Here's The Steam Controller's Final Design

[RedDeadFred]

I'm confused. Wasn't the point of this thing to be having the comfort of a controller while having the flexibility of a keyboard? I'm only seeing two more buttons than a 360 controller... Maybe it's the angles in the pictures and there's actually more. Can anyone confirm this?

Besides the buttons issue, the thing doesn't really look very comfortable to hold. Those track pads look they'll really be a pain on the thumbs with prolonged use. I mean, if you're regularly having to go to the edge of those pads, you're going to be contorting your thumbs pretty often.

Maybe I just don't get it but I don't really see any reason to use this over a 360 controller and definitely not over a keyboard and mouse.

 

[Objectable]

You ever see that episode of Voyager where Tuvok and Neelix accidentally get fused inside of the transporter and create an unholy abomination to God and science alike?
Casue that's the vibe i'm getting here

 

[CrystalShadow]

snip

Oh, I know. You're actually 'preaching to the choir' on this one. I've done my fair share of 'homework' on properly config'ing mice.

I was just saying that the track-pads on the Steam controller are far more high resolution than the sorts of track-pads we tend to come across on other devices, and their tracking speeds and range of motions are variable.

I spoke with someone who owned one of the early beta builds of the controller. He said that, even in its early state, the pads were highly adjustable to suite a wide range of input needs. From 1-to-1 tracking like one would want on a mouse, to 'momentum' based tracking like you'd see with a track-ball. It could even be configured to emulate an analog stick. (though he saw little point to it)

Well, that makes sense given the huge variety of things you're expected to be able to use it for...
So, that's a good sign at least that someone that had a prototype found it worked quite well.

 

[flying_whimsy]

I really would have preferred a d-pad to the analog stick. That bums me out to the point where I am reconsidering getting one. Granted, if the touchpad works just as well might be worth the wait. I was so excited for this stupid controller.

I'm just glad to have some competition for the xbox controller that doesn't require unofficial 3rd party drivers. Too bad they didn't release last year when I actually needed a new pc controller >:[

 

[RolandOfGilead]

I'm glad they kept the two circular trackpads, I thought they got rid of those.
Question is, will I be able to use it with all the games that require a xbox360 controller?

 

[Hairless Mammoth]

Agh, I was worried they would go this route. Quite a bit of the games I own on Steam work best with a good d-pad (PS3 controller for me). Right now, I don't see a point in getting this, since my controller needs are fulfilled and, if I really want to play a Kb and M centric game on the big screen, I'll set up a way to make a Kb and M work away from my desk. I'll wait for reviews and personal opinions of random Joes like us (as one should with any game related thing these days) before I truly decry it as an over-invested attempt to shoehorn two distinctively different control schemes.

 

[FPLOON]

Meh... I'll still buy it... unless I can't play my Lemmings on it... That's my personal dealbreaker right there...

But seriously, I'll still buy this controller just because it looks cool to me... <color=white>and it's Steam-compatible and shit...

 

[Rack]

It's interesting if nothing else, having played around using the trackpad on my keyboard with my thumb I can see it being more accurate than an analogue stick without losing too much in comfort. Mind you for living room options I've settled on a mouse + keypad which feels like a much better compromise for playing rts and shooters than this.

 

[War_Dyn27]

I saw a video of the Steam controller from the Verge where they were demoing the controller using System Shock 2, which, as anyone who has played SS2 could tell you is a game that would never work on a regular controller thanks to its intricate interface. The person testing the controller seemed to have a positive reaction, so colour me enthused. Plus it's going to be cheaper than the ps4 and xbox 1 controllers at $50 USD.

 

[zumbledum]

That and I have no real idea who this controller is for, but I guess I cannot really say much more without actually using it.

Its just better , or that's the feedback i've heard, more precise controls for movement more mouse like aiming, just plain better.

Not many people that arent old farts like me realise PC players never used to use mice, doom was played with just a keyboard, i remember one summer holiday when everyone was back from uni and we had a 6 player lan set up in my house for the brand new release of quake and some serious multi player death matching, I unhappy with the speed of aiming in this new truly 3d battlefield noticing the mouse option decided to use it. and it was horrible.. at first. a week later and everyone else had been forced into going through the same learning process.

"if" this controller delivers, and from the reports ive read its going to. when people have to start using it to be competitive in COD and FIFA it could well be the same level of revolution for consoles as the mouse was for PC.

 

[War_Dyn27]

Hm, doesn't dropping that d-pad mean playing games like street fighter is pretty much impossible?
And how about platformers?
I kinda need the precision of a d-pad for something like super metroid and that touch "d-pad" just gives off touchscreen vibes for me.
That is it looks awkward and floaty and not something you'd want to play mario with.

