# jittery motion with filloval **URL:** \ **Category:** Uncategorized\ **Created:** [January 14, 2009, 3:48pm UTC](https://psychtoolbox.discourse.group/t/jittery-motion-with-filloval/2042 "2009-01-14T15:48:34Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![Allen\_Ingling](https://avatars.discourse-cdn.com/v4/letter/a/a183cd/32.png) [@Allen\_Ingling](https://psychtoolbox.discourse.group/u/Allen_Ingling)\ **Post date:** [January 14, 2009, 3:48pm UTC](https://psychtoolbox.discourse.group/t/jittery-motion-with-filloval/2042/1 "2009-01-14T15:48:34Z")
Hello everyone. My goal is to have a circle, drawn by screen('FillOval'), move according to a predefined random trajectory on the screen. Pretty simple eh? The problem is the circle jitters / vibrates /jumps every now and then as it is moving on its trajectory. the funny thing is if I use the same code but replace FillOval with FillRect, the motion is smooth! Any tips on why this is happening and how to prevent it ? This simple code will be used as part of a more complex project involving alpha blending. Thanks in advance ---
**Author:** ![Allen\_Ingling](https://avatars.discourse-cdn.com/v4/letter/a/a183cd/32.png) [@Allen\_Ingling](https://psychtoolbox.discourse.group/u/Allen_Ingling)\ **Post date:** [January 15, 2009, 4:05pm UTC](https://psychtoolbox.discourse.group/t/jittery-motion-with-filloval/2042/2 "2009-01-15T16:05:25Z")
right now all I want to accomplish is Screen(window,'FillOval', [0 0 255], mask{frame}); flipTime = Screen(window, 'Flip'); Mac Pro OS X 10.5.3 2 X 3.2 Quad Core intel graphics cards: ATI Radeon HD 2600 XT NVIDIA GeForce 8800 GT > --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mario Kleiner" \ wrote: > \> > \> What do you mean with "jitters/vibrates"? > \> > \> A "jump" would mean for me that you miss video refresh > \> intervals, so the motion is not smooth, maybe because > \> drawing the filled oval is too slow on your graphics > \> hardware? What graphics card, operating system do > \> you have? Example code? > \> > \> -mario > \> > \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> \> > \> \> Hello everyone. > \> \> > \> \> My goal is to have a circle, drawn by screen('FillOval'), move according to a predefined > \> \> random trajectory on the screen. Pretty simple eh? The problem is the circle jitters / > \> vibrates > \> \> /jumps every now and then as it is moving on its trajectory. the funny thing is if I use > \> the > \> \> same code but replace FillOval with FillRect, the motion is smooth! > \> \> > \> \> Any tips on why this is happening and how to prevent it ? > \> \> > \> \> > \> \> This simple code will be used as part of a more complex project involving alpha > \> blending. > \> \> > \> \> Thanks in advance > \> \> > \> ---
**Author:** ![mariokleiner](https://avatars.discourse-cdn.com/v4/letter/m/13edae/32.png) [@mariokleiner](https://psychtoolbox.discourse.group/u/mariokleiner)\ **Post date:** [January 16, 2009, 12:22am UTC](https://psychtoolbox.discourse.group/t/jittery-motion-with-filloval/2042/3 "2009-01-16T00:22:02Z")
David Fencsiks tip is a good one. Provide the size of the biggest oval you're going to draw, ie. if the max diameter is say 100 pixels, do a > \> Screen(window,'FillOval', [0 0 255], mask{frame}, 100); If the resolution of your display is high and the size of your ovals is small then the GPU has to do much more work than neccessary, and you can cut down by providing the correct maximum value. if your animation loop really just looks like... for frame=1:something Screen(window,'FillOval', [0 0 255], mask{frame}); flipTime = Screen(window, 'Flip'); end ... then Filloval is too slow and providing