizzy, recovery options out of hitstun, control during hitstun, hitfreeze, super armor, no hitstun, and blockstun.

The single biggest thing hitstun does to a game is it allows one character’s attacks to interrupt another character’s. This means a lot. It means that the first person to attack will win, which means you don’t always want to throw out an attack because your opponent might be attacking first. It also means you can’t just attack repeatedly, because eventually your opponent will interrupt you, assuming you can’t infinite them.

Advantage time or frame advantage is basically the difference between when you recover from your attack, and your opponent recovers from hitstun. If you recover first, then you are said to be plus on hit. This is noted as +# where the number is the amount of frames you’re up on your opponent. If you recover second, you’re minus on hit, noted as -#. Being plus means you get to act first, minus means you get to act second. The more plus you are, the sooner you can act, which means you have a higher chance of winning if both of you attack at the same time, and you can afford to throw slower moves since they effectively have less startup time relative to your opponent. If you’re minus then it means your opponent can act that way towards you. Going minus means surrendering initiative to your opponent. If you’re minus enough then it is even possible that you could be hit back by your opponent. Personally I consider it a flaw to allow a move to be minus enough on hit that you can be punished by your opponent. This is fine on block, but on hit it’s unacceptable. If you’re plus enough on hit to have enough advantage time to fit the startup of another move in, then that’s a combo. If you’re plus enough on hit or block to deny your opponent enough time to start up any of their moves, then that’s a frametrap.

Next up is pushback. Pushback is related to frame advantage because if you push something far away enough, then frame disadvantage becomes safe. I suggested with a theoretical 3d zelda that it should have low hitstun and high pushback, so you’re giving initiative to the monsters with each successful attack, but you’re never unsafe to their attacks. The more pushback there is, similarly it becomes harder to follow up when you get frame advantage. This can help keep combo systems in check.

A basic example of this is the smash bros knockback/hitstun system. In Smash Bros, attacks do a certain amount of knockback based on the attack strength and percent. The number of frames of hitstun is the knockback value times 0.4. This means you’re gaining more advantage as you push opponents further away. Eventually you have a ton of advantage time, but can’t catch up with them.

The exception to the rule in Smash Bros is meteors or spikes into the stage. These knock off 20% of the knockback, but keep the hitstun the same. So in Smash the general key to building combos is to at first hit opponents with attacks that have a quick recovery at low percent, then ones with good knockback and good recovery at mid percent, then ones with weak knockback to keep opponents close to you at high percents, then kill them with something really strong.

Knockdown isn’t generally considered a state of hitstun, but it’s pretty similar in a lot of ways. Players or enemies aren’t allowed to act during this state, and it’s generally harder to attack them. In some games they’re entirely invincible until they get up (dark souls, most fighting games). In some games they can be hit, but they take less damage and aren’t sent into a new stun state, effectively having super armor. In some games they can be combo’d off the ground by certain moves, which may have penalties to combo damage or generally be less convenient. In some games the attacking player is allowed to perform special finisher moves on knocked down opponents. In some games the player (or enemy) who is knocked down has multiple options for getting up, like rolls, or getup attacks. Knockdown generally serves as a longer special state of hitstun where opponents can’t do as much, and you can use them being knocked down as setup time to hit them when they get up.

Meaties are when you hit someone with the late part of an attack on purpose. In most games, attacks do the same amount of hitstun no matter what part of them hits, so hitting with a later frame of the attack means you get more advantage time. This is especially noticeable with rushing attacks where a character moves across space with a hitbox out. By hitting with the late portion of an attack, it’s possible to do new combos, or make attacks safe on block with good timing or spacing. Slow moving projectiles can have an effect similar to meaties at a range by allowing the user to recover and move up behind the fireball before it hits.

Multihits are related to meaties, because they’re kind of the opposite of them. Multihit attacks hit with many hitboxes over a period, each one dealing its own hitstun. Multihit moves get a consistent amount of hitstun no matter what part of the attack hits, but they keep the opponent locked down for their duration. This means multihit moves can never be unsafe by hitting with an early portion of the attack, and they usually go on long enough to give the player time to confirm they’re hitting. Fox’s dair versus falco’s dair is a good comparison for meaty versus multihit, except that Falco’s dair is one of the rare examples of a hitbox that stays out for a long period that actually decreases in hitstun the longer it stays out. Jigglypuff’s pound might be a better example.

