>

> >

> >

> > Sure pathfinding and on line of sight work fine in real time.

> >

> > Use the bullet behavior so it detects collisions when you release the boomerang but use on line of sight and pathfinding to bring it back to you.

> >

> > If it has you locked in on line of sight just have it return to your position. If not find a path and clear obstacles until it has line of sight.

> >

> > It should always return along the X axis towards you so include that to keep it moving.

> >

> I thought pathfinding had to calculated and couldn't work very efficiently on an every tick basis, which is what this would need to be? The solids might not be moving, but the player would be, so wouldn't pathfinding be really heavy on performance? I'm still having trouble working out how to incorporate this in my example

>

Well what I suggested is using the bullet behavior. That is used to keep the boomerang flying back towards the player while on line of sight and pathfinding is used to change the bullets course as needed to get it back to the player so they are checking position every tick but not making adjustments every tick and should't slow down performance.

If I get some time I will try to make a capx with an example of what I mean but I am just going off what I see in your example so it may not work for your game.

Sounds like it could work really nicely, wish I knew where to start! I might have a cack-handed go at trying to figure it out...

Sure pathfinding and on line of sight work fine in real time.

Use the bullet behavior so it detects collisions when you release the boomerang but use on line of sight and pathfinding to bring it back to you.

If it has you locked in on line of sight just have it return to your position. If not find a path and clear obstacles until it has line of sight.

It should always return along the X axis towards you so include that to keep it moving.

I thought pathfinding had to calculated and couldn't work very efficiently on an every tick basis, which is what this would need to be? The solids might not be moving, but the player would be, so wouldn't pathfinding be really heavy on performance? I'm still having trouble working out how to incorporate this in my example

> No takers? :/

>

Well, I still think using the bullet behavior and On Line Of Sight and Pathway behaviors might be better to get that boomerang to return to you instead of falling off the screen but do what you are doing and maybe someone can get that boomerang to return every time.

Best!

Say again? So, can pathfinding be used in realtime?

gamedev middleware where alternatives are all around.

My guess is they want to break out of this area with C3?

The mindset of some people here seems to be that they want everything for free, or they don't really understand the value and cost of maintaining a product, and providing support, updates, new features, bug fixes.

I absolutely don't believe that's true at all, I believe my exact words the week or two before the announcement was "I'm ready to pay for C3 if it comes out on Wednesday". I can't believe they charged so little for the "lifetime" updates to C2, but the majority of people here just want to voice their dissatisfaction with a lock-out subscription, especially for what looks like mostly potential C2 tweaks that have been hoarded for Construct Cloud, and a bunch of bleeding-edge features that no-one really wants.

I'm getting tried of posts implying most users here want a free lunch. I love open-source, but I'd never expect to get the same experience as I would with proprietary software that costs, but fundamentally there's a problem with subscriptions charges for users, and I think that's something most developers haven't anticipated yet.

No takers? :/

Mmmm, open sauce spaghetti, and meatballs.

What exactly is it that you dislike about open-source? Half of the web is open-source (and nearly every server, certainly the stable ones, are running Linux).

>

> >

> >

> > I won't even pretend to know you or your intentions. You said "show us" some vids, which means you really want to see more, even though you clearly state you're not into the sub, and have said everything from Scirra being cynical to lambasting the blogs by throwing up "meh" images, to referring people to competitors, to even more stuff that I don't care to go back and quote you on. Sure, the sub isn't for me either, and Unity is a great option for those who have more experience, but dang give it a rest already. Everyone on this forum knows your discontent with C3. How many more times must your repeat yourself in every thread?

> >

> > Now you go and make a thread asking to see vids, not for yourself you see, but "for others..." So what do I think you're doing it for? Ammunition to use for more criticism. Maybe I'm wrong. Sure looks like it.

