Digital Joysticks provide better control than Analog Joysticks

[Divya16]

This really should be simple to see. Both types of control have their place. To argue that only one is better in every case is just being stubborn. And you know it - there's no harm in admitting you are just trying to get a rise out of all of us.

 

But the facts show that digital joysticks provide superior control. No need to dismiss that the sun rises in the east under the rug if it's a fact.

[potatohead]

Actually, the facts DON'T do that.

 

Right now, you (two) are the only ones asserting that to be true. Nobody else is.

 

Refuted.

 

Now, if you are willing to admit that, we can qualify the "better" control bit, and actually have a discussion that's resolvable. As it stands right now, the only matter under discussion is whether or not you can come to acceptance on being very solidly refuted, and how nice you are about that.

 

If you need to feel better denying that, by all means do so. It's been very entertaining so far, but make no mistake, the record here shows you getting your ass kicked huge! And that record shows you asked for it, and were kind of shitty about it when the refuting started to look serious.

[kool kitty89]

The gravis in the picture in post #1 was the main analog joystick used. It allows for self-centering to be on or off. It allows for tightness, calibration, etc. Plus, the buttons are easily accessed and programmable and easily accessible like Atari 2600 joystick. It's not like you are trying to jump in donkey kong and having to press the button on the top of the stick which is already in motion. I did play around with a few other analog sticks but gravis by far gave the best results so that was the one used.

That sampling is WAY to small... and you used a flight stick, a controller designed for a very specific genre and not particularly well suited to many others.

I wouldn't use a flight stick to control a FPS, 3D platformer, or racer if I had a better option (ie a proper gamepad -though I prefer keyboard+mouse in some cases, and some racers are better with a wheel). For some cases, an "analog" gamepad is also quite good for flight sims though depending on the case it's inferior to a flight stick. (for more console oriented games, a short throw analog gamepad is far better, but sim type stuff fares better with a flight stick -especially if you need precision 4 axis control for roll/pitch/yaw and the throttle, hell I can even muddle through with digital control for some cases though not for others -for very arcade style games with forgiving/simple controls with no stalling, etc, digital can be acceptable -or some simple sim style games like the original wing commander, though I think Origin went out of their way to make it work well for analog -ideal- digital/cursor keys, and even mouse control -probably the best mouse control scheme for a 1st person flight/space sim I've ever used)

 

You also obviously failed to compare a wide enough sample of software and hardware platforms (you're missing well over 20 years of stuff there, and you'd need PC games and modern console games from 1996 or later to address that side of things).

Anyone not reasonably experienced across the board with modern and classic games from a full breadth of generations lacks the perspective to make any such claims on general control mechanisms.

 

 

If you chose an extremely simple, primitive game for a comparison, you'll also find that 1 button could be "superior" to a controller with a few (or many) buttons as you'd ignore games that absolutely require many button inputs. (or key inputs, but using a gamepad/joystick with many built-in buttons can often be preferable by far -games using a mouse favor keyboard a bit more, but gamepad/joystick is great to have with many buttons -joy to key is great for converting old keyboard specific games to comprehensive joysticks/gamepads).

Hell, there's games that choke with as many as 7/8 buttons (have to rely on combos), like the console ports of Wing Commander.

 

I've got several games where I max out the 12 buttons (plus 4 analog axes and 8 direction hat) on my Logitech Wingman pro joystick and STILL have to supplement some functions with the keyboard (obviously the most-used stuff is on the stick).

 

The compression algorithm applies here. There's no reason to have 12+ buttons if most games only use 1 or 2. Complexity affects control as well. The less the user has to think about in relation to the controller, the better. If most of your image is gray-scale 0..15, saving it as 24-bit is a waste of space. Just save as 4-bit nibbles and make some exceptional pointers where the content differs. So for joysticks, just have one or two buttons and if there are a few games that use 10+, use some auxiliary keypad (like Star Raiders) as there is no reason to add the confusion for all games. Some games purposely use the extra buttons for stuff that's unnecessary for the gameplay.

A game with few buttons on a controller with many buttons is just as simple as on a controller with few buttons... you'd either map those few controls to the most accessible buttons (ie on the Jaguar you'd use the 3 face buttons) and/or have custom control options for the user to set the preferred control maps (there's far too many games with sub-par control schemes that would strongly benefit from that as PCs often allow).

 

Case in point, even if you go back to really old systems, you've got cases where you's want to have easy access to a keypad for added control inputs, but as with analog control, it's newer games (PCs started before consoles) that really need a lot more buttons as well as analog control. (many needing 4 to 6 analog axes and several buttons) OTOH, the whole keypad thing never really caught on for console gaming (and the potential for cleaner control in PC ports was never taken advantage of), but many buttons persist and with full multi-press combinations rather than restricted single-key press of some older keypad controllers (albeit the Jaguar had multi-key support too).

 

It's much better to have more than you need than not enough... and a well designed controller will also make the most used buttons obvious so even people who are easily confused non-gamer/"casual" types won't be too confused in figuring out the controls when only a couple buttons are needed. Not only that, but having more buttons avoids the need for combos (or at least as many combos) so it simplifies control as well -in cases where it would otherwise become overly awkward and possibly require clunky menus for games needing fast action.

 

 

For the record, pure 2D oriented game consoles were pushing 8 buttons standard (or as a prominent accessory) from 1990/91 onwards (more so in 1993 with SFII pushing the TG-16/PCE and Genesis/MD to have "6-button" controllers -technically 8 buttons in both cases just like the SNES)

 

 

 

Having more functionality on the stock controllers means developers have far more freedom in what they can add while obviously retaining the support for simpler stuff not making use of that functionality.

 

With fewer buttons it can even be harder for a user to learn the controls without a manual or in-game directions (an often criticism of more complex games)... I had that problem with some VCS games that I got loose before I started poking around online: I didn't know that pressing down in asteroids would do hyperspace for example and a couple other cases like that where a 2nd button would have been obvious. (and with a 2nd button it's far tougher to accidentally use in-game) Some other games got cuts due to such, like Missile command only having 1 platform on the VCS, 5200, and original A8 version (there was the "+" version with 3 platforms though).

