Digital Joysticks provide better control than Analog Joysticks

[atariksi]

...You also mentioned the Playstion gamepad with the "keys" for the D-pad... which is incorrect. It has full 8-way control like any other d-pad... it's just a bit funny looking (and I personally don't like it at all as your thumb bottoms out against the plastic too much). It's a single piece thumbpad but only the 4 tips protrude through the casing, but they're not separate buttons or keys and thus you are limited to the normal 9 states. (opposed to the c-buttons on the N64 for example where all 16 states are possible)

The picture of the Playstation control is all digital. I took it apart and there's no 8-way control-- just arrow keys like a keyboard. Still better than an analog joystick, but not as good as the one with the stick.

It works exactly the same way as 8-way joysticks and d-pads on others. 4 switches and 9 states used (no press, up, dn, left, right, up+left, up+right, dn+left, dn+right) And yes, the arrow keys on a keyboard work similarly as well except they're separate buttons, not restricted by a stick/pad to only allow 9 states but all 16. (just as using the SNES's face buttons as a second 8-way direction input for robotron, or the same for the N64's c-buttons)

It would have been unusable if Sony had made it only work in 4 directions. (in which case you'd only need 3 switches anyway, 5 states)

 

The CX-40 operates in exactly the same fashion: 4 switches in a diamond pattern... the only difference is they're metal dome switches rather then rubber dome switches. (granted arcade joysticks tend to use heavier duty switches, often metal contact strips/plates, but the principal is still the same) The VCS sticks with the Flasback II do use rubber dome switches too for that matter, though the control tends to be a bit sloppy compared to the CX-40. (let alone CX-10)

 

Unless there are some oddly resigned joysticks that wastefully use 8 switches internally.

 

All I was stating there was it's easier to use with a lever (stick) to get to the buttons especially the diagonals and that digital would still be preferred over analog. However, the type of switches does make a difference in the bounce as I have noticed that even at 1Mhz sampling some digital joysticks show no bounce.

 

Koolkitty89: I was going to reply to your post #238 but looks like you got the quotes interleaved improperly so you need to fix those as some of the names shown up quoting are incorrect people.

[ledzep]

Your mocking me not knowing the controls of Star Wars arcade was clearly irrelevant to refute my controlled experiment. One thing is for sure, you aren't the only person who compares apples and oranges to try to establish how the analog joysticks provide superior control. That disqualifies you and others who do the same from refuting things that are comparing apples with apples.

 

Oh no no no, I'm qualified. I can drive a truck. Ask your relief pitcher.

 

As for the mocking, you deserved it. Let me give you an example of this -

 

If you have never eaten an authentic Italian pizza in Italy (by the way, I have) then you are completely unqualified to judge whether a Chicago-style pizza or a New York-style pizza is closer to the Italian ideal. Get it? If you don't know what the original standard (Star Wars yoke) is, you cannot evaluate alternates (digital joystick, analog joystick) to that standard. That is why you were mocked. And you will continue to be mocked until you try out that Star Wars "contraption" for yourself. Gather some experimental data in order to make this useless comment from you

 

"No, I have played Star Wars with analog joystick and digital joystick. They both don't do the job well. It takes too long to move around with digital joystick like in breakout and analog joystick is missing the accuracy. I haven't seen the contraption in the arcade, but a trackball would do well for Star Wars or a mouse neither of those are analog. A joystick can mimic the trackball for some things like using a tap vs. a full hold."

 

actually worth the electrons it's written on. Why? Because it might be the case that that an analog or digital joystick does do the job well or it might be the case that a trackball would not do the job well. How would you know until you have actually played Star Wars the way it was intended to be played, experiencing the most accurate version? In the same way in my example, without actually eating an authentic Italian pizza in Italy how can your opinion about neither American offshoot "doing the job well" receive any reaction other than mockery?

 

Your comment above about the Star Wars arcade game certainly provides further evidence of your inability to conduct a competent experiment and your willingness to make uneducated claims about games and controllers.

 

I did read it and my experiment still shows how you do know the digital joystick states 100% a priori whereas with analog you don't. No need to know at the nanosecond level as neither software nor computer is sampling at that rate nor needs to. Human interface devices usually sample in milliseconds. In fact, I already answered this earlier regarding some joystick switches that have bounce-- it's in the order of a few microseconds and joysticks get read instantly at usually a slower sampling rate. Analog joysticks however get read using slower sampling so once again its your analog joysticks that would fail that test. In this joystick control experiment, we're only concerned that the software we interface our joystick to gets the state that we want it to get without waiting for feedback. In this "simplified experiment" the software is a BASIC program and your analog joystick once again fails although it's a slower program. So obviously, it'll fail on faster sampling program (like examples given in post #114,137).

 

All your experiment description shows is that you want analog joysticks to work harder than digital joysticks to test the same abilities.

 

If your goal is to evenly test a digital joystick against an analog joystick and your standard for testing the digital joystick is reading the output of the 4 cardinal positions (buttons or switches) that the digital joystick spits out then your test program for the analog joystick must match the requirements for your test program for the digital joystick.