the maxdiameter parameter is the best you can do to cut down drawing time. Do you get many skipped frames? Ovals are expensive to draw, because their round shape must be approximated by lots of small line segments or filled triangles, so this is quite a bit of load in some cases. The 'DrawDots' method is efficient for circles, but the upper limit on NVidia and ATI hardware is around 63-64 pixels diameter, on Intel hardware even lower (around 10 or so). The other way to do it would be to draw the set of ovals into a set of Offscreen windows, then 'DrawTexture' those to the screen, as demonstrated in "DrawingSpeedTest". -mario > --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> > \> right now all I want to accomplish is > \> > \> > \> Mac Pro > \> OS X 10.5.3 > \> 2 X 3.2 Quad Core intel > \> graphics cards: > \> ATI Radeon HD 2600 XT > \> NVIDIA GeForce 8800 GT > \> > \> > \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mario Kleiner" \ wrote: > \> \> > \> \> What do you mean with "jitters/vibrates"? > \> \> > \> \> A "jump" would mean for me that you miss video refresh > \> \> intervals, so the motion is not smooth, maybe because > \> \> drawing the filled oval is too slow on your graphics > \> \> hardware? What graphics card, operating system do > \> \> you have? Example code? > \> \> > \> \> -mario > \> \> > \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> \> \> > \> \> \> Hello everyone. > \> \> \> > \> \> \> My goal is to have a circle, drawn by screen('FillOval'), move according to a > predefined > \> \> \> random trajectory on the screen. Pretty simple eh? The problem is the circle jitters / > \> \> vibrates > \> \> \> /jumps every now and then as it is moving on its trajectory. the funny thing is if I > use > \> \> the > \> \> \> same code but replace FillOval with FillRect, the motion is smooth! > \> \> \> > \> \> \> Any tips on why this is happening and how to prevent it ? > \> \> \> > \> \> \> > \> \> \> This simple code will be used as part of a more complex project involving alpha > \> \> blending. > \> \> \> > \> \> \> Thanks in advance > \> \> \> > \> \> > \> ---
**Author:** ![coder244](https://avatars.discourse-cdn.com/v4/letter/c/cc9497/32.png) [@coder244](https://psychtoolbox.discourse.group/u/coder244)\ **Post date:** [January 19, 2009, 5:01pm UTC](https://psychtoolbox.discourse.group/t/jittery-motion-with-filloval/2042/4 "2009-01-19T17:01:37Z")
forget the last post, this is the output from testCircle.m ( I had changed graphics cards to test it) PTB-INFO: This is the OpenGL-Psychtoolbox for Apple OS X, version 3.0.8. (Build date: Aug 15 2008) PTB-INFO: Type 'PsychtoolboxVersion' for more detailed version information. PTB-INFO: Psychtoolbox is licensed to you under terms of the GNU General Public License (GPL). See file 'License.txt' in the PTB-INFO: Psychtoolbox root folder for a copy of the GPL license. OpenGL-Extensions are: GL\_ARB\_transpose\_matrix GL\_ARB\_vertex\_program GL\_ARB\_vertex\_blend GL\_ARB\_window\_pos GL\_ARB\_shader\_objects GL\_ARB\_vertex\_shader GL\_ARB\_shading\_language\_100 GL\_EXT\_multi\_draw\_arrays GL\_EXT\_clip\_volume\_hint GL\_EXT\_rescale\_normal GL\_EXT\_draw\_range\_elements GL\_EXT\_fog\_coord GL\_EXT\_gpu\_program\_parameters GL\_EXT\_geometry\_shader4 GL\_EXT\_transform\_feedback GL\_APPLE\_client\_storage GL\_APPLE\_specular\_vector GL\_APPLE\_transform\_hint GL\_APPLE\_packed\_pixels GL\_APPLE\_fence GL\_APPLE\_vertex\_array\_object GL\_APPLE\_vertex\_program\_evaluators GL\_APPLE\_element\_array GL\_APPLE\_flush\_render GL\_APPLE\_aux\_depth\_stencil GL\_NV\_texgen\_reflection GL\_NV\_light\_max\_exponent GL\_IBM\_rasterpos\_clip