This seems to be geared towards egoshooters more than anything else and I'm not so sure if stick+touchpad comes close to keyboard and mouse.

according to a video from Kotaku, the pads have 8-directional digital click inputs.

 

[Strazdas]

For instance, since I have one in front of me right now, there are several properties of it that stand out.
2. the distance it moves depends on the speed of your finger movement across the pad.
3. If you hit the ends of the pad (or get near them), you lift your finger, move it back across, and slide it further.
- Which is a smaller scale equivalent to what you do when you run out of space with a mouse.

Number 2 is a subtle point that I never fully appreciated until just now. In fact, I thought you were wrong until I just confirmed it with my own mouse. Previously I had assumed the cursor distance was in direct proportion to the mouse distance, irrespective of speed. Thank you for that.

It is called mouse acceleration. It is a turn-offable function in OS. Many shooters also disable this for more precise aiming. its great for navigating your OS, not so much for FPS.

That really depends on the type of game.
But for many games the OS settings are irrelevant anyway, because the game does it's own calculations.

I've done enough game programming to know that. Yes, you can read the values the OS gives you, but more often than not it's better, if creating a game to work with the raw data the mouse sends out.

And that data is a series of packets describing the change of position over time. 99% of the time you can't apply this directly, especially in a game, because there is an absurd mismatch between the resolution of the mouse and what you are trying to control with it.

This is made much worse by the wide range of mice with huge differences in dpi settings (or equivalent, depending on how the mouse works)

Non-linear motion is one option among many, which indeed you may not want in some situations, but then again, may appreciate in others.
It depends...

This is not entirely correct. any game that uses mouse uses OS calcuation for mouse input with one exception. The exception is using RAW input and do your own mouse driver support. Very few games do this. Most games instead use windows API and send in commands to disable things like Mouse Acceleration while you play which windows restore once you quit the game.

Raw is great if you can code your mouse driver well. most dont bother and use windows settings to disable all fancy windows stuff and leave only basic input.

Ironically, windows do not support this outside of fullscreen game, so you cannot truly force mouse acceleration off in your OS. yes, even if you disable it in settings.

I'm confused. Wasn't the point of this thing to be having the comfort of a controller while having the flexibility of a keyboard? I'm only seeing two more buttons than a 360 controller... Maybe it's the angles in the pictures and there's actually more. Can anyone confirm this?

No, the point of this controller was to show the wonders of mouse precision controls to unfortunates who can only play with controllers.

 

[Kahani]

Not many people that arent old farts like me realise PC players never used to use mice, doom was played with just a keyboard, i remember one summer holiday when everyone was back from uni and we had a 6 player lan set up in my house for the brand new release of quake and some serious multi player death matching, I unhappy with the speed of aiming in this new truly 3d battlefield noticing the mouse option decided to use it. and it was horrible.. at first. a week later and everyone else had been forced into going through the same learning process.

Indeed. I think it's especially bad for console gamers who have had controllers of pretty similar design for quite a while now with no real option to try different control schemes, but even with PCs a lot of people don't seem to realise just how much things have changed in a very short time before and, importantly, how easy and quick it is to get used to it when the change is actually beneficial. I absolutely hated mice when they were first introduced to gaming, I generally just found myself looking at the floor or ceiling and could hardly play games at all. But that was before mice were used much at all, with DOS still the main platform for gaming. As mice became more widespread and people actually learned to use them, it quickly became obvious how superior they were to just using a keyboard.

And of course it's not just mice. Trackpads, tablets, touchscreens, trackballs, those stupid nipple things, and plenty of other input methods have turned up over the years. Some of them aren't exactly great in many situations, but none of them are that hard to get the hang of using. And this isn't even some revolutionary new control scheme, it's just a slight variation on a very well known controller layout. Maybe it will be good and maybe it will suck, but the majority of the hate seems to be simply that it's different and might take a bit of getting used to, and that's really just silly. Just because you're used to one particular control scheme doesn't mean it must be the best, and just because you can't immediately use a new one as well as your old one doesn't mean it's not worth the really quite minor effort of learning it.

"if" this controller delivers, and from the reports ive read its going to. when people have to start using it to be competitive in COD and FIFA it could well be the same level of revolution for consoles as the mouse was for PC.

I very much doubt it. As noted above, this is actually a fairly minor variation on an existing input device. It's still the same basic controller with the same basic buttons in largely the same places. Even if it does turn out to be better in some ways than existing controllers, it's very much an evolutionary change, not a revolutionary one. It's more like the addition of scroll wheels to mice than the introduction of mice themselves.

 

[CrystalShadow]

For instance, since I have one in front of me right now, there are several properties of it that stand out.
2. the distance it moves depends on the speed of your finger movement across the pad.
3. If you hit the ends of the pad (or get near them), you lift your finger, move it back across, and slide it further.
- Which is a smaller scale equivalent to what you do when you run out of space with a mouse.

