Forums > Jitter

Lumakey shader slower than jit.lumakey with noaccel

May 28, 2008 | 3:56 pm

I’m working through some OpenGL stuff on this machine:
2 X 2Ghz Mac Pro, 1 Gig RAM
OS 10.4.8
NVIDIA GeForce 7300 GT, 256 VRAM

When I load bball.mov and blading.mov into the patches below, I get around 75fps for the GPU version and around double that for the CPU version with noaccel turned on. Apparently upgrading to 10.4.9 improves OpenGL performance on this machine– as it’s a borrowed computer I’m holding off if it’s not going to make a difference.

lumakey_CPU

max v2;
#N vpatcher 375 79 898 454;
#P toggle 27 170 15 0;
#P window setfont "Sans Serif" 9.;
#P newex 53 199 29 196617 gate;
#P user jit.fpsgui 53 226 60 196617 0;
#P message 349 204 48 196617 binary 1;
#P flonum 365 157 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 365 176 46 196617 luma $1;
#P flonum 320 157 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 320 176 37 196617 tol $1;
#P flonum 239 158 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 239 177 44 196617 fade $1;
#P message 427 75 27 196617 stop;
#P flonum 455 55 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 455 75 42 196617 rate $1;
#P message 395 75 28 196617 read;
#P newex 311 127 144 196617 jit.qt.movie 320 240;
#P newex 239 201 76 196617 prepend param;
#P newex 317 256 32 196617 print;
#P message 168 202 67 196617 getparamlist;
#P newex 21 145 70 196617 jit.window fa;
#P newex 158 279 182 196617 jit.gl.videoplane fa @scale 1.333 1. 1.;
#P newex 21 123 121 196617 jit.gl.render fa @ortho 2;
#P newex 21 79 58 196617 t b b erase;
#P newex 158 229 162 196617 jit.gl.slab fa @file co.lumakey.jxs;
#P message 274 75 27 196617 stop;
#P flonum 302 55 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 302 75 42 196617 rate $1;
#P message 242 75 28 196617 read;
#P flonum 62 37 35 9 0.5 0 1 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P toggle 21 37 15 0;
#P newex 21 58 51 196617 qmetro 2;
#P newex 158 127 144 196617 jit.qt.movie 320 240;
#P connect 2 0 1 0;
#P connect 1 0 9 0;
#P fasten 9 2 10 0 74 105 26 105;
#P connect 9 0 10 0;
#P connect 30 0 29 0;
#P connect 29 0 28 0;
#P connect 3 0 1 1;
#P connect 0 0 29 1;
#P fasten 9 1 0 0 50 117 163 117;
#P fasten 5 0 0 0 307 117 163 117;
#P fasten 7 0 0 0 279 117 163 117;
#P fasten 4 0 0 0 247 117 163 117;
#P connect 0 0 8 0;
#P fasten 13 0 8 0 173 223 163 223;
#P fasten 15 0 8 0 244 223 163 223;
#P connect 8 0 11 0;
#P connect 22 0 21 0;
#P fasten 25 0 15 0 370 197 244 197;
#P fasten 23 0 15 0 325 197 244 197;
#P connect 21 0 15 0;
#P fasten 27 0 15 0 328 219 328 196;
#P connect 6 0 5 0;
#P connect 16 0 8 1;
#P fasten 9 1 16 0 50 117 316 117;
#P fasten 17 0 16 0 400 117 316 117;
#P fasten 20 0 16 0 432 117 316 117;
#P fasten 18 0 16 0 460 117 316 117;
#P fasten 8 1 14 0 315 251 322 251;
#P connect 24 0 23 0;
#P connect 26 0 25 0;
#P connect 19 0 18 0;
#P pop;

lumakey_GPU

max v2;
#N vpatcher 395 65 918 440;
#P toggle 57 136 15 0;
#P window setfont "Sans Serif" 9.;
#P window linecount 1;
#P newex 57 157 29 196617 gate;
#P message 25 284 55 196617 noaccel $1;
#P toggle 25 266 15 0;
#P flonum 311 159 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 311 184 39 196617 lum $1;
#P flonum 398 159 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 398 184 42 196617 fade $1;
#P flonum 353 159 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 353 184 35 196617 tol $1;
#P newex 166 232 58 196617 jit.window;
#P newex 184 184 61 196617 jit.lumakey;
#P user jit.fpsgui 57 188 60 196617 0;
#P message 427 75 27 196617 stop;
#P flonum 455 55 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 455 75 42 196617 rate $1;
#P message 395 75 28 196617 read;
#P newex 311 127 144 196617 jit.qt.movie 320 240;
#P message 274 75 27 196617 stop;
#P flonum 302 55 35 9 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P message 302 75 42 196617 rate $1;
#P message 242 75 28 196617 read;
#P flonum 62 37 35 9 0.5 0 1 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P toggle 21 37 15 0;
#P newex 21 58 51 196617 qmetro 2;
#P newex 158 127 144 196617 jit.qt.movie 320 240;
#P connect 2 0 1 0;
#P connect 22 0 23 0;
#P connect 25 0 24 0;
#P connect 24 0 13 0;
#P connect 3 0 1 1;
#P connect 0 0 24 1;
#P connect 1 0 0 0;
#P fasten 4 0 0 0 247 117 163 117;
#P fasten 7 0 0 0 279 117 163 117;
#P fasten 5 0 0 0 307 117 163 117;
#P connect 14 0 15 0;
#P connect 23 0 15 0;
#P connect 0 0 14 0;
#P connect 20 0 14 0;
#P connect 16 0 14 0;
#P connect 18 0 14 0;
#P connect 8 0 14 1;
#P connect 6 0 5 0;
#P fasten 10 0 8 0 460 117 316 117;
#P fasten 12 0 8 0 432 117 316 117;
#P fasten 9 0 8 0 400 117 316 117;
#P connect 1 0 8 0;
#P connect 21 0 20 0;
#P connect 17 0 16 0;
#P connect 19 0 18 0;
#P connect 11 0 10 0;
#P pop;


May 28, 2008 | 4:35 pm

your gpu patches will not run faster than your machine’s display refresh rate. my guess is that you’re looking at a machine with a 75hz rate, yes?


May 29, 2008 | 6:42 am

Sure it can, just disable vsync on the jit.window, and you’ll get very high fps. But for all intents and purposes, what Josh said.


Viewing 3 posts - 1 through 3 (of 3 total)