Request Notes Toward a More Competent "Call Attendant" Feature

Sean DonCarlos

Moderator
Staff member
Mar 17, 2012
4,293
0
So I was playing a real Monster Bash table the other day and got a ball stuck. After about 7 ball searches, the game started tallying my bonus and served my next ball to the plunger. I eventually knocked the first ball loose and the game resumed normal behavior. Later on, I found out this is the "chase ball" feature that some newer machines have to prevent a stuck ball from rendering the machine inoperable.

These forums have plenty of posts about epic games in TPA that had to be abandoned because a ball or balls were stuck somewhere and the Call Attendant feature did not resolve the issue. Based on the behavior of the real table "chase ball" feature, and given the number of issues that arise concerning balls in the wrong place (like CV's multiball issues, RBION's "pinball missing" issues, etc.), I'd like to see if FarSight would consider adding a similar feature in TPA:

  1. Player gets ball stuck somewhere, goes to menu, selects "Call Attendant", but it doesn't work or says that the balls are not stuck.
  2. Player, determining that it is not possible to continue normal play, selects a new "Reset Table" option. Game counts down from 15 to prevent accidental activation and to allow cancellation if the ball does free itself, and also warns player that current ball will be lost when timer hits zero. (For why losing the ball is necessary, see below).
  3. When countdown expires, game removes all balls from the playfield and places them in the trough. In games with physically locked balls, game then removes enough balls from the trough to replace any balls player had locked. Game then tells ROM emulator that the ball has drained down the center and to begin bonus count. Games with special mechanics at this point (like Twilight Zone's autolauncher refilling the gumball machine) take care of their business, and the ROM takes care of any cleanup necessary to get into a consistent state for the start of the next ball.
  4. Game serves next ball to plunger if one is available, otherwise goes to the "game over" mode (including entering initials and/or posting for high score, etc.).

The loss of current ball is necessary to prevent players from abusing this feature if they are about to drain, etc. It also is more accurate in comparison to the real tables' "chase ball" feature, which also causes loss of current ball.

While I obviously believe that FarSight should fix issues that cause stuck balls, multiple balls that can't escape the plunger lane, etc., I think that implementing this feature would reduce the impact such bugs have on players' games until they can be fixed, and would reduce player wrath by allowing them to at least continue/finish their game when such an error occurs.
 

Matt McIrvin

New member
Jun 5, 2012
801
0
PHoF (on the Wii, at least) had a common bug in which a fast-moving ball would simply fly through a wall and vanish into the aether. In that case, for The Williams Collection they had a serviceable hack in which the game would automatically detect that this had happened and just give you a ball at the plunger, without penalty. The feature was completely automatic, so it couldn't be abused (easily, at least).

I don't know if this is actually what's happening with the disappearing balls on RBION, but it sure seems similar in that the ball apparently isn't stuck but is somehow not there.

"Call Attendant" also worked fine in PHoF; I remember getting balls stuck on Jive Time (sometimes in places where they'd rack up enormous scores!) Either the feature in PA is buggy, or what's happening is some new problem specific to PA's engine.
 

Sean DonCarlos

Moderator
Staff member
Mar 17, 2012
4,293
0
PHoF (on the Wii, at least) had a common bug in which a fast-moving ball would simply fly through a wall and vanish into the aether. In that case, for The Williams Collection they had a serviceable hack in which the game would automatically detect that this had happened and just give you a ball at the plunger, without penalty. The feature was completely automatic, so it couldn't be abused (easily, at least).
TPA will do that now. About one in three games I will send a ball at speed through the bumpers on CV and it will exit stage right. Almost immediately (before the errant ball has left the screen) a new ball will autofire from the plunger.

My suggestion is more for situations where balls are stuck on the table but are in endless motion, which neither TPA's nor PHOF's attendant will fix. CV's multiball issues (where 2-3 balls bounce endlessly on the plunger) and RBION's ball-not-quite-locked-but-still-jiggling issue (I've also had one get stuck behind the vari-target and bounce endlessly) were more what I was thinking of correcting with this new feature.
 

jaredmorgs

Moderator
Staff member
May 8, 2012
4,334
3
Removing all balls from the table and resetting locked balls during a game will cause issues with the game logic. On CV, it already tries to eject locked balls that are reset to the trough at the start a new game.

This is probably a request for another thread, but tables that can manage the state of locked balls should be allowed to continue to manage the balls between games. The ball state of locked balls should be remembered, instead of reset programmatically.

When a game ends on most physical pinball machines, all balls are returned to the trough during the match sequence. On all games Farsight has made, this behavior is not present.
 

Sean DonCarlos

Moderator
Staff member
Mar 17, 2012
4,293
0
Removing all balls from the table and resetting locked balls during a game will cause issues with the game logic.
Not necessarily. The emulator framework could query the ROM to see how many balls it thinks are locked and replace that many balls before serving a new ball to the player. The ROM doesn't actually know where the balls are (this is true for a real pinball table as well), it just receives inputs such as "switch 15 is closed, switch 29 is open", etc., and reacts accordingly. As long as FarSight was careful not to have the framework send the ROM erroneous signals when replacing balls, I don't think it would cause a problem.
 

Epsilon

New member
Apr 19, 2012
144
0
One way or another, this needs to get fixed. In a game where online leaderboards are pretty much the primary purpose in terms of long-term replay value and competition, having great scores get completely wiped out because the game can't recover from its own bugs is completely unacceptable. Since the odds of these bugs occurring are greater the longer a game goes on, who knows how many of the potential best scores have never actually been recorded?

At the very least, the game needs to at least have an "End Game" menu option that allows the player to simply record the current score as a completed game, perhaps with a prompt when the game is ball searching for more than a minute and never finding anything. It would be far from ideal, since who knows how many more points could have been scored if the game continued properly, but at least it wouldn't torpedo what could possibly have been someone's personal best score.

The downside of doing something like Visual Pinball's "add ball" command is far outweighed by the critical failure of having high-scoring games vanish completely into the ether. Worrying about bugs related to that compared to the utterly game-breaking one we're discussing would be like worrying about getting a splinter when you have a gunshot wound to the chest.
 
Top