next »

[qMopey]

Got a SAT collision resolution working, which I was testing with a triangle and a box. Had been bugging me more than it should've. The collider on the character is the same size as the sprite in this case, so it's colliding on the lower left corner there. Also, I haven't even tested rotated sprites yet, but should work

Now, to build some prototype gameplay

EDIT: that printf from shader code is awesome 

If it remains buggy, maybe you could give this a try: https://github.com/RandyGaul/tinyheaders/blob/master/tinyc2.h

[ZebraInFlames]

Got a SAT collision resolution working, which I was testing with a triangle and a box. Had been bugging me more than it should've. The collider on the character is the same size as the sprite in this case, so it's colliding on the lower left corner there. Also, I haven't even tested rotated sprites yet, but should work

Now, to build some prototype gameplay

EDIT: that printf from shader code is awesome 

If it remains buggy, maybe you could give this a try: https://github.com/RandyGaul/tinyheaders/blob/master/tinyc2.h

Yeah, I saw that earlier, nice stuff.

[ferreiradaselva]

How's everyone's progress?

[JWki]

How's everyone's progress?

Shitty. Lost direction and got pretty much nothing done in the last two weeks. Also have to prioritise working on, well, work stuff and uni is starting again so will probably not be very productive in the next few weeks either. All in all not a very happy programmer.

[gimymblert]

the last 5% is 95% I'm stuck too in that planet generator lol, going back soon to it and restart on unity from blitz ...

[InfiniteStateMachine]

How's everyone's progress?

blehhhh

[ferreiradaselva]

:/ that sucks

the last 5% is 95% I'm stuck too in that planet generator lol, going back soon to it and restart on unity from blitz ...

They say that the last 10% is like other 90%, so you just need to finish the other 45%

Edit: oh, now i see your statement says that the last 5% is 95%, i got it wrong the first time.

[ferreiradaselva]

I decoupled my math source from my engine and made a separated pure C lib with it:

https://github.com/ferreiradaselva/mathc

[oahda]

How's everyone's progress?

Got a little commission, so bye C++ for a mo' and hello JavaScript with Vue.js and Firebase. Seems I picked good tools! Modern in the good way (reactive and so on) and don't need to worry about user safety myself but also lightweight and without all that NPM stuff of nightmares (modern in the bad way); both of them are just includes. Everything is going very smoothly so far!

[powly]

How's everyone's progress?

something's cookin', not sure if it'll be cool in the end or not. at least another idea finally (somewhat) tested out

[ferreiradaselva]

something's cookin', not sure if it'll be cool in the end or not. at least another idea finally (somewhat) tested out

That is hecking cool. How are you doing it? Geometry/tessellation shader?


----

I got 16 stars on github, that's the most stars I got ever.

[powly]

something's cookin', not sure if it'll be cool in the end or not. at least another idea finally (somewhat) tested out

That is hecking cool. How are you doing it? Geometry/tessellation shader?


----

I got 16 stars on github, that's the most stars I got ever.

Two passes of compute and a trivial “paste this on screen” vert+frag pair. The geometry is just a ray marched sdf with a single diffuse bounce per pixel, reconstructing a somewhat okay AO from that data is the interesting thing here.

Which project? Let’s make that 17 :9

[ferreiradaselva]

A math C lib https://github.com/ferreiradaselva/mathc

Two passes of compute and a trivial “paste this on screen” vert+frag pair. The geometry is just a ray marched sdf with a single diffuse bounce per pixel, reconstructing a somewhat okay AO from that data is the interesting thing here.

I never used a compute shader. I will give a try soon.

[powly]

A math C lib https://github.com/ferreiradaselva/mathc

Two passes of compute and a trivial “paste this on screen” vert+frag pair. The geometry is just a ray marched sdf with a single diffuse bounce per pixel, reconstructing a somewhat okay AO from that data is the interesting thing here.

I never used a compute shader. I will give a try soon.

Cool! There are a lot of alternatives but this looks relatively different so there might be actual use cases.

You definitely should, compute is the best. It opens lots of possibilities for more efficient implementations with the possibility to do persistent threading and shared memory operations. As a potentially helpful pointer/shameless self plug, I recently released a thing that implements printf in glsl. It's proven personally quite useful for debugging more complex stuff where outputting a single state doesn't quite cut it.

[Ordnas]

How's everyone's progress?

This week I was full of work, but the next one I will resume my 3rd person action game, made in Unreal Engine 4. If my task plan is correct, I should release my first "trailer" video after my 15th working hour from now.

[ferreiradaselva]

I recently released a thing that implements printf in glsl. It's proven personally quite useful for debugging more complex stuff where outputting a single state doesn't quite cut it.

Oh, I cloned that repo, hahhaa. I didn't know you made it.

This week I was full of work, but the next one I will resume my 3rd person action game, made in Unreal Engine 4. If my task plan is correct, I should release my first "trailer" video after my 15th working hour from now.

  Let us know.


I'm happy bc I'm at a point I'm only polishing my gamedev framework.

[qMopey]

c strings to integers, very satisfying 

[Zireael]

Happy because I finally got the car AI weaving less after 1,5 days of poking about (I have one function that is basically "car on rails" and one other which makes it more realistic (i.e. not stuck to the rails).

Next step, I guess I'll work on offsetting the found path so that it's in the middle of my lane, because then it'll stop crossing the lanes sometimes. (Which is fine for a racer, but not so much for traffic)

[ferreiradaselva]

c strings to integers, very satisfying 

Niiiiice!

Happy because I finally got the car AI weaving less after 1,5 days of poking about (I have one function that is basically "car on rails" and one other which makes it more realistic (i.e. not stuck to the rails).

Next step, I guess I'll work on offsetting the found path so that it's in the middle of my lane, because then it'll stop crossing the lanes sometimes. (Which is fine for a racer, but not so much for traffic)

Heck yeah! What are you using for physics? Also, is it 2D or 3D?

I was just working on that menu again, and I added some animation using easing functions, moon going up using cubic_ease_out() and stars distance from the moon using back_ease_out():

[Zireael]

Happy because I finally got the car AI weaving less after 1,5 days of poking about (I have one function that is basically "car on rails" and one other which makes it more realistic (i.e. not stuck to the rails).

Next step, I guess I'll work on offsetting the found path so that it's in the middle of my lane, because then it'll stop crossing the lanes sometimes. (Which is fine for a racer, but not so much for traffic)

Heck yeah! What are you using for physics? Also, is it 2D or 3D?

I'm using Godot, which comes with their own physics for now (although vehicle code is almost a direct copy of Bullet's). It's 3D and the only reason why I haven't started a thread is, it doesn't look nice so far. Waiting for Godot 3.0. on that count.