4 ms·

I'm biased since I'm an EE, but to me the real magic in game controllers is the mechanical/materials/manufacturing engineering that goes into making the buttons

I'm biased since I'm an EE, but to me the real magic in game controllers is the mechanical/materials/manufacturing engineering that goes into making the buttons feel right and have consistent responsiveness. This is what most third-party controller manufacturers seem to get wrong, even (or perhaps especially) when they clone the basic shape and layout of a well-known controller. That said, one of the cool things about how they designed the NES and SNES pads electrically/logically is that an SNES pad will actually work with an NES if you make an adapter for the different connectors, since it's the same protocol with a different number of data bits clocked. But the best part is that they didn't simply add X, Y, L, and R at the end of the bit stream for the SNES pad; instead they transposed it so that SNES Y is NES B and SNES B is NES A, which translates the common idioms used in action games on the two systems.

derefr 11y ago
> instead they transposed it so that SNES Y is NES B and SNES B is NES A That confuses the development process; they likely were using NES controllers during the development of the SNES hardware for some time, before creating a SNES controller. (Similar things can be observed in more modern Nintendo SDKs: before the Wii U GamePad was finalized, for example, the Wii U SDK console unit just exposed a Wiimote extension port, and expected a Wii Classic Controller to be plugged into it. You could then treat this in software—at least for HID purposes—as if it was the GamePad.) So it was more that for a while they had two face buttons, A and B, and then when they decided to add two more, the A and B buttons got re-labelled "B" and "Y" on the controller and in the SDK. (You can observe many games released early in the SNES lifecycle that were programmed under the assumption of only two face-buttons; these games either leave the "A" and "X" buttons functionless, or just mirror the functions between "X"<->"Y" and "A"<->"B".) --- As an aside, I like to look at the Nintendo controller bitplane as existing in a sort of "order of complexity" for the development of arcade hardware—which Nintendo has always targeted their platform to alongside home consoles[1]. For example, the first button in the bitplane—0x1—is always "Select". This makes very little sense (hell, what game even uses that button?[2]) until you realize that, in arcade machines, this is the "button" triggered by inserting a coin![3] After that, you get A and B, because some games just need button inputs and nothing more. Then, the directional inputs, because the next level of complexity up is an arcade cabinet with a joystick. And then the rest of the buttons, in an arbitrary order. In more recent SDKs, you'll sometimes see L and R inserted early on in the bitplane, earlier even than the directional inputs. Why? Because those buttons are for a simpler kind of arcade machine: they're flipper-signals for pinball machines! (I'm kind of surprised Nintendo cares about pinball machines, given that pinball isn't really a thing in Japan and pachinko—which is—doesn't use flippers. My guess at this point is that Nintendo IRD has a tradition of writing quick one-off pinball prototypes to test their hardware.) --- [1] Nintendo VS., Ultra 64, Triforce, etc. Nintendo loves arcades! Though they seem to get more subdued about this fact over time; their newer arcade platforms are effectively white-labelled embedded-hardware SDKs for third-parties to use, rather than a side-effect of the creation of first-party cabinets. [2] The Nintendo SRD "lotcheck" and Nintendo QA "burn-in test" ROMs for each console use Select and Start as their only functional buttons. (I would bet that, in the earliest console generations, before they had millions of mass-manufactured controllers laying about, the hardware developers would just have a breadboard with two resistors for Select and Start, with leads going through the 74LS165 and then over to the console. You'd just tap some wires together to simulate a button-press. Note that if all you need is Select, you don't even need the 74LS165; it doesn't matter if you're sending 0x11111111 if the console is only looking for (i & 0x1). You can accomplish that by just sticking wires directly into the console.) [3] Run any arcade-platform ROM in an emulator; you'll have to press Select to get it going.
kozukumi 11y ago
Couldn't agree with you more! I am amazed at the precision of modern controllers considering how complex they are. The fact I can pick up any Dual Shock or Xbox 360 controller and it feel just like every other one is pretty mind blowing. Sure over time they develop their own little habits but out of the box they are almost clones (with the exception of ones that should have failed QA but somehow shipped).
- georgemcbay 11y ago
  "The fact I can pick up any Dual Shock or Xbox 360 controller and it feel just like every other one is pretty mind blowing." I totally agree with the gist of your statement but as someone who has an absurd amount of Xbox 360 controllers I'm positive that they changed the components at least twice in the standard controllers (not including the obvious redesign of the gray special D-pad controller or all the various special editions) in ways (weight distribution, outer materials) that you can totally feel without taking the controller apart. I haven't researched this to know if this was due to changes over time or different factories or what but of the 9 or 10 controllers I had over the years (used on both the actual Xbox 360 and a PC) there were 3 distinct different types of feel that were subtle but obvious if A/B testing.
- roflchoppa 11y ago
  This actually drives me nuts, when i used to play MLG Halo3, my buddies always had these abused controllers, the fact that it was not EXACTLY like my controller I had at home made me play worse. It was the right analog controller that was 'loose'
ssalazar 11y ago
Absolutely. A big pet peeve of mine is a mushy D-pad and the lack of mechanical separation of each direction on the D-pad. The OEM SNES D-pad and Playstation Dual Shock D-pads get this right; the standard XBox/360 gamepad completely miffed this, making playing throwback/retro 2D games that used the D-pad for control a total drag.
- vcarl 11y ago
  Part of the reason a lot of D-pads sucked is because Nintendo had a patent on it, it expired in 2005. https://www.google.com/patents/US4687200 https://www.google.com/patents/US4687200
  - fdsarewq 11y ago
    maybe they should "mickey mouse"that shit ..
    - cryowaffle 11y ago
      This is patents, you are referencing copyright.
      - fdsarewq 11y ago
        ha...youre probably right...but actually i was referencing a bad joke / metaphor...think mickey might be a trademark though EDIT : eh...stupid question anyway ..

