jbejarano
New member
- Jul 6, 2012
- 893
- 0
I've seen interesting points of view on these fora about where randomness can be inserted into ball movement physics and where it shouldn't be. There are those who think absolutely every movement should be as regular as clockwork, and those who enjoy a bit of unpredictability tossed in all over the place.
First of all, I'm wondering if indeed there is any "randomness" (i.e. effects partially governed by pseudo-random numbers genereated within the application) anywhere within The Pinball Arcade. Or, are the "random" effects simply a high degree of precision calculated as a result of a spinning round object hitting another round object. Ask any baseball fan. A spinning round object (the ball) being hit by another round object (the bat) creates all kinds of unpredictable effects that can easily be perceived as random and can have dramatically different outcomes on the game (fair or foul, caught or in play, etc.). FarSight would have to weigh in on whether pseudo-random numbers play a role in determining their physics (if that isn't a corporate secret).
Next, if indeed psuedo-random numbers play a role in governing the physics of the ball, where would it be appropriate that they do so?
The best objection to the use of "randomness" that I've heard would be on balls entering, traveling along or exiting ramps, loops, orbits, habitrails, channels, turnarounds, inlanes, outlanes, etc. Even in real life, the behavior of a ball should be reasonably predictable as it travels along and interacts with these features. Yes, the spin of the ball and its momentum may cause it to have small variability, but there is hardly the need to add extra unpredictability to enhance gameplay.
The best argument FOR the use of "randomness" would be eject holes and auto-plunges. In these cases, if they produce a ball that ALWAYS behaves the same way, it seems to detract from the realism and the enjoyment of the game. Think what it would be like if an auto-plunged ball in Medieval Madness ALWAYS hit the exact same rollover for the "Skill Shot". Not much skill involved, is there? Or think of the eject hole near the top of Scared Stiff. This doesn't ALWAYS behave the same way, but imagine if it ALWAYS went back up to the deadheads. These days (at least on an iPad), it already does it enough to seriously menace the enjoyment of the game. It would be a travesty if it did that even more reliably. Remember that when a ball enters an eject hole or is about to be auto-plunged, it typically has lost any of its spin or momentum, so you couldn't rely on that to provide any variability to the ball's action. (EDIT: In response to a post below, I thought of the example of the top eject hole in Firepower. The ball ALWAYS comes back through the "R" rollover. There's no way it should do that so consistently. This is an instance where a soupcon of randomness could be helpful.)
The other reason to defend the use of pseudo-random numbers in eject holes and auto-plunges has to do with the mechanical nature of them. It's quite plausible that a mechanical feature on a real table will not fire off with precisely the same force every time whereas its digital re-creation might. This seems like a reasonable justification for adding "randomness". The same argument could even be extended to slingshots, jet bumpers and other features that reflexively react to the ball by pushing it away with mechanical force.
Where this argument gets murkier is with sharp corners and posts. These features don't have a mechanical element to them in real life, so the justification mentioned in the above paragraph doesn't apply. However, a ball coming off of them can't react too predictably either as they certainly don't in real life. In real life, the precision of the ball and the sharp corner or post is down to the quantum mechanical level. I would hardly expect FarSight's physics engine to calculate precision to that level. However, the precision should be very small indeed. And, to the extent that behavior within one angular "pixel" of this precision of the digital object isn't calculated, it's reasonable to add a pseudo-random number to resolve any plausible action that would be calculated within that lowest precision. The better the precision, the better the game and the less need for that type of "below the grain" randomness. However, this precision is a matter for the physics engine software developers to decide on a platform-by-platform basis.
While not strictly a "round object" phenomenon, I think this same principle applies to manual plunges as well. The greater the precision of the plunge (subject to UI limitations), the better the game, but as above, randomness could account for variability at a grain lower than that precision.
Then we have the flippers. Like the eject holes, auto-plunges, slingshots, and jet bumpers, they are mechanical devices, so one could argue that randomness is justified. However, these specific devices are special in that they are the primary things that the player uses to interact with the game. Unpredictable behavior from the flippers, may make users think the game is merely a game of chance, and discard it. A bit more predictability from the flippers gives the player a feeling of control that is quite important for the user experience. Perhaps randomness at a level lower than the precision of the digital objects as described above could be justified, but giving the player a sense of control over the flippers is far more important than that of other features of the table. I'd hesitate to use "randomness" with them.
The last point would be the nudges. Right now, the nudges seem to be a bit binary. Either you nudged or you didn't and the nudge always seems to be about the same amount. I doubt that players can nudge with quite that level of precision in real life, so some randomness does seem warranted here. However, I'm sure some expert players will disagree and claim to be able to nudge real life tables with very high precision. For touch nudging, I think some randomness is needed so as to not allow players to begin to use nudging in as reliable a fashion as the flippers. But, for nudging initiated by accelerometers in mobile devices or console controllers, some less binary concept of nudging seems like it could provide a better level of control. Even then, a bit of randomness below the level of the UI's nudging precision could be warranted.
I would especially love anyone with knowledge of the physics engine (and the authority to speak about it) to chime in with the whys and wherefores of randomness being inserted into the physics algorithms.
