Valve replaces Steam Controller touchscreen with new analog face buttons | Hacker News Reader

Valve replaces Steam Controller touchscreen with new analog face buttons

arstechnica.com

88 pointsaudiophilip_66 comments

I love the fact that they responded to feedback and adjusted their design accordingly. While the original concept of a programmable touch screen interface in the center of the controller was pretty cool, in practice it was unwieldy.

Controllers are meant to be interacted with by touch. You should never have to look at it. A touch screen interface is completely useless for this type of interaction.

The entire point of the clickable touchscreen, as I recall, was that resting your finger on it would bring up an overlay on the monitor, mirroring what was shown on the touchscreen.

You would then move your finger to hover over what you wanted to select, then click down the touchscreen. All without ever having to look at the actual touchscreen, since it was all mirrored up on the monitor.

That use case would be easily covered by the touchpads and buttons though - your argument for a touchscreen is so the user wouldn't ever have to look at it; so why does it ever need to emit an image?

I imagine they would replace the touchscreen with a touchpad like they had on the prototypes sent to the 300 testers.

Correct. They then realized that they could take the screen part of it out, and then they realized that they already have a touchpad on the left and right sides.

The team went over the full story in their SteamDevDays talk: https://www.youtube.com/watch?v=UfN5WK7OzU8

On the touchscreen specifically:

https://www.youtube.com/watch?v=UfN5WK7OzU8&t=11m12s

that sounds weird/pointless. i wonder if they tried using the touchscreen as an absolute-position locator for the cursor. that would have been legit, assuming the screen had that kind of fidelity in the first place... which it probably didn't.

You finger doesn't have that kind of fidelity

there is always the center of the contact area, which is a single point. it would certainly take getting used to though, and the data would probably have to be run through some kind of annealing

This is the reason I opted not to get Lenovo's new X1 Carbon ultrabook; it replaces the function keys with a programmable touchscreen like that, which I think inhibits usability.

http://shop.lenovo.com/us/en/laptops/thinkpad/x-series/x1-ca...

Damn, I like everything else about that product too.

What a shame.

My hunch is that they removed it because of cost, not usability. That touchscreen seems like an expensive part to put in a $30-50 controller. This is all just my guess, though.

Now that the original d-pad patent has expired, a controller without a good d-pad is a non-started IMHO. Analog controls are great, and this new design looks cool, but for certain uses, a nintendo style d-pad has basically proven to be the best 8-direction digital interface ever conceived...and it's amazing how much the experience suffers when it's just a little bit off.

Looks like they want their controller to be fully symmetric so it could be reprogrammed for left-handed users without making compromises.

What makes a controller right handed?

Asymmetry. It's not an accident that movement is on the left and aiming is on the right, both on controllers and PC. On PC, though, you can move the mouse to the left side of the keyboard and rebind your movement keys if you want to.

I've never heard a left-handed complaining about the traditional gamepad layout.

now they will.

And yet, many popular console FPSes have optional "southpaw" layouts that switch the left and right sticks.

Kid Icarus on 3ds was completely unplayable as a lefty.

any have stats of the left handed population in the world?

11% overall.

http://en.wikipedia.org/wiki/Handedness

Quote: "Globally, roughly 12% of men and 10% of women are left-handed."

I came here to post this, but with "Sega" instead of "Nintendo". The Genesis/Saturn D-pad is still king, as far as I'm concerned. Easier to roll the pad than a Nintendo one and it isn't a cross-shaped cutout from the plastic so diagonal presses are more comfortable.

I guess it is subjective but for me, being a cross-shaped cutout is precisely what makes the Nintendo D-pads superior. It means that you have obvious tactile feedback to tell "diagonal" from "non-diagonal" directions. In Sega-like D-pads, I often find myself pressing a diagonal when I wish to press a non-diagonal, and vice versa; that's practically impossible in Nintendo controllers.

I've heard that before, but Sega D-pads have a raised cross with rockers in the corners. I grew up with both, which might be part of it, but I'm not sure how you'd press a diagonal accidentally.

This isn't for games that would use a d-pad. That's what an Xbox 360 controller is for.

This is for games that would otherwise require a mouse and keyboard.

Shouldn't the goal be to have a controller that's good for most games, though? Why have several lying around?

The goal of this controller is to be good with most Steam games, which have been designed for computers. Traditionally, console controllers predate the games, the Steam box is trying to be a "console" that goes the other way.