Some games have special states for hitstun, like in some fighting games such as guilty gear, skullgirls, and third strike, you can hit your opponent reeling back, which makes them vulnerable to throws and command throws, which they normally aren’t during hitstun. Another example would be the dizzy state, where upon hitting an opponent enough, a meter fills up that when full causes the opponent to enter a prolonged state of hitstun, giving you enough time to set up whatever attack you want before they recover. If the opponent is player controlled, then usually they’re allowed to mash to speed this animation up. In Dark Souls, some bosses have special states similar to dizzy once they take enough damage, like the iron golem who can be knocked off balance, or a bunch of bloodborne bosses who will let you visceral attack them once hit enough. Some bosses also have tails that can be cut, triggering a special hitstun when they are. Guard Crush is similar, except it’s triggered when an opponent guards too many attacks, or has a specific guard breaking attack performed on them. This sometimes also awards bonus damage to the attacker (like in dark souls 1) or allows a special followup (dark souls 2 and 3). In the Souls games, The Last Blade 2, Nioh, Soul Calibur 1, and Guilty Gear Xrd you can parry your opponent to put them into a special parried state, which sometimes allows specific followups and in the case of Soul Calibur 1 and Xrd, they can parry you back during this state, at risk of being punished. Guilty Gear has a couple more special hitstun states in the form of the ground slide and wall stick. In some games you can set up special hitstun states that make the enemy weak to certain attacks, like freezing an enemy in metroid prime, then shattering them with a missile.

By far the best known type of special state for hitstun is juggles. Juggles are like an extended state of hitstun, except they have more timing involved than normal hitstun because you gotta time your attacks to hit the enemy on their way down. Juggles can be affected by enemy weight, gravity, or air drift. Juggles can have a lot of different potential trajectories and make use of things like ground bounces or wall bounces. Characters can even follow those being juggled into the air in some games to wail on them up there. The dynamics of a juggle are typically that you need to keep them in the air, or they simply fall into a knockdown or ukemi. Juggles are limited in some games by the natural ending of combos and progressive gravity. In some games characters are capable of “air teching” juggles to return to a neutral state. In God Hand, enemies will flip out of juggles at a slightly randomized angle.

Recovery options out of hitstun are relatively unexplored for most single player games except for the god hand example above. In a number of fighting games (smash, darkstalkers, marvel, blazblue, SFV) you can tech knockdowns to get up faster or roll in different directions. These typically have different amounts of invincibility associated with them. In God Hand, enemies can also block during normal hitstun, but cannot during counterhit/guard break hitstun.

In Bayonetta, enemies are allowed to attack directly out of hitstun, unless it’s a juggle or witch time. In many action games it’s possible to tech the landing or do a getup attack when you’re knocked down. The more recent soulsborne games let you roll out of knockdown, but have less invincibility on knockdown in general. Arc Sys games allow players to air tech from juggles, but the combo can continue if they don’t as an invalid combo. The Smash Bros example is interesting, because it’s possible to beat every option on reaction in what’s called the reaction tech chase. The option your opponent will go for is clear on frame 19 of the animation, allowing you to jab reset if they missed the tech, grab on neutral tech, and dash left or right and grab if they rolled. This might be an interesting avenue for a single player game to pursue in their hitstun design, allowing enemies different ways out of hitstun that sit at the periphery of human reaction time, so you can follow up your combos if you react perfectly. This could keep combos more active and engaging. Some attacks in some games can deal “hard knockdown” which is untechable.

Hitfreeze or hitstop, or hitpause is a short delay that happens at the moment of impact to both the attacker and target before hitstun plays out. In most games this isn’t very serious. In Smash bros you can actually jiggle and move yourself around during this period. In a lot of older games hitfreeze is applied instead of hitstun and only to the damaged enemy. Meaning that hitting an enemy will temporarily freeze it, but will not interrupt what they were doing. This can have a hitstun-like effect, but since it does not interrupt the enemy’s action, you are more vulnerable to what they do. This works well in older games and ones styled like older games because they were more about enemy movement patterns with collision damage on touching you rather than telegraphed attacks, so you can attack them to freeze them temporarily and hold off their assault for a bit in something like a pseudo-combo. The bottle in Castlevania 3 inflicts so much hitfreeze it can stunlock enemies.

