The original has a transistor that energizers the coil. I would guess on the remake there is software emulation in the path which is much slower.

I just did this on my SW LE and Hobbit to get a feel for what Stern and JJP games are doing.

SW LE: Has a delay of about 2-3 ms without much variation. It was hard to measure the timing exactly because the coil pulses rather than being all or nothing.

Hobbit: Essentially no delay similar to my original AFM. Perhaps this is why some say JJP games feel like original B/W?

Might get around to posting some tracings but need to play some pins.

Baseballs and hockey pucks can go 100 MPH in professional games. That is 4.5 cm/ms. Swing with a 5 ms delay and you are off by more than 20 cm.

Triggering off the switch signal from the opto board which is activated when the flipper button is pressed.

Of the two games with remakes available I own a AFM orig and MMr LE so I guess if there are two camps I have one foot in each. I raised this issue because I felt it might be a major reason why many (including me) feel the remakes have a different feel. Clearly some (mostly remake owners) disagree and feel this timing issue is not important. I personally believe the timing is important but don't know for sure and have enjoyed reading much of the analysis and opinion in this thread. If the timing issue is not important to some then nothing here should impact their enjoyment of these remakes. My hope is that by raising this issue CGC will provide a fix of some kind for those that want one. If not perhaps a third party fix will be possible in the future. This would of course benefit remake owners the most but also the hobby as a whole.

The Bally/Willlams flipper coils have a high power winding that initially fires and a lower power hold winding. These coils are the same part on both the original and remake. My probe had trouble staying on when connected directly to the coil so I used a alligator clip and connected the Fluke probe to a wire coming off that.

EDIT: Initially I was not actually recording from the power coil but the voltage difference between the power and hold coils. The timing issue remains the same. Updated pics and tracings will be posted.

Yes, but that is not what I'm showing here. I measured the timing between the flipper opto board and the solenoid. The timing difference would be the same regardless of button, spring, who or what is pushing it, etc.

Thanks WOZ and Pinzap. You are absolutely correct. I was recording the difference between the power and hold coils on both AFM and MMr LE. Because the coils both are initially energized and there is a significant resistance difference between them I was recording a signal here that I used to determine the timing. I repeated the measurements this morning on the correct leads. The timing issue remains the same. Will upload more pics when I get a chance.

There have been two important issues which have been raised since I started this thread

1. The opto switch on the flippers has 2 opts -SW1 and SW2. In the previously analysis I used SW1 as a trigger. SW1 and SW2 likely trigger at very close to the same time but may not be exactly the same. I'm not sure how these games work if SW1 and SW2 are triggered at different times. One possibility is that which ever switch is triggered first is used, if this is what happens or if only SW1 is used my previous analysis is valid. But what if only SW2 is used? Perhaps SW1 triggers slightly sooner and the coil is synchronous with SW2. Here I repeated the experiment using SW2 as the trigger. The results are the same. I should mention that I cannot monitor SW1, SW2, and the coil at the same time because my scope only has 2 channels.

2. More importantly it has been correctly pointed out that my leads were previously measuring the difference between the power and hold coils rather than the power coil and +50V. As it turns out both coils are initially energized and there is a resistance difference between them that creates a large signal so I picked up the timing correctly in the earlier experiment. However, I've repeated my recordings measuring the difference between power and ground and show some images here. For completeness I also measured the latency using the ground and hold coil (two outer terminals) and found the same latency variability.

I've attached some pictures that show channel A (trigger) attached to SW2 and ground at the opto board. I attached the connector to do the experiments and it is shown here with the connector off. Another picture shows channel B attached to the right flipper coil. Similar to before the shortest delay I measured was about 1.5 ms. One picture shows this type of delay. Notice that late in the tracing (about 4 ms after the coil fires) you can see the coil voltage going back to zero consistent with the limit switch being hit and only the hold coil engaging. Notice that SW2 has a bit more noise in the signal that the earlier tracings using SW1. I'm not sure if this is real noise in the signal or maybe I just didn't have a solid connection to the pin. In any case this is not enough to influence the timing. I also including a tracing were the delay is much longer at 5 ms.

So in summary, although I didn't actually record from the power coil in the earlier analysis the results are the same when I do.