Even if you do release accessory controllers with added features, they're much less likely to get support than standard controllers. (the paddle controllers on the VCS were standard... the keypads got modest support and the driving controllers got only the original launch game even though they were released at the beginning of the VCS's life and offered interesting control capabilities)

 

 

 

 

 

 

 

That being staid, for the limited selection of comparisons you've made, I'd agree that the conclusions are valid in the context of those specific comparisons (is with the specific controllers and games compared). Furthermore, I'd agree that the majority of games from the era being generally looked at would favor a good quality switch based digital controller over an analog joystick, though there are exceptions to that: and cases made much more extreme by the programmign techniques used as well as the specific controller implementations. (a gamepad would be a more even comparison in many cases -as would the vectrex controller which is not ideal for 4/8-way games but still works miles better than pretty much any other analog joystick of the time and even works better for a number of analog-specific applications where quick response as well as high precision are needed -similar context to modern short-throw analog thumbsticks)

Edited December 21, 2010 by kool kitty89

[+Stephen]

But the facts show that digital joysticks provide superior control. No need to dismiss that the sun rises in the east under the rug if it's a fact.

You opinion shows that you can control a digital stick better. Your experiment that one person did also shows this. Keep arguing this non-fact. I'll refute every post you make just to prove I can be as stubborn as you. I have nothing but free time for the next two weeks, so keep it up.

[+poobah]

As someone who designs and executes experiments for a living, I humbly suggest that it is you, atariksi, who does not understand what a controlled experiment entails. Can we see your test plan, hmmmmmm?

Just for your information as I don't think some of these other people can understand this simple point:

 

If I apply skill S0 to game g with digital joystick D and analog joystick A then S0(g)*A+S0(g)*D is the experiment. Guess what-- if someone else with skill S1 uses same game g and digital joystick D and analog joystick A then S1(g)*A+S1(g)*D then those subjects S0..Sn get factored out. It's called the distributive property in simpler mathematics. Your skill applies equally to both games so you don't need hundreds of subjects to determine the results. Although other people can repeat the experiment for themselves. Newton performed the control experiment himself-- didn't depend on other people doing the samething for him to do the control experiment. The target is to determine which joystick provides better control not how various people fare at various games.

 

I don't know where to begin on your little equation, practically every aspect of it is incorrect, so I'll go with the most glaring math example: You can't factor out your skill variables. Although your notation uses the same names, they are in fact different variables and cannot be factored out, a common error in flawed proofs.

 

No test plan? (do you know what a test plan is?)

 

You are the idiot. If you can prove mathematically, you don't need to waste your time with experimenting. If I can prove that if 2(X*X) = 50 then X must be 5 or -5 then the experiment can only prove the SAME result. Experiments can only show a subset of what's proven mathematically or logically. Mathematics and logic apply for all time and in all cases.

Only if X is a real scalar.

Engineers often 'prove' stuff in MATLAB, only to have it fail miserably in the real world.

Would you care to use math/logic to tell me how big my monitor is, or how many keys on my keyboard?

Using deductive logic, you go to the spec or schematic rather than do experimental measurements. Mathematical proofs are always deductive so they cannot be disproven experimentally. Once you prove the formula for the solution to a quadratic equations, it's always going to be true.

 

The spec for my monitor has +/- tolerances, the # of keyboard keys varies depending on sub-model. You *must* measure to get the exact value.

Incidentally, all deductive proofs are built on base assumptions. If your assumptions are flawed, your proof is invalid. (I'll refer you to an excellent text The Deductive Foundations of Computer Programming, Manna & Waldinger for a much better treatise on the subject.)

 

You are mentally retarded.

Well, there's a well thought out, rational argument.

You took it out of context. He was calling me retarded earlier so it's right back at him since I refuted his point.

 

Seriously? You are going to go with "Well, he called me names first"?

 

Nope, all I said was I ran hundreds of experiments. The data that was submitted was for the most popular games that most people here are familiar with. It's sufficient to show my point.

No, it is not.

It sure is. The total number of experiments is aribtrary as long as you have sufficient number to show the original hypothesis-- namely that digital joysticks provide better control than analog joysticks and with no data coming showing the opposite.

 

And you do not have a sufficient number. You have vastly undersampled the "joystick operators" variable (among others).

 

I won the HSC using an analog joystick, beating out many competitors using digital joysticks. That is a single data item that disagrees with your hypothesis. (You may recall that disproving a "for all" hypothesis merely requires a single instance where it is false. - I again refer you to the text I mentioned earlier.)

 

And before you try to claim my skill invalidates the statement... (though you were the one trying to mathematically factor out skill.) You would need to establish that my superior skill overcame the handicap posed by the analog stick. In fact, I would have to be the most skilled player in the contest by at least the analog handicap, which you claim is huge. (and truly, Fandal and BountyBob are much better at these games than I am.)

[atariksi]

As someone who designs and executes experiments for a living, I humbly suggest that it is you, atariksi, who does not understand what a controlled experiment entails. Can we see your test plan, hmmmmmm?

Just for your information as I don't think some of these other people can understand this simple point:

 

If I apply skill S0 to game g with digital joystick D and analog joystick A then S0(g)*A+S0(g)*D is the experiment. Guess what-- if someone else with skill S1 uses same game g and digital joystick D and analog joystick A then S1(g)*A+S1(g)*D then those subjects S0..Sn get factored out. It's called the distributive property in simpler mathematics. Your skill applies equally to both games so you don't need hundreds of subjects to determine the results. Although other people can repeat the experiment for themselves. Newton performed the control experiment himself-- didn't depend on other people doing the samething for him to do the control experiment. The target is to determine which joystick provides better control not how various people fare at various games.

 

I don't know where to begin on your little equation, practically every aspect of it is incorrect, so I'll go with the most glaring math example: You can't factor out your skill variables. Although your notation uses the same names, they are in fact different variables and cannot be factored out, a common error in flawed proofs.

 

No test plan? (do you know what a test plan is?)

I'm so sorry, but the skill is of the same person using both types of joysticks and is thus factorable. When I played Miner 2049er with analog joystick and scored average of 35K and scored average of 48K with digital joystick-- I applied the same skills using those two joysticks. My skills didn't change. The controlled variable is ONLY the joysticks, and thus the conclusion from that experiment is that digital joystick provides better control. And I already listed five (5) items that back me up logically (inherently flawed analog joysticks). FYI, S(Miner2049er)*A < S(Miner2049er)*D, so given you have some skills to play the game (i.e, S(g)!=0), you divide by S(g) and thus A<D. QED.