 

Fine. Milliseconds then. The requirements don't change, though. In order to accurately test a person's ability to know "100% a priori" what state the digital joystick is in you must factor out the states of the other portions of the chain. You must show that the person being tested can distinguish between the state of the digital joystick, resistance in the cable wiring, millisecond when the console's controller-reading register receives the input from the joystick, millisecond when the program reads that register and millisecond when the program updates the video display. It's amazing that you would ignore all those variables. Ah yes, I forgot, you have no idea what the scientific method is.

 

Still waiting on a competently-worded hypothesis. Should I hold my breath?

 

Wait, do you even know what "a priori" means? I'm surprised no one brought this up before but now that you have demonstrated time and time again that you just string words together without making sure you're saying what you think you're saying, it needs to be asked. So, what does it mean? -

 

a pri·o·ri (ä pr-ôr, -r, pr-ôr, -r)

adj.

1. Proceeding from a known or assumed cause to a necessarily related effect; deductive.

2.

a. Derived by or designating the process of reasoning without reference to particular facts or experience.

b. Knowable without appeal to particular experience.

3. Made before or without examination; not supported by factual study.

 

Well, now, that makes things a lot clearer, haahaha. This explains why you can't figure out how to correctly construct and conduct an experiment and why you have no data to present to support your theory! You reject facts and experience! And here we were giving you the benefit of the doubt. I wonder why you stated that it was a "scientific fact", then. Why confuse the issue when your whole approach to life is to make sweeping definitive statements based on ZERO experience or facts? Seems like a lot of extra work. Maybe you suffer from some really specific form of Tourette's where you randomly spit out sciency phrases instead of curse words? "A priori"! "Thus"! "Logic"! Would fit the facts. You know, facts. Facts.

 

You can stop speculating. I ran the test and offered you the simulator and you declined. So don't complain.

 

No you didn't. If you ran the test then you'd have data to present. At the least you'd have one of your precious screen grabs showing you deleting the data you recorded from your hard drive. Do you have even that? No. Simulators of joysticks are worthless when discussing the actual devices (digital and analog joysticks) to test that are physically available. Tell me, in your most a priori way, is a chip emulator the same thing as the actual chip? Do Atari 2600 emulators handle games 100% the same way as the real hardware does? There are dozens of threads here that refute that assumption.

 

Your sidekick brought up the scientific method. I have explained what that entails and what is required to satisfy it. You have woefully failed to. You need to communicate the results. Is that not clear? I'll make it clearer, then.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Nope, I clearly wrote: "Make sure to take into account the thresholds for all the games, the various levels you use for in-between features, etc." Only get to the threshold value so we know exactly when the direction change occurs for pac-man. But it does follow that if you can get to those exact threshold values, you can get to any of the other values.

 

Nope, you cannot lean back and forth between an experiment to test one thing and an experiment to test all things simultaneously. Scientific method, remember that? You have to lock down all the variables and then change one at a time. So if you want to lock down the variables for the digital joystick (cardinal directions) you must do the same for the analog joystick (ignore in-between states) for this test. Now, there's nothing stopping you from running another test (except for the fact that you haven't run any experiments and do not intend to) on the analog joystick to see how it works with in-between values and then compare that to the digital joystick's ability or inability to do the same but you can only deal with that separately, in a different experiment that you won't dare conduct.

 

It's no bias. It just so happens that it's a fact that analog joysticks are hybrids between paddles and digital joysticks. They tried to get functions of both and got the control of neither. Paddles provide better control than analog joysticks and digital joysticks provide better control than analog joysticks. And given trackballs and mice to deal with two-dimensional speed motion, I see no use for analog joysticks.

 

Whoa whoa whoa! Did you mention paddles? Paddles?!? What is this, Apples and Oranges Hour? Why bring up paddles in a discussion about digital vs. analog joysticks? You've made it clear that no one is supposed to do that. That includes you. Amazing, you and Divya16 crying gallons of tears about how I confused people by introducing paddles (while ignoring your identical confusing move of including buttons) and now here you are saying it's a fact that analog joysticks are hybrids between paddles and digital joysticks. A fact! What, do you think you're the only one in charge of what is and isn't a fact? Forget that. Your adherence to this "a priori" nonsense shows how little you care for facts.

 

But I'm glad you agree with me that paddles are part of the discussion. But I must correct you there, slightly, it's POTs, not paddles. Paddles are fair game (because they're single-axis POT controllers) so long as you continue to include buttons in your poll, though. Get rid of that choice, aprioriksi.

[atariksi]

 

I know you can't claim which is better because you don't even address my arguments but want people to blindly accept your views. We already know no controller is ideal for every situation-- post #1. Digital joysticks provide better control than analog controllers is the point. Go experiment for yourself. Or better understand the logic and science behind why analog controllers give inconsistent readings, require thresholds, calibration, etc. The bounce/micro or nano-effects are present in both controllers and software never sees it. At least for the experiment given above, you won't see any noise show up on the digital data. Go try sampling it at various rates that HIDs do and see for yourself. Only you want to try to confuse people by brining up these things as if they make a big difference.