I like a game where I feel as if I'm in control, but where unpredictable things can and do happen and I have a chance to try to meet those challenges. What are your thoughts? Where do you come down on randomess versus clockwork-like predictability? Did I miss anything?
First of all, I'm wondering if indeed there is any "randomness" (i.e. effects partially governed by pseudo-random numbers genereated within the application) anywhere within The Pinball Arcade. Or, are the "random" effects simply a high degree of precision calculated as a result of a spinning round object hitting another round object. Ask any baseball fan. A spinning round object (the ball) being hit by another round object (the bat) creates all kinds of unpredictable effects that can easily be perceived as random and can have dramatically different outcomes on the game (fair or foul, caught or in play, etc.). FarSight would have to weigh in on whether pseudo-random numbers play a role in determining their physics (if that isn't a corporate secret).
Next, if indeed psuedo-random numbers play a role in governing the physics of the ball, where would it be appropriate that they do so?
The best objection to the use of "randomness" that I've heard would be on balls entering, traveling along or exiting ramps, loops, orbits, habitrails, channels, turnarounds, inlanes, outlanes, etc. Even in real life, the behavior of a ball should be reasonably predictable as it travels along and interacts with these features. Yes, the spin of the ball and its momentum may cause it to have small variability, but there is hardly the need to add extra unpredictability to enhance gameplay.
The best argument FOR the use of "randomness" would be eject holes and auto-plunges. In these cases, if they produce a ball that ALWAYS behaves the same way, it seems to detract from the realism and the enjoyment of the game. Think what it would be like if an auto-plunged ball in Medieval Madness ALWAYS hit the exact same rollover for the "Skill Shot". Not much skill involved, is there? Or think of the eject hole near the top of Scared Stiff. This doesn't ALWAYS behave the same way, but imagine if it ALWAYS went back up to the deadheads. These days (at least on an iPad), it already does it enough to seriously menace the enjoyment of the game. It would be a travesty if it did that even more reliably. Remember that when a ball enters an eject hole or is about to be auto-plunged, it typically has lost any of its spin or momentum, so you couldn't rely on that to provide any variability to the ball's action. (EDIT: In response to a post below, I thought of the example of the top eject hole in Firepower. The ball ALWAYS comes back through the "R" rollover. There's no way it should do that so consistently. This is an instance where a soupcon of randomness could be helpful.)
The other reason to defend the use of pseudo-random numbers in eject holes and auto-plunges has to do with the mechanical nature of them. It's quite plausible that a mechanical feature on a real table will not fire off with precisely the same force every time whereas its digital re-creation might. This seems like a reasonable justification for adding "randomness". The same argument could even be extended to slingshots, jet bumpers and other features that reflexively react to the ball by pushing it away with mechanical force.
Where this argument gets murkier is with sharp corners and posts. These features don't have a mechanical element to them in real life, so the justification mentioned in the above paragraph doesn't apply. However, a ball coming off of them can't react too predictably either as they certainly don't in real life. In real life, the precision of the ball and the sharp corner or post is down to the quantum mechanical level. I would hardly expect FarSight's physics engine to calculate precision to that level. However, the precision should be very small indeed. And, to the extent that behavior within one angular "pixel" of this precision of the digital object isn't calculated, it's reasonable to add a pseudo-random number to resolve any plausible action that would be calculated within that lowest precision. The better the precision, the better the game and the less need for that type of "below the grain" randomness. However, this precision is a matter for the physics engine software developers to decide on a platform-by-platform basis.
While not strictly a "round object" phenomenon, I think this same principle applies to manual plunges as well. The greater the precision of the plunge (subject to UI limitations), the better the game, but as above, randomness could account for variability at a grain lower than that precision.
Then we have the flippers. Like the eject holes, auto-plunges, slingshots, and jet bumpers, they are mechanical devices, so one could argue that randomness is justified. However, these specific devices are special in that they are the primary things that the player uses to interact with the game. Unpredictable behavior from the flippers, may make users think the game is merely a game of chance, and discard it. A bit more predictability from the flippers gives the player a feeling of control that is quite important for the user experience. Perhaps randomness at a level lower than the precision of the digital objects as described above could be justified, but giving the player a sense of control over the flippers is far more important than that of other features of the table. I'd hesitate to use "randomness" with them.
The last point would be the nudges. Right now, the nudges seem to be a bit binary. Either you nudged or you didn't and the nudge always seems to be about the same amount. I doubt that players can nudge with quite that level of precision in real life, so some randomness does seem warranted here. However, I'm sure some expert players will disagree and claim to be able to nudge real life tables with very high precision. For touch nudging, I think some randomness is needed so as to not allow players to begin to use nudging in as reliable a fashion as the flippers. But, for nudging initiated by accelerometers in mobile devices or console controllers, some less binary concept of nudging seems like it could provide a better level of control. Even then, a bit of randomness below the level of the UI's nudging precision could be warranted.
I would especially love anyone with knowledge of the physics engine (and the authority to speak about it) to chime in with the whys and wherefores of randomness being inserted into the physics algorithms.
I like a game where I feel as if I'm in control, but where unpredictable things can and do happen and I have a chance to try to meet those challenges. What are your thoughts? Where do you come down on randomess versus clockwork-like predictability? Did I miss anything?
Last edited:
