[ux-app]

It's always a tradeoff between responsive movement and realistic movement. RDR2 has very realistic animations, with the tradeoff being slightly "floaty" controls.

Personally I prefer snappy movement i.e. when I press left the screen character moves left immediately. A more realistic looking animation system introduces a delay while you wait for the feet animation to "catch up" to player input.

[jordanthoms]

I guess the problem there is pretty fundamental - in reality you'd be tensing muscles and shifting your weight etc before the snappy movement, but the game only knows that you want to move when you move the stick or push a button - so it either has to show that realistic motion after your button press and introduce latency, or sacrifice the realism in the animations.

[lukeholder]

Correct, this is an issue in nba2k but it’s become a fundamental part of the game. As they made movements more realistic they introduced a delay. So when you try and get around you opponent you direct half a second before you move, but your defender also has to move their player - not in reaction to the screen but also in anticipation of your move in order to defend you. I actually think this is more like real life which makes the game better, but a lot less frantic than basketball arcade games of the past.

[josephg]

Guild wars 2 does something similar. When you’re running around on your own, the game responds instantly. But all of the in-game mounts take a moment to react to your steering. Responsiveness, jump height, horizontal speed and turning radiuses all differ massively depending on the mount you’re using. As a result, long distance navigation is a complex puzzle requiring you to choose a good mount and a good route at the right time and manage your energy bars and cooldowns. Do you try to hop up this ledge with the griffin you’re already on, or take the time to swap to the springer and clear it in a big charged bounce? Was there a better way around this ledge? It’s shockingly fun.

[mitthrowaway2]

This is also why 2D jump platformers like Megaman have triangular jump trajectories instead of parabolic trajectories. For snappy controls you leave the ground at the instant of the downpress and peak when you let go of the button. As the game can't know how high you intended to jump at the time the jump begins, the trajectory can't be parabolic. (At least, not on the way up.)

You could instead have the button downpress "charge up" energy and then begin the jump when the button press ends, which would allow for more realism, but also introduce a delay.

[ux-app]

>I guess the problem there is pretty fundamental

yes absolutely, you can't have both realism and snappy response times.

[_kulang]

Considering you walk around and don’t have a problem with it, I’d wager that putting a more realistic walk model into a game and changing the control scheme to be a bit more natural would lead to a learnable system for control that would achieve both goals. Say ZMP

[itishappy]

> Considering you walk around and don’t have a problem with it...

I think humans are worse than this than you might think. Give it a try! Walk around and try to change directions on the fly, maybe have a friend shout suggestions to you. I can make maybe 2 or 3 adjustments per second, which gives about 400ms response time. That's about where RDR2 sits, and players criticized it for being unresponsive.

[JohnBooty]

Tangential but a lot of sports drills are along these lines.

A great tennis drill is when your coach hits the ball to you and randomly yells "LEFT!" "RIGHT!" or "MIDDLE!" after you've begun your swing. Then you have to hit the ball in that direction. Helps hone your reaction time and helps you to have a "neutral" swing that doesn't telegraph your intentions to an opponent.

One can imagine variations in many sports.

[Aeolun]

But if you take 400ms to respond, and the game takes 400ms to respond after that, the actual response delay is 800ms, which is quite a lot.

[Retric]

People take ~150 ms to react and press a key, the ~400ms in running/walking is because you need to first shift your feet into a new orientation and let gravity change your momentum. The physical distance signals need to travel down the length of your spine, and the need to move your feet is larger distance before any change can occur etc.

Watch a sprinting football player dodge. Their feet go in the opposite direction as they want to move the same way you move an inverted pendulum. If you want the top of an inverted pendulum to move left you move your hand to the right. It still looks very fast because you don’t see the initial ~200ms delay between deciding to doge and the point when their feet start to respond.

Most martial arts will teach their own type of footwork optimized for the style, but it’s common to use a shuffling motion which keeps people’s feet close to the ground. It allows for a more rapid change in direction but is slower and less effective than normal walking if you actually want to get somewhere. Fencing and Kendo want more mobility where wrestling and Judo wants more stability etc.