Super armor allows characters to ignore hitstun, especially if it’s only for one hit, or below a certain threshold. In most fighting games, super armor works per-hit, so you absorb a certain number of hits before taking hitstun. Skullgirls is a notable exception, as well as smash bros which has multiple types of super armor that all do not follow the per-hit rule, knockback threshold and knockback subtraction armor. Armor that can take an infinite number of hits is called heavy armor. The souls series came up with their own concept of armor called poise, turning hitstun into something similar to a dizzy meter. Once the invisible poise meter is depleted, the target will take hitstun until the meter refills, and the combo ends.

Some attacks, particularly ranged attacks, deal no hitstun, like fox’s lasers and FANG’s poison ball. This is kind of interesting for low startup low damage attacks in systems that normally have hitstun (unlike say FPS games which do this, but don’t normally have hitstun, so there’s no contrast there).

In some games, like most modern fighters after SF3, counterhits, meaning attacks that interrupt an enemy’s startup or active frames, deal extra frames of hitstun. This means that new combo options open up if you can interrupt an opponent’s attack. This is primarily useful for frametrap strings, because the intent is to catch the opponent mashing buttons during the string, at which point the frametrap suddenly becomes a combo and it can serve as a confirm into a longer sequence. In God hand, counter hits carry special hitstun and juggle properties depending on the move. In SFV, certain normals can crush counter opponent’s attacks for bigger combo damage. In Souls games counter hits have no special effect on hitstun, but deal more damage. In Blazblue there are Fatal Counters and Counter Hit Carries. Fatal Counters add 3 frames of hitstun to every subsequent attack in the combo, making comboing off any move more easily. Counter Hit Carries are a property of most multihitting moves, making every single hit of the multihit a combo so there’s frame advantage at the end of the multihit sequence on the last hit, making it more easy to combo off that. In SFIV the level 1 focus attack will cause crumple on counterhit (another special extended hitstun state).

Blockstun is like hitstun except obviously you block the attack. A lot of the same rules of frame advantage apply, though the rules for what you can do out of block may be different per game, like in Smash it takes 15 frames to drop a shield, however you can grab out of it or jump out of it immediately, and your opponent is almost always negative, so the different options out of shield can be a big deal. In games like dark souls, different shields have different amounts of deflection, inflicting special recovery animations on opponents who attack. This is also found in Last Blade 2 for sword attacks versus punches or kicks. In some fighting games there are special blocks like instant blocks in guilty gear which reduce the amount of blockstun taken so attacks can be more readily punished on block, or faultless defense which increases the amount of blockstun along with pushback and prevents chip damage. Smash has similar to the faultless defense with its light shields that increase blockstun and pushback, also its perfect shields which can be canceled directly into actions or in later smash games negate shield drop time.

I think that’s everything I can say on hitstun. All of these are different factors that can be tweaked, tuned, combined, or experimented with to make different effects.

Saurian Dash did an amazing exposition on hitstun states for Transformers Devastation that shows some of how this information can be applied practically to a game. He calls these hit reactions, and later refers to “hitstun” as a special type of hitstun, akin to dizzy in fighting games.

This video shows how hitstun states can be varied between enemies to give them variety.

Favorite Difficulty Modes

Celia WagarLeave a comment

What are your favorite hardest difficulty modes from the games you’ve played and why?

Thief’s Expert Mode (made the AI smarter and added more objectives), and I guess DMC4’s Legendary Dark Knight (also heaven and hell, and hell and hell)? Revengeance might count too (I prefer Very Hard personally, but Revengeance had 1000% enemy aggression, plus powered up perfect parries). I remember Order of Ecclesia’s hard, locked to level 1 mode to be really fun (NG+, enemies had more health). European Extreme in MGS3. Crysis’s “Delta” difficulty deserves a shout-out for the soldiers speaking in Korean. God Hard in Vanquish is also cool, but really I just love the name.

