SEGA D-pad analysis

Anyone who has read my previous posts knows I’ve been obsessed with game controller design for a long time. My obsession has led me to three overall questions/conclusions:

  1. The Japanese SEGA Saturn controllers (and the genuine SEGA-produced Genesis 6-button controller modeled directly after its internal design) have the best D-pad ever designed — but why?
  2. The “transforming d-pad” on the new 360 controllers works bettter but still has some problems — why?
  3. Convex-shaped analog thumbsticks are superior to concave-shaped analog thumbsticks — but why?

I have analyzed each of these topics with OCD-like attention and have come up with some explanations/answers that are actually very interesting. This post will address points 1 and 2, because they are related.

The one remaining problem with the “transforming D-pad” on the new xbox360 controllers is that it is still a tip-prone teeter-totter. Meaning, as you are trying to cleanly switch from pressing “left” to “right”, it’s all too easy for the d-pad to accidentally “tip” up or down enough to erroneously activate one of those vertical directions. This is what still infuriates many players of games like PacMan CE DX, which require accurate split-second 4-direction controls.

SEGA solved this problem in the 1990s with their D-pad design for the 6-button Genesis controller (same design used in the Japanese Saturn controller). They eliminated the tip-prone nature of the d-pad by eliminating the central pivot point altogether! See the following illustration:

I bought an old 6-button Genesis controller (the good SLS-manufactured ones, not the crappy knock-off ones manufactured by Jalesco or whoever it was and rebranded with the SEGA logo) off eBay and took it apart. I also bought off eBay a pair of the new (2009) SLS-manufactured and reissued-only-overseas “Saturn” controllers that are USB for use with PCs, and took one of them apart. The controllers are cosmetically different on the outside and have a slightly different feel, but they use the same D-pad design as illustrated above. The Saturn USB controllers, when used with emulators like MAME or KEGA Fusion, are far more accurate and “less frustrating”-feeling than the D-pads on either the PS2 or Xbox360 controllers I also hook up to my PC. The SEGA pads truly are the best design, even after all these years.

Which leaves me wondering, why don’t all the modern game makers use that design? Surely whatever patent(s) (if any) SEGA held on the design have expired by now. Maybe it’s just because the engineers at Microsoft and Sony and various third-party controller makers are unaware of this design.

This entry was posted in Games. Bookmark the permalink.

7 Responses to SEGA D-pad analysis

  1. Wyatt says:

    I’ve been thinking a lot about this lately. Why don’t the designers now use a 6-button genesis D-Pad? The Genesis controllers were always some of my favorites, and some of the D-Pads today just don’t hold up (coughXBoxcough). Nice blog, by the way.

    • c0d3h4x0r says:

      Thanks for the compliment!

      The answer to your question might lie somewhere in patent law. Companies treat patents as if they were just trading cards, but they’re only required by law to license patents deemed “essential” for the industry. And the rules about what can and can’t be patented may mean that specific aspect of the design is locked up as part of some larger design patent. For example, I suspect the “transforming” aspect of the D-pad on the newer Xbox360 controllers was just a gimmick to enable patent protection on other smaller (yet important) aspects of the revised design that wouldn’t be patentable on their own.

      Or it might be just be a data-driven decision. Most modern gamers rarely/never use the D-pad, and of the minority who do, some think the D-pad they have is just fine. So it may be a small percentage of users that would motivate making any kind of change, and the cost of acquiring patent licenses or manufacturing changes may just not make financial sense.

      Or it might be a durability issue. One downside of the SEGA D-pad design is that it involves plastic-sliding-against-plastic, which will wear down over time faster than other designs. And planned obsolescence of the hardware doesn’t work if you’re selling the hardware at a loss and subsidizing it through software sales.

      • Wyatt says:

        I think that it would be the second reason you brought up, if any, mainly because of the fact that a big company like Microsoft or Sony would rather not fix something they don’t think is broken, mainly because it would cost too much money. Not that I have anything against the companies, it’s just that, as you said, it wouldn’t make sense financially. I’ve always thought that the D-Pads from the older game systems were better (i.e. the SNES controller, or the 6-Button Genesis). What I find funny, though, is that the patent on Sega controllers may not be over yet (I’m not sure), but the default 3-Button controller’s D-Pad and the XBox 360’s D-Pad have an extremely similar design.

        I’d like to also say that I appreciate you taking your time to respond to my comment.

  2. sianagearz says:

    Apropos forgotten designs. SEGA Dreamcast had the most unusual triggers and analog sticks. They did not consist out of potentiometers or have any contact surfaces at all, so there’s one less thing to wear down with time – of course you can break anything by applying too much force, but i never managed to. They consist out of magnets and Hall effect digital sensors. Of course, no steel screws near Hall sensors, brass screws only.

    The sensor based design freed up plenty of room to build a force modulating rig into the stick. There’s always a centering spring, right. So in this design, the centering spring presses against a plate, which travels against a shaped plate such that the height of the spring plate depends on the stick position. The spring is relaxed along 8 principal directions and tightened elsewhere, so that you have this tactile feedback on whether you’re hitting the principal directions. If you apply more downforce, you have full analog control, and if you apply less, this assembly nudges you along the principal directions, which is useful for so many 2D style games.

    Of course one downside is that there is no push button functionality in the stick, but honestly this was never such a good idea in the first place, it’s just way awkward to use – i wish those were simply extra buttons, and by the way, there’s simply too much unused convenient space on the rear of the controller where your fingers naturally are anyway – i’d like shoulder buttons to be on the bottom too! The bowl design of Xbox sticks makes using push button easier than on other controllers, but has precision and wear drawbacks. The Dreamcast gamepad came with a 4-directional pad which wasn’t very well made, but this didn’t matter a terrible lot because of how evolved the analog stick was, so it served as an extra set of rarely used buttons.

    One other curious detail of Dreamcast pad is how convenient it is regardless of thumb length and hand size, because of how tall and straight it is and the controls are essentially nearly the same distance from the sides, so you just slide your hand up or down a bit to chose the controls. as opposed to spraining your thumb trying to reach for that stick in the middle of the controller, like Xbox and Playstation. Really think of the children! (and girls with small hands)

    There were apparently at least 3 iterations of the 8-way D-pad on SEGA Genesis. You only mention two of them, the simplistic kind like in most controllers out there, and the one where the faceplate forms a spherical guide for the D-pad to travel upon. These designs form a logical progression.

    If you think of the simplest possible design, you have 4 buttons connected by a common haptic surface. This design fails because pressing the surface down is unstable and potentially presses down too many underlying buttons. If you attempt to improve this design, you add a centre post to the haptic surface, which prevents too many buttons to be pressed down at the same time. Assume that this post went all the way down to where it touches the PCB underneath, forming a pivot. This however might not provide for sufficient travel of the surface to activate the buttons, especially for diagonal configurations, where a larger disc edge travel is required to achieve the same travel at activation points. So the post is shortened (by at least 1mm in case of Xbox360), such that the surface floats, and the user provides the travel by pressing the whole D-Pad surface down, and lowering the stability of the design and introducing a compromise. Simply lengthening the post a bit should resolve some of the confusion, but it could be a tough balancing act.

    That’s where the second SEGA design (newer generation 3-button) comes in. The D-Pad post goes through the PCB to the opposite (rear) side with a really long post which has an anchor point in the rear housing. On that rear of the PCB, there’s the buttons and a circular rig that presses them, instead of them being on the face side, like on any other controller. The D-Pad no longer floats, and the rig has enough travel to press down any of the buttons or the desirable 2-button combinations securely.

    The third design (6-button) that you mention is an evolution of this design. It makes the post and the pivot virtual and really long, by providing a guide for the D-Pad in the face plate, and the radius of the guide determines the virtual position of a stable pivot.

    Now think about the difference between the outer shape of the SEGA controllers and the Xbox controller. SEGA controllers are really compact, and the PCB is barely beneath the face plate of the controller. Xbox 360 has a bulky controller. 20mm from PCB to face plate, 22mm from PCB to the edge of the D-Pad well. That’s rather generous already to make a good compromise classic design. If Microsoft didn’t employ evil screws, i’d certainly try a number of things already. I wonder why this is done though, i mean SEGA never had any qualms about their controllers being disassembled, not even as much as a security sticker, not even on the extremely evolved Dreamcast controller.

    I think one of the reasons why the Xbox360 design is the way it is is the diminutive size of the D-Pad well, which was chosen so for placement issues and because it’s a control of secondary importance nowadays, so they largely had to limit the lateral travel and emphasize vertical travel instead, with the stability compromise this entails.

    • c0d3h4x0r says:

      Very thoughtful and informative post. Thanks!

      I’m familiar with the “second rev” design of the Sega Genesis D-pad. It was definitely a novel approach, but it unfortunately located the pivot point (center of the virtual sphere of movement) behind the PCB, resulting in too much side-to-side sway in the D-pad, making it more (not less) similar to the xbox360’s “teeter-totter” problem.

      I never owned or played a Dreamcast, but your description of its controller’s analog stick design sounds fascinatingly unique! I may track one down just to see how it feels and take it apart šŸ™‚

      The ideal D-pad design is one which uses 8 distinct micro-switches (with tactile “click” feedback) for symmetrical feel in all 8 directions, and which places the center of the virtual sphere above the D-pad rather than below it. What you want is for all other sides of the D-pad to “slide up” (lift) when you press a desired direction, and the only way to accomplish that is by having the D-pad live inside the concave bottom of the virtual sphere.

      The PS2 Dual Shock D-pad design comes close to this (but doesn’t go quite far enough); I highly recommend taking one apart to see how that design works. Also, I’ve watched action videos of the D-pad on the new Xbox One Elite Wireless Controller, and they clearly show the opposite side of the pad lifting up in the way I describe, so it will be very interesting to see how that one works out.

    • Commenter says:

      The problem with placing the controls the same distance from the side (as in your “think of the children” comment) is that it’s hard to use both effectively.

      That was not really an issue in the days of the Dreamcast. Back then, the analog stick and D-pad were mainly viewed as alternate movement options. Most games at the time used only one or the other, sometimes allowing you to choose whichever you preferred. The N64’s multi-pronged design even forced this.

      Modern games, however, often make extensive use of both the analog stick AND the D-pad for different functions, usually with the D-pad acting as additional buttons. Having them at an angle as in modern controllers makes both controls reachable without having to re-position your hand.

  3. jBren says:

    Hey guys just wanted to drop a line and say, as someone who is utterly fascinated/ obsessed with controller design, this was really cool to read. Good information and like minds. I feel baffled by the majority who seemingly have no interest or preference for control. The addition that defines the video game as something more than other media is this tactile interactivity.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s