 

You are the idiot. If you can prove mathematically, you don't need to waste your time with experimenting. If I can prove that if 2(X*X) = 50 then X must be 5 or -5 then the experiment can only prove the SAME result. Experiments can only show a subset of what's proven mathematically or logically. Mathematics and logic apply for all time and in all cases.

Only if X is a real scalar.

Engineers often 'prove' stuff in MATLAB, only to have it fail miserably in the real world.

Would you care to use math/logic to tell me how big my monitor is, or how many keys on my keyboard?

Using deductive logic, you go to the spec or schematic rather than do experimental measurements. Mathematical proofs are always deductive so they cannot be disproven experimentally. Once you prove the formula for the solution to a quadratic equations, it's always going to be true.

 

The spec for my monitor has +/- tolerances, the # of keyboard keys varies depending on sub-model. You *must* measure to get the exact value.

Incidentally, all deductive proofs are built on base assumptions. If your assumptions are flawed, your proof is invalid. (I'll refer you to an excellent text The Deductive Foundations of Computer Programming, Manna & Waldinger for a much better treatise on the subject.)

We're not talking about the base assumptions. We're talking about mathematical proofs. The proofs are ALWAYS true. If I gave you an Atari 800 computer w/GTIA and you tried to figure out the functionality experimentally, you may not even figure out half of what it's capable of, but if you went to the schematic and deduced it, you have proven it for ALL CASES and for ALL TIME. Logic/math is ALWAYS the superior method over limited experimentation. And experimentation is always limited since we don't have infinite time to do it.

 

You are mentally retarded.

Well, there's a well thought out, rational argument.

You took it out of context. He was calling me retarded earlier so it's right back at him since I refuted his point.

 

Seriously? You are going to go with "Well, he called me names first"?

Nope, he deserved it more since he was mistaken. You took it out of context by selectively quoting it without its context. I don't call people names but if they use that sort of arguing then I have the right to show that their own reasoning applies to them more.

 

Nope, all I said was I ran hundreds of experiments. The data that was submitted was for the most popular games that most people here are familiar with. It's sufficient to show my point.

No, it is not.

It sure is. The total number of experiments is aribtrary as long as you have sufficient number to show the original hypothesis-- namely that digital joysticks provide better control than analog joysticks and with no data coming showing the opposite.

 

And you do not have a sufficient number. You have vastly undersampled the "joystick operators" variable (among others).

Nope, I have taken joysticks that are very common and pretty much optimal. If I took some specialized or abnormal joysticks, then the data wouldn't be as good and could be uncertain to draw the conclusion. Given the specs for the various joysticks, there's no reason to think the results will be different for other joysticks.

 

I won the HSC using an analog joystick, beating out many competitors using digital joysticks. That is a single data item that disagrees with your hypothesis. (You may recall that disproving a "for all" hypothesis merely requires a single instance where it is false. - I again refer you to the text I mentioned earlier.)

 

And before you try to claim my skill invalidates the statement... (though you were the one trying to mathematically factor out skill.) You would need to establish that my superior skill overcame the handicap posed by the analog stick. In fact, I would have to be the most skilled player in the contest by at least the analog handicap, which you claim is huge. (and truly, Fandal and BountyBob are much better at these games than I am.)

 

Sorry, but your skill has to apply to the game for both joysticks. If you don't know how to use one of the joysticks or you have someone else play the digital part, the results aren't controlled. There are three things that take place in most of these games-- (1) the computer algorithm of the game gets to you and you lose a life, (2) your skill level wasn't sufficient to escape losing a life, or (3) the controller didn't do what you expect it to. I'm only interested in item (3) which occurs in the games. Thus, the skill level doesn't play a role at all as long as you can use both joysticks and can play the game. The failure rate narrows down the experiment to item (3). For example, in jumping in the ledges in the Donkey Kong pie screen, there's a high failure rate for analog joysticks.

 

You are never going to get a controlled experiment just by playing with one joystick. Nor can you draw any conclusion about either joystick. It doesn't matter how the others fair-- their skill level is different so you have the following: S0(g)*A vs. S1(g)*D. You can't draw any conclusion.

[atariksi]

The gravis in the picture in post #1 was the main analog joystick used. It allows for self-centering to be on or off. It allows for tightness, calibration, etc. Plus, the buttons are easily accessed and programmable and easily accessible like Atari 2600 joystick. It's not like you are trying to jump in donkey kong and having to press the button on the top of the stick which is already in motion. I did play around with a few other analog sticks but gravis by far gave the best results so that was the one used.

That sampling is WAY to small... and you used a flight stick, a controller designed for a very specific genre and not particularly well suited to many others.

Hold it right there. I said, I DID NOT use the flight stick. I tried a few other analog sticks, but gravis was giving the best control and more ways to configure it. Oh, that flight stick does have some advantages in that it uses a digital interface and speed up the sampling rate and make the samples more exact. Microsoft seemed to know some of the problems with the analog interface. I'll give you time to rethink your reply given this before I continue my reply.

[JamesD]

Medication tests aren't the samething since you are interested in what effects it has on various patients. In this case you are interested in what effects the two types of JOYSTICKS have on various games. As long as you know how to use both joysticks (which isn't a big deal), that's all that matters. If I play pac-man with both joysticks 100 times, then the skill level I apply for the game is the same-- only the controllers would make the difference. In fact, I can play better now than I did a few years but that still doesn't affect the relationship. The experiment activity is S(Pac-man)*A + S(Pac-man)*D. I guess you can use a scaling factor A and D since the specifics of the function S(g) can encompass everything else. It's an expression of logic in mathematical terms.

It's been mentioned before, but it needs to be mentioned again: This test is only valid for the person performing the test. If this wasn't true, then under the same conditions, different players would get the same score. Medication tests are similar in this way: the effect of the person is key. Everyone's body is different in the way they respond to medication; everyone's gaming ability is different in the way they respond to different joysticks.

 

You really should think about it before you reply. The test is valid for the joysticks not for the person as the skill isn't that important. The score's value isn't important; it's the comparison. Think about it some more or better yet, do the experiment.

 

You were wrong before and are wrong again.

Pay no attention to that man behind the curtain.

[JamesD]

Here is a math formula for you.