One of my favorites though is one I was developing for Dark Souls 1. I called it, “Hokuto No Souls”. I made it by combining the enemy aggression mods, the permanent gravelord mod, NG+7, and a mod I made that edited your fist damage to kill any enemy instantly. I was going to edit the weapon swap mod to force you to only use fists (and a claw in your offhand, both for variety and because lefthanded punches are really fast). This would be packaged in addition to a God Hand hud that I made by ripping graphics from God Hand. I also wanted to edit the damage of the weapon so it would scale with level so I could have it so bosses need to be hit a couple times before dying, and later areas wouldn’t totally outclass you.

So the idea is that everything kills you in one hit. You kill everything in one hit. You have only your fists. There are more enemies, and they’re all aggressive.

I planned on doing an LP of that when I finished it, but I never got around to it due to not knowing enough about coding in Cheat Engine.

I love it when games do something creative with their difficulty levels. Imagine a game that had a slightly different gimmick on each difficulty level, so you actually wanted to replay them all.

Definitions: Fairness and Challenge

Celia WagarLeave a comment

You’ve got a solid definition for “depth.” Can you give similar definitions for “fairness” and “challenge”?

Sure thing.

Fairness is perceptual, people have differing standards of fairness. In single player games fairness is typically a matter of accurately communicating to the player what they’re engaged in, preventing them from getting blindsided, or letting them know that they may get blindsided and how to prevent it. Fairness is typically perceived as a matter of clarity. For that purpose many single player games even overcorrect, adding extra buffers or leniencies to tilt things in the player’s favor, because player perception is not always perfect, and they can miss an input in the space of a frame or two and think they got it when they didn’t. Many games make enemy hurtboxes smaller for this purpose.

In multiplayer games, there’s a similar impetus on communicability for fairness, but more than that it comes down to fairness between players and there’s even more widely varied standards there. This is typically a matter of balance, but players can perceive a lot of things as unfair, leading to scrub mentality. They can get hung up on individual moves that they think are unfair. They can get hung up on other players being better than them. They can get hung up on which characters are good or bad. It’s a mess. Is it fair that players can master certain techniques and win way more consistently? Is it fair to use unintended tricks? Is it fair to use unreactable mixups? Is it fair to have randomness affect the outcome of a match? Is it fair for a player who puts in less time to beat someone who puts in more time? Is it fair that some characters are better than others? A lot of these questions will be answered differently by different people.

Fairness means treating players without bias and preventing injustice. In Singleplayer contexts, preventing injustice is a matter of figuring out what players perceive as injust or imposing your definition of just upon them (“average human reaction time is about 15 frames, if this move’s startup is 15 frames long, you have no right to complain”). Ultimately you’re working to satisfy your players, so you need to be in tune with their perceptions to make them feel like the game is fair.

Challenge is a lot simpler. It’s a task that one is inconsistent at completing, where consistency is dependent on the individual’s skill. It may be used to refer to a specific instance of a difficult task in a game (eg. a reaction challenge).

Games for Learning about Depth

Celia Wagar21 Comments

What games would you recommend for learning game design and depth in different genres (games for FPS, Action, Fighting, Strategy, Puzzle, RPG, etc.)

I think you gotta play a little of everything, good and bad. It pays to see games that screw up too. I think analyzing Nier was interesting in part because it’s so clearly flawed.

FPS:
Doom (great enemy variety, great level design, alright weapons), Blood (one of the best doom derivatives), Quake (3d successor to doom, awesome movement, so-so weapons and enemy variety, but still good compared to modern shooters), Unreal (I dunno, supposed to be good), Serious Sam 1 & 3 (good enemy design in the absence of good level design), Tribes (cool movement system, amazing emphasis on large maps), Desync (great enemy/weapon variety, very focused on combat encounters, weapon combos), Crysis Warhead (best in the series, nice suit abilities, nice levels, decent enemy AI)

Fighting:
SF2, SF3, SFV, KoF 98, 2002, XIV, Garou, Last Blade 2, Guilty Gear AC+R, GG Xrd, Marvel 2/3, Skullgirls, Vampire Savior, Melee, Divekick,

RPG:
Pokemon (lots of configurable parts, every monster you encounter is made from commonly accessible parts), Shin Megami Tensei Nocturne/Digital Devil Saga (press turn system is brilliant, strong emphasis of buffs/debuffs), All the Mario RPG games (Timed Hits yo, and a couple other nice things), TWEWY (alternative approach to RPGs from the the ground up, tons of interlinking systems), Penny Arcade RPG (unique approach to ATB systems and realtime action queuing), Zeboyd Games RPGs (interesting choices at every step), Tales of Symphonia/Abyss (I think these are the best in the series, I don’t really know, action combat with a fighting game inspiration), Megaman Battle Network (deck building, unique grid based combat system)

RTS/tactics: (I’m weak in this category and haven’t played a lot of the games I’m recommending)
Starcraft Brood War/Starcraft 2 (I recommend brood war because it’s good, though unless you have someone who knows how to play that you can springboard off of, you won’t get much out of it, 2 for contrast and because it’s also good, but less so), Supreme Commander Forged Alliance (I believe this is the best version of supcom, I’m currently playing this), Company of Heroes 2, Dawn of War, Dungeon Keeper, Warcraft 3, Warcraft 2 (for contrast, the two games are significantly different), Age of Empires 2, Homeworld, Command and Conquer (Red Alert 2 or Generals), X-COM, a fire emblem game, advance wars.

Stealth:
Thief 1 & 2 (great emphasis on lighting levels and floor surfaces, great level design, slightly collectathon-like regrettably), Metal Gear Solid 3 (the deepest stealth game), Mark of the Ninja (one of the most versatile stealth games around, second deepest perhaps), Monaco (gets the interesting part of running away from guards completely right, does alright at everything else), Hitman (disguises), Splinter Cell (I dunno).

Platformer:
Mario 64 (has a ton of different options for movement and levels that allow you to take advantage of them), Mario Sunshine (Same, but slightly different), all the mainline Super Mario Bros games (1, Lost Levels, 3, World, NSMBW) Yoshi’s island, Kirby Canvas Curse (unique as hell, one of the best kirby games), Ducktales, STREEMERZ, Bubble Bobble, Sonic (pick one), a donkey kong country game, Megaman 2, 3, 9 (solid design), Megaman X1, 2, 3, Megaman Zero (I don’t know which to recommend), Castlevania 1/3 (great level design with simplistic limitations), Order of Ecclesia (nonlinearity and complex melee platformer combat), Ninja Gaiden 1-3 (great simple fast design), Cave Story, Kero Blaster, Demon’s Crest, Metal Slug, Contra, probably a dozen NES and SNES games.

Metroidvania:
Metroid, Super Metroid, Metroid Zero Mission, AM2R, Ori and the Blind Forest (tons of movement mechanics that all have interaction with each other), La Mulana, Battle Kid 2, Megaman ZX, ZX Advent.

Top down 2d action:
Zelda, Link to the Past, Link’s Awakening, Oracle of Seasons, Ys Origin (like a 3d zelda), Ys 1 & 2 (Bump system!), Lara Croft and the Guardian of Light (tons of unexpectedly 3d object interactions), Hotline Miami (stealth and mixed action),

Beat Em Up:
Devil May Cry 3/4 (command moves for days, tons of recombineable moves), Bayonetta (Dodge offset, great enemy designs), Ninja Gaiden Black/Sigma/2 (enemies that want to kill you so hard, great use of blocking and dodging in one system), Transformers Devastation (culmination of everything platinum, 3rd person shooting, unique vehicle dodge system and vehicle attacks),

Racing:
Mario Kart DS (my favorite mario kart, best physics), F-Zero GX (deepest racing game), Wipeout, Trackmania, Need For Speed (dunno which one)

Puzzle:
Antichamber (metroidvania puzzler with a funky layout, and nice unique puzzle mechanics), Portal 1 (lets you place portals in a ton of places, has multiple solutions to every puzzle, great speed tech), Professor Layton (just a ton of nice puzzles of all different varieties, not really deep necessarily), The Witness (interesting approach to puzzles even if it doesn’t work out all the time)

That’s all I can think of. Notably this is not just a “my favorite games” list.