Celia Wagar's CritPoints

What do you think of special moves in fighting games that damage the player when they use them?

I’m not a fan of mechanics that push both players closer to losing simultaneously, negative feedback. Mechanics like this could also be called Collateral Damage or Mutually Assured Destruction. The idea is, you’re taking damage, but hopefully you deal more than you take, and even though you’re wrecking yourself, you will push the game closer to ending faster and be on top when it does end.

You can think of moves like this dealing effectively less damage, since they’re damaging you at the same time, so you’re not gaining as big a lead over your opponent by using them. So if I deal 4 points of damage to you, and 1 to myself, I only really dealt 3 points of damage to you, but moved the game 1 point closer to overall conclusion. This can tend to make games more swingy, because it functions a bit like negative feedback. It also makes characters with this trait effectively weaker than characters without it, unless their damage is compensated for the damage they deal to themselves.

In a game like Smash where the amount of damage you can take isn’t capped, and isn’t directly tied to losing, this can work out slightly better, because you can continue to use the self-destructive moves ad infinitum, just taking on additional damage, instead of outright killing yourself. I take this tactic with Snake a lot. He can pull out grenades and go suicide bomber with them, and you blow yourself up a lot, but hopefully you blow the enemy up more often. Also when coming down, you can pull out a grenade and if the opponent hits you, it will cause you to drop the grenade. If they hit the grenade too, then it will explode, damaging both of you, and overriding the knockback of their move, usually for a lower knockback, which can help you survive a lot of hits. In this way you can evade death at great personal cost, while dealing damage to the opponent as well. Since there isn’t a fixed percentage that you die at in smash, you can abuse this a fair amount, and use it to turn the tables on your opponent (though of course, you come out a lot worse in that trade than they do).

Overall though, I think adjusting payoffs in this way doesn’t contribute significantly to the depth of a game. It doesn’t strongly differentiate the game states in my opinion the same way changing the physical characteristics of moves does. I don’t think Pichu is significantly more interesting for damaging himself, I think it’s largely pointless. Even if it is a cool touch for characters like Tien in DBFZ (and it works mildly better in DBFZ, because killing a character is a much bigger benefit than just getting a health advantage, so it can be a tradeoff of orthogonal assets, rather than strictly victory points).

How much of a problem is extended travel time in games? For example: lack of fast travel in the beginning of Dark Souls, or traveling to a location in an open world? Is it just filler?

I think this is a false premise.

Travel time can be a pain, but I think the better metric is, time spent without a challenge of some kind, or even more accurately, time spent without an interesting decision, because obviously things like configuring equipment, leveling up, and other such clerical tasks aren’t challenging, but they involve interesting decisions.

I have stated many times that I really like that there’s no fast travel at the start of dark souls. I like that you need to pathfind your way around a complex series of areas that have new connections opening up as you go, so trying to find the optimal route is an interesting decision in-of itself, and you need to do things like fight enemies on the way there which is an interesting series of decisions, and given you’re pathfinding, you end up generally going through the areas in an order you didn’t go through it the first time, so it’s slightly different each time, rather than just repeated content.

I think open world games suffer a little more in this respect because the world is not dense enough with challenges. Traveling from place to place takes a lot of time, and you don’t do much along the way, which can be boring. This is where lessons from platformers or racing games could probably factor in more.

What genre, if any, do you think benefits most from an open world?

The open world genre.

I mean, open world is kind of its own genre at this point, right? Among the big entrants we have RPGs, like Skyrim and the Witcher 3, “Stealth” like Asscreed and MGSV, and a ton of 3rd person shooters, like Just Cause, Watchdogs, GTA, Far Cry, MGSV, Red Dead Redemption, plus a lot of others, and we have sandbox games like Minecraft and No Man’s Sky.

I don’t think any genre benefits particularly more than any other genre from being combined with open world. It’s just a content presentation style really. Probably games that are more sandboxy like minecraft or breath of the wild benefit a bit more, but I mean, being sandboxy comes with being an open world game to a degree.

What do you think of the “bloom” mechanic in FPS games?

I dislike randomized bullet spray in general, so naturally I don’t like Bloom either.

For those not aware, Bloom is when a gun’s bullet spread area increases over time as it’s fired, resetting when you stop firing. It’s supposed to encourage burst shots as opposed to spray fire. I don’t understand the intention of doing this, except maybe some sense of realism. It seems like whatever it accomplishes could be accomplished a different way than this.

In regards to “bloom” in FPS or randomized bullet spread, what would be your alternative? I assumed it was for skill purposes. My assumption was that instead of your gun being a laser beam, you now have to control the recoil in order to win the gun fight.

Bloom is not Recoil. It’s about the cone of fire expanding over time, not the cursor moving upwards over time.

Bloom does add a skill element, burst firing the weapon instead of holding the trigger down, but it’s questionable if adding a skill element like that is the actual intent of bloom, because that’s a pretty lame and linear skill element to add, requiring people to tap the button instead of hold it down.

