I have never been able to get good at using analog sticks with FPSs. I'm so bad at it that if the game doesn't have auto-aim, I basically can't enjoy the game because I am sucking so badly.
The problem for me is that the analog sticks are velocity-based; ie. you don't get to control the position directly, you can only control the first derivative of the position, the velocity. That means instead of just moving the controller where you want to go, you have to move in the direction you want to go and hold it there for just the right amount of time.
I'm amazed that so many people can do this so well, because no matter how much time I play I still suck at it. It just feels so clumsy.
So the idea of being able to use a handheld controller that isn't velocity-based really appeals to me, though I do worry a bit that the active area is so small.
I can't believe that the trackball (very old tech might I say) hasn't replaced the analog sticks yet!
People have already modded controllers with a trackball, and it works great for the most part, bringing the best of both worlds together, yet Sony and Microsoft still insist on using analog sticks...
The reason is because it doesn't work. How do you walk forward continuously for a long stretch? Do you roll it to some position and keep it there? Do you keep rolling?
Yet you have two such analog sticks. But usually only one is used the way you describe, to keep walking continuously. The other is usually used to rotate a bit, to look around you. So it seems having an analog stick on the left of the controller and a track ball on the right would be nice.
The rollers for the tracking ball would collect a lot of dust. Could be one reason they opted out of it. Cleaning the rollers for the track ball every other week to get a smooth feeling would quickly become bothersome, contrary to the nice feeling of relaxation or thrill the console is supposed to give you.
Granted, it doesn't reach console levels of use on most days, but I only clean it once every month or so, and even then it's easy - Logitech has a simple system to pop the ball out and cleaning it is very easy and fast.
Using a trackball is definitely not a major chore.
I used to be a serious CS player and when I use analog sticks I just feel physically handicapped. I don't like that feeling. I've tried and tried but reached a plateau on my analog skill that just I can't get past. It takes the joy out of any console FPS game and some over-the-shoulder RPGs. I think this controller can make me enjoy those games again.
Been a while since I played much with a K+M, but with analog sticks I find that my aiming technique involves a mixture of both moving and aiming. Gives a bit more control and stops me from being a motionless target at any point.
Spot on. What I find amazing is how people use the nubs on Lenovo laptops. Those joysticks give you control of not the position, not the velocity, but the acceleration - the second derivative.
It's actually a useful thought to bring in. You can actually prove, mathematically, that controlling a system becomes "more difficult" in a few important ways when your input is to a derivative of the system rather than the system itself.
I was really skeptical about this and I'm getting more and more skeptical.
1. See how he always lift his right hand thumb to reproduce the same movement, sometimes three times, just to go through the screen (in Civ5).
2. Also the fact that buttons get mapped like a "pie chart" on the trackpads could have been a good idea, but we need buttons. We forget the luxury of having buttons with all those new tactile screens because we want more tactile surface, but for a controller we hold in our hands, we need some feedbacks bellow our fingers. This is really a bad idea...
3. I'll have to admit that it looks like a better alternative than joysticks for FPS games.
EDIT: As someone else pointed out, if you get two controllers : a valve one for FPS, RTS, games that works well with it, and a Xbox/PS4 controller for the rest, then it's a very good thing.
What exactly are you skeptical of? The skepticism previously was that this controller would be unusable. This video shows a person playing PC games with the controller rapidly and with skill.
I still want to see a platformer and a fighting game since the best mapping is a little unclear. (Though these a) won't be showstoppers and b) are likely to be workable.) Humans are adaptable and if this controller is reasonably usable in all the main gaming genres then very quickly you'll see people get really, really good at using it.
i play for leisure with a mouse on a pc and I'm not a pro or anything.
Yet, i'm way smoother, faster and accurate at the same time. I mean I can headshot pretty often within 50ms. This guy needs a full second on NON-MOVING targets in counterstrike. He would have zero chance against a mouse user. zero.
Remember that this guy played with the controller for a long time and his job is to make it look like as good as possible.
So yeah, the controller seems ABSOLUTELY fine for a controller (and probably more precise in some cases than a "regular" controller), but please, let's not fool ourselves, this is NOT a PC gaming experience - i.e. this is very far from the level of control a MOUSE provides.
It doesn't show a person playing PC games rapidly - it shows, for example, a common Civ5 action (moving cursor to a position in menu and clicking) being slow and awkward, requiring multiple separate finger movements just to push the cursor from middle of screen to side.
It's better than any other couch-options, but the controls in Civ still looked frustrating instead of fun.