[hoseja]

Likely not solvable until we have a BCI working.

[notjoemama]

I played RDR2 and was very happy to walk away from the molasses like movement once finished. It was realistic in both its behavior and cadence and I can emphatically say, give me fantasy (better/faster) movement in a playable fantasy world. Imagine waiting for a real time washer and dryer cycle in a video game because it’s “realism”. There are limits…

[simbolit]

Have you played "Death Stranding"?

There you 'actively' walk and 'manually' keep balance. It's an interesting experience, but it makes walking a conscious act, it becomes something you do.

Arguably that is less realistic than just moving an analog stick, for most people walking is just telling your body to move in direction x.

[beenBoutIT]

Amazing game. Goes in the rare category of truly singular games that have no peer. Manhunt(PS2 version) is another.

[xeromal]

I really enjoyed that game but I made the mistake of trying to finish my highway before beating what I think was the final boss so I burned out and never finished it.

[firecall]

Of course there is already measurable lag from when your brain tells your hand and fingers to move, to when they actually move!

An interesting personal anecdote I have:

I damaged my back with bulging disc, which caused horrific sciatica nerve pain.

When I was able to walk again, I had some nerve damage.

This meant I had lag in my left leg!

I’d tell the leg to move instinctively when walking and there would be a delay. The entire walking movement of the leg was present, but just with a noticeable lag!

It was weird!

Eventually it healed fully, or I adjusted. Not sure which! :-)

[GreymanTheGrey]

Complete tangent, but how did you treat your injury? Coming back from bulging discs and sciatica is rare, to my understanding.

[firecall]

I was fortunate in that I was very fit and generally physically capable with access to excellent medical care.

I had a couple of MRIs to see what was going on.

Essentially I’ve got degenerative disc disease.

My specialist was the head of the Spinal Care unit at the Royal Adelaide Hospital.

So I trusted his advice.

He said that surgery could very well make it worse, and that over time it would likely begin to heal itself.

I had a couple of steroid injections and took mild painkillers and anti inflammatories to manage the pain for a year or so.

The initial incident was well over five years ago.

Ive had reviews since, and the advice is still the same.

Surgery is a last resort. Just keep mobile and don’t do anything to make it worse.

The MRIs show that the disc that is jamming into my nerve has slowly begun to reduce.

So I force myself to remain mobile by continuing to walk everywhere, and have a standup desk for work as I literally couldn’t sit down for months!

I pretty much now never lift anything.

Your body cant process pain and motion at the same time. Which is a simplistic explanation. So walking is an excellent path to recovery.

Before the injury I’d back squat 130kilos, and strict press 95, could run a marathon and had a resting heart rate of 42bpm.

It flares up from time to time, and is never truely better.

I’m never pain free, I just learn to ignore the pain and keep moving!

I could talk for hours about my journey LOL

[askonomm]

Yeah a friend of mine had this and he had to get surgery to not lose his left leg to permanent numbness due to the disc pressing on the nerves. I know one other person with the exact same issue and solution, and am also under the impression that without surgery only bad things happen.

[ncallaway]

> Considering you walk around and don’t have a problem with it

Your brain is hiding a lot from you.

[kylebenzle]

I can't see how zero moment point (ZMP) is relevant here? I think what you are saying is why don't we use the same algorithms to control game characters as robots? If its a robot or game character the issue is the same. You can try and predict the next movement but if you are wrong you again have the same problem and the response time suffers even more.

https://en.wikipedia.org/wiki/Zero_moment_point

[4hg4ufxhy]

It would still feel unresponsive in comparison. Just like normal physics would feel if you got a taste of breaking them in real life.

[ux-app]

>changing the control scheme

Controllers use analog input for the sticks. I guess you could create a system which uses a dead zone where the player "signals" their intention by partially moving the stick in their desired direction.

