360 - Bug Funhouse. 464m. Ball one. No balls inplay or by plunger. Call attendant not working.

Ark Malmeida

New member
Apr 3, 2012
360
0
Bugs require a certain series of events to occur. Because each person plays differently, the events may not happen that are necessary to reproduce the bug. The reason some people see the same bugs more often, most likely has to do with play style and reaction times. They hit the same shots a lot, etc.

When this sort of thing happens, any idea where the ball was last? Did the ball(s) in play drain and then it said they were lost, did they go into a trap, if so which ones, etc?

In situations on the two tables that I've had this happen on - Ripley's (multiple times) and Cirqus Voltaire it has always been during multiball and appears to always have been one of the balls going down a trap/scoop and then disappearing. In RBON, the ball has gone into the main scoop (the amazing events scoop?) during the standard multiball and in CV it went into the sideshow scoop during juggler multiball. It's almost as if the ball somehow falls through the geometry and off of the table, which would fit with what you were saying about it being 'dead'.

Hope that helps.
 

Mike Reitmeyer

FarSight Employee
Mar 13, 2012
1,735
1
I think it would still be worthwhile to implement some user-activated feature to reset the balls to a known state. As more tables are released, each will bring the possibility of obscure bugs that will be difficult to anticipate, reproduce and fix. Having some sort of "table recovery" option in place for use when unexpected weirdness occurs would lessen the impact of these bugs on players, even if we can't identify what causes a particular issue yet. I could see where this feature might be difficult to implement, as the ROM might get confused if (for example) balls were suddenly "reset" and disappeared from physical locks, but I think the idea at least worth looking into. Of course, I just play the tables, not develop them, so feel free to disregard me if I'm heading completely off base!

Even if we just had an "end game in progress" option (with appropriate safeguards against accidental activation!) so that we could force a malfunctioning game to a semi-graceful conclusion and preserve whatever scores/goals we had attained up to the point when the problem occurred, I think players would be happier.

The problem is we're emulating real life, so if we do something like, remove a ball from a trap and stick it in the ball trough, the rom gets confused. In real life that can't happen, so it starts doing things like spitting multiple balls in the plunger, or breaking in some other way.

Whats worse, is every tables is different. So it's unfortunatly not a simple thing to fix. Still working on some solutions.

In situations on the two tables that I've had this happen on - Ripley's (multiple times) and Cirqus Voltaire it has always been during multiball and appears to always have been one of the balls going down a trap/scoop and then disappearing. In RBON, the ball has gone into the main scoop (the amazing events scoop?) during the standard multiball and in CV it went into the sideshow scoop during juggler multiball. It's almost as if the ball somehow falls through the geometry and off of the table, which would fit with what you were saying about it being 'dead'.

Hope that helps.

Every bit of information helps. If the ball falls off the table, gravity continues to pull it down. When it hits a certain distance down the ball is reset to the plunger automatically.

Balls in traps are considered "trapped" not "dead". So it's quite possible you are seeing a differnent situation than the one we saw here. If I were to guess it sounds like the ball is trapped, but the switch for the trap didn't get set properly.

This info does give me some things to look into. Thanks.
 

Sean DonCarlos

Moderator
Staff member
Mar 17, 2012
4,293
0
The problem is we're emulating real life, so if we do something like, remove a ball from a trap and stick it in the ball trough, the rom gets confused. In real life that can't happen, so it starts doing things like spitting multiple balls in the plunger, or breaking in some other way.

Whats worse, is every tables is different. So it's unfortunatly not a simple thing to fix. Still working on some solutions.
I was afraid of that. I don't know how much control you guys have over the ROM itself - if there are things you can do to "unconfuse" it or if it is more or less a black box from a coding perspective. Sorry if I sounded like I was trying to tell you how to do your job; that was not my intent at all.
 

MichRX7

New member
Aug 2, 2012
76
0
Thanks for the time Mike. As a Director of IT with a coding background I see your struggle with confirming bugs every day of the week. While we have a much smaller user base than you I find it important to educate that base on best practice for my team to track down a bug. While screenshots, in most cases, are probably out of the question, what would you like to see from us whenwe submit issues on this forum (outside of just giving as much detail as possible). I'd say that would lead to the most constructive reporting for your team.
 

Matt McIrvin

New member
Jun 5, 2012
801
0
Every bit of information helps. If the ball falls off the table, gravity continues to pull it down. When it hits a certain distance down the ball is reset to the plunger automatically.

Balls in traps are considered "trapped" not "dead". So it's quite possible you are seeing a differnent situation than the one we saw here. If I were to guess it sounds like the ball is trapped, but the switch for the trap didn't get set properly.

This info does give me some things to look into. Thanks.