 

What does make a difference to software and you see it in the sampling is stuff like the flawed/inconsistent readings from analog joystick controllers.

 

Some typos in my last part of reply to Poobah:

 

That should be "Digital joysticks provide better control than analog joysticks" not analog controllers. Same with another line that should read analog joysticks although its true for analog controllers as well. And it should be "bringing" not brining. Anyway here's some more data measuring extremes on Wico: 2/3..227/228 with center toggling between 112..115 (this is after calibration). Another Wico: 20/23..187/188 with center toggling between 113..115. Not only is the range inconsistent but the values at extremes are also not fixed nor the value at center. Nothing to do with bounce here, but actual measurable data. You won't see any of that in digital joystick data. Poobah's smokescreen of bringing up microsecond/nanosecond effects is not going to EQUATE a digital joystick to an analog one.

[ledzep]

See post #114 where I gave examples of games where there's a higher failure rate using analog joysticks vs. digital joysticks for same games played on same system. There's 4 ways people take a view:

 

(1) logical deduction/mathematical analysis (undeniably true)

(2) experimental data (which leads to the scientific fact idea)

(3) mental speculation-- something out of the blue according to emotional attachment, or mistaken views with no evidence

(4) blind following the blind-- because most people seem to be doing that so that must be right

 

You are in the catagory of (3). You just speculated something without any evidence. I have proof from (1) and (2). I can also speculate my views but that's all subjective. You are free to have your subjective views, but they don't hold up against the experiment/logic presented in this thread. Your example of cars doesn't apply either-- another misanalogy.

 

Oh, sweet! You claim to have proof.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Otherwise, as even Divya16 can tell you, you are in the category of (3).

[atariksi]

Your mocking me not knowing the controls of Star Wars arcade was clearly irrelevant to refute my controlled experiment. One thing is for sure, you aren't the only person who compares apples and oranges to try to establish how the analog joysticks provide superior control. That disqualifies you and others who do the same from refuting things that are comparing apples with apples.

 

Oh no no no, I'm qualified. I can drive a truck. Ask your relief pitcher.

 

As for the mocking, you deserved it. Let me give you an example of this -

Well, then I guess we can't discuss the topic if you want to talk about arcade controls and how they are superior or inferior to digital joysticks.

 

"No, I have played Star Wars with analog joystick and digital joystick. They both don't do the job well. It takes too long to move around with digital joystick like in breakout and analog joystick is missing the accuracy. I haven't seen the contraption in the arcade, but a trackball would do well for Star Wars or a mouse neither of those are analog. A joystick can mimic the trackball for some things like using a tap vs. a full hold."

 

actually worth the electrons it's written on. Why? Because it might be the case that that an analog or digital joystick does do the job well ...

You are misreading. If I want to compare control of analog joystick vs. digital joystick, I don't need to study the Star Wars arcade controls. I have both types of joysticks in several varieties and I can experiment without involving other unrelated factors.

 

Your comment above about the Star Wars arcade game certainly provides further evidence of your inability to conduct a competent experiment and your willingness to make uneducated claims about games and controllers.

It only provides evidence that YOU don't know what a controlled experiment is. You have to stick to the subject which is analog joystick vs. digital joystick.

 

All your experiment description shows is that you want analog joysticks to work harder than digital joysticks to test the same abilities.

No, I don't want it to work hard. It's by nature harder to get to exact states on the analog joystick. Heck, if you can deterministically get to exact states on the analog joystick with exact center, left, right, and whatever in-between states you wanted, you could wipe out the digital world and build analog devices with as much *CONTROL* as digital devices.

 

Fine. Milliseconds then.

Hey, we finally agreed on something. It took us like 256 posts.

 

a pri·o·ri (ä pr-ôr, -r, pr-ôr, -r)

adj.

1. Proceeding from a known or assumed cause to a necessarily related effect; deductive.

2.

a. Derived by or designating the process of reasoning without reference to particular facts or experience.

b. Knowable without appeal to particular experience.

3. Made before or without examination; not supported by factual study.

 

Well, now, that makes things a lot clearer, haahaha. This explains why you can't figure out how to correctly construct and conduct an experiment and why you have no data to present to support your theory! You reject facts and experience!

Stop the word jugglery. From context it's clear I meant knowing beforehand without relying on feedback. So that would be definition 2b and it's also deductive since what you think you did with the joystick is what actually happened.

 

No you didn't. If you ran the test then you'd have data to present.

Hey just make those moves that I presented in post #114 and get the statistics. I can post the statistics, but it's better to understand the science of why there's a higher failure rate so you don't have to go through thousands of games. For example, I can repeat the experiment of jumping over the oil barrel in Donkey Kong with analog joystick and digital joystick a hundred times and give you a percentage.

 

At the least you'd have one of your precious screen grabs showing you deleting the data you recorded from your hard drive. Do you have even that? No. Simulators of joysticks are worthless when discussing the actual devices (digital and analog joysticks) to test that are physically available. Tell me, in your most a priori way, is a chip emulator the same thing as the actual chip? Do Atari 2600 emulators handle games 100% the same way as the real hardware does? There are dozens of threads here that refute that assumption.