He was explicit in that he was demonstrating 1:1 legacy mouse support, presumably there are other options for non-legacy games (and perhaps legacy games too?)
For example, my macbook track pad takes velocity into account. Something like that could make a big difference for covering large distances. I'm withholding judgment until I try one.
But then I play FPS with a trackball now, so it doesn't seem so alien to me.
He lifts his finger just as much in Portal play as he does in Civ. It's incredibly fast. The ergonomics of this thing make it pretty hard to say that it will be frustrating to track the cursor across the screen just because you have to lift your finger. Think about how you use a trackball. This is basically a trackpad device with the thumb ergonomics of a trackball.
The last time I touched (or saw) a trackball was a dozen years ago, so it's hard to say - but on any acceptable trackpad its dpi should allow to move across the screen in a single motion, shouldn't it?
I.e., if I move my finger across the whole steam controller right pad, then the sensitivity should be high enough to move across the whole screen in that motion while still allowing enough accuracy to click where I want.
If, for example, some pad was an inch-wide and with 1000 dpi resolution, then it couldn't ever be configured to cover a 2000 pixel wide screen that way - we don't know the specs of the controller elements yet.
With some acceleration it could - finer control with small movements, but quick movements provide more travel. Not sure how well this would work, but I like Apple's implementation on their trackpad (I realise I may be in the minority here!).
It's an example of a control scheme the controller drivers allow you to come up with in order to map to a game's expected mouse-and-keyboard inputs. If a game developer wants to explicitly support the controller, then even better schemes can be designed.
> The skepticism previously was that this controller would be unusable.
Who really thought Valve making a controller would make an unusable one? This was never a concern. The concern here is that it might work well with FPS, RTS etc... (or good enough)... but might not work as good as button based controllers for other genres.
The controller is not competing with a mouse! Valve is trying to replace the standard console controller and it looks like a large improvement over the thumbstick.
If by ratcheting you mean the repeated movements to look around, that is very similar to moving a mouse with a low sensitivity settings.
Of course this is meant to compete with a mouse. It's intended to let people play mice based games, such as FPS and RTS with a controller.
It's intended to take the gaming pc (and by, extension steam) out of the man-cave and into the living room to compete with consoles. This requires providing people with an ability for people to play FPS and RTS decently with a controller, rather than a mouse, or they are going to stick with their man caves and mice.
It's not competing with a mouse, it's competing with "a mouse and solitude in a man cave", since these are coupled together due to the fact a mouse can't be used in the living room practically. In other words, if this is close enough to the mousing experience to make the net experience of playing PC games in the living room better than the same experience with a mouse in solitude, it will be a hit.
2. It looks like the pads are "pressable", so whilst they can react to touches, you can configure them to only react to presses. Not quite the same as buttons, but it might be good enough.
You might be interested in looking at Tommy Refenes thoughts on his testing of the steam controller prototype. He discusses using it to play his difficult platformer Super Meat Boy:
If you were to ask me to choose between Steam Controller and a 360
controller, I would choose 360. Don’t take that as slight to the controller
though because it’s more about the comfort of familiarity over
functionality. I would choose a 360 controller because I have several
thousand hours experience using it, however if tomorrow all game controllers
were wiped off the earth and the only option was the Steam Controller, I
don’t think this would be a bad thing. In fact, I don’t think gaming would
miss a beat.
1. It's like having your mouse on a tiny mousemat. As someone who once had a tiny mousemat, it works great for turn based games but anything real time is no fun at all. You can simulate this feeling right now by using a round beermat.
2. Something as simple as putting one tiny raised dot at each of the 4 main positions would make a huge difference.
I couldn't watch the entire video because of my connection but note that he did say it was in "legacy" mode, i.e. the game has not been modified at all for the controller and the controller has likely not been changed from whatever it's default configuration is. If you use a mouse or trackpad on their default configuration things will generally be difficult if they're not what you're used to. I've gotten so accustomed to the acceleration on my trackpad that there are many things I can do on it without thinking that I would struggle with on my trackball or laser mouse. I agree that this particular set up as they've shown looks very awkward. What could make it great though– assuming no modification to the game– is to treat it like an inverted ThinkPad trackpoint. If it feels just right and the acceleration is either customizable or perfectly tuned, I think it could be a lot of fun and avoid the trackpad/mouspad-too-small issue.
> 2
I imagine this will come up during the beta and probably result in something close to what you suggest. Either a dot like on F/J keys or a small dimple.