 

outrageous claim * pathetic support = fail squared

[potatohead]

fail^2 = EPIC FAIL.

 

Post #1 is the fail. All else is just entertainment.

[atariksi]

It's been mentioned before, but it needs to be mentioned again: This test is only valid for the person performing the test. If this wasn't true, then under the same conditions, different players would get the same score.

Go back to the F=ma analogy discussed earlier. Be consistent and say the samething about that. And your second sentence doesn't follow at all. I guess it's already answered earlier. You can only prove things experimentally for yourself. But there's NO WAY you can prove it's ONLY valid for the person performing the test. In fact, the opposite follows. If I perform the experiment that shows digital joysticks provide better control then I have NO REASON to think it won't work for others as the skills don't play a role. Just to PROVE my point using a simpler example: play Pac-man with paddles and then play with a digital joystick. Use all the skills you want, you will get inferior results with the paddles. And given the inferior control, only conclusion you can draw is that others will also experience similar failure. To say otherwise is just mental speculation (drivel).

 

Medication tests are similar in this way: the effect of the person is key. Everyone's body is different in the way they respond to medication; everyone's gaming ability is different in the way they respond to different joysticks.

 

People have gaming skills and people have ability to use controllers. Lets not mix the two. The ability to use joysticks isn't a big deal. Medication analogy doesn't apply at all.

 

Just repeating the same points (as this was refuted earlier) without taking the facts into account won't help you.

[atarian63]

uh.. no... :roll

 

Gaming de evolution/ Joystick (Atari) >Gamepad (Nintendo)> analog joystick> Kinect (MS)

bottom reached.

Which is your opinion, not a fact.

On a retrogaming site, it's easily in the majority. The simple FACT is analog is imprecise and therefore hard to adjust to. I guess if you are younger and grew up using an imprecise device you may think it's great as that is what you know.

Classic arcades are better with digital.

 

Not imprecise, but FAR more precise, but accuracy is an issue. Of course, that's assuming you're not talking about analog vs digital, but wide range (high precision) vs restricted low precision (high accuracy) controls. (the N64 controller is not analog at all for example -unless you count 3rd parties)

 

There's many, many applications where added precision is really necessary for the best gameplay, especially in many 3D genres, but other cases where low-precision high accuracy control is preferable. In both cases, there's other issues tied to form factor (human interfacing), quality of a product, programming, etc.

Tons of all digital controllers that have shitty accuracy and are worse than a decent quality medium/short throw analog joystick/thumbstick. (like some crappy 3rd party gamepads, the Intellivision disc, etc -to a lesser extent the colecovision controller, at least for any games needing the buttons -for stick only games you can get by pretty well using your thumb)

 

 

For most console games who grew up in the 90s, 8-way digital directional control was the norm, only in the tail end of the 90s did "analog" really come into play (and only standard on the N64). Prior to that it was pretty much limited to PC gaming and no consoles had had analog on standard controllers since the vectrex (real analog or "analog" like digital like the N64), though you did have accessories like the rare AX-1 gamepad for the genesis (intended for After Burner and Space harrier iirc -to cater to the analog arcade controls) and the flight stick on the 3DO and saturn. (and the racing wheel) "Analog" thumbsticks didn't appear on consoles until 1996. (N64, Saturn 3D controller, PS1 Dual Analog)

Point taken, I was mostly referring to 80's arcade classic period but the 90's had plent as well. I have many arecades that have analog controls of a wide variety like Star Wars (though I dont like that control) STUN runner, Vindicators etc. Some iterations just have too much control, I spend way too much time dealing with an analog controller rather than just playing the game. Main reason I returned HALO after buying it. Bad enough with 1 analog but both... Yeah it's the norm now so I dont play any of that type of game.

[+5-11under]

It's been mentioned before, but it needs to be mentioned again: This test is only valid for the person performing the test. If this wasn't true, then under the same conditions, different players would get the same score.

Go back to the F=ma analogy discussed earlier. Be consistent and say the samething about that. And your second sentence doesn't follow at all. I guess it's already answered earlier. You can only prove things experimentally for yourself. But there's NO WAY you can prove it's ONLY valid for the person performing the test. In fact, the opposite follows. If I perform the experiment that shows digital joysticks provide better control then I have NO REASON to think it won't work for others as the skills don't play a role. Just to PROVE my point using a simpler example: play Pac-man with paddles and then play with a digital joystick. Use all the skills you want, you will get inferior results with the paddles. And given the inferior control, only conclusion you can draw is that others will also experience similar failure. To say otherwise is just mental speculation (drivel).

 

Medication tests are similar in this way: the effect of the person is key. Everyone's body is different in the way they respond to medication; everyone's gaming ability is different in the way they respond to different joysticks.

 

People have gaming skills and people have ability to use controllers. Lets not mix the two. The ability to use joysticks isn't a big deal. Medication analogy doesn't apply at all.

 

Just repeating the same points (as this was refuted earlier) without taking the facts into account won't help you.

Sorry, I'm not going to find the F=ma analogy discussed earlier. I've no time for that. If I haven't stated my points on that, then others have, more eloquently.

 

To flip the coin, there's no way you can prove it's valid for others performing the test. It's your "experiment", and you should be the one to prove it's valid for others. Write up that report, and lay it bare, for all to see.

 

I'll play Pac-Man with paddles when you play WaveRace 64 with a digital (i.e. 9 states) joystick. Deal? I had an Etch-A-Sketch when I was a kid, so I'm guessing I might actual do okay with the paddles.

 

"The ability to use joysticks isn't a big deal", eh? Then what are we doing here, ~40 pages later?

[+poobah]

 

I'm so sorry, but the skill is of the same person using both types of joysticks and is thus factorable. When I played Miner 2049er with analog joystick and scored average of 35K and scored average of 48K with digital joystick-- I applied the same skills using those two joysticks. My skills didn't change. The controlled variable is ONLY the joysticks, and thus the conclusion from that experiment is that digital joystick provides better control. And I already listed five (5) items that back me up logically (inherently flawed analog joysticks). FYI, S(Miner2049er)*A < S(Miner2049er)*D, so given you have some skills to play the game (i.e, S(g)!=0), you divide by S(g) and thus A<D. QED.

Skill at a game with controller A != skill with controller B, they are dependent. You can't factor it.