Simulators and not emulators. Simulator is suppose to exactly mimic the event. Paddles are not part of the dicussion but used an example to show that even amonst analog controller there's some that offer better control...

More later...I have to run to the JFK airport.

[+Stephen]

Well, then I guess we can't discuss the topic if you want to talk about arcade controls and how they are superior or inferior to digital joysticks.

 

But why - it's an analog controller. Don't want to talk about one that is superior to a digital

[+remowilliams]

To claim otherwise, or assign them blanket superiority, and ignore 30 years of engineering is just foolish.

 

Nearly all of this entire thread is foolish. And a giant waste of perfectly good electrons.

 

I think I'll ask Al if we can have a Pointless 'Discussion' forum so we can move crap like this into it.

[Divya16]

Well, then I guess we can't discuss the topic if you want to talk about arcade controls and how they are superior or inferior to digital joysticks.

 

But why - it's an analog controller. Don't want to talk about one that is superior to a digital

 

I don't mind discussing it. To start with, how do you end up with so many variety of arcade controls at home? Does anyone sell these as separately useable controllers? I'm pretty sure a plane cockpit would be a perfect flight simulator control-- better than an analog joystick.

 

@potatohead: electric cars have been around but sometimes for nontechnical reasons people choose other things that aren't the best technically.

[potatohead]

Well, then I guess we can't discuss the topic if you want to talk about arcade controls and how they are superior or inferior to digital joysticks.

 

But why - it's an analog controller. Don't want to talk about one that is superior to a digital

 

I don't mind discussing it. To start with, how do you end up with so many variety of arcade controls at home? Does anyone sell these as separately useable controllers? I'm pretty sure a plane cockpit would be a perfect flight simulator control-- better than an analog joystick.

 

@potatohead: electric cars have been around but sometimes for nontechnical reasons people choose other things that aren't the best technically.

 

So, you will concede the entire thread then??? LOL!!!

[Divya16]

See post #114 where I gave examples of games where there's a higher failure rate using analog joysticks vs. digital joysticks for same games played on same system. There's 4 ways people take a view:

 

(1) logical deduction/mathematical analysis (undeniably true)

(2) experimental data (which leads to the scientific fact idea)

(3) mental speculation-- something out of the blue according to emotional attachment, or mistaken views with no evidence

(4) blind following the blind-- because most people seem to be doing that so that must be right

 

You are in the catagory of (3). You just speculated something without any evidence. I have proof from (1) and (2). I can also speculate my views but that's all subjective. You are free to have your subjective views, but they don't hold up against the experiment/logic presented in this thread. Your example of cars doesn't apply either-- another misanalogy.

 

Oh, sweet! You claim to have proof.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Show. Us. The. Data.

 

Otherwise, as even Divya16 can tell you, you are in the category of (3).

 

Anyone who has played some same games with both analog and digital joystick has some experimental data to go on.

 

I am sure Atari/Amiga/Sega/etc. had some good basis for sticking with digital joysticks rather than going with the PC standard of analog joysticks.

 

@potatohead: it's not my thread to concede nor am I conceding but it's interesting subject to discuss arcade controls.

[potatohead]

given the similar style of commentary...

[+poobah]

Whenever you want to evaluate ANYTHING for a purpose, you have to DEFINE a proper, consistent and readily observable METRIC. In this case one needs a metric for "better control". In 9 pages, no one has defined such a metric (likely because given the range of uses for a joystick, there are a huge range of metrics with conflicting outcomes)

See post #114 where I gave examples of games where there's a higher failure rate using analog joysticks vs. digital joysticks for same games played on same system. There's 4 ways people take a view:

 

(1) logical deduction/mathematical analysis (undeniably true)

(2) experimental data (which leads to the scientific fact idea)

(3) mental speculation-- something out of the blue according to emotional attachment, or mistaken views with no evidence

(4) blind following the blind-- because most people seem to be doing that so that must be right

 

You are in the catagory of (3). You just speculated something without any evidence. I have proof from (1) and (2). I can also speculate my views but that's all subjective. You are free to have your subjective views, but they don't hold up against the experiment/logic presented in this thread. Your example of cars doesn't apply either-- another misanalogy.

 

Please document where you (or anyone else) in the thread outlined the specific metrics used to evaluate the controllers.

 

I humbly suggest that it is you who fall into "atariksi category #3." You make wild speculation "its a scientific fact" and emotionally attack anyone who disagrees with you. You have been repeatedly asked for evidence meeting scientific rigor to support your statement, yet, oddly, refuse to provide it.

 

Other notes:

Scientific FACTS are rare beasts, and nearly impossible to prove for all instances.

Scientific theories are plentiful (though often incorrect).

For any given scientific experiment (assuming the experiment itself is valid) to be accepted as verifying a theory, it must be repeatable by many scientists under a variety of conditions.

Rarity or non-rarity doesn't help you here. It's perfectly repeatable what I presented here and many have already repeated it to some extent at least.

I am unaware of anyone repeating your experiment and obtaining the same results. Given that you have failed to define a metric for controller evaluation, I'm not sure how anyone could.