Haptics don't allow for many touch instincts, like edge-detection and depth of depression.
It's better than no acknowledgement, but it's dramatically different.
As an exaggerated counterpoint: imagine operating a buttonless, flat "haptic" vibrating keyboard in a moving vehicle, and contrast with a conveniental keyboard.
I agree the Civ5 demo didn't look good, but did you watch all the way through to the Papers, Please demo? I think using both pads to control the mouse cursor could end up being really good. In fact, it could be even better than a real mouse in some ways, since you sometimes have to lift a real mouse but you shouldn't ever have to lift both thumbs at once.
> I think using both pads to control the mouse cursor could end up being really good.
I don't know. I've tried using two hands on a big trackpad once and it just gets confusing. Maybe with more practice but I felt like it was more a brain confused problem than a material one.
Certainly there would be a learning curve, but in a well-designed game learning the input device is part of the fun. I think there's nothing about controlling a mouse pointer with both thumbs that would make it inherently less learnable than e.g. traditional gamepad FPS controls (which are anything but intuitive).
True, it look playable compared to what a joystick could offer. The real problem is going to be with platformers, fight games, third person games who relies heavily on buttons.
Its just a matter of using the right tool for the job. Plug in an xbox 360 controller for games where it makes the most sense. Use this steam controller for any game which you'd normally use mouse and keyboard for but want to play from the couch.
It would be a nice tweak if when you put your thumb all the way to the edge you would get decaying momentum on the cursor while you moved your thumb back into place to swipe the mouse forward again. And if you didn't go all the way to the edge it would stay as it was in the demo.. with some haptic feedback to let you tell the difference. I guess it would ultimately make it feel like having a thumb trackball.
There is no reason a touch pad can't be as accurate and natural as at least a trackball, which could already probably let you play Civ from your couch.. or quake
but we need buttons. We forget the luxury of having buttons with all those new tactile screens because we want more tactile surface, but for a controller we hold in our hands, we need some feedbacks bellow our fingers. This is really a bad idea...
Until I actually have the controller in my hand, I can't make that assumption at all.
I feel it was a somewhat poor demonstration of the controller. Each demo was set up for a game without, or as if the game did not have, controller support. Games with controller support would make it feel more like a traditional console controller. No idea why did not show that aspect of it.
I'm kinda baffled by all the people in these threads talking about platformers or fighting games. That's completely missing the point, which is very obviously not to replace traditional gamepads, but to complement them.
Even talking about competitive FPSes or DOTA sort of gets it wrong. Among the few games Valve prominently displayed in their announcement were Europa Universalis and Football Manager, turn-based mouse-driven games. That is where this controller should shine. It allows you to sit back and relax while playing thoughtful games that would be impossible with the typical gamepad.
Valve's main objective was probably to replicate the keyboard/mouse combo rather than improve or alter the standard joystick gamepad (like the X360 Controller). In the end, it looks like a controller halfway between these 2 types of input.
The main problem with a controller-based "mouse" (and it shows in the video) is that you control the cursor with only your thumb rather than your full hand gently set down on your desktop. I guess we'll never obtain a perfect alternative to the mouse. And the Steam controller will probably leave both joystick users and mouse users somehow frustrated because it tries to be both at the same time. But considering the variety of Steam's library, it was probably Valve's only solution. Anyway, like this comment, everything still remains speculation until you've tried it yourself (for a reasonable amount of time).
There would be two disadvantages compared to traditional mouse-and-keyboard: 1) a low number of total possible bindings, and 2) a lower range of possible movements compared to a mouse.
1) is straightforward, if you don't have enough input buttons you're out of luck. A problem for games like SC2 and WoW.
2) this problem forces you to use acceleration for input, which is commonly accepted to ruin precision (reproducibility) because you lose the easy mapping between inches and pixels.
I'll use Team Fortress 2 (an FPS) as an example for this. First, if 180 degrees takes more than 1 swipe, the controls are effectively unusable. The demonstration indicates this may be the case.
You need to be able to flip 180 degrees to check your back, as well as make other large and fast movements like those required by rocket jumping. At the same time, you also need the ability to make very small aim adjustments. It's guaranteed that the trackpad will be weak somewhere along this variety of movements.
It's unsurprising that it will be a poor replacement for mouse and keyboard, but existing controllers have demonstrated that people are fine with sacrificing resolution of controls for other advantages like relative simplicity and easily playing with friends.
