DEFENDER 'REVERSE' Glitch

[smartbomb2084]

I'm sure all of you die hard DEFENDER players have encountered this frustrating phenomenon. You know, you are having an 'out of body experience' while playing the game of your life and just as you go to hit REVERSE to blast a Baiter or Swarmer the button doesn't work with one strike...sometimes.  It is also in the MAME version as well.

It also seems to happen with the SMARTBOMB button too.

I read somewhere that DEFENDER only polls the control panel switches once every half of video picture generated.
That is, after odd raster lines are drawn, poll switches, after even raster lines are drawn, poll switches. With 60 frames per second drawn that is 120 switch polls per second and yet is still not fast enough to recognize every button push.

When you play STARGATE the REVERSE button works flawlessly every time.

I was wondering if a Capacitor could be added to the switch to help the CPU 'see' the switch closure just like BALLY did on early solid state pinballs because they thought the CPU might bee too slow for reactionary soleniods i.e. slingshots and pop bumpers to seem like they were immediately activated by the ball.

Since DEFENDER doesn't use a SWTCH MATRIX like a pinball I am thinking this is not possible... or is it?

[smartbomb2084]

SWITCH BOUNCE?  Yeah you wish!  Maybe you outta read the CLAY GUIDES that you often BASH a little closer Sonny!!!

[TOK]

I'm going to try to pay attention to this when I play again... I've always attributed the close calls that result in my death to slow reflexes.

[smartbomb2084]

Maybe not 1/10th the time but it wouldn't be hard to do. I haven't done it yet because of fear of the unknown as in I don't want to cause any damage to my board. Plus I am not entirely sure how to wire it or what type/value of cap to use since the game doesn't use a switch matrix.

Maybe after a few beers I might get enough courage to try it.

[boardjunkie]

The cap will make the switch appear to be closed longer to the PIA chip. But, you will be creating an RC time constant (with the PIA circuitry) that may make the switch closure take longer to be recognized due to how fast/slow the cap charges.

Can't hurt to try it tho. I'd start with a 47u and see how that works. If that doesn't "stick" the switch long enough, try a 100u.

This was done on alot of Midway games either from the factory or as a service bulletin due to them missing coin switch closures.

[boardjunkie]

They were there because of random coin switch trips being missed. If you build a simple circuit to watch the switch closure on a scope, its *very* fast. Micro switches tend to be that way 'specially when used with a rotary trip wire. Early game mfgr's just didn't take that into account when they sent the games out the door.

For switch debouncing, small values are used. Usually around .1u or around there. But its really the same idea....it just sticks the switch closed for the specified time created by the cap discharge so the multiple closures created by contact bounce don't get seen by the switch input circuitry.

[smartbomb2084]

How about helpin' out a struggling maroon here...  Do I want to use an electrolytic, tantalum, or disc capacitor?

Should I use a polarized or non-polarized cap and how do I wire it?  Inline with the strobe or inline with the return or some other way?

I always thought diodes were added to the switch matrix to eliminate 'Phantom' switch closures?

[boardjunkie]

Defender doesn't use a matrixed switch setup. Its all discrete inputs. A 10 or 16v lytic will be fine, but you may have trouble finding such small values that low a voltage. It doesn't care what the voltage rating of the cap is since you are only dealing with 5v, so a 50 or 100v cap works just as well as anything. Use yer meter to find the high side of the switch line (+5v) and connect the pos end of the switch to that. Neg side goes to the other (low) switch wire.

[boardjunkie]

Could....if they're there. Williams didn't fit those IIRC. But that may be too large a value anyway . I want him to start with the smaller value first just to see how that works.

[boardjunkie]

And....make sure the switch contacts are clean and polished.

[smartbomb2084]

THANKS for the hook up instructions.   

I went with the 47U/16V cap and you can see the switch stay on longer in 'Switch Test' mode.  When you play the game the REVERSE works more reliably the FIRST time you hit it but a second quick pressing gets ignored because the switch is on TOO long.

So I guess a 100U cap would be too big so I have a 22U/16V cap to try instead but it is gonna have to wait 'til tomorrow as I am out of time for today.

 

[boardjunkie]

Right ahn. Just keep trying different values until you find one that works the way you want it to.

[smartbomb2084]

Tried the 22U cap today and it is definitely better than the 47U.  A super fast second push still gets ignored but not as often and single quick swipes are still a lot better.

So in the end, like everything else in life, it is a trade off.

Just to make it official I am going to try a 10U cap next.