Basically, imagine trying to create a "controller" that works well with Civilization, and you're on the right track. (That isn't the only example, but it's a nice, concrete one to try to think about.)

The xbox 360 controller d-pad is pretty bad.

There are millions of great controllers out there with good D-pads you can buy. There are a tiny hadnful that have a decent pointing-device (eg. the Wiimote), and none that come even close to comparing to a mouse.

That's why this pad is important. Anything to do with d-pads is just going to be gravy.

> There are millions of great controllers out there with good D-pads you can buy.

I'm open to suggestions. The Xbox 360 controller I have hooked to my PC is about the best I can find, and the d-pad on it is pretty terrible, but just on this side of usable.

I have drawer full of other old controllers as well because the d-pads were virtually unusable.

The original NES d-pad, which Nintendo has wisely brought forward on all their consoles, is simply the best 8-directional digital control interface ever designed -- it's so good it absolutely ended the age of joysticks as a home control interface.

A few comments here mention the Sega pads which are a very close second IMHO.

About the only device I've found with a reasonable d-pad alternative is my GP2x Wiz, which is weird at first, but ends up being a pretty good control interface. Even Sony's d-pad alternatives, which are better than the 360s, barely compare.

Since it's now basically free to use, there's very little reason for everybody to not just put the Nintendo d-pad on their stuff.

> none that come even close to comparing to a mouse.

True, but a mouse is a terrible interface for platformers, fighters, and basically the entire genre of gaming genres from pre-N64 pre-dual shock times.

That's why, if you're going to offer a controller with a new style of analog interface intended to replace a mouse, the complementary 8-way digital parts better be rock solid awesome.

The OpenPandora Dpad is excellent, however.

Get the Saturn USB re-release from about 2006. They are white with blue buttons (ugly, I know -- they should have kept the black and grey). They have a "Play Sega" sticker in the middle. Do NOT get the "saturn-style" knock-offs.

I got a few new off of eBay for about $6 a piece, they are great for classic / MAME / shmups.

Saturn D pad is better than nintendo D pad.

Don't forget that the two touchpads can still be configured to work as dpad + buttons.

They throw around the word 'analog' a lot, but I'm not seeing how that applies here. Typically, buttons are a digital affair. Do these buttons have continuous states, like a trigger? The article never explains this.

It usually means pressure-sensitive buttons. For example, a car driving game might use them to give the player more control over throttle and braking.

IIRC the PS2's dual shock 2 had pressure sensitive buttons. It wasn't a feature used by a lot of games. Nowadays only the shoulder triggers are analog I think.

It's a shame, really. For example, Metal Gear Solid 2 used this to great effect. If I recall correctly pressing square on the control would raise your weapon but not automatically fire it until you actuated the button all the way? It was finicky, but very cool.

I hated how they worked. The buttons did not have enough travel in them to be comfortable to use in this way plus the way the thumb sits on the buttons it is not a very comfortable way to control pressure like with your fingers on the shoulder triggers. I am glad the Dual Shock 4 got rid of them and just has two decent shoulder triggers.

For automatic weapons, yes, but it was even neater for handguns. Square raised the weapon, releasing quickly fired, easing off slowly eased the hammer back. It was a really neat trick.

I always thought more games should have used the gamecube shoulder buttons' feature. Analog and then a click before the bottom which allows it to be used for subtle followed by accurate/sudden actions. I can only think of two games that used it... Mario Sunshine (button was probably included just for it) and Eternal Darkness. Does anyone recall if a shooter did this for aiming on gamecube? I didn't play many shooters on there.

ps3 does too. on GT5 amount of pressure controls amount of throttle.

I'm imagining a big reason for "pressure sensitive" is that it has no thumbstick. Obviously the intent is that the left touchpad is to substitute for a thumbstick, but some stubborn gamers will insist on using the d-pad-ish buttons. So say you're playing a game designed for a thumbstick on the left button-pad, and the game requires a light-touch to "walk" like assissin's creed or Mario 64? The pressure-sensitive buttons provide that.

They probably just mean "physical". I've noticed "Analog" gets abused in that way a lot.

No, they mean "analog," as in the Xbox 1 and PS2 controllers. You get different results when pressing a button lightly or firmly.

Pressure sensitive, as a peer post says.

The original Xbox buttons were also pressure sensitive. Developers did not use this feature much, and it was eliminated in the 360 controller. Only a couple of the original Xbox games couldn't be ported to the 360 because of the lack of analog buttons.