 

atariksi has offered a theory (that he represents as fact) without a proper experiment (with appropriate controls) with results that have not been verified by the community at large.

No, you presented your mental speculation and are expecting people to accept you as some authority and blindly take as "fact". You have not countered anything that I presented in this thread. Perhaps, you will come back a few days from now and blurt out the samething again.

Actually, I am an authority on the subject.

 

Nonetheless, this is about your statement, not mine (in fact, I've made no claims as to which control may be superior)

You present a theory as fact, you produce no proper experiment, and no verification.

You can deny, redirect, obfuscate, call-names, whatever but you can't escape that fact.

 

Going out on a limb (and hopefully not feeding atariksi too much), EVERY controller is analog is some way. Digital sticks don't magically change state instantaneously. (there are no square waves in the 'real' world). Switching time, debounce, threshold voltage, force needed to overcome switch dome resistance, transistor rise time, hold time for a signal to get latched, etc.

That's NOT even the point of this thread. If you even cared to read this thread, the POST right before yours I addressed the debounce. How can you miss that? Were you in a rush or just felt emotionally biased to state something against my proven fact? I sampled the joystick via the parallel port at 2Mhz and I can measure the bounce. It's defintion is a digital joystick-- duh, despite have those characteristics. Need help find an engineering book.

 

Also, it has been proffered that knowing the exact state of the controller is a defining characteristic.

That is silly and irrelevant. The effect of the controller on the system is what matters.

In real systems, we study the impulse response of a signal in a system, not the signal itself.

 

Oh, I see you are going to tell us what's irrelevant. FYI, that's NOT the defining characteristic-- that's just your speculation. The focus is which provides better control. From that you can derive that one that you know the exact state of leads to better control. You are mixing up electrical engineering analog signal analysis with clearly defined digital signals.

So you are saying the effect of the controller on the system is NOT the defining characteristic???

 

Each of the effects mentioned (of which you claim to have examined one) increase the decoupling of the exact state of the controller from the exact state of the system. You have repeatedly expressed the opinion that a user knows the exact state of a digital joystick, thus making it better. I now claim that a user does NOT care about the exact state of the controller, but only the state of the system. When I am driving a car (for real or in a game), and wish to turn the vehicle, I have no interest in the exact position of the steering controller. I care only that the car turns appropriately. If I need to turn more, I move the controller more, and vice versa. I do NOT think "I need the pot on the steering axis to equal a measurement of 147 in the computer's steering axis coordinate system"

 

I was not aware that digital signals were electrically different from analog ones...

Please feel free to show me the difference.

 

Let us look at the input port for a moment... If it is NOT latched, one could conceivably close a switch contact and then reopen it (a 'tap' if you will), and have the system miss it because the port wasn't sampled during the closure time. On the flip side, if the port IS latched, you may 'tap' the controller immediately after sampling and not have it recorded by the system until the next sample interval, thus delaying the 'tap' by one sample time. You can apply these type of effects to each part of the system. In essence, regardless of the controller, the physical movement and the impulse effect on the system is decoupled to some extent, making exact knowledge of controller state meaningless.

As I said, the bounce is detectable on some joysticks in order of a couple of microseconds and has no effect on sampling done at millisecond range. What matters is the software gets the state that you want it to get. Go compute the probability of what you wrote happening first before making it some sort of highlight of your post.

Assuming a normal distribution, the average delay should be 1/2fs, where fs is the effective sampling rate.

And I agree, what matters is the software gets the state you want it to get.

 

Most games are a feedback loop, the game displays a state, the user manipulates the controller, the software uses the input (and other things) to generate the new state. A user is concerned with changing the state of the system ("I need to move my pacman left") not the controller ("I need to move the controller"). It is a subtle, but important distinction.

 

Simulators: Please. Simulators predict results, they do not produce them. No simulator can exactly model a given system. Using a simulator does not prove a theory. It DOES produce an expected result that needs to be verified by actual experimentation.

I already answered this last time you came by. I actually biased the simulator in favor of analog joystick by using fixed thresholds instead of variable. Simulators can be exact as well. I have no idea which dictionary you are reading.

No you didn't, which is why I repeat it.

In fact, your admission that your simulator is intentionally biased invalidates your results.

Repeat after me: "No simulator is exact." Show me ANY simulator, I will give you a parameter it doesn't model.

[Divya16]

...You also mentioned the Playstion gamepad with the "keys" for the D-pad... which is incorrect. It has full 8-way control like any other d-pad... it's just a bit funny looking (and I personally don't like it at all as your thumb bottoms out against the plastic too much). It's a single piece thumbpad but only the 4 tips protrude through the casing, but they're not separate buttons or keys and thus you are limited to the normal 9 states. (opposed to the c-buttons on the N64 for example where all 16 states are possible)

The picture of the Playstation control is all digital. I took it apart and there's no 8-way control-- just arrow keys like a keyboard. Still better than an analog joystick, but not as good as the one with the stick.