Number 2 is a subtle point that I never fully appreciated until just now. In fact, I thought you were wrong until I just confirmed it with my own mouse. Previously I had assumed the cursor distance was in direct proportion to the mouse distance, irrespective of speed. Thank you for that.

It is called mouse acceleration. It is a turn-offable function in OS. Many shooters also disable this for more precise aiming. its great for navigating your OS, not so much for FPS.

That really depends on the type of game.
But for many games the OS settings are irrelevant anyway, because the game does it's own calculations.

I've done enough game programming to know that. Yes, you can read the values the OS gives you, but more often than not it's better, if creating a game to work with the raw data the mouse sends out.

And that data is a series of packets describing the change of position over time. 99% of the time you can't apply this directly, especially in a game, because there is an absurd mismatch between the resolution of the mouse and what you are trying to control with it.

This is made much worse by the wide range of mice with huge differences in dpi settings (or equivalent, depending on how the mouse works)

Non-linear motion is one option among many, which indeed you may not want in some situations, but then again, may appreciate in others.
It depends...

This is not entirely correct. any game that uses mouse uses OS calcuation for mouse input with one exception. The exception is using RAW input and do your own mouse driver support. Very few games do this. Most games instead use windows API and send in commands to disable things like Mouse Acceleration while you play which windows restore once you quit the game.

Raw is great if you can code your mouse driver well. most dont bother and use windows settings to disable all fancy windows stuff and leave only basic input.

Ironically, windows do not support this outside of fullscreen game, so you cannot truly force mouse acceleration off in your OS. yes, even if you disable it in settings.

That's dp rather questionable given that mouse acceleration in the windows api is only applied to direct mode (where the api call returns a cursor location), rather than relative mode. ( where it returns the coordinate delta values)

And using direct mode for situations where there is no cursor onscreen (like say, when controlling the targeting reticule in an fps) is totally absurd.

While it's true that in windowed mode there are a lot of extra complications (which explains why historically, many games didn't support it well, if at all)

Among these many are graphics limitations. There's no ability to do page flipping in windowed mode. You cannot set the resolution or colour depth. (a big pain if your engine isn't coded for every possible colour depth).
The front buffer contains the entire screen, not just your game's window. The title bar counts as part of the window... You ideally need to be able to handle window move and resize events. (a surpring number of games can't deal with that)

All common stuff for general windows programming, but a headache for games.

Mouse behaviour also varies depending on if you use the windows api calls, directInput or XInput to handle things. Then it also varies on what kind of data you ask for...

But why you'd use screen coordinates outside of situations in which you have a cursor onscreen...

Well, anyway, what makes you so sure about what most games do anyway?
My knowledge comes from programming books and api documents.

That doesn't tell me what any given game does, but it does tell me some methods would be rather dumb to use in games...

 

[Strazdas]

That's dp rather questionable given that mouse acceleration in the windows api is only applied to direct mode (where the api call returns a cursor location), rather than relative mode. ( where it returns the coordinate delta values)

And using direct mode for situations where there is no cursor onscreen (like say, when controlling the targeting reticule in an fps) is totally absurd.

While it's true that in windowed mode there are a lot of extra complications (which explains why historically, many games didn't support it well, if at all)

Among these many are graphics limitations. There's no ability to do page flipping in windowed mode. You cannot set the resolution or colour depth. (a big pain if your engine isn't coded for every possible colour depth).
The front buffer contains the entire screen, not just your game's window. The title bar counts as part of the window... You ideally need to be able to handle window move and resize events. (a surpring number of games can't deal with that)

All common stuff for general windows programming, but a headache for games.

Mouse behaviour also varies depending on if you use the windows api calls, directInput or XInput to handle things. Then it also varies on what kind of data you ask for...

But why you'd use screen coordinates outside of situations in which you have a cursor onscreen...

Well, anyway, what makes you so sure about what most games do anyway?
My knowledge comes from programming books and api documents.

That doesn't tell me what any given game does, but it does tell me some methods would be rather dumb to use in games...

acceleration is the most noticable but not the only fancypants stuff windows do to your cursor. That being said i dont think many games use direct mode anyway, more using the windows api with commands to disable acceleration.

and yes windowed gaming is entirely different beast in terms of support there.

My knowledge comes from observing how games so stuff and experimenting. i read quite a lot about it (when i want to understand how something works i go on research binge, wasting a lot of time and ealrning a lot of useless info) but i never made a game myself

It would be cool if noone used direct mode though, however appears that even games like Far Cry 4 fail at that. though FC4 seems to have a shitton of technical problems, but of course being ubisoft they tried to hide them.