GL\_SGIS\_generate\_mipmap GL\_ARB\_imaging GL\_ARB\_point\_parameters GL\_ARB\_texture\_env\_crossbar GL\_ARB\_texture\_border\_clamp GL\_ARB\_multitexture GL\_ARB\_texture\_env\_add GL\_ARB\_texture\_cube\_map GL\_ARB\_texture\_env\_dot3 GL\_ARB\_multisample GL\_ARB\_texture\_env\_combine GL\_ARB\_texture\_compression GL\_ARB\_texture\_mirrored\_repeat GL\_ARB\_shadow GL\_ARB\_depth\_texture GL\_ARB\_fragment\_program GL\_ARB\_fragment\_program\_shadow GL\_ARB\_fragment\_shader GL\_ARB\_occlusion\_query GL\_ARB\_point\_sprite GL\_ARB\_texture\_non\_power\_of\_two GL\_ARB\_vertex\_buffer\_object GL\_ARB\_pixel\_buffer\_object GL\_ARB\_draw\_buffers GL\_ARB\_shader\_texture\_lod GL\_EXT\_compiled\_vertex\_array GL\_EXT\_framebuffer\_object GL\_EXT\_framebuffer\_blit GL\_EXT\_framebuffer\_multisample GL\_EXT\_texture\_rectangle GL\_ARB\_texture\_rectangle GL\_EXT\_texture\_env\_add GL\_EXT\_blend\_color GL\_EXT\_blend\_minmax GL\_EXT\_blend\_subtract GL\_EXT\_texture\_lod\_bias GL\_EXT\_abgr GL\_EXT\_bgra GL\_EXT\_stencil\_wrap GL\_EXT\_texture\_filter\_anisotropic GL\_EXT\_secondary\_color GL\_EXT\_blend\_func\_separate GL\_EXT\_shadow\_funcs GL\_EXT\_stencil\_two\_side GL\_EXT\_depth\_bounds\_test GL\_EXT\_texture\_compression\_s3tc GL\_EXT\_texture\_compression\_dxt1 GL\_EXT\_texture\_sRGB GL\_EXT\_blend\_equation\_separate GL\_EXT\_texture\_mirror\_clamp GL\_EXT\_packed\_depth\_stencil GL\_EXT\_bindable\_uniform GL\_EXT\_texture\_integer GL\_EXT\_gpu\_shader4 GL\_EXT\_draw\_buffers2 GL\_APPLE\_flush\_buffer\_range GL\_APPLE\_ycbcr\_422 GL\_APPLE\_vertex\_array\_range GL\_APPLE\_texture\_range GL\_APPLE\_float\_pixels GL\_ATI\_texture\_float GL\_ARB\_texture\_float GL\_ARB\_half\_float\_pixel GL\_APPLE\_pixel\_buffer GL\_APPLE\_object\_purgeable GL\_NV\_point\_sprite GL\_NV\_register\_combiners GL\_NV\_register\_combiners2 GL\_NV\_blend\_square GL\_NV\_texture\_shader GL\_NV\_texture\_shader2 GL\_NV\_texture\_shader3 GL\_NV\_fog\_distance GL\_NV\_depth\_clamp GL\_NV\_multisample\_filter\_hint GL\_NV\_fragment\_program\_option GL\_NV\_fragment\_program2 GL\_NV\_vertex\_program2\_option GL\_NV\_vertex\_program3 GL\_ATI\_texture\_mirror\_once GL\_ATI\_texture\_env\_combine3 GL\_ATI\_separate\_stencil GL\_SGIS\_texture\_edge\_clamp GL\_SGIS\_texture\_lod PTB-INFO: You are using a multi-display setup (2 active displays): PTB-INFO: Please read 'help MultiDisplaySetups' for specific information on the Do's, Dont's, PTB-INFO: and possible causes of trouble and how to diagnose and resolve them. PTB-INFO: According to the operating system, some of your connected displays do not seem to PTB-INFO: be switched into mirror mode. For a discussion of mirror mode vs. non-mirror mode, PTB-INFO: please read 'help MirrorMode'. PTB-INFO: OpenGL-Renderer is NVIDIA Corporation :: NVIDIA GeForce 8800 GT OpenGL Engine :: 2.0 NVIDIA-1.5.36 PTB-INFO: Renderer has 512 MB of VRAM and a maximum 493 MB of texture memory. PTB-Info: VBL startline = 1050 , VBL Endline = 1050 PTB-Info: Measured monitor refresh interval from beamposition = 16.678533 ms [59.957311 Hz]. PTB-Info: Will use beamposition query for accurate Flip time stamping. PTB-Info: Measured monitor refresh interval from VBLsync = 16.679081 ms [59.955340 Hz]. (50 valid samples taken, stddev=0.033410 ms.) PTB-Info: Reported monitor refresh interval from operating system = 16.666667 ms [60.000000 Hz]. PTB-Info: Small deviations between reported values are normal and no reason to worry. I am using a high resolution Im not entirely sure about the missed frames, is there a way to check > --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mario Kleiner" \ wrote: > \> > \> Hmm, > \> > \> your