 

3rd time: Let's see your test plan.

 

The spec for my monitor has +/- tolerances, the # of keyboard keys varies depending on sub-model. You *must* measure to get the exact value.

Incidentally, all deductive proofs are built on base assumptions. If your assumptions are flawed, your proof is invalid. (I'll refer you to an excellent text The Deductive Foundations of Computer Programming, Manna & Waldinger for a much better treatise on the subject.)

We're not talking about the base assumptions. We're talking about mathematical proofs. The proofs are ALWAYS true. If I gave you an Atari 800 computer w/GTIA and you tried to figure out the functionality experimentally, you may not even figure out half of what it's capable of, but if you went to the schematic and deduced it, you have proven it for ALL CASES and for ALL TIME. Logic/math is ALWAYS the superior method over limited experimentation. And experimentation is always limited since we don't have infinite time to do it.

You can't have a proof without base assumptions. For instance, your 'equation' above assumes that multiplication can be distributed over addition in your system.

If logic and math were always superior, you could tell me the exact size of my monitor without measuring, which you cannot. QED

By your reasoning, we would never have to conduct experiments.

 

You are mentally retarded.

Well, there's a well thought out, rational argument.

You took it out of context. He was calling me retarded earlier so it's right back at him since I refuted his point.

Seriously? You are going to go with "Well, he called me names first"?

Nope, he deserved it more since he was mistaken. You took it out of context by selectively quoting it without its context. I don't call people names but if they use that sort of arguing then I have the right to show that their own reasoning applies to them more.

 

No. there's no room for that sort of thing in a rational discussion.

 

And you do not have a sufficient number. You have vastly undersampled the "joystick operators" variable (among others).

Nope, I have taken joysticks that are very common and pretty much optimal. If I took some specialized or abnormal joysticks, then the data wouldn't be as good and could be uncertain to draw the conclusion. Given the specs for the various joysticks, there's no reason to think the results will be different for other joysticks.

 

By "joystick operators" I mean various participants in the experiment using the controllers. Your 'experiments' only involve you, so any conclusion can only apply to you.

 

I won the HSC using an analog joystick, beating out many competitors using digital joysticks. That is a single data item that disagrees with your hypothesis. (You may recall that disproving a "for all" hypothesis merely requires a single instance where it is false. - I again refer you to the text I mentioned earlier.)

 

And before you try to claim my skill invalidates the statement... (though you were the one trying to mathematically factor out skill.) You would need to establish that my superior skill overcame the handicap posed by the analog stick. In fact, I would have to be the most skilled player in the contest by at least the analog handicap, which you claim is huge. (and truly, Fandal and BountyBob are much better at these games than I am.)

 

Sorry, but your skill has to apply to the game for both joysticks. If you don't know how to use one of the joysticks or you have someone else play the digital part, the results aren't controlled. There are three things that take place in most of these games-- (1) the computer algorithm of the game gets to you and you lose a life, (2) your skill level wasn't sufficient to escape losing a life, or (3) the controller didn't do what you expect it to. I'm only interested in item (3) which occurs in the games. Thus, the skill level doesn't play a role at all as long as you can use both joysticks and can play the game. The failure rate narrows down the experiment to item (3). For example, in jumping in the ledges in the Donkey Kong pie screen, there's a high failure rate for analog joysticks.

 

You are never going to get a controlled experiment just by playing with one joystick. Nor can you draw any conclusion about either joystick. It doesn't matter how the others fair-- their skill level is different so you have the following: S0(g)*A vs. S1(g)*D. You can't draw any conclusion.

 

We have that their skill level is greater than mine.

ie: S(g_poobah) < S(g_fandal), so for your equality to hold, A is necessarily greater than D. (not that I agree with any of this alleged math you are trying to pull, but your own argument falls apart in this case)

[atariksi]

It's been mentioned before, but it needs to be mentioned again: This test is only valid for the person performing the test. If this wasn't true, then under the same conditions, different players would get the same score.

Go back to the F=ma analogy discussed earlier. Be consistent and say the samething about that. And your second sentence doesn't follow at all. I guess it's already answered earlier. You can only prove things experimentally for yourself. But there's NO WAY you can prove it's ONLY valid for the person performing the test. In fact, the opposite follows. If I perform the experiment that shows digital joysticks provide better control then I have NO REASON to think it won't work for others as the skills don't play a role. Just to PROVE my point using a simpler example: play Pac-man with paddles and then play with a digital joystick. Use all the skills you want, you will get inferior results with the paddles. And given the inferior control, only conclusion you can draw is that others will also experience similar failure. To say otherwise is just mental speculation (drivel).

 

Medication tests are similar in this way: the effect of the person is key. Everyone's body is different in the way they respond to medication; everyone's gaming ability is different in the way they respond to different joysticks.

 

People have gaming skills and people have ability to use controllers. Lets not mix the two. The ability to use joysticks isn't a big deal. Medication analogy doesn't apply at all.

 

Just repeating the same points (as this was refuted earlier) without taking the facts into account won't help you.

Sorry, I'm not going to find the F=ma analogy discussed earlier. I've no time for that. If I haven't stated my points on that, then others have, more eloquently.

Only points that were refuted. Newton conducted the experiment and he can say it's good for others based on his tests just like I can. But YOU CAN NEVER SHOW that it's only good for me since to do that you would have to go and try it for every person in existence. So your remark is just speculation (drivel).

 

To flip the coin, there's no way you can prove it's valid for others performing the test. It's your "experiment", and you should be the one to prove it's valid for others. Write up that report, and lay it bare, for all to see.

It's not flip the coin. To say something so specific from an experiment like "only you" implies knowledge of all other people in existence. I hope that's clear now.

 

I'll play Pac-Man with paddles when you play WaveRace 64 with a digital (i.e. 9 states) joystick. Deal? I had an Etch-A-Sketch when I was a kid, so I'm guessing I might actual do okay with the paddles.

Did you even notice the point being made-- that the CONTROLLER makes a difference that doesn't depend on the skill of how you play the game. You are stretching your imagination if you think you can do as good in pac-man with a paddle as with a digital joystick. You need to try it out and not speculation like you do most of the time. It's obvious for people who have used paddles and joysticks. It's easy to try if you can't think it out-- just hook up two paddles to Atari 5200 version of pac-man.