In regards to the 180 degree spin, this demonstration shows exactly the opposite of what you say. At about 1:34 (elevator in portal 2) he performs a 180 with a single swipe from left to right.
Having played all StarCraft games so far, somewhat well, I can get by with maybe 8 buttons as Protoss: A, H, E, P, B, V, G, Z. Other races might need other buttons, but 8 is reasonable. If I remember correctly, the Steam Controller has 16 buttons. The other functionality is 1 click away and is (usually) not hat time sensitive, if you are not from Korea :-) obviously it's not for competitive players.
Also, flipping the camera 180 degrees can be achieved with a button press. I think I remember Need For Speed having this feature, though they didn't allow you to flip the car, just the camera. Anyway, I think pressing a button to flip 180 is an acceptable solution given the constraints of the controller.
Yep, I just wanted to explain why it isn't :) Having to click rather than press a key is a big downside, but if I remember correctly SC2 already has a keybind mode that limits you to a subset of the full qwerty keyboard. World of Warcraft is a good example of a PC game that uses a huge swathe of keyboard, players often have 30-60 unique keybinds.
My point about the 180 was that you need a wide range of different motions. Sometimes it'll be 170, sometimes 145, etc, sometimes only 2.4 degrees, and if a mouse user uses 1 swipe when you need 2 or 3 you're toast. A common tactic for Scouts in TF2 is to stay in the air above your head, which from experience is basically impossible to beat on a trackpad.
I'm looking forward to the Steambox/Steam controller, if only because I have quite a few Steam games accumulated and the biggest reason for me to avoid consoles is that I'd be forced to buy multiple copies of games.
"I guess we'll never obtain a perfect alternative to the mouse"
Because Valve's first effort doesn't completely replace a mouse? That seems a bit defeatist.
"control the cursor with only your thumb rather than your full hand "
Control with the thumb vs hand isn't the problem so much as that thumbs are not pointy like mouse lasers. Valve should look at adding a tiny fingertip digitizer.
Of course they are still lacking proper text entry and sitting out motion input (i.e. the big innovation in the last 10 years), voice input (the big innovation over the last 5 years) and telepresence isn't going to be a core feature unlike all the competition. And of course they are going to be the most expensive option with the least bang for the buck.
> I guess we'll never obtain a perfect alternative to the mouse.
Is that really what they're going for?
I don't think you should view the Steam Controller as the controller to end all controllers. From that point of view, I'm sure it will fail, and people will be disappointed.
Instead, it's a controller that makes PC games playable from the couch. It won't be quite as good as a mouse and keyboard for games that were designed around those. It probably won't be quite as good as a standard console controller for games that were designed around those (you don't have the physical buttons to roll your fingers over to hit particular combos quickly). But it should make both types of games playable on a single controller.
And once it's out, who knows. Maybe people will be able to design games specifically for this type of controller, and there will be some games that are best played on it.
That was my thought as well. Basically the mouse has a huge range of motion and you can very aggressively move it and execute commands with buttons while moving. That doesn't work on track pads or presumably on this demo yet. Moves that would be one mouse stroke from left corner to right corner, become three or four "strokes" of the touch pad. And if you make the touch pad sensitive that this is a single stroke, then fine movement is compromised.
I keep hoping for something like the nunchuck controller on the Wii where the nunchuk can keep track of its absolute coordinates and you then can do fine movement in the center of the box and large movement on the outer edges.
Looks like what I expected. When it comes to FPS games, it's difficult to know his skill level, but it's probably not something you'd use competitively. Many times better than an analog stick though, so I'd use it for casual gaming.
Yea, I can't remember where I saw it but i think they tried pitting top COD players from console vs PC and even with auto aim assist on PC player destroyed console players. Mouse is just too accurate of a pointing device compared to joysticks (physical or virtual).
It also reminds me about how CS:GO was originally supposed to allow true cross-platform play (PC against console) but Valve scrapped that support because they couldn't work out a reasonable way to patch on consoles due to "patch scheduling" on PS3.
The demonstrated FPS aim-and-shoot was very slow, it's not clear if it's because they couldn't do it better because of the controller, or didn't do it faster because they did everything slowly to be clearly visible how it happens.
A full speed try-to-do-the-best-you-can FPS aiming demo would be welcome.
I suspect part of it is familiarity and practice, too -- I'm very quick and precise with the mouse, but I use it all day long, and have been gaming with mice for years. It looks like the touchpad resolution is quite good, and it's not "jumpy" like some touchpads tend to be, so it might be that with enough practice, this could be as usable as a KB+M setup.