code is missing. Just put it into the "Files" section > \> of the forum. > \> > \> However, i tested drawing a 'FillOval' within a > \> 'FrameRect', passing the same rect parameter as > \> you described. I tested in an animation loop, > \> on a MacPro like yours with NVidia 8800 under > \> OS/X 10.5.6, and on a MacBookPro with ATI X1600 > \> und 10.4.11. > \> > \> I can't reproduce the "jitter", the oval always nicely > \> fits into the framed rectangle with no wobble whatsoever, > \> i don't get any skipped frames at 90 Hz refresh, with > \> or without the 'maxdiameter' parameter at a resolution > \> of 1600 x 1400 pixels and the drawing time is about > \> 0.4 msecs for an 800 x 800 pixels oval, way too short > \> to cause timing problems. > \> > \> So either: > \> > \> a) You found an operating system bug in your OS and > \> an upgrade would fix it. Rather unlikely but not impossible. > \> > \> b) Something else is causing massive timing problems? > \> > \> c) Your code does something unhealthy? > \> > \> d) Its not real, but some kind of artifact of motion > \> perception? Have yout tried playing it at a low redraw > \> rate or as sequence of still images to see if there is > \> really something in the picture that causes this? > \> > \> I think of things like this "Brillenputzillusion" (look > \> at the frame of the left glass): > \> > \> \<[http://www.yakachu.de/temp/brillenputzillusion1.avi](http://www.yakachu.de/temp/brillenputzillusion1.avi)\> > \> \<[http://www.yakachu.de/temp/brillenputzillusion2.avi](http://www.yakachu.de/temp/brillenputzillusion2.avi)\> > \> > \> For b/c we need a minimal piece of code that shows > \> the problem. Does PTB report lots of skipped frames/ > \> missed flip deadlines at the end? Do you use very > \> high resolution and very high refresh rate? > \> > \> For d) Did you test on a CRT or on a flat panel? > \> > \> -mario > \> > \> > \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> \> > \> \> THanks for all the replies so far. > \> \> > \> \> The UpToMaxDiameter setting does not solve my problem. > \> \> > \> \> I tried using offscreen windows as in the DrawingSpeedTest example but this is not > what > \> I > \> \> want to achieve. This is my goal: > \> \> For each frame (monitor refresh) a NEW circle is drawn on the screen, its position > \> \> incremented slightly, this gives the illusion a Single circle without a trace/trail is > moving > \> \> around the screen. > \> \> > \> \> The moving circle will act as an aperture for complex moving dots. So using drawDots > \> \> directly is not an option. My code is complex but here is a simplified version of what > I'm > \> \> doing: > \> \> > \> \> for i =1: numberOfFrames > \> \> > \> \> Screen(window,'FillOval', [0 0 0 255], mask{i},maxSize); > \> \> > \> \> end > \> \> > \> \> mask is essential a cell containing 4 column vectors for each new position at each > \> frame, > \> \> this part works as I mentioned before using FillRect gives a filled rectangle moving > \> around > \> \> the screen smoothly. I know its a speed issue with fillOval, so the question is, is there > a > \> \> way to work around it? > \> \> > \> \> here is some code ( its alot more complex) > \> \> > \> \> > \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mario Kleiner" \ wrote: > \> \> \> > \> \> \> David Fencsiks tip is a good one. Provide the size > \> \> \> of the biggest oval you're going to draw, ie. if > \> \> \> the max diameter is say 100 pixels, do a > \> \> \> > \> \> \> \> Screen(window,'FillOval', [0 0 255], mask{frame}, 100); > \> \> \> > \> \> \> If the resolution of your display is high and the > \> \> \> size of your ovals