 

"The ability to use joysticks isn't a big deal", eh? Then what are we doing here, ~40 pages later?

You don't need much skills or training to use a digital or analog joystick. I see you missed this easy point as well. Learning to master games is a longer time training event whereas using joysticks being discussed isn't.

[atariksi]

 

I'm so sorry, but the skill is of the same person using both types of joysticks and is thus factorable. When I played Miner 2049er with analog joystick and scored average of 35K and scored average of 48K with digital joystick-- I applied the same skills using those two joysticks. My skills didn't change. The controlled variable is ONLY the joysticks, and thus the conclusion from that experiment is that digital joystick provides better control. And I already listed five (5) items that back me up logically (inherently flawed analog joysticks). FYI, S(Miner2049er)*A < S(Miner2049er)*D, so given you have some skills to play the game (i.e, S(g)!=0), you divide by S(g) and thus A<D. QED.

Skill at a game with controller A != skill with controller B, they are dependent. You can't factor it.

 

3rd time: Let's see your test plan.

I guess we disagree regarding game skills and using joysticks. They are separate. A new gamer can easily learn to use the joysticks but his game mastery requires quite some time. I don't know for what you are asking "test plan". If it's for these joysticks, they were tested with hundreds of games with some of them listed in post #114.

 

We're not talking about the base assumptions. We're talking about mathematical proofs. The proofs are ALWAYS true. If I gave you an Atari 800 computer w/GTIA and you tried to figure out the functionality experimentally, you may not even figure out half of what it's capable of, but if you went to the schematic and deduced it, you have proven it for ALL CASES and for ALL TIME. Logic/math is ALWAYS the superior method over limited experimentation. And experimentation is always limited since we don't have infinite time to do it.

You can't have a proof without base assumptions. For instance, your 'equation' above assumes that multiplication can be distributed over addition in your system.

If logic and math were always superior, you could tell me the exact size of my monitor without measuring, which you cannot. QED

By your reasoning, we would never have to conduct experiments.

You can conduct experiments whenever you want. My point is that if you have logic/mathematical proofs, they are superior to experiments. That's why I listed those catagories earlier:

 

(1) Logic/Math

(2) Experiments

(3) Mental Speculation/Emotional

(4) Blind following the blind

 

They are in order of importance. For example, 5-11's claim that "it's only valid for me" is catagory (3). He has not conducted any experiment nor has he any logical/mathematical proof. Going back to your analogy of the monitor, the spec. does tell you the exact size as that's based on what the manufacturer of the monitor used. If you measure it experimentally, you may or may not get it to the exact size of the manufacturer especially if it was some irrational number like PI*8 by PI*4. See even in your own analogy, it favors the logic/math over experiment. And I gave you my own analogies-- trying to determine the functionality of the GTIA chip, trying to prove the quadratic equation, etc. It's based on assumptions which are accepted by both parties as obvious. For example,

 

All men are mortal.

Socrates is a man.

Socrates is mortal.

 

I think you got your "quotes" intermingled below.

 

...You are never going to get a controlled experiment just by playing with one joystick. Nor can you draw any conclusion about either joystick. It doesn't matter how the others fair-- their skill level is different so you have the following: S0(g)*A vs. S1(g)*D. You can't draw any conclusion.

 

We have that their skill level is greater than mine.

ie: S(g_poobah) < S(g_fandal), so for your equality to hold, A is necessarily greater than D. (not that I agree with any of this alleged math you are trying to pull, but your own argument falls apart in this case)

 

Did fandal even participate first of all? Secondly, skill level inequality you state is speculative. Maybe he isn't as good at some games. However, I agree if you can establish S(poobah) < S(fandal) and fandal uses digital joystick and you use analog joystick for the same game and you played enough times, that would be okay.

 

My experience with HSC is that I sometimes play a game only once since I don't like the game and some others I play many many times so there's that uncontrolled factor which wouldn't exist if you went by the same person using both joysticks.

[JamesD]

Only points that were refuted. Newton conducted the experiment and he can say it's good for others based on his tests just like I can. But YOU CAN NEVER SHOW that it's only good for me since to do that you would have to go and try it for every person in existence. So your remark is just speculation (drivel).

Ah Newton.

Newton presented his methodology as a set of four rules for scientific reasoning.

These rules were stated in the Principia and proposed that (1) we are to admit no more causes of natural things such as are both true and sufficient to explain their appearances, (2) the same natural effects must be assigned to the same causes, (3) qualities of bodies are to be esteemed as universal, and (4) propositions deduced from observation of phenomena should be viewed as accurate until other phenomena contradict them.

Link

 

Take a close look at #4. The winner of the games competition is an example of a phenomena that contradicts your propositions. Even Newton says you are wrong.

 

You didn't even show your results are true for every person on this board let alone every person on the planet.

<edit>

Oh, and lets not forget every game.

Edited December 22, 2010 by JamesD

[potatohead]

Let's keep focus.

 

The item to be refuted is:

 

Digital Joysticks give BETTER control than Analog.

 

Under what conditions? For whom? Which games?

 

There is no qualification! FAIL. Refuted (easily)

 

Now, these other mini-discussions where we are drilling down to specific cases are GOOD discussions. There exists some potential for consensus, and or just some great tech discussion, which is all good.

 

, BUT

 

Validating those doesn't validate:

 

1. Initial premise, which is deeply flawed, and not something that is possible to agree on, without also accepting a LOT of assumptions. It's like a fucking push poll, or some other twisted advocacy form. Shit.

 

2. Claims of ignorance, speculation, and other rather ugly personal attacks, delivered by our "Mr. Scientist", who didn't bother to formulate a challenge that's possible and practical. (if it were, we would not be here so many posts later)

 

This one is particularly galling because a quick look at the comment history of every contributor here, minus the female mystery persona, reveals people who will easily admit a point taken fairly, and solid conduct at most times.

 

Notably, Atariski DOES NOT DO THAT, EVEN WHEN CONFRONTED WITH COMPELLING INFORMATION. That's insulting to everybody here, and it's the primary reason why several of us are entertaining this discussion. If we are to talk of accounting and consideration for people, that absolutely must be resolved, or we are complete fools for allowing otherwise. (thus, my post) This, for lack of a better word, is just shitty, and not something that should slide, given the absolute refusal and brutal attacks seen when it's obvious it should be done.

