DSP Robotics Support

Please have a look at those two pictures:

trigger issue.png
trigger issue2.png

In the first one, index is set to 1, the corresponding trigger switch opens and the values pass through. In the second one, index is set to 2, and again the corresponding trigger switch opens and the values pass through.
That's the behavior I wanted.

But what you can also see is that in the second picture the trigger switch for index 1 is correctly closed (set to false) - and that's producing a trigger! It is breaking the purpose of the trigger switch. In my example I now get two coord pairs being set (the previous one because of the to-false-trigger, and the correct actual one).

Does anybody have an idea how I could prevent the coords being sent when the trigger switch is set to false (Maybe blocking the trigger in a way, that it doesn't disturb the already working circuit)?

tulamide wrote:Maybe blocking the trigger

You should get a result with the "trigger blocker" prim.
Trog wrote a great tutorial about it all that explains it all real well.
http://www.dsprobotics.com/support/viewtopic.php?f=3&t=292#p765

billv wrote:

tulamide wrote:Maybe blocking the trigger

You should get a result with the "trigger blocker" prim.
Trog wrote a great tutorial about it all that explains it all real well.
http://www.dsprobotics.com/support/viewtopic.php?f=3&t=292#p765

Yes, that's a bible for me!
But in this case it's a special case. That's why my sentence was longer as you quoted As you can see, I already did my job right, no unneccessary triggers arrive at the input. But an additional one is created when switching the trigger switch to false! So, although the prims bool is false a trigger is sent forward. How the heck should I correct that error? I can't differentiate between correct and false triggers after the trigger switch (that's the reason why I use them, to prevent false triggers). It's a nightmare I'm struggling with for a week now. Have tried totally different circuits, other prims, it just doesn't work. And the one prim that was designed to prevent false triggers generates a false trigger itself!

So, I'm really desperate. If anyone can create a module that takes an index input from 0-3 and outputs the input coords to the correct two of 8 outputs, with the rule that only one index may sent triggers at any time - then please help me

Or if somebody could give a specific hint - I listen

Trigger switch is actually almost the same thing as trigger blocker, it allows triggers to go "passively" (some folks call them backward triggers). Anything after - may "suck" triggers and data through them so to speak.

Try to use one of these two solutions (sorry - typo; fixed); inverting bools you can handle on your own .

And see how it works.

tester wrote:Trigger switch is actually almost the same thing as trigger blocker, it allows triggers to go "passively" (some folks call them backward triggers). Anything after - may "suck" triggers and data through them so to speak.

Try to use one of these two solutions (sorry - typo; fixed); inverting bools you can handle on your own .

And see how it works.

Thanks, tester.

The main problem is, the values that are outputted are used to draw points. There's nothing connected to the outputs that will ask via backward triggers. That's why I can't use trigger blocker here (no asking from forward = nothing gets through). So it's important that the values are triggered, when needed. The trigger switch does it just as wanted - with the one exception that it sends 1 trigger when switching from true to false. That's also the only trigger I need to prevent from propagating forward. Without this trigger the points are drawn correctly. But with that bool to false trigger sent, two points are drawn at the same position, where one of these points loses its correct position. And if you do this game a few times, finally all four points are drawn at exactly the same (always the fourth) position. And all of that just because of ONE trigger sent erroneously...

@tester
You've got a pm

I admit - I did not understand (and still don't). But when it comes to true/false triggers, it is possible to generate triggers and boolean states separately. Then everything becomes a matter of trigger order in the rest of the design. You can decide what to trigger first by using sample and hold prim to cascade triggers.

Problems will emerge only in ambiguous designs.

Hope this times it helps a little bit, because I'm not able to consume long texts and large schematics today.

Is what you looking for: index should open (0 to 1) comparators before x,y data arrives, but should close comparators (1 to 0) after data arrived? (or in opposite manner?)

S+hold the trigger blocker and trigger it via
bool to true. Should prevent that exception.

tester wrote:Is what you looking for: index should open (0 to 1) comparators before x,y data arrives, but should close comparators (1 to 0) after data arrived? (or in opposite manner?)

index should open before data arrives and close the path before data arrives. That's what it's doing when using the trigger switch (except for ..., well you know )

billv wrote:S+hold the trigger blocker and trigger it via
bool to true. Should prevent that exception.

I tried that. Maybe I've built it wrong, but now even the true value doesn't get through

I'm close to giving up. So I will post the whole schematic (with the changes made that billv proposed). Don't bother downloading if you don't want to help. It isn't working yet, and as soon as it is working, I will post it on the forums anyway

Not at my cpu...will check later...