I recently saw something like this on Black Hole. A ball was locked in the capture hole on the upper level, but no new ball appeared at the plunger, and the game was therefore stuck in an unplayable state (with call attendant not working). What's potentially interesting is that, since you can see the locked ball in that case, I know that it wasn't that the locked ball somehow disappeared. It was there in the lock; the lock just didn't have the consequences it was supposed to.

Edit: Unfortunately, I don't remember whether there was another ball locked in the lower-playfield captive hole, which might be relevant to the question of whether the problem was something like a ball missing from the trough.
 
Last edited:

Mike Reitmeyer

FarSight Employee
Mar 13, 2012
1,735
1
Thanks for the time Mike. As a Director of IT with a coding background I see your struggle with confirming bugs every day of the week. While we have a much smaller user base than you I find it important to educate that base on best practice for my team to track down a bug. While screenshots, in most cases, are probably out of the question, what would you like to see from us whenwe submit issues on this forum (outside of just giving as much detail as possible). I'd say that would lead to the most constructive reporting for your team.

Utlimatly video capture is the best, but I understand that it's just not feasible for most players. Screenshots can sometimes be helfpul, because its a visible way of seeing what someone is describing.

The more information you can give the better. Sometimes something that would seem insignifigant is really important to finding a bug. Things like which balls are trapped and in which traps, where you in multiball. Is it the first game, did you quit a previous game and if so where were the balls when you quit?

Quitting the table can be a big issue. Real pinball tables don't have a quit button (just a power off button) so making sure every state is correct is tough.

Of course, telling us which platform it happens on, and in the case of consoles which resolution are you using (mostly for graphics type bugs) can also help.

I was afraid of that. I don't know how much control you guys have over the ROM itself - if there are things you can do to "unconfuse" it or if it is more or less a black box from a coding perspective. Sorry if I sounded like I was trying to tell you how to do your job; that was not my intent at all.

I didn't take it that way at all. We do our best to not mess with the rom at all, cause we dont' have access to the real code that the rom was built on, it would be hard to fix anything in it without breaking other things.

I recently saw something like this on Black Hole. A ball was locked in the capture hole on the upper level, but no new ball appeared at the plunger, and the game was therefore stuck in an unplayable state (with call attendant not working). What's potentially interesting is that, since you can see the locked ball in that case, I know that it wasn't that the locked ball somehow disappeared. It was there in the lock; the lock just didn't have the consequences it was supposed to.

Edit: Unfortunately, I don't remember whether there was another ball locked in the lower-playfield captive hole, which might be relevant to the question of whether the problem was something like a ball missing from the trough.

Thanks, that's helpful. A few of these comments are making me think we might have issue with some traps getting confused on what the ball is doing.

I also understand sometimes it's hard to give details cause you are just playing and it breaks. It's hard to think back what all happened before that unless you were specifically testing it to break it.
 

Mark Miwurdz

New member
Apr 7, 2012
684
0
Just happened again. Both times it was hitting the extra ball lane to make it 11.45. Nothing too important, then.
 

Carl Spiby

New member
Feb 28, 2012
1,756
0
Mike, Why is the self-test part of the ROM missed out on TPA tables? Wouldn't running that on load be a good way of making sure all the switches and balls are accounted for?
 

DopedToInfinity

New member
Mar 31, 2012
440
0
Mike, Why is the self-test part of the ROM missed out on TPA tables? Wouldn't running that on load be a good way of making sure all the switches and balls are accounted for?

While watching the PAPA tutorial I saw the ball search kick in after a minute oF inactivity, but I cradled the ball for over 3 minutes on iPad 2 iOS 5.1.1 with no ball search.

I believe this table is scripted (very well I might add!) but it is still scripted. No ball search, overlapping music tracks, and mystery mirror not showing the bonus long enough (fixed).

I still love playing it, but I'm after authenticity. Can't wait for proper ROM emulation, Gorgar, BH & BK are crying out for Proper ROM emulation.
 

Mike Reitmeyer

FarSight Employee
Mar 13, 2012
1,735
1
Mike, Why is the self-test part of the ROM missed out on TPA tables? Wouldn't running that on load be a good way of making sure all the switches and balls are accounted for?

We save out an emulation state, so that the user can start playing immediatly, and so that we always start in a known state. The problem here is the bug appears to be a logic issue, not necessarily a rom issue. Ball is considered dead, but the switches on the table don't match. Seems like a switch isn't being set properly.

While watching the PAPA tutorial I saw the ball search kick in after a minute oF inactivity, but I cradled the ball for over 3 minutes on iPad 2 iOS 5.1.1 with no ball search.

I believe this table is scripted (very well I might add!) but it is still scripted. No ball search, overlapping music tracks, and mystery mirror not showing the bonus long enough (fixed).

