game-editor.com

Kodo wrote:After a quick look this is looking really great, I haven't exported a version to test on my iPod yet (would that work yet if I tried?)


I've noticed that if an objects scale is changed after it has been created the bounding polygon used for collision is not updated. Possibly something like b2UpdateDynamicBody/b2UpdateDynamicCircle methods would solve this if possible? Then if for some reason we changed the scale or rotation in an event we could also recreate the collision polygon manually. It would make the system a bit more flexible, but as it is now it is already very usable!

This already has huge potential for new games, I've got one in mind already
Amazing work you've done here Akr!

Kodo,

This is a very good comment. Right now I am not sure how to handle this. Do u think it makes sense to scale and/or even rotate collision polygons when
sprites are transformed?

Thx

Do u think it makes sense to scale and/or even rotate collision polygons when
sprites are transformed?

I only noticed because I was scaling an object in one of your demos geds up and down slightly and noticed that the physics objects didn't react, which looked quite odd. I did try manually rotating a physics object (in the draw event) as well with the same result. To make the system as flexible as possible, to allow for any game design scenario it probably would be a good thing to be able to scale and rotate collision polygons when the sprites are transformed. But in my mind these would be bonus features, you could still do an awful lot with the system as it is now, so maybe these things could be added in a later update (1.51 maybe). I'd like to hear what others think about this as well!

I suggest you make "Link transformation to collision mask" check box

Yeah, I would prefer physics on manipulated objects as well, but already this version is a lot better than the previous. I've only had a little bit of time to mess around with it, but definately better, especially with speed. I still maintain that the Set and Get functions should be actor specific and/or actor variables for them, but as that isn't quite what this preview is for, it can wait. It would make implementing some of my ideas a little easier though.

Thanks for that great feedback. I appreciate that. I had a chat with makslane second half last year. My first approach was with actor variables.
I found this cooler. I agree. Just do a angle++ and sprite rotates is very nice. But makslane told me not to do. The actor varialbles are a special
section in ge which is very sensitive and shouldnt be modified.
I have to say. The prototype with actor varialbles worked for that purpose. But incursion didnt work any more after that.
The actor variables are implemented in a very optimized special way. Maybe Makslane can give a comment on that. Would be helpful. I will ping him.

Hmm... that's unfortunate. I guess though that would explain why there isn't a zdepth actor variable. It would be very convenient if it had that kind of support. Well, hopefully makslane can help, otherwise it'll be like my zdepth problems all over again...

Whats the problem with zdepth?

Oh, it's just that there is no zdepth variable, and no getZDepth function, so I either have to create and constantly update a self created actor variable zdepth, or create a function to determine zdepth. It's probably only annoying because a single actor variable would fix it, but it is just one of the many things on my list of desired gE features.

Then its time to discuss that. The goal for 1.5 is not to bring fancy features only. We can implement the getZDepth(). No poblem.

How does your list of desired functions look like? I am would be highly interrested.

I posted the list here. http://game-editor.com/forum/viewtopic.php?f=3&t=10190 It isn't just functions, and sorry for the messy appearance. Tomorrow I'll probably go through it and fix it up.

skydereign wrote:I posted the list here. http://game-editor.com/forum/viewtopic.php?f=3&t=10190 It isn't just functions, and sorry for the messy appearance. Tomorrow I'll probably go through it and fix it up.

This list is impressive and too much for the next version. We should keep everything in mind and prioritize.

What about a poll. Cant u just create a poll with the top 10 features of your list and post it in the forum? Would be interrested to get feedback.

Before we need to ensure that we have the features defined we want to run the poll on.

Will do, though to be honest the ones that I feel are most necessary are probably not quite what everyone else wants. Most of the more important ones for me I think are bugs created from more obscure script based issues, like the actor variable global code problem... I'll see if I can come up with an adequate 10. Too bad I can't list them all and let people pick 10 of their favorites from the list.

akr wrote:Thanks for that great feedback. I appreciate that. I had a chat with makslane second half last year. My first approach was with actor variables.
I found this cooler. I agree. Just do a angle++ and sprite rotates is very nice. But makslane told me not to do. The actor varialbles are a special
section in ge which is very sensitive and shouldnt be modified.
I have to say. The prototype with actor varialbles worked for that purpose. But incursion didnt work any more after that.
The actor variables are implemented in a very optimized special way. Maybe Makslane can give a comment on that. Would be helpful. I will ping him.

If I understood you right, you thought of making changes to current actor vars, like angle.
I understand that current ones can't be modified, but how about creating new ones instead?
Like animAngle and with this you could rotate animation while angle would control only angle of directional_velocity. I hope this made sense.

skydereign wrote:Will do, though to be honest the ones that I feel are most necessary are probably not quite what everyone else wants. Most of the more important ones for me I think are bugs created from more obscure script based issues, like the actor variable global code problem... I'll see if I can come up with an adequate 10. Too bad I can't list them all and let people pick 10 of their favorites from the list.

I know that poll is limited to 10 and this is pretty small. I can make a poll of much more on my website.

Windows version added today. Windows community: Let us know your thoughts!