It works exactly the same way as 8-way joysticks and d-pads on others. 4 switches and 9 states used (no press, up, dn, left, right, up+left, up+right, dn+left, dn+right) And yes, the arrow keys on a keyboard work similarly as well except they're separate buttons, not restricted by a stick/pad to only allow 9 states but all 16. (just as using the SNES's face buttons as a second 8-way direction input for robotron, or the same for the N64's c-buttons)

It would have been unusable if Sony had made it only work in 4 directions. (in which case you'd only need 3 switches anyway, 5 states)

 

The CX-40 operates in exactly the same fashion: 4 switches in a diamond pattern... the only difference is they're metal dome switches rather then rubber dome switches. (granted arcade joysticks tend to use heavier duty switches, often metal contact strips/plates, but the principal is still the same) The VCS sticks with the Flasback II do use rubber dome switches too for that matter, though the control tends to be a bit sloppy compared to the CX-40. (let alone CX-10)

 

Unless there are some oddly resigned joysticks that wastefully use 8 switches internally.

 

The cx-40 does NOT operate the same way as a d-pad. The thumb is harder to move around especially diagonally than using a stick. Try lifting a trunk with your thumb vs. your hand.

[+5-11under]

I think the problem with this thread lies at the beginning. Two quotes:

 

"It's a scientific fact; let's see who can refute it", and "I will provide more arguments as needed and later".

 

The first quote is the thread/topic description, and the second quote is from the first post of the thread. The first brings on the challenge to disagree, and the second gives me the impression that this will be an argument rather than a discussion. The discussion is worthwhile; the arguing is lame. For instanced, ignoring the fact that this may be outside the scope of this thread, however much I'd love to discuss measurement of switch debounce in the analog and digital realms, there's no way I'm going to, because it'll just turn into an argument, it seems. Resistance to the original opinion is futile, apparently.

 

Ciao,

5-11under

[kool kitty89]

All I was stating there was it's easier to use with a lever (stick) to get to the buttons especially the diagonals and that digital would still be preferred over analog. However, the type of switches does make a difference in the bounce as I have noticed that even at 1Mhz sampling some digital joysticks show no bounce.

Right, some d-pads make for easier diagonals than others for sure (many prefer the Sega ones, especially for fighting games where you have a lot of combo moves that require accurate circular motion). You also have some gamepads with tumbsticks or optional thumbsticks to have more lever action. (some gravis gamepads, 7800 pads, and early SMS pads had screw points for that)

 

The bounce would depend on the switch type and additional springs/buffers (rubber boot, etc) with a number of variabled. But rubber dome switches with a very short throw (especially with a d-pad) should tend to have less bounce in general compared with metal switches. (with more specific cases on various switch types in general)

[NorthPole(heeHAW)]

I think the problem with this thread lies at the beginning. Two quotes:

 

"It's a scientific fact; let's see who can refute it", and "I will provide more arguments as needed and later".

 

The first quote is the thread/topic description, and the second quote is from the first post of the thread. The first brings on the challenge to disagree, and the second gives me the impression that this will be an argument rather than a discussion. The discussion is worthwhile; the arguing is lame. For instanced, ignoring the fact that this may be outside the scope of this thread, however much I'd love to discuss measurement of switch debounce in the analog and digital realms, there's no way I'm going to, because it'll just turn into an argument, it seems. Resistance to the original opinion is futile, apparently.

 

Ciao,

5-11under

 

 

For the above reason I think this thread should have been locked ages ago.

[atariksi]

I think the problem with this thread lies at the beginning. Two quotes:

 

"It's a scientific fact; let's see who can refute it", and "I will provide more arguments as needed and later".

 

The first quote is the thread/topic description, and the second quote is from the first post of the thread. The first brings on the challenge to disagree, and the second gives me the impression that this will be an argument rather than a discussion. The discussion is worthwhile; the arguing is lame. For instanced, ignoring the fact that this may be outside the scope of this thread, however much I'd love to discuss measurement of switch debounce in the analog and digital realms, there's no way I'm going to, because it'll just turn into an argument, it seems. Resistance to the original opinion is futile, apparently.

 

Ciao,

5-11under

 

You forgot to address any of the points like previously. There's no problems in the thread. It's just that some people come in and get emotional from time to time. As long as the emotionally biased people don't stick around and just come and go, it's okay with me.

 

It's a scientific fact because I have performed the experiment as well as those that I know here in my area. They know there's some obvious flaws in the analog joysticks. However, rather than turn a blind eye to the rest of the world, I left it open for others to refute it. I prefer the method of logic/mathematics. To me it's a far gone conclusion, analog joysticks have no use given the existence of digital joysticks and paddles, mice, etc. since all those give better control.

 

I'm all ears to logic, mathematics, or even experiments to prove/disprove what I stated in this thread.

[atariksi]

I think the problem with this thread lies at the beginning. Two quotes:

 

"It's a scientific fact; let's see who can refute it", and "I will provide more arguments as needed and later".

 

The first quote is the thread/topic description, and the second quote is from the first post of the thread. The first brings on the challenge to disagree, and the second gives me the impression that this will be an argument rather than a discussion. The discussion is worthwhile; the arguing is lame. For instanced, ignoring the fact that this may be outside the scope of this thread, however much I'd love to discuss measurement of switch debounce in the analog and digital realms, there's no way I'm going to, because it'll just turn into an argument, it seems. Resistance to the original opinion is futile, apparently.

