my particle system
Community Forums/Showcase/my particle system
| ||
| just got this particle system working, I think it is really fast (at least compared to all the others that I have used). Could somebody download this and try it out, and post what the fps was? links removed |
| ||
| Looks like FPS is locked to monitor refresh rate. I get 60 fps for 8400 particles. P4 3mhz, gfx440mx. XP Pro. Good speed. I get similar frame rates using my own single surface library with no external dll's. Stevie |
| ||
| 18fps 8400Particles Sempron 3100+ SYS M760 |
| ||
| 60 FPS 8400 particles |
| ||
| 59 fps 8300 - 8900 particles |
| ||
| 70 FPS Also tried it after turning off Vsync in the drivers, but got the same result! |
| ||
| I don't quite know what is happening...there is no frame limit code at all anywhere...just a flip true. Stevie G: How do you do that??? The best I was able to get without a dll with 8000 triangles was 34 fps, and this runs at 68! |
| ||
| flip true does sync to 60Hz on most systems. Use Flip false instead for fps calculation. |
| ||
| 37FPS, which is good on my antique GF1 GTS 32 MB and P4 2.4 / 1GB/ WinXP |
| ||
| I cant see any link for download |
| ||
| sorry about that...now there are two tests, one with about 30,000 and the other with about 8,000. I also made a few adjustments, as well as made it so it now flips false, and does not lock the frames per second down. |
| ||
| The download seems to have been removed. |
| ||
| GRRRR...I am getting annoyed at that website, does anyone know of a better site to host things on??? It is up again, but who knows how long it will stay up... On a seperate note, I have encountered a slightly annoying bug. The program runs fine in the IDE, but when you compile it into an exe, the program is getting messed up. What does blitz do different from when you push the compile and run button to when you push the compile into an exe button? Here is an example. Both versions work with the IDE, but neither work when creating an exe (but they don't work in different ways). One uses the TormVector command with matrices from MiniB3D and the other use a tformvector that I came up with using quaternions. I have another version using my version of matrices, but it does the same thing. I am stumped as to why this is happening... (see link above) |
| ||
| You can remote link either according to the anglefire message. |
| ||
| Okay...how should I upload things? edit: I think it is working now...who knows for how long. particles is the demos, while particle demo is the demo that ilustrates a problem. |
| ||
| Don't know if you still wanted testing :) Just ran your "particles_small" demo and was getting 50/51 fps with ~9900 particles. Unable to run the "particles_prob" code. (Core2Duo 2.4G, Geforce7600) |
| ||
| What is wrong with the particles_prob? In the blitz3d programming section I have a thread about it and it's problems...any input on that would be helpful. Thanks for doing the test, I might make a version that has about 33000 particles soon, but I am trying to get this problem sorted out first. |
| ||
| Oh I think when I tried to run the code it errored with "could not find Init_Triangles()"? |
| ||
| Thats strange...It works fine here. Did you unzip it first? Anyways, there is a new demo to try out using just over 33000 particles, please try it out. |
| ||
| Large: >33000 particles, 16 FPS Small: circa 9900 particles, 51 FPS |
| ||
| LETO -> make sure you copy the particles.decls into your blitz3d\userlibs folder before running /StuC |
| ||
| Small: ~9900 particles, ~18PFS on ATI RADEON 9600 series card with 1.6GHz AMD |
   |