is small then the GPU has to do > \> \> \> much more work than neccessary, and you can > \> \> \> cut down by providing the correct maximum value. > \> \> \> > \> \> \> if your animation loop really just looks like... > \> \> \> > \> \> \> for frame=1:something > \> \> \> Screen(window,'FillOval', [0 0 255], mask{frame}); > \> \> \> flipTime = Screen(window, 'Flip'); > \> \> \> end > \> \> \> > \> \> \> ... then Filloval is too slow and providing the > \> \> \> maxdiameter parameter is the best you can > \> \> \> do to cut down drawing time. > \> \> \> > \> \> \> Do you get many skipped frames? > \> \> \> > \> \> \> Ovals are expensive to draw, because their > \> \> \> round shape must be approximated by lots of > \> \> \> small line segments or filled triangles, so this > \> \> \> is quite a bit of load in some cases. > \> \> \> > \> \> \> The 'DrawDots' method is efficient for circles, but the > \> \> \> upper limit on NVidia and ATI hardware is around > \> \> \> 63-64 pixels diameter, on Intel hardware even > \> \> \> lower (around 10 or so). > \> \> \> > \> \> \> The other way to do it would be to draw the > \> \> \> set of ovals into a set of Offscreen windows, > \> \> \> then 'DrawTexture' those to the screen, as > \> \> \> demonstrated in "DrawingSpeedTest". > \> \> \> > \> \> \> -mario > \> \> \> > \> \> \> > \> \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> \> \> \> > \> \> \> \> right now all I want to accomplish is > \> \> \> \> > \> \> \> \> > \> \> \> \> Mac Pro > \> \> \> \> OS X 10.5.3 > \> \> \> \> 2 X 3.2 Quad Core intel > \> \> \> \> graphics cards: > \> \> \> \> ATI Radeon HD 2600 XT > \> \> \> \> NVIDIA GeForce 8800 GT > \> \> \> \> > \> \> \> \> > \> \> \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mario Kleiner" \ wrote: > \> \> \> \> \> > \> \> \> \> \> What do you mean with "jitters/vibrates"? > \> \> \> \> \> > \> \> \> \> \> A "jump" would mean for me that you miss video refresh > \> \> \> \> \> intervals, so the motion is not smooth, maybe because > \> \> \> \> \> drawing the filled oval is too slow on your graphics > \> \> \> \> \> hardware? What graphics card, operating system do > \> \> \> \> \> you have? Example code? > \> \> \> \> \> > \> \> \> \> \> -mario > \> \> \> \> \> > \> \> \> \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> \> \> \> \> \> > \> \> \> \> \> \> Hello everyone. > \> \> \> \> \> \> > \> \> \> \> \> \> My goal is to have a circle, drawn by screen('FillOval'), move according to a > \> \> \> predefined > \> \> \> \> \> \> random trajectory on the screen. Pretty simple eh? The problem is the circle > \> \> jitters / > \> \> \> \> \> vibrates > \> \> \> \> \> \> /jumps every now and then as it is moving on its trajectory. the funny thing > is > \> if I > \> \> \> use > \> \> \> \> \> the > \> \> \> \> \> \> same code but replace FillOval with FillRect, the motion is smooth! > \> \> \> \> \> \> > \> \> \> \> \> \> Any tips on why this is happening and how to prevent it ? > \> \> \> \> \> \> > \> \> \> \> \> \> > \> \> \> \> \> \> This simple code will be used as part of a more complex project involving > \> alpha > \> \> \> \> \> blending. > \> \> \> \> \> \> > \> \> \> \> \> \> Thanks in advance > \> \> \> \> \> \> > \> \> \> \> \> > \> \> \> \> > \> \> \> > \> \> > \> ---