This would be incredibly cumbersome though, and the payoff would not be worth it.

[Arrath]

Full body control input it is!

Like those arcade shooters where you duck and lean to dodge bullets, or a full fledged whole-body VR setup on an infinite omni-directional treadmill.

[viraptor]

There's a good workaround for this presented some time ago at GDC (although it was still simplified). In Overgrowth, if I remember correctly, your body core is responsive, even if your legs have to catch up. You lean, because you'll stay running in that direction. Which is actually closer to the realistic movement.

[thomastjeffery]

To elaborate on this, there is a third aspect being glossed over: the correctness of the animation itself.

They could alter the animation such that feet don't slide across the ground and keep responsive movement. The result would be a worse quality animation, because the movement of the legs would not appear to be pushing the rest of the body around. Instead, it would look more like the feet are following the rest of the body retroactively, while holding onto the ground.

A good example of this is Factorio's spidertron. When the spidertron moves, the legs follow with a walking motion that perfectly tracks the ground below. In this case, it's a great-looking tradeoff, probably because there are so many legs, and not much animation done to the body itself.

[fennecfoxy]

As mentioned in other comments, I think a big thing is the difference between the older way of doing it ie you have a character object that you move along a vector and the animation is supplementary to this to make it look like they're walking, vs animating the walk and then having the animation/movement of the character itself actually move the character through the world (which almost nobody seems to do).

[thomastjeffery]

The third option is inverse kinematics: you move the world across the character, and the character reacts by moving its feet to positions that make sense.

[fennecfoxy]

Oh yeah for sure, but I'm thinking more for the realism of the simulation, we don't move like that irl so it would make sense to simulate roughly how we do when it comes to games.

I guess it's all trade offs at the end of the day, dev effort vs game style vs priorities.

[darzu]

I’ve always preferred the Wolfire approach[0]: animation should “do no harm” to player control, it’s only there to add flavor, never at the cost of game responsivity

[0] https://youtu.be/LNidsMesxSE?si=W7xnQfXt5ulfHklR

[ordinaryradical]

His proposition of a Hippocratic oath for animation is a beautiful and succinct way to frame a game design thesis. Great talk.

[larschdk]

I think there is also a difference between realistic and believable. The original Half-Life had IMHO more believable leg/feet movement. In HL, the feet were much more "stuck to the ground" compared to most modern games where they just slide around.

It appears much more believable to me than the character pathing demo, where the character moonwalks 1/2 the time. The entire pedal structure is extremely stiff compared to how humans move (feet, thigh and torso can all turn nearly 90 degrees, but they hardly turn at all in the demo). The other demos are better, but their bodies still appear stiff, like they are suffering from hernia.

[iamcreasy]

> Personally I prefer snappy movement

I do too, but also think it greatly depends on the game. For example, Hollow Knight designed to have snappy response to player input from the start and I loved it. In RDR2, I find the floaty behaviour adds another layer of realism.

[formerly_proven]

> It's always a tradeoff between responsive movement and realistic movement. RDR2 has very realistic animations, with the tradeoff being slightly "floaty" controls.

In GTA V characters have two different animation modes, the realistic one based on Euphoria when using the third-person camera and the "do what the darned keys are saying right now" when using first-person. Always seemed like a sensible compromise to me, though first-person movement is particularly snappy and direct, more so than pretty much any other FPS, which typically still have some inertia.

[fennecfoxy]

I don't see how that enables foot sliding unless it's a lazy solution tho? A no foot-sliding solution can be snappy, you just have to increase animation speed so the character is "ready" to execute the players input faster.

I think the real issue is the difference between moving the entire character as an object, with the walking animation being supplementary to that vs the walking animation being central to the character whose object moves because of the animation.

But for a game like this, arcadey instant direction change type movement doesn't really seem warranted either.

[henriquecm8]

If you look at npc locomotion, there's no foot slide, and for a character there is, but the character feet are out of screen most of the time when walking. I've played the whole game and never noticed this.