> >

> Nope, just a thought. I haven't actually knocked them for good ideas, and I've frequently been optimistic the web-browser tech will be stable and reliable when others have doubted it (even though I would never need or use it). I'm just looking for a good feature or marketing tactic, and I've yet to see one I've been satisfied with, and I mean to voice my discontent as a ways to (hopefully) add to the dissonance that might eventually lead to some idea of reform on their strategies.

>

> I'm also keen to suggest ideas that might improve their marketing and/or the product.

>

Well that's all fine and good. I'm not here to attack you. I'm only saying what I see. I'm with you about certain things though. I'd really like to see more performance-based blogs and more in-depth stuff about exporting and the overall optimizations that have been made to make it run well in the browser. Creating a prototype on my mobile when I'm chilling at the park or something sounds fun to me. Not practical mind you, but the idea of making a game on the go rather than playing a game on the go seems more productive. I want to know more stuff about that, how the performance will be overall, how smoothly the apk generation will be, yada yada.

Actually, while I did post the "meh" meme, I think probably my favorite feature I've seen so far (and most certainly worthy of opening the blog post) was the rounded corners.

I often felt like I was going to catch something on the sharp point of those hard edges.

I won't even pretend to know you or your intentions. You said "show us" some vids, which means you really want to see more, even though you clearly state you're not into the sub, and have said everything from Scirra being cynical to lambasting the blogs by throwing up "meh" images, to referring people to competitors, to even more stuff that I don't care to go back and quote you on. Sure, the sub isn't for me either, and Unity is a great option for those who have more experience, but dang give it a rest already. Everyone on this forum knows your discontent with C3. How many more times must your repeat yourself in every thread?

Now you go and make a thread asking to see vids, not for yourself you see, but "for others..." So what do I think you're doing it for? Ammunition to use for more criticism. Maybe I'm wrong. Sure looks like it.

Nope, just a thought. I haven't actually knocked them for good ideas, and I've frequently been optimistic the web-browser tech will be stable and reliable when others have doubted it (even though I would never need or use it). I'm just looking for a good feature or marketing tactic, and I've yet to see one I've been satisfied with, and I mean to voice my discontent as a ways to (hopefully) add to the dissonance that might eventually lead to some idea of reform on their strategies.

I'm also keen to suggest ideas that might improve their marketing and/or the product.

> C'mon Ash and Tom, put yer makeup on and make us some vids! Seriously, dev vlogs could be a good marketing strategy (not that I'd go for subscription myself, but for others...)

>

If you're not going for a C3 sub, like you've said countless times already, why even start a thread again? For the benefit of others? Give me a break.

Well then, what am I doing it for? You tell me.

C'mon Ash and Tom, put yer makeup on and make us some vids! Seriously, dev vlogs could be a good marketing strategy (not that I'd go for subscription myself, but for others...)

Zebbi

You could try to tune the parameter of bullet behavior (max speed, or max rotate angle), then find out that the LJpotential behavior won't work well anymore. It needs tune the parameter A,B,m,n, too.

A possible solution is,

1. tune the behavior of the target object (bullet behavior in this case)

2. tune the parameter A,B,m,n of LJpotential behavior to match your requirement

Yes, that's the hard part, I barely know what I'm adjusting with LJ! Is there some alternative to using "rotate 10 degrees"? I tried an anglediff idea, but it didn't work either.

Cool!

Are you using bullet for the boomerang/knife?

I am thinking when it hits an obstacle have it arc upward like a boomerang would and that would give it a better chance to find a path. Maybe make it move back along the X coordinate until it finds that path.

Not sure how your obstacles will be arranged but looks like if it goes up after a collision it will find a path back.

I'm using the car, as it had a good built-in bounce and turned nicely when you use a second object pinned on top to disguise the angling of the sprite. It's just hard to find a decent way of finding it's way home in a highly dynamic way (since pathfinding tends to be turn-based.) If I could work LJ out, it should be perfect as I could reject the solids, but I'm awful at setting it up properly, and the only reason my first example (and the gif) looks that good is because it didn't hit any solids!

Any boids experts want to have a go?