**Author:** ![coder244](https://avatars.discourse-cdn.com/v4/letter/c/cc9497/32.png) [@coder244](https://psychtoolbox.discourse.group/u/coder244)\ **Post date:** [January 19, 2009, 2:39pm UTC](https://psychtoolbox.discourse.group/t/jittery-motion-with-filloval/2042/5 "2009-01-19T14:39:10Z")
okay, I've finally put some code up (simplified) theres two files a .mat file you need to load and a testCircle.m . It might be my random position generation thats wrong but if you try the same data with fillRect and fillOval you get smooth motion in the first. If anyone tries and has any thoughts like me know. Thanks again! > --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mario Kleiner" \ wrote: > \> > \> Hmm, > \> > \> your code is missing. Just put it into the "Files" section > \> of the forum. > \> > \> However, i tested drawing a 'FillOval' within a > \> 'FrameRect', passing the same rect parameter as > \> you described. I tested in an animation loop, > \> on a MacPro like yours with NVidia 8800 under > \> OS/X 10.5.6, and on a MacBookPro with ATI X1600 > \> und 10.4.11. > \> > \> I can't reproduce the "jitter", the oval always nicely > \> fits into the framed rectangle with no wobble whatsoever, > \> i don't get any skipped frames at 90 Hz refresh, with > \> or without the 'maxdiameter' parameter at a resolution > \> of 1600 x 1400 pixels and the drawing time is about > \> 0.4 msecs for an 800 x 800 pixels oval, way too short > \> to cause timing problems. > \> > \> So either: > \> > \> a) You found an operating system bug in your OS and > \> an upgrade would fix it. Rather unlikely but not impossible. > \> > \> b) Something else is causing massive timing problems? > \> > \> c) Your code does something unhealthy? > \> > \> d) Its not real, but some kind of artifact of motion > \> perception? Have yout tried playing it at a low redraw > \> rate or as sequence of still images to see if there is > \> really something in the picture that causes this? > \> > \> I think of things like this "Brillenputzillusion" (look > \> at the frame of the left glass): > \> > \> \<[http://www.yakachu.de/temp/brillenputzillusion1.avi](http://www.yakachu.de/temp/brillenputzillusion1.avi)\> > \> \<[http://www.yakachu.de/temp/brillenputzillusion2.avi](http://www.yakachu.de/temp/brillenputzillusion2.avi)\> > \> > \> For b/c we need a minimal piece of code that shows > \> the problem. Does PTB report lots of skipped frames/ > \> missed flip deadlines at the end? Do you use very > \> high resolution and very high refresh rate? > \> > \> For d) Did you test on a CRT or on a flat panel? > \> > \> -mario > \> > \> > \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> \> > \> \> THanks for all the replies so far. > \> \> > \> \> The UpToMaxDiameter setting does not solve my problem. > \> \> > \> \> I tried using offscreen windows as in the DrawingSpeedTest example but this is not > what > \> I > \> \> want to achieve. This is my goal: > \> \> For each frame (monitor refresh) a NEW circle is drawn on the screen, its position > \> \> incremented slightly, this gives the illusion a Single circle without a trace/trail is > moving > \> \> around the screen. > \> \> > \> \> The moving circle will act as an aperture for complex moving dots. So using drawDots > \> \> directly is not an option. My code is complex but here is a simplified version of what > I'm > \> \> doing: > \> \> > \> \> for i =1: numberOfFrames > \> \> > \> \> Screen(window,'FillOval', [0 0 0 255], mask{i},maxSize); > \> \> > \> \> end > \> \> > \> \> mask is essential a cell containing 4 column vectors for each new position at each > \> frame, > \> \> this part works as I mentioned before using FillRect gives a filled rectangle moving > \> around > \> \> the screen smoothly. I know its a speed issue with fillOval, so the question is, is there > a > \> \> way to work around it? > \> \> > \> \> here is some code ( its alot more complex) > \> \> > \> \> > \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mario Kleiner" \ wrote: > \> \> \> > \> \> \> David