 

Ciao,

5-11under

 

 

For the above reason I think this thread should have been locked ages ago.

 

I don't think you understood what he wrote. I already discussed bounce and it applies to both analog and digital controllers and doesn't show up in the experiments performed. It's a futile point but I have discussed it. So he's mistaken like he was previously when he went on about various analog controllers before he even read post #1 apparently.

[atariksi]

Whenever you want to evaluate ANYTHING for a purpose, you have to DEFINE a proper, consistent and readily observable METRIC. In this case one needs a metric for "better control". In 9 pages, no one has defined such a metric (likely because given the range of uses for a joystick, there are a huge range of metrics with conflicting outcomes)

See post #114 where I gave examples of games where there's a higher failure rate using analog joysticks vs. digital joysticks for same games played on same system. There's 4 ways people take a view:

 

(1) logical deduction/mathematical analysis (undeniably true)

(2) experimental data (which leads to the scientific fact idea)

(3) mental speculation-- something out of the blue according to emotional attachment, or mistaken views with no evidence

(4) blind following the blind-- because most people seem to be doing that so that must be right

 

You are in the catagory of (3). You just speculated something without any evidence. I have proof from (1) and (2). I can also speculate my views but that's all subjective. You are free to have your subjective views, but they don't hold up against the experiment/logic presented in this thread. Your example of cars doesn't apply either-- another misanalogy.

 

Please document where you (or anyone else) in the thread outlined the specific metrics used to evaluate the controllers.

 

I humbly suggest that it is you who fall into "atariksi category #3." You make wild speculation "its a scientific fact" and emotionally attack anyone who disagrees with you. You have been repeatedly asked for evidence meeting scientific rigor to support your statement, yet, oddly, refuse to provide it.

Read post #1 where my experiment of getting consistently higher scores using digital joysticks vs. analog joysticks on the same games. I have like 30 different systems at home-- some analog and some digital. I am quite used to all of them and all their joysticks so nothing emotionally against any of them. The higher scores relate to the high failure rate in parts of the games where the analog joysticks are highly vulnerable and I tried to capture some of those screen shots. It's a metric. Once again, I prefer the logic/science behind why things happen rather than just piling up tons of data. I would be perfectly happy if the experiment had shown analog joysticks give better control as I still maintain and use both types of systems and will continue to do so. I guess there are other reasons for using the systems besides just controllers. I still use PCs although I know their analog joysticks are inferior to Atari joysticks. So you're wrong and witnesses around me and my customers can attest to the fact that I support various platforms and use them.

 

atariksi has offered a theory (that he represents as fact) without a proper experiment (with appropriate controls) with results that have not been verified by the community at large.

No, you presented your mental speculation and are expecting people to accept you as some authority and blindly take as "fact". You have not countered anything that I presented in this thread. Perhaps, you will come back a few days from now and blurt out the samething again.

Actually, I am an authority on the subject.

 

Nonetheless, this is about your statement, not mine (in fact, I've made no claims as to which control may be superior)

You present a theory as fact, you produce no proper experiment, and no verification.

You can deny, redirect, obfuscate, call-names, whatever but you can't escape that fact.

 

You are the one name-calling by declaring my stuff wrong/emotional/etc. when it's a perfectly repeatable experiment. How is it a theory when I drew a conclusion based on the observation and measurements. It's a theory for you because you have yet to perform the experiment or study the science behind it.

 

So you are saying the effect of the controller on the system is NOT the defining characteristic???

 

Each of the effects mentioned (of which you claim to have examined one) increase the decoupling of the exact state of the controller from the exact state of the system. You have repeatedly expressed the opinion that a user knows the exact state of a digital joystick, thus making it better. I now claim that a user does NOT care about the exact state of the controller, but only the state of the system. When I am driving a car (for real or in a game), and wish to turn the vehicle, I have no interest in the exact position of the steering controller. I care only that the car turns appropriately. If I need to turn more, I move the controller more, and vice versa. I do NOT think "I need the pot on the steering axis to equal a measurement of 147 in the computer's steering axis coordinate system"

 

I was not aware that digital signals were electrically different from analog ones...

Please feel free to show me the difference.

All you did was give an example where you rely on feedback and then generalize from a specific case that user does NOT care about the exact state of the controller. But if you did know the exact state of the controller, you can still achieve the same result and go to an exact point directly and a priori (without waiting for feedback). That sums up the better control idea that I am trying to portray. Yes, I know all digital signals also are in one sense analog in the electrical terms but I stated that software doesn't have to deal with that level of electronics. If I can tap my digital joystick left and cause the the figure on the screen to move left one pixel (without even looking) that's better control than if I move the joystick and wait for figure to move before letting go and hoping it only went one pixel.