Switch attachment (resized).jpg
Coil attachment (resized).jpg
MMr LE long (resized).jpg
MMr LE short (resized).jpg

My game already has the power turned up (I think to +2). I don't think the power has any effect on the latency but I haven't checked. I would guess that the power influences either the duration the power coil is energized or the voltage to the coil.

This is a question unrelated to the latency issue I have raised, but it is a good question. From what it says in the manual the coil won't burn out even if the EOS switch doesn't work. My guess is that the system limits the time that the power coil is energized even if the switch isn't triggered. It is also possible that changing the time the power coil is energized is used to modulate the power. When I get some time I'll look at this and compare the original and remake.

8 nano seconds = 8 billionths of a second. We are talking about a variable delay as long as 5 thousandths of a second - literally almost 1 million times longer. Since the delay is unpredictable simply pushing the button earlier won't help.

Thanks, this is correct. The earlier post has been edited to indicate orange is +50V.

I'm not sure either. I've seen a couple original B/W games where one of these optos stopped working and the game still played normally but there was a credit dot to indicate a bad opto. Both flipper optos from both sides go to the CPU board on the originals, so I always figured it was so if one went bad the game would still work until the operator could fix it. I've not sure what happens if they are both triggered at slightly different times which is why I repeated the analysis using SW2.

I PMed Doug Skor at CGC about the issue and he got back to me. He said their engineers are looking at it and they are confident they can fix it if necessary.

What I don't understand is all the AFMr and MMr owners who are vocally claiming this isn't an issue or isn't important to them. Like most companies CGC might only put resources into fixing this if it is important to their customers. I hope other remake owners will support me and others in the hobby who would like to see this fixed.

soren I moved the pre trigger level to get a better look at this short time course artifact. However I agree that the measurements on the original AFM need to be revisited, especially since I was not directly measuring the power coil initially and later repeated this measures on the MMR LE. Another user asked me to do this via PM. I won't get a chance to do it until tonight, but will post some tracings when I do.

I'm not lying if that is what you're asking? Is it possible some of these measurements have artifacts or could have been done better - yes. I have repeated the measurements on the MMR LE and found the same thing. I plan to repeat them on AFM. I've been honest in presenting what I did, even showing pictures of exactly where the probes were placed. If there is skepticism, I would encourage others to repeat these measurements on their own games.

I used a Fluke Scopemeter 123. I think it is a high quality instrument. It has previously been mentioned by an earlier poster in this thread that he doesn't believe in digital scopes and an analog scope should be used. I don't have an analog scope, but again I would encourage people who don't believe these results or think they should be done differently to repeat them with equipment that meets their standards.

Shipping would be less for you to send the scope to me

@cgc_dduba : Thanks for looking into this. I will do what you suggest on my original AFM tonight and post the results.

woz : I forgot to mention that I after I did this I got shocked and the meter caught on fire... Actually I didn't use the common (center port) at all but had differential leads on both A & B. Anyway I'll do it the way @cgc_dduba suggested tonight with a common ground.

You did. My earlier response was: "Is it possible some of these measurements have artifacts or could have been done better - yes. I have repeated the measurements on the MMR LE and found the same thing. I plan to repeat them on AFM. I've been honest in presenting what I did, even showing pictures of exactly where the probes were placed. If there is skepticism, I would encourage others to repeat these measurements on their own games."

I hope you will find it in your heart to forgive me for making an honest mistake.

I feel bad about it this thread and wish I had checked my results better before posting. I think probably when I originally tested AFM the ground wire lost contact on the opto board. That was a difficult connection on pins that were close together and I had to put the connector back on after I hooked the wires around the pins. When the coil fired it pulled enough power to make it look like the opto was firing at the same time once the ground was lost.

I've updated the original post to say the original claim that AFM did not have a delay were in error and both games have a similar variability. Like many I still think the originals have a different flipper feel than the remakes, but with respect to the flipper timing they are similar. I have changed the title of the thread to "Remakes and originals have similar flipper delays"

I think this has spurred some interesting discussion, and I thank all - no almost all - that participated.

It seems to show this in the tracings I've posted earlier. Multiple people have mentioned it to me between this thread and PM. Given my past experience I don't want to jump to any quick conclusion about this but I'll take a look at it in more detail.