Fencsiks tip is a good one. Provide the size > \> \> \> of the biggest oval you're going to draw, ie. if > \> \> \> the max diameter is say 100 pixels, do a > \> \> \> > \> \> \> \> Screen(window,'FillOval', [0 0 255], mask{frame}, 100); > \> \> \> > \> \> \> If the resolution of your display is high and the > \> \> \> size of your ovals is small then the GPU has to do > \> \> \> much more work than neccessary, and you can > \> \> \> cut down by providing the correct maximum value. > \> \> \> > \> \> \> if your animation loop really just looks like... > \> \> \> > \> \> \> for frame=1:something > \> \> \> Screen(window,'FillOval', [0 0 255], mask{frame}); > \> \> \> flipTime = Screen(window, 'Flip'); > \> \> \> end > \> \> \> > \> \> \> ... then Filloval is too slow and providing the > \> \> \> maxdiameter parameter is the best you can > \> \> \> do to cut down drawing time. > \> \> \> > \> \> \> Do you get many skipped frames? > \> \> \> > \> \> \> Ovals are expensive to draw, because their > \> \> \> round shape must be approximated by lots of > \> \> \> small line segments or filled triangles, so this > \> \> \> is quite a bit of load in some cases. > \> \> \> > \> \> \> The 'DrawDots' method is efficient for circles, but the > \> \> \> upper limit on NVidia and ATI hardware is around > \> \> \> 63-64 pixels diameter, on Intel hardware even > \> \> \> lower (around 10 or so). > \> \> \> > \> \> \> The other way to do it would be to draw the > \> \> \> set of ovals into a set of Offscreen windows, > \> \> \> then 'DrawTexture' those to the screen, as > \> \> \> demonstrated in "DrawingSpeedTest". > \> \> \> > \> \> \> -mario > \> \> \> > \> \> \> > \> \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> \> \> \> > \> \> \> \> right now all I want to accomplish is > \> \> \> \> > \> \> \> \> > \> \> \> \> Mac Pro > \> \> \> \> OS X 10.5.3 > \> \> \> \> 2 X 3.2 Quad Core intel > \> \> \> \> graphics cards: > \> \> \> \> ATI Radeon HD 2600 XT > \> \> \> \> NVIDIA GeForce 8800 GT > \> \> \> \> > \> \> \> \> > \> \> \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mario Kleiner" \ wrote: > \> \> \> \> \> > \> \> \> \> \> What do you mean with "jitters/vibrates"? > \> \> \> \> \> > \> \> \> \> \> A "jump" would mean for me that you miss video refresh > \> \> \> \> \> intervals, so the motion is not smooth, maybe because > \> \> \> \> \> drawing the filled oval is too slow on your graphics > \> \> \> \> \> hardware? What graphics card, operating system do > \> \> \> \> \> you have? Example code? > \> \> \> \> \> > \> \> \> \> \> -mario > \> \> \> \> \> > \> \> \> \> \> --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "coder244" \ wrote: > \> \> \> \> \> \> > \> \> \> \> \> \> Hello everyone. > \> \> \> \> \> \> > \> \> \> \> \> \> My goal is to have a circle, drawn by screen('FillOval'), move according to a > \> \> \> predefined > \> \> \> \> \> \> random trajectory on the screen. Pretty simple eh? The problem is the circle > \> \> jitters / > \> \> \> \> \> vibrates > \> \> \> \> \> \> /jumps every now and then as it is moving on its trajectory. the funny thing > is > \> if I > \> \> \> use > \> \> \> \> \> the > \> \> \> \> \> \> same code but replace FillOval with FillRect, the motion is smooth! > \> \> \> \> \> \> > \> \> \> \> \> \> Any tips on why this is happening and how to prevent it ? > \> \> \> \> \> \> > \> \> \> \> \> \> > \> \> \> \> \> \> This simple code will be used as part of a more complex project involving > \> alpha > \> \> \> \> \> blending. > \> \> \> \> \> \> > \> \> \> \> \> \> Thanks in advance > \> \> \> \> \> \> > \> \> \> \> \> > \> \> \> \> > \> \> \> > \> \> > \>