They're not trying to emulate mouse + keyboard precision neither. I don't think you could have played like that with a Xbox/PS controller (I know I couldn't have, it would have looked more smooth but not as precise for headshots).
For me it seems like FPS with this controller will be somewhere between Xbox/PS3 controller and mouse+keyboard.
I'd like to see a video that I can compare with, for example, controlling the same game on a good laptop touchpad - the current video is weak even compared to that; if they can do better, then they should make a video that proves it.
Basically, give it to a professional players for a week, then record them playing at his best. If they have any chance against another team, then I'd consider that a success.
They wouldn't, one because it's not precise enough, and two because there's no way a week is going to compare to a lifetime, and thousands of hours in a singular game, of using a mouse.
I was surprised to see the negative comments on the article page. I'm personally excited for it. As I think it will open up "couch" gaming to more genres and it looks like a pretty decent controller when compared against other controllers. I don't think it will be a true mouse + keyboard replacement but will do great for big screen HTPC / SteamMachine type setups.
Basically we still don't know enough to know how good it's going to be. Having used the next gen controllers at PAX, it's going to very hard to be as good at they are at being a controller.
Valve's controller does seem to be a good 1/2 step for mouse/keyboard replacement, eg very good accuracy, but maybe not the best of all worlds. Also able to do controller based games but not quite as good as a real controller.
to me it's a sign that they hope to help developers support console versions by making a control system that will work well for games designed for mouse and keyboard, if not as well as a mouse and keyboard. But to be honest, standard controllers on games not designed for them are quite bad. Moving a "mouse" pointer replacement with a controller is not good. So I think it will be good in that regard. It'll also be better than a keyboard and have MUCH better support than a logitech controller for games that want controllers.
So I see it as a good middle ground. And I think this will allow game devs to support both more easily. It's not the tactic I would take to make games work on both, but I think it's a good approach even if it isn't going to be the best of either world.
However, if it IS as good as the next gen console controllers and it does better than them at mouse/keyboard then it's a big win.
With as few moving parts as they typically have, console controllers already experience durability issues for less... hygienic, players. With trackballs that issue is compounded, even if you make them easily removable so you can clean them (just having to clean them at all would be more of a hassle than people are use to).
Logitech's optical trackballs are pretty bulletproof. No moving parts; they use a special dot pattern to detect how the trackball's being moved around.
Likely for the same reason that mouse balls went out of favor -- they get gummed up with dust and gunk from your fingers, and are a PITA to maintain and keep clean.
It would be nice if a single two-handed controller or even some type of two element controller (ala. wii) could somehow approach the accuracy of a mouse and a keyboard.
This clearly does not appear to be at the same accuracy and speed level as the combination of a mouse and keyboard. There may be a console controller that can approach, equal or exceed the speed and accuracy of a mouse and keyboard, but this doesn't seem to be it.
Addition: I suspect that every steambox will have USB ports available for whatever controller, keyboard and mouse floats your boat.
Looks like a great alternative to the Xbox PC controller, but like everyone has been thinking, this is in no way a serious replacement for a mouse. Especially if you're a competitive gamer.
The biggest problem is not the controller itself but the expectation that people that use this will be able to compete at all with those using a keyboard & mouse. The only reason FPS games currently work on consoles is because everyone is essentially equally handicapped. Even a casual player using a mouse dominates versus game pads.
Maybe his aiming in a fps is not up to competitive level of quake/cs, but this is definitely a step up from controllers for these types of games.
Hopefully, this thing will bring the skill of being both fast and precise to consoles. It will be good for consoles, and good for gaming in general.
I wanna aim with my finger and with zero latency. Heck it's probably possible to hack something like that already in fact. Not sure if we'd be as good as with the mouse as you will lose awareness (hand in front of you = can't see as well)
Though it's heavily touch-driven, the new Steam controller does have several physical buttons, as well as tactile feedback (ridges) on the circular touch controls. Not having to look at the screen to use the controls radically changes the experience.
The problem for me is that the analog sticks are velocity-based; ie. you don't get to control the position directly, you can only control the first derivative of the position, the velocity. That means instead of just moving the controller where you want to go, you have to move in the direction you want to go and hold it there for just the right amount of time.
I'm amazed that so many people can do this so well, because no matter how much time I play I still suck at it. It just feels so clumsy.
So the idea of being able to use a handheld controller that isn't velocity-based really appeals to me, though I do worry a bit that the active area is so small.