Messages

Uhm, I talked a bit with CW yesterday, I think the easiest way right now would be to run the game under the VS debugger. With it we can see how AGS is allocating ram across things to see what's going on - why we get this Out of Memory, maybe something else is using the RAM.

About adding more information when things like this happen, I don't know if CW had any additional thoughts.

8318416=2079604*4=1922*1082*4

It looks like it's a GUI that is the size of the screen that has a nonzero transparency is the specific case reported, but I wonder how is the memory usage in general.

Just looking into cool new ways to use AGS    but I think I won't be able to hit my MAGS goal, so probably will not be doing anything now, maybe in the future.

Our dll comes from SDL2 releases from their page here:

https://github.com/libsdl-org/SDL/releases

It's the asset that ends in VC.

It appears the sample rate would be slightly wrong when using the Windows in some case and the distortion is almost imperceptible except by being slightly slower from icculus description above. I had not notice it in AGS before, but thought maybe it could be relevant.

@Crimson Wizard I wonder if it could be a problem in SDL2: https://github.com/libsdl-org/SDL/issues/6326#issuecomment-1267599472

Theoretically the resampling change could mean maybe it works when using the 2.26.1 SDL dll? (https://github.com/libsdl-org/SDL/releases/download/release-2.26.1/SDL2-2.26.1-win32-x86.zip)

I showed the video above to someone and they asked me about the effects on the graves. It appears the screen distortion effect can be dependent on VSync being active or not in some computers to render properly, I had no idea. I know some computers runs AGS games slower when using VSync, so perhaps a thing to test when using winsetup here.

@Khris it will take me a bit to go back to this, I am trying to look into some other things right now for a game idea. Thank you for the module you made.

Haven't yet advanced much, but testing around with lots of slices, above is an example from Magica Voxel. Need to properly figure out the camera stuff, but it's been promissing.

Edit: oh, apparently I can use the camera height of twice the size, in the room, and by keeping the viewport the same, I can make the image squashed on vertical, which gives the illusion of it being rotating in a perspective like look. Very interesting!

Edit2: UHM, if I map x,y,z to cylindrical coordinates I can position something in a rotated world and still have game controls that make sense!

Holy cat! Someone played my game on YouTube! Kinda fascinating to watch! The player also picked up the SNES graphics intention!

(I think you forgot to link where to download the game!   )

Yeah @newwaveburritos ! I love this stuff because it's the minimum scope sort of thing.

Trying to figure a good strategy for MAGS for me, I think it's something like so:

• Keep Scope Small
• Design the game so it can be developed interactively
• Finish a prototype of the vision in the first week
• Get testers early
• Abandon your game if needed

But unfortunately this month I don't think I will make it, so I am probably on last item already.

I have one idea @ZapZap , but I will only have the last two weeks to code and I am not sure it's a good idea though.

Spoiler

but it would require Magica Voxel

[close]

@lorenzo, here is the kitten when I was submitting the game on the forums.



@heltenjon, I am pleased with the gameplay specifics feedback, I think it helped to balance so it's playable, and I don't think I would be able to polish the forest design towards orienting the player in it without your help! I think it helped a lot with shaping it and your feedback was amazing as you are very patient to write things very specifically.

@Shadow1000 thanks for the play time information! This is something I had no idea since I knew the map beforehand and I asked the testers to play too many times until they eventually knew too the map.

@Creamy thanks for playing! I wanted to play more with the title screen but this was the idea I ended up going for - wanted it to be demosceney.

This is a great theme! Also, unexpected crime scene at the dinosaur is a new one, thanks for it.

[count = 1;]

Can I ask one more thing about the code? In line 176, inside the if clause, you do count = 1;, instead of an increment. Now I tried to do an increment, and visually it looked the same. But this means only 1 point per pass is more constrained than lc? Now I mention this because if this is true, I wonder if it's an opportunity for optimization.

Edit: Nevermind, now I found two lines up.

Edit2: on line 112 to 119, not sure you know, but you can use alt+shift to select the x and y stuff in the vertical block to paste on top of line 154 to 161.

I am having a hard time figuring out these steps in your code

Observation:
Find a wave element with the minimal nonzero entropy. If there is no such elements (if all elements have zero or undefined entropy) then break the cycle (4) and go to step (5).
Collapse this element into a definite state according to its coefficients and the distribution of NxN patterns in the input.
Propagation: propagate information gained on the previous observation step.

Sorry if I am an idiot, but how do you name entropy?

I'm currently checking all pixels each pass though so ideally there should be a list of blank ones that slowly gets reduced to zero.

I am trying to figure kind of this, essentially the ones that are left. 

@Khris it's amazing!



I can force at least 4 passes in my CPU, per frame, so I can do it 4x faster than the gif.

I thought maybe GetPixel and SetPixel are too slow... And they are, so I made a hacked version where they are slightly faster, using a hack for set pixel get pixel internally.

Code: ags

((int*)ds->GetAllegroBitmap()->line[y])[x]

Still, because you used GetPixel and SetPixel quite smartly, they don't appear to matter that much currently.

Now I was curious about why the sequence is 8 pixels instead of 9 for 3x3 tile, and wondered if this could matter for the missing dots in the final image. I hacked around the code for this but apparently it doesn't matter, so my guess is you have the center pixel stored separately to save ram. I will read the code more carefully later. 

Amazing @Khris !!! I am really impressed! Usually once one get into a code that at least works, getting it optimized and cleaning up is the easy part - if you share yours I wouldn't mind scrubbing it's edges.

But anyway, great work, really impressed! 

Reading the original code today here: https://github.com/mxgmn/WaveFunctionCollapse

There's an explanation of the algorithm in the readme I think I am almost understanding it. I think I will try later to base it from here.

I tried to look for tutorials but most of what I found uses a pre-made library and it's more about using and experimenting than implementing the algorithm.