 

[CrystalShadow]

That's dp rather questionable given that mouse acceleration in the windows api is only applied to direct mode (where the api call returns a cursor location), rather than relative mode. ( where it returns the coordinate delta values)

And using direct mode for situations where there is no cursor onscreen (like say, when controlling the targeting reticule in an fps) is totally absurd.

While it's true that in windowed mode there are a lot of extra complications (which explains why historically, many games didn't support it well, if at all)

Among these many are graphics limitations. There's no ability to do page flipping in windowed mode. You cannot set the resolution or colour depth. (a big pain if your engine isn't coded for every possible colour depth).
The front buffer contains the entire screen, not just your game's window. The title bar counts as part of the window... You ideally need to be able to handle window move and resize events. (a surpring number of games can't deal with that)

All common stuff for general windows programming, but a headache for games.

Mouse behaviour also varies depending on if you use the windows api calls, directInput or XInput to handle things. Then it also varies on what kind of data you ask for...

But why you'd use screen coordinates outside of situations in which you have a cursor onscreen...

Well, anyway, what makes you so sure about what most games do anyway?
My knowledge comes from programming books and api documents.

That doesn't tell me what any given game does, but it does tell me some methods would be rather dumb to use in games...

acceleration is the most noticable but not the only fancypants stuff windows do to your cursor. That being said i dont think many games use direct mode anyway, more using the windows api with commands to disable acceleration.

and yes windowed gaming is entirely different beast in terms of support there.

My knowledge comes from observing how games so stuff and experimenting. i read quite a lot about it (when i want to understand how something works i go on research binge, wasting a lot of time and ealrning a lot of useless info) but i never made a game myself

It would be cool if noone used direct mode though, however appears that even games like Far Cry 4 fail at that. though FC4 seems to have a shitton of technical problems, but of course being ubisoft they tried to hide them.

eh. fair enough. Game engine and windows programming is the closest thing I have to a formal qualification. Which isn't saying much, because I failed all my studies (if I hadn't I'd now have an undergraduate physics degree with a minor in either programming or japanese, a tafe diploma in programming, and a degree in game development from a specialist training centre)

Anyway, the windows api is bizarre. The official unit of measuring mouse movements is the 'mickey' (no, I'm not making that up).

Well, mouse and keyboard input were for a while, things for which using directInput didn't really get you much. Might as well use the windows api. You need it anyway, just to get a windows program to even run.

I do like to mess around sometimes. I've found it amusing to write breakout clones using nothing but gdi rectangle drawing calls and some basic code.

You can do that in a few hours...

A version using directX is pretty simple too, but takes a lot more setup code.

(in my experience a rectangle aligned with sides aligned to the screen axes is the simplest geometric shape to code. There's almost nothing to it. Only thing simpler is writing Single pixels to a buffer)

 

[vallorn]

Sweet...and thank god! I have games on steam that I'm struggling with because the interface would suit a controller better!

Heh. As someone playing Dark Souls with the default (no DS fix because masochism) Keyboard and Mouse set up. This controller looks like it would be perfect for me.

 

[Vigormortis]

I'm confused. Wasn't the point of this thing to be having the comfort of a controller while having the flexibility of a keyboard? I'm only seeing two more buttons than a 360 controller... Maybe it's the angles in the pictures and there's actually more. Can anyone confirm this?

Button counts, minus directional digital 'clicks':

Xbox 360 controller - 13

Steam Controller - 18

Button counts, plus direction digital 'clicks':

Xbox 360 controller - 17

Steam Controller - 34

Besides the buttons issue, the thing doesn't really look very comfortable to hold. Those track pads look they'll really be a pain on the thumbs with prolonged use. I mean, if you're regularly having to go to the edge of those pads, you're going to be contorting your thumbs pretty often.

I've yet to hear a single complaint of this type from those I've spoken to who've used the controller - even for extended periods.

Besides, the track-pads are configurable, so you can set the tracking resolution high enough to never need to go to the edges.

Maybe I just don't get it but I don't really see any reason to use this over a 360 controller and definitely not over a keyboard and mouse.

It's not meant to replace a keyboard and mouse.

Personally, I'm sick of the 360 controller. It's too small, the sticks are abysmal (and their dead-zones wear out in no time), the d-pad is a joke, and the analog triggers are not nearly as accurate as they could be. I can't wait to have a decent replacement for it, and the Steam Controller looks to be that replacement.

And even if it's not, I'll be out only fifty bucks.

 

[Vicarious Reality]

I tried shoving a console controller into my computer 15 years ago, it was funny

I use a nordic SteelSeries 6G v2 and a Logitech G600 MMO with a CRT non-widescreen

 

[Robert Marrs]

I want one really bad. Even if its a bit awkward at first the thought of being able to play mouse and keyboard games from my bed is worth it.