What's a use case for having these on a Steam controller? Most PCs don't have analog buttons so I don't see why a controller made to play PC games would have them, though I guess the feature doesn't hurt.

Yeah, the touchscreen seemed like a solution looking for a problem. The Wii-U pretty-well demonstrated that gamers aren't really that interested in having a screen nested in their controller, and it's already got a good pointing-device in the thumbpads. The only use for the touchscreen is private information display... and I think the SteamBox is going to be predominantly a single-player console so I doubt there would be enough usage of this feature to justify this.

I mean, Nintendo has demonstrated a handful of games that use private information for exciting effect, but not enough to justify the cost of a screen in your gamepad. Simple good old-fashioned reliable face-buttons will get far more use.

It's remarkable to witness a company that responds to criticism instead of powering through it via reeducation campaigns and hype.

Reeducation and hype is the middle course. Punishment, banning and denial is the far end of the spectrum. Recent situations I can think of include forum bannings, post deletions, account closings and statements denying there is a problem.

This may not be as sexy/novel as the previous design, but it looks supremely more functional. I'm glad to see them using empirical results from testing rather than trying to push some grand vision on the hapless gamers.

The people who use these controllers are going to be at a massive disadvantage compared to those playing with keyboard and mouse. Valve needs to conduct more widespread testing with a trackball variety. That would allow controller users to have a more even match.

The intent is that the touchpad function as a trackball, and until I hear otherwise I'm going to consider the possibility that the touchpad can function as a trackball pretty well - the motion is similar, all your missing is the haptics of the thing physically moving under your thumb.

Either way, there are no pro-gamers in the twitch-oriented genres that use trackballs. So no matter what, the gamepad/trackball is going to be a handicap. I think we're all just hoping for "good enough", not "as good as a mouse". I'd settle for "good enough". The gap between a mouse and a thumbstick is such a massive ocean that a substantial improvement, even if it doesn't reach "as good as mouse" would be worth buying.

Trackpads can't mimic the precision (a dot the size of a pinhead vs a fat thumb). They can't get rid of their terrible z-axis problem (far worse than mice -- and not present in trackballs). They also can't handle inertia well due to lack of feedback (When you use a trackball, you become very adept at flinging the ball and stopping it at the exact position you want).

As to pro-gamers, some of us remember when over half of the pro gamers used nothing but trackballs (in preference to optical or mechanical mice). The short story here is that trackballs didn't keep up with the newer laser technology needed for better resolution, so the pros moved on to mice. There are now some trackballs that use laser technology, but everyone (for better or worse) has moved on.

They have already tested trackballs

http://www.youtube.com/watch?v=UfN5WK7OzU8#t=206

Trackpads have a major issue. They are not accurate. If you heighten the sensitivity, they have severe z-axis problems (worse than any mouse) and decreasing the sensitivity makes motions slower and less responsive. Because of the size of the target (the human finger) there is no definitive "this is where I am".

Several other of their ideas are in error. Larger trackballs are not needed (logitech and kesington both make popular trackballs with small radii that perform well. The weight of the ball is also a red herring as it is possible to make lighter balls if needed.

The "issue" of cleaning is minimal and a non-issue overall (ask any trackball user). The idea that circuit space is constrained is also a non-issue. Modern trackballs use a laser to read the ball (a feature of trackballs is no z-axis movement) The circuitry to deal with this is small as is the circuitry for much of the rest of the device.

The real argument is that the material cost is higher, so they'd rather use inferior design and bump up profit margins. Most gamers will pay over $50 for a mouse (with some of us giving over $100 for our mice or trackballs). I'd rather have a quality device and pay a little more.

Have you used a Steam Controller?

What do you mean by z-axis movement?

This isn't a product for high-level competitive gaming. This is a product that allows the average gamer to play the vast catalog of steam games from the couch.

The Steam controller gets less interesting with each new announcement.

Well, the twin touchpads were always the most exciting part, and they're still here for now. Hopefully they'll survive 'til the end, and not get replaced with another XBox pad clone.

I'm not familiar with the original design. Can someone post a before/after picture here?

It can be seen in this article.

http://arstechnica.com/gaming/2014/01/steam-controller-impre...

They've got pictures of the original design still up on the controller's webpage: http://store.steampowered.com/livingroom/SteamController/

There's a small picture to the right of the article text near the top of the article that shows how it used to look.

That d-pad design is horrid.

Reply on news.ycombinator.com