 

 

Let us look at the input port for a moment... If it is NOT latched, one could conceivably close a switch contact and then reopen it (a 'tap' if you will), and have the system miss it because the port wasn't sampled during the closure time. On the flip side, if the port IS latched, you may 'tap' the controller immediately after sampling and not have it recorded by the system until the next sample interval, thus delaying the 'tap' by one sample time. You can apply these type of effects to each part of the system. In essence, regardless of the controller, the physical movement and the impulse effect on the system is decoupled to some extent, making exact knowledge of controller state meaningless.

As I said, the bounce is detectable on some joysticks in order of a couple of microseconds and has no effect on sampling done at millisecond range. What matters is the software gets the state that you want it to get. Go compute the probability of what you wrote happening first before making it some sort of highlight of your post.

Assuming a normal distribution, the average delay should be 1/2fs, where fs is the effective sampling rate.

And I agree, what matters is the software gets the state you want it to get.

Thanks, we agree here.

 

No you didn't, which is why I repeat it.

In fact, your admission that your simulator is intentionally biased invalidates your results.

Repeat after me: "No simulator is exact." Show me ANY simulator, I will give you a parameter it doesn't model.

 

The joystick is a simple device so it can be exactly mimicked. Intentionally biasing data to favor the loser (analog joystick) doesn't invalidate the result. I can just as easily get the exact threshold of the games I use and plug those into the simulator and have it exactly mimic the joystick in that way. And for some games, I do know the thresholds, but didn't want to bother disassembling games and figuring out each one's thresholds.

[atariksi]

Simulators and not emulators. Simulator is suppose to exactly mimic the event. Paddles are not part of the dicussion but used an example to show that even amonst analog controller there's some that offer better control...

More later...I have to run to the JFK airport.

 

Apparently, there were two lanes blocked for construction near the Whitestone bridge while coming back from JFK so I won't have time to finish your reply tonight as I lost time in traffic.

[+Stephen]

It's a scientific fact because I have performed the experiment BIG SNIP

There is the entire problem, coupled with the fact that you could have 10 million pieces of evidence to the contrary. You feel that your opinion is the only one in the world that matters and nothing will ever change that. Congratulations on having such a wonderfully huge ego.

 

Guess what - I just did an experiment myself. I prefer KitKats to all other chocolate bars. Therefore, I claim that it's a scientific FACT that KitKat is the best chocolate in the world. Have fun refuting this. You can't argue with science.

[Divya16]

It's a scientific fact because I have performed the experiment BIG SNIP

There is the entire problem, coupled with the fact that you could have 10 million pieces of evidence to the contrary. You feel that your opinion is the only one in the world that matters and nothing will ever change that. Congratulations on having such a wonderfully huge ego.

 

Guess what - I just did an experiment myself. I prefer KitKats to all other chocolate bars. Therefore, I claim that it's a scientific FACT that KitKat is the best chocolate in the world. Have fun refuting this. You can't argue with science.

 

What evidence to the contrary do you have that you can't get exact states on analog joysticks whereas you can on digital joysticks? One piece of contrary evidence hasn't been given and you are posting as if it's 10 million than mixing things up with an analogy that doesn't even apply. It's not an opinion that people have better control using digital music/video to edit things exactly rather than in analog form. You can only perform the experiment for yourself not for others.

 

As far as bounce goes, the way I see it given the slow scanning of analog joysticks, it's more susceptible to errors there as well.

[+Stephen]

It's a scientific fact because I have performed the experiment BIG SNIP

There is the entire problem, coupled with the fact that you could have 10 million pieces of evidence to the contrary. You feel that your opinion is the only one in the world that matters and nothing will ever change that. Congratulations on having such a wonderfully huge ego.

 

Guess what - I just did an experiment myself. I prefer KitKats to all other chocolate bars. Therefore, I claim that it's a scientific FACT that KitKat is the best chocolate in the world. Have fun refuting this. You can't argue with science.

 

What evidence to the contrary do you have that you can't get exact states on analog joysticks whereas you can on digital joysticks? One piece of contrary evidence hasn't been given and you are posting as if it's 10 million than mixing things up with an analogy that doesn't even apply. It's not an opinion that people have better control using digital music/video to edit things exactly rather than in analog form. You can only perform the experiment for yourself not for others.

 

As far as bounce goes, the way I see it given the slow scanning of analog joysticks, it's more susceptible to errors there as well.

Chewbacca defense. It's an undisputable scientific fact that KitKat is the best chocolate in the world. I have the evidence to back this up. Where is your evidence to refute this?

[+orpheuswaking]

Chewbacca defense. It's an undisputable scientific fact that KitKat is the best chocolate in the world. I have the evidence to back this up. Where is your evidence to refute this?

 

But is that the British Kit-Kat or the American Kit-Kat... Becuase I tell you sir, that the British Kit-Kat is vastly superior in all ways to the subpar American Kit-Kat, It's a scientific fact!.

 

oh and who needs joysticks? I control my games with my MIND!!!!!

[potatohead]

Exact states is done with code, and a range check. Come on man, this is not hard. A range then, is "on", and another range is "off", done! I'm sure people here could suggest all kinds of options. Maybe just use a single value, and use greater than, less than for one and zero binary states. One value for each direction, and hmmm that begins to look a lot like the code checks for the digital sticks, does it not?