 

I don't like shitty, and when somebody is arrogant enough to ask for it, THEN GET SHITTY? Yeah, game on!

 

3. What the fuck is it with the spare persona? Is it two people, one, what?

 

Those things must be resolved here. A burden was put onto people, and a lot of shit was delivered to those people, all with the implied claim of purity, ethics, and other lofty things, simply not demonstrated in the discussion at hand. That is simply not ok.

 

Atariski, you have no authority on these matters, being no different from anybody else here, and failing to give everybody else here the same consideration you claim is due. You owe people for that, and this eye for an eye crap, mixed with your arrogant indignity at some of us dishing it out your way hard, is NOT OK.

 

You need to own up to the fact that you did not qualify your initial statements, then proceeded to imply that everybody who brought that to light was ignorant, stupid, insulting, you name it.

 

Fuck that!

 

If you don't own that, I will personally begin to make it very, very ugly for you. No holds barred ugly. Do you understand that?

 

You asked for the challenge, then got pissed and weaseled out every way possible, nailing most of the items on the lists, most of the time, at most of the people.

 

The fact that you got your ass handed to you isn't anyone's problem, as we all simply were doing it AT YOUR REQUEST!

 

Do you withdraw that request now? Does it make sense to admit the initial statement was not qualified, continuing on to those cases where it is possible to reach some agreement and have a rational debate, sans your usual garbage? Scratch that, you can include the garbage, and it will be entertaining, so long as the point of discussion is one with sufficient qualification to be productive, which the current one is not.

 

For the readers at home, what is happening right now is a combination of "distract and persuade", where sub-points, similar to the flawed one are being raised, politely discussed, so Atariski can build repore with others, in a attempt to win the hearts and minds, to avoid accounting for the shit delivered early on, and to avoid having to account for being completely bested at his own challenge.

 

Petty on my part? Absolutely!

 

He asked for it, and I fully intend on delivering on that, until such time as he withdraws that request.

Edited December 22, 2010 by potatohead

[kool kitty89]

The gravis in the picture in post #1 was the main analog joystick used. It allows for self-centering to be on or off. It allows for tightness, calibration, etc. Plus, the buttons are easily accessed and programmable and easily accessible like Atari 2600 joystick. It's not like you are trying to jump in donkey kong and having to press the button on the top of the stick which is already in motion. I did play around with a few other analog sticks but gravis by far gave the best results so that was the one used.

That sampling is WAY to small... and you used a flight stick, a controller designed for a very specific genre and not particularly well suited to many others.

Hold it right there. I said, I DID NOT use the flight stick. I tried a few other analog sticks, but gravis was giving the best control and more ways to configure it. Oh, that flight stick does have some advantages in that it uses a digital interface and speed up the sampling rate and make the samples more exact. Microsoft seemed to know some of the problems with the analog interface. I'll give you time to rethink your reply given this before I continue my reply.

 

The rest still applies... unless I'm mistaken and there's a dozen or more controllers you also compared with several other systems and a massive array of games. (all played to the extent of being competently familiar)

[atariksi]

By "joystick operators" I mean various participants in the experiment using the controllers. Your 'experiments' only involve you, so any conclusion can only apply to you.

One more point regarding this (although already refuted in my reply to 5-11 and my previous reply to you), you can repeat the experiment but to be a valid experiment, that's not a necessity to have many participants repeat it to begin with. Take the easier example of playing pac-man, donkey kong, etc. with paddles and digital joystick. It's quite easy to see that the problems will be present for others as well. It's speculation to think -- "my experiment only applies to you." There's a principle of induction involved in experimentation whereas the logic/math proofs are deductive. A big difference. There are many scientific experiments performed where the number participants do not have to be large. I can do the F=ma experiment myself.

 

We have that their skill level is greater than mine.

ie: S(g_poobah) < S(g_fandal), so for your equality to hold, A is necessarily greater than D. (not that I agree with any of this alleged math you are trying to pull, but your own argument falls apart in this case)

 

I just looked up one of your games that you won -- pastfinder and there was no fandal playing nor anything in that thread suggesting that inequality. In general, a controlled experiment has only one variable that is the target for the experiment and you are playing around with skill level differences as well as joystick differences. That complicates things. And some games show higher failure rate with analog joysticks than others. You want the games and joysticks to be the prominent thing of the experiment not the person who plays them.

[+poobah]

I don't know for what you are asking "test plan". If it's for these joysticks, they were tested with hundreds of games with some of them listed in post #114.

When real scientists conduct an experiment, they write a test plan. It documents exactly what actions will be done, what measurements will be taken, how those measurements will be taken, etc. Details about how cables are connected, configuration of test equipment, etc. are all outlined. A 'simple experiment' can easily have a 40 page+ test plan.

 

The purpose is to force the researcher to design a proper experiment, ensure that the experiment is conducted as designed, and most importantly provide for peer review.

 

With the test plan, another interested individual can duplicate the exact testing conditions and can independently verify the results. If you are going to offer your conclusions up for review (say in a published paper, or on AA), you need to provide the reader with the ability to verify your claims.

 

You can conduct experiments whenever you want. My point is that if you have logic/mathematical proofs, they are superior to experiments. That's why I listed those catagories earlier:

 

(1) Logic/Math

(2) Experiments

(3) Mental Speculation/Emotional

(4) Blind following the blind

 

They are in order of importance. For example, 5-11's claim that "it's only valid for me" is catagory (3). He has not conducted any experiment nor has he any logical/mathematical proof.

 

5-11's analysis is correct (he's analyzing the limited data you provided about your experiment) Your experience with a joystick is can't be applied to the general population. It is only valid for YOU. Several possible explanations for the phenomena you observed have been reported. (this is the peer review part). In order for you to defend your assertion, you must find a way to extend your observation to the general populace. Normally you would accomplish this by repeating your experiment across a larger set of the population. Repeatability is fundamental to the scientific method.

 

Going back to your analogy of the monitor, the spec. does tell you the exact size as that's based on what the manufacturer of the monitor used. If you measure it experimentally, you may or may not get it to the exact size of the manufacturer especially if it was some irrational number like PI*8 by PI*4. See even in your own analogy, it favors the logic/math over experiment.