My alternative would honestly be just having the gun be a laser beam, with a fall-off in damage over a range, so that you can’t use an assault rifle as a sniper weapon. Barring that (if you REALLY wanted it to look like a gun and have the bullets spray everywhere so it looks somewhat realistic), deterministic bullet spread, similar to Counter Strike, so that skilled players could functionally have the gun work like a laser beam anyway. Star Citizen is doing a neat thing with recoil based on procedural animations, but I don’t really expect that to be much different functionally than random recoil or spread.

If you wanted the minor skill element of pacing your shots, you could have damage decrease as you fire longer. There’s always a deterministic way to recreate these behaviors without introducing randomness, but for some reason I always get ridiculous amounts of pushback on removing this type of randomness. We happened to make these effects in mimicking real life weaponry, and post-hoc we’re defending the benefits of designing weapons this way. We could achieve these same benefits without invoking the random number generator, but people complain that it’s unrealistic, that players won’t like that type of behavior, that it won’t actually accomplish the same purpose, etc.

I don’t see the point in having some of your shots randomly miss, and therefore your DPS randomly being higher or lower sometimes. I don’t see the point in being able to compensate for this by pacing your clicks.

Honestly, I’d like to be able to just point the gun at people and hold down the trigger while keeping the reticule on the target. I don’t get why we need all these other things getting in the way of that. I don’t get why we need to randomly vary how much damage you do. I get that bullet spread makes shots less accurate over a distance, so you do less DPS the further they are, but you could do that just be reducing damage based on distance. It’s a very indirect method of accomplishing the objective.

Having randomized bullet spray, or bloom, or iron sights, or some such just means that you need to do some extra little thing to maximize your DPS, like burst fire depending on range, raise your sights, and it doesn’t add any significant extra layer of decision-making to the game, it just means that you need to do a slightly more complex motion to optimize your DPS, and RNG will come down and screw you over in varying amounts.

Is there a name for ‘reverse Bloom’? Weapons that gain accuracy from sustained fire? Can’t recall ever seeing it in a game.

I can’t think of any games that do that either, but it sounds like it could be cool, reward people for commitment. I still hate anything with randomized bullet spread though. Honestly, just increase the damage of the weapon over time. Don’t bother with bullet spread. There’s literally no good reason to add bullet spread to a game. There is no gameplay benefit achieved with bullet spread that cannot be achieved by deterministic means.

Do you think that AI in competitive vidya will ever get better at simulating playing against a human opponent? Is that even something that should be strived for? I’ve been getting into fighting games and am really realizing how entirely different it is playing against the CPU and other players.

That’s tough. I think it depends on the game. For fighting games, I think it’s hard to say if it will ever reach that point.

Sethbling did an interesting proof of concept here with neural networks being trained to copy his style of play, with the fitness criteria being how closely it matched the actions he’d take in a given scenario, so you end up with an AI that plays very similarly to a human player.

Of course this is a lot easier because Mario Kart is what I call an “Efficiency Race”. It’s a game about efficiency, more than a game about Rock Paper Scissors. There’s a tiny bit of RPS in the use of items and passing opponents, but not much.

I doubt the ability of AI to play like a human in that they can really play RPS in a complex game with multiple RPS loops and we the same way a human would. Or to be able to learn and adapt to a player over the course of a few games. I feel more like it would just have an expert policy and do the best thing each time, rather than play with the mentality of a human player. Also it’s hard to say if AI trained on imitating human behavior would really be capable of being “read” the same way as a human.

It would probably work pretty okay for FPS games, racing games, maybe RTS?

It’s hard to really say how this stuff will turn out, whether AI will develop patterns that can be read, and attempt to read your patterns in turn, or whether it could be programmed to play with a personality, like intentionally repeating the same move that the opponent loses to a lot.

Then again, there was Killer Instinct, which gave this a real shot with its Shadow AI System, which I haven’t personally tried, despite owning the game, but apparently it worked really well.

This isn’t a pure machine learning solution, a lot of the structure of the AI, and the tweaking of various variables was done by hand (essentially, the structure the AI learns by was designed by people who knew what it should look like), but still, it’s really impressive the types of results they got. It took very observant players to know all the tendencies they should copy for this. If we got this in more games, if these replaced the standard fighting game AI, then I think fighting games would be a lot better for it. I don’t know how hard that is, but this sounds from the talk like it was thrown into KI unceremoniously, as a cheap side-mode instead of a major developmental focus.

AI like this could theoretically be deployed to train new players. Imagine training an AI to do nothing but move forward and sweep at different ranges to teach players to punish on block? Or only throw fireballs and anti-air to teach them to beat that? Or only jump in to teach them to anti-air?

There’s a big potential here, but it’s hard to say if we’ll ever see a feature like this ever again.