Make score a global variable and don't reset your variables on layout change.

I found this one interesting enough to try and create:

Try here

One way to do it would be by spawning an object at it's bottom corner, the right corner when moving right and the left corner when moving left. Pin your 32x32 block to it and rotate the spawned object 180 degrees.

Also you probably shouldn't deactivate the group has no target, for there are still enemies that don't have target, right?

No, that's not it, That's an event without any actions attached, so essentially useless for it doesn't do anything.

Right click on the x<charavters.x and add another condition enemy boolean has target and do the same for the x>characters.x

That should help with having only having the enemies with target set to true follow the player.

But in this case I think adding a for each family condition to the two target detection events would work.

Oh, on second thought, adding a condition to the has target events is necessary. Add a condition has target to those two events and the above mentioned for each shouldn't be necessary..

The has target events now affect all family members when the group is activated.

You can still reference your objects by their own name even though they are in a family.

So you could have events that are the same for all family members and events that are just for enemy 1, or enemy2, or enemy3 sprites.

I can't open your capx, for it uses a third party plugin, but I understand what you mean.

There are multiple ways to do this, but here's an easy one (It's not the best solution, but it will work).

Create a global variable called TopSpeed

Set it to the original bullet behaviour.

System compare two values : bullet.speed < TopSpeed

• add 1 to bullet speed.

You should do that for each of your objects having bullet behaviour

Then add the event to change the variable TopSpeed to a higher number when collecting a boost.

This blogpost has information on wasting precious memory with graphics.

And this topic has some nice info about reducing image impact on performance.

I believe the 4 stands for R,G,B and Alpha.

Some simple questions to help you along:

Do you really need the sprites to be this big?

Could you work with repetition instead of using new sprites?

Can the animations be shortened without instantly looking bad?

Even large game-making-companies use techniques like repetition to reduce the amount of graphic memory used by the game. It's just one of those things gamemakers should take into account.

If you have the map, and the framework to import positions of the car in realtime, it should be possible.

There is a difference between a saved image and a rendered image. Saved images are compressed to take up the least amount of space possible.

While being rendered the image is not compressed. so one 512x512 image will take up 512x512x4 = 1048576bytes of graphic memory.