If the spec says 27 inches, and I measure 26.9, the exact size is 26.9, period, end of story. Measurement produces the true value, the spec is only an approximation. Your logic/math argument would require ever single monitor of this model to be the exact same size, but that is impossible to accomplish, there will be some variances. math/logic fails here.

 

And I gave you my own analogies-- trying to determine the functionality of the GTIA chip, trying to prove the quadratic equation, etc. It's based on assumptions which are accepted by both parties as obvious. For example,

 

All men are mortal.

Socrates is a man.

Socrates is mortal.

 

And therein lies the fundamental flaw in your arguments. Your assumptions are accepted by YOU, but not anyone else. removing those assumptions destroys your proof. If your assumptions are not accepted, you must prove them as well.

 

...You are never going to get a controlled experiment just by playing with one joystick. Nor can you draw any conclusion about either joystick. It doesn't matter how the others fair-- their skill level is different so you have the following: S0(g)*A vs. S1(g)*D. You can't draw any conclusion.

 

We have that their skill level is greater than mine.

ie: S(g_poobah) < S(g_fandal), so for your equality to hold, A is necessarily greater than D. (not that I agree with any of this alleged math you are trying to pull, but your own argument falls apart in this case)

 

Did fandal even participate first of all? Secondly, skill level inequality you state is speculative. Maybe he isn't as good at some games. However, I agree if you can establish S(poobah) < S(fandal) and fandal uses digital joystick and you use analog joystick for the same game and you played enough times, that would be okay.

 

My experience with HSC is that I sometimes play a game only once since I don't like the game and some others I play many many times so there's that uncontrolled factor which wouldn't exist if you went by the same person using both joysticks.

 

Replace fandal with bountybob, or any of a number of HSC participants who are better than me at these games.

In order for your argument to hold, on any game where I won (or even tied), A is necessarily greater than D, and we only need one case where that is true to disprove your assertion. Since we have many instances where this is true, your assertion is refuted.

[+5-11under]

It's been mentioned before, but it needs to be mentioned again: This test is only valid for the person performing the test. If this wasn't true, then under the same conditions, different players would get the same score.

Go back to the F=ma analogy discussed earlier. Be consistent and say the samething about that. And your second sentence doesn't follow at all. I guess it's already answered earlier. You can only prove things experimentally for yourself. But there's NO WAY you can prove it's ONLY valid for the person performing the test. In fact, the opposite follows. If I perform the experiment that shows digital joysticks provide better control then I have NO REASON to think it won't work for others as the skills don't play a role. Just to PROVE my point using a simpler example: play Pac-man with paddles and then play with a digital joystick. Use all the skills you want, you will get inferior results with the paddles. And given the inferior control, only conclusion you can draw is that others will also experience similar failure. To say otherwise is just mental speculation (drivel).

 

Medication tests are similar in this way: the effect of the person is key. Everyone's body is different in the way they respond to medication; everyone's gaming ability is different in the way they respond to different joysticks.

 

People have gaming skills and people have ability to use controllers. Lets not mix the two. The ability to use joysticks isn't a big deal. Medication analogy doesn't apply at all.

 

Just repeating the same points (as this was refuted earlier) without taking the facts into account won't help you.

Sorry, I'm not going to find the F=ma analogy discussed earlier. I've no time for that. If I haven't stated my points on that, then others have, more eloquently.

Only points that were refuted. Newton conducted the experiment and he can say it's good for others based on his tests just like I can. But YOU CAN NEVER SHOW that it's only good for me since to do that you would have to go and try it for every person in existence. So your remark is just speculation (drivel).

 

To flip the coin, there's no way you can prove it's valid for others performing the test. It's your "experiment", and you should be the one to prove it's valid for others. Write up that report, and lay it bare, for all to see.

It's not flip the coin. To say something so specific from an experiment like "only you" implies knowledge of all other people in existence. I hope that's clear now.

 

I'll play Pac-Man with paddles when you play WaveRace 64 with a digital (i.e. 9 states) joystick. Deal? I had an Etch-A-Sketch when I was a kid, so I'm guessing I might actual do okay with the paddles.

Did you even notice the point being made-- that the CONTROLLER makes a difference that doesn't depend on the skill of how you play the game. You are stretching your imagination if you think you can do as good in pac-man with a paddle as with a digital joystick. You need to try it out and not speculation like you do most of the time. It's obvious for people who have used paddles and joysticks. It's easy to try if you can't think it out-- just hook up two paddles to Atari 5200 version of pac-man.

 

"The ability to use joysticks isn't a big deal", eh? Then what are we doing here, ~40 pages later?

You don't need much skills or training to use a digital or analog joystick. I see you missed this easy point as well. Learning to master games is a longer time training event whereas using joysticks being discussed isn't.

First of all, write the experiment/study report. This is so passionate for you, it should be an enjoyable experience.

Regarding the "experiment" and the individual performing the experiment. I've said my points; you've said yours, and I have no new information to add. I will underline this, though: Just remember that the individual is a key variable in the "experiment", that cannot be ignored. I hope that's clear now.

I actually think I could do quite well with the paddles. By the way, thanks for letting us discuss paddles again, because paddles are awesome. I like the idea of separating the X from the Y, as far as control goes. Paddles provide better control than digital joysticks. I will provide more arguments as needed and later.

[potatohead]

The Atariski formula

 

1. Look for faults in others

 

2. Ignore own faults, invoking any plausible reason

 

3. Make others the subject, while claiming purity, or implying authority

 

4. Wash, rinse, repeat.

[NorthPole(heeHAW)]

The Atariski formula

 

1. Look for faults in others

 

2. Ignore own faults, invoking any plausible reason

 

3. Make others the subject, while claiming purity, or implying authority

 

4. Wash, rinse, repeat.

 

 

Where's the 'emotion'?

[+Stephen]

The Atariski formula

 

1. Look for faults in others

 

2. Ignore own faults, invoking any plausible reason

 

3. Make others the subject, while claiming purity, or implying authority

 

4. Wash, rinse, repeat.

You forgot :

1 - Chewbacca defense / straw-man argument

2 - Ignore evidence in posts that counter his arguments

3 - Believe that his opinion is fact and everyone who disagrees is wrong and or stupid

4 - Rely on logic more than Admiral Spock while supposedly ignoring emotion

5 - MOST IMPORTANT - he is never wrong, has never been wrong, and will never be wrong