I still love playing it, but I'm after authenticity. Can't wait for proper ROM emulation, Gorgar, BH & BK are crying out for Proper ROM emulation.

Funhouse is definitely rom emulated. The reason you don't see a ball search is because on the real table it doesn't search if the ball is cradled. Because you can cradle a ball to check scores and other info on the backglass displays.
 

Mike Reitmeyer

FarSight Employee
Mar 13, 2012
1,735
1
Is ROM emulation actually on the cards for these tables?

I'm hoping to redo BK and Gorgar since they are Williams System chips which we will have working in the future. We're pretty close to finishing System 11. For BH, I'm not sure what the plan is for Gottlieb emulation yet.
 

DopedToInfinity

New member
Mar 31, 2012
440
0
We save out an emulation state, so that the user can start playing immediatly, and so that we always start in a known state. The problem here is the bug appears to be a logic issue, not necessarily a rom issue. Ball is considered dead, but the switches on the table don't match. Seems like a switch isn't being set properly.



Funhouse is definitely rom emulated. The reason you don't see a ball search is because on the real table it doesn't search if the ball is cradled. Because you can cradle a ball to check scores and other info on the backglass displays.

Cheers Mike, I did notice the scores and info display while I was cradling the ball. Looking forward to the new tables and thanks for keeping us in the loop. Your insider knowledge is appreciated :)
 

laughing_lunatic

New member
Jun 6, 2012
359
0
yeah, I use the cradling the ball to see where I'm at in the game all the time... or just to "pause" the game for a few seconds while I reach for a drink/smoke/whatever...
 

LocustGod

New member
Aug 10, 2012
10
0
I've had this happen to me once, getting the ball stuck someplace in the top-right during Bride of Pinbot.

Wouldn't it just make sense to have an in-game timer programmed in, so that if there's no table activity, the ball is taken off the table and put into the plunger after xx amount of time elapsed? Seems like a simple fix to me.

Edit: I guess I should have read what Mike said. It's probably tricky to implement new code into an emulated Rom without introducing game-breaking bugs
 
Last edited:

Sean DonCarlos

Moderator
Staff member
Mar 17, 2012
4,293
0
I've had this happen to me once, getting the ball stuck someplace in the top-right during Bride of Pinbot.

Wouldn't it just make sense to have an in-game timer programmed in, so that if there's no table activity, the ball is taken off the table and put into the plunger after xx amount of time elapsed without any on-table activity? Seems like a simple fix to me.
That's what Call Attendant does now, but it doesn't work in all situations. I (and others) have suggested a more forceful version that would cause a stuck ball to be destroyed and re-served to the plunger, but this would often confuse the ROM running the table, as it is not expecting balls to suddenly vanish and reappear elsewhere. This leaves the table in an inconsistent state, which in turn can lead to various bugs and general weirdness.

FarSight does not have as much control over the ROM in each table as I thought, so it's not just a matter of telling the ROM to "forget that ball, it's in the plunger now".
 

Mark Miwurdz

New member
Apr 7, 2012
684
0
I'm dying to post a decent score. I would obliterate that top score without this glitch.

I had a plan to get a 100m score then let the balls drain. Then 200m and let the balls drain... etc.

Haven't made it to 200m yet without the glitch kicking in. Beautiful.
 

Luke

New member
Aug 13, 2012
1
0
Hey Mike, I don't know how MS views this kind of thing, but in a future update could you put an 'easter egg' of sorts in where if you hold something like the Back + Start button down simultaneously for 15 seconds, it would pop up a simple but useful set of debugging info we can use to post here? Like X/Y/Z of the ball, the state of traps/sensors/states, etc.?

I've only had a few freakish behaviors that were actually caught by existing contingency code (like the ball flying through the side of the table and returning after a second or two) but I can imagine when it gets perpetually lost, there's gotta be a few bits of info we could retrieve for you to help track down the cause.
 

Chris6210

New member
Jul 27, 2012
9
0
When this kind of thing has happened to me (on Ripley's, I think, which is an emulated table), the table software actually goes into ball-search mode. I think it'd be reasonable for TPA to assume a ball really has gotten stuck or gone missing if the table itself is in ball search: no need for a complex algorithm; the table devs have already done the hard part for you.

Of course, that strategy would depend on the table logic being emulated, which not all of them are. And it depends on whether it's easy for TPA to tell that the emulated software has gone into ball search.

I have recently played Ripley's IRL and it done this to me for real, it timed down through 2 timers and started searching for the ball with all the clickers and what not. Exactly like it has done to me in-game (and obviously you and more), but the point is it's a real occurance and thats the whole reason I bought this Pinball game over all others.
 

Members online

Members online

Top