derefr 11y ago

> instead they transposed it so that SNES Y is NES B and SNES B is NES A That confuses the development process; they likely were using NES controllers during the development of the SNES hardware for some time, before creating a SNES controller. (Similar things can be observed in more modern Nintendo SDKs: before the Wii U GamePad was finalized, for example, the Wii U SDK console unit just exposed a Wiimote extension port, and expected a Wii Classic Controller to be plugged into it. You could then treat this in software—at least for HID purposes—as if it was the GamePad.) So it was more that for a while they had two face buttons, A and B, and then when they decided to add two more, the A and B buttons got re-labelled "B" and "Y" on the controller and in the SDK. (You can observe many games released early in the SNES lifecycle that were programmed under the assumption of only two face-buttons; these games either leave the "A" and "X" buttons functionless, or just mirror the functions between "X"<->"Y" and "A"<->"B".) --- As an aside, I like to look at the Nintendo controller bitplane as existing in a sort of "order of complexity" for the development of arcade hardware—which Nintendo has always targeted their platform to alongside home consoles[1]. For example, the first button in the bitplane—0x1—is always "Select". This makes very little sense (hell, what game even uses that button?[2]) until you realize that, in arcade machines, this is the "button" triggered by inserting a coin![3] After that, you get A and B, because some games just need button inputs and nothing more. Then, the directional inputs, because the next level of complexity up is an arcade cabinet with a joystick. And then the rest of the buttons, in an arbitrary order. In more recent SDKs, you'll sometimes see L and R inserted early on in the bitplane, earlier even than the directional inputs. Why? Because those buttons are for a simpler kind of arcade machine: they're flipper-signals for pinball machines! (I'm kind of surprised Nintendo cares about pinball machines, given that pinball isn't really a thing in Japan and pachinko—which is—doesn't use flippers. My guess at this point is that Nintendo IRD has a tradition of writing quick one-off pinball prototypes to test their hardware.) --- [1] Nintendo VS., Ultra 64, Triforce, etc. Nintendo loves arcades! Though they seem to get more subdued about this fact over time; their newer arcade platforms are effectively white-labelled embedded-hardware SDKs for third-parties to use, rather than a side-effect of the creation of first-party cabinets. [2] The Nintendo SRD "lotcheck" and Nintendo QA "burn-in test" ROMs for each console use Select and Start as their only functional buttons. (I would bet that, in the earliest console generations, before they had millions of mass-manufactured controllers laying about, the hardware developers would just have a breadboard with two resistors for Select and Start, with leads going through the 74LS165 and then over to the console. You'd just tap some wires together to simulate a button-press. Note that if all you need is Select, you don't even need the 74LS165; it doesn't matter if you're sending 0x11111111 if the console is only looking for (i & 0x1). You can accomplish that by just sticking wires directly into the console.) [3] Run any arcade-platform ROM in an emulator; you'll have to press Select to get it going.

kozukumi 11y ago

Couldn't agree with you more! I am amazed at the precision of modern controllers considering how complex they are. The fact I can pick up any Dual Shock or Xbox 360 controller and it feel just like every other one is pretty mind blowing. Sure over time they develop their own little habits but out of the box they are almost clones (with the exception of ones that should have failed QA but somehow shipped).

georgemcbay 11y ago
"The fact I can pick up any Dual Shock or Xbox 360 controller and it feel just like every other one is pretty mind blowing." I totally agree with the gist of your statement but as someone who has an absurd amount of Xbox 360 controllers I'm positive that they changed the components at least twice in the standard controllers (not including the obvious redesign of the gray special D-pad controller or all the various special editions) in ways (weight distribution, outer materials) that you can totally feel without taking the controller apart. I haven't researched this to know if this was due to changes over time or different factories or what but of the 9 or 10 controllers I had over the years (used on both the actual Xbox 360 and a PC) there were 3 distinct different types of feel that were subtle but obvious if A/B testing.
roflchoppa 11y ago
This actually drives me nuts, when i used to play MLG Halo3, my buddies always had these abused controllers, the fact that it was not EXACTLY like my controller I had at home made me play worse. It was the right analog controller that was 'loose'

ssalazar 11y ago

Absolutely. A big pet peeve of mine is a mushy D-pad and the lack of mechanical separation of each direction on the D-pad. The OEM SNES D-pad and Playstation Dual Shock D-pads get this right; the standard XBox/360 gamepad completely miffed this, making playing throwback/retro 2D games that used the D-pad for control a total drag.

vcarl 11y ago
Part of the reason a lot of D-pads sucked is because Nintendo had a patent on it, it expired in 2005. https://www.google.com/patents/US4687200 https://www.google.com/patents/US4687200
- fdsarewq 11y ago
  maybe they should "mickey mouse"that shit ..
  - cryowaffle 11y ago
    This is patents, you are referencing copyright.
    - fdsarewq 11y ago
      ha...youre probably right...but actually i was referencing a bad joke / metaphor...think mickey might be a trademark though EDIT : eh...stupid question anyway ..