# Radeon 9200 and Mac OS 9 ptb **URL:** \ **Category:** Uncategorized\ **Created:** [July 10, 2006, 7:23pm UTC](https://psychtoolbox.discourse.group/t/radeon-9200-and-mac-os-9-ptb/1662 "2006-07-10T19:23:46Z")\ **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:** [July 10, 2006, 7:23pm UTC](https://psychtoolbox.discourse.group/t/radeon-9200-and-mac-os-9-ptb/1662/1 "2006-07-10T19:23:46Z")
Hi, Does anyone have experience using the Radeon 9200 in OS 9? I'm trying to get it to work on an old mac using the lastest os 9 version of psych. toolbox, and I cant get the supposed 10- bit DACs to function-- TwoBitFlickerTest fails, and when I measure successive gray levels (same nunber, between 0-1023, sent to all 3 DACs) with a good light meter I only see changes in luminance every 4 levels, as if its ignoring the last two bits of the values I give it. Anyone work through this before? Thanks, Steve ---
**Author:** ![denispelli](https://avatars.discourse-cdn.com/v4/letter/d/cdc98d/32.png) [@denispelli](https://psychtoolbox.discourse.group/u/denispelli)\ **Post date:** [July 11, 2006, 4:51pm UTC](https://psychtoolbox.discourse.group/t/radeon-9200-and-mac-os-9-ptb/1662/2 "2006-07-11T16:51:10Z")
dear steve yes, i worked on that. All the more modern Radeon cards have 10 bit DACs, but not all the OS 9 drivers actually pass 10-bits to the lookup table. (They're made by ATI.) At one point I had a long talk with one of the head engineers and it's my impression that all the drivers issued since then do have 10-bit support. One important difference turned out to be between drivers issued by ATI (retail cards) or by Apple (built-in). Their contract with apple specified that upgrades to the drivers for built-ins would ONLY be changed in response to requests (and presumably payment) from apple. ATI quickly added 10 bit support for all the retail cards, but i don't think that this happened for all the drivers for the built-in cards. thus, i suggest you make sure you've got the latest driver, and then, if your test indicates absence of 10-bit support, believe the result and move on. all my notes about this are here: [http://psychtoolbox.org/mac.html#drivers](http://psychtoolbox.org/mac.html#drivers) i'd welcome any new information you think we should add to that page. i strongly suggest running ScreenTest for a thorough characterization of the card. best denis On Jul 10, 2006, at 3:23 PM, sengelweb wrote: > > Hi, > > > > Does anyone have experience using the Radeon 9200 in OS 9? I'm trying to get it to work on > > an old mac using the lastest os 9 version of psych. toolbox, and I cant get the supposed 10- > > bit DACs to function-- TwoBitFlickerTest fails, and when I measure successive gray levels > > (same nunber, between 0-1023, sent to all 3 DACs) with a good light meter I only see > > changes in luminance every 4 levels, as if its ignoring the last two bits of the values I give it. > > > > Anyone work through this before? > > > > Thanks, > > > > Steve > > > > . > > > > ![](https://global.discourse-cdn.com/free1/uploads/psychtoolbox/original/1X/2daeaa8b5f19f0bc209d976c02bd6acb51b00b0a.gif) ---
**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:** [July 12, 2006, 1:53pm UTC](https://psychtoolbox.discourse.group/t/radeon-9200-and-mac-os-9-ptb/1662/3 "2006-07-12T13:53:04Z")
Hi Denis, Thanks for the tips. So, here is some information if you want to add it to the table on the web page: We have the ATI Radeon 9200 (PCI) in a G4 running OS 9 with the latest driver downloaded from ATI's support page. The card has functional 10 bit DACs according to ScreenTest, ClutTest and LoadClutTest, but TwoBitFlickerTest and our own photometric measurements indicate that the last 2 bits are not actually affecting the displayed color. Below is the output of some of the tests. One thing I couldnt tell from the web page: Are the OS X drivers generally better at supporting 10 bits? Best, Steve ----- ScreenTest \*\*\* 's Power Mac G4/400, Mac OS 9.2.2 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* G4, 400 MHz, memory bus 100 MHz, 63.060 Mflop/s Psychtoolbox 2.55, 12 March 2005, Matlab 5.2.0.3084 FileSharing is on. Screen saver present, now disabled. \*\*\* Screen 0 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* "ATY,Bugsy\_A" (.Display\_DualHead version 1.0.1f58) in slot SLOT-C 10 bit dacs. 1152x870 73 Hz. (56,60,65,67,70,72,73,75,76,85,90,100,120,150,160,200 Hz avail) Prefs: cscGetClutBehavior, cscGetNextResolution, DipPriorityAfterSetClut, MinimumSetClutPriority 2. - - - - - - - - - - - - - - - - - - - - - - - - - pixel size 8 16 32 bits pages 1 1 1 CopyWindow (ie CopyBits) 86 88 88 MB/s CopyWindow (ie CopyBits) VRAM 1539 1520 1524 MB/s SetClut suppresses ints. for 0.0 0.0 0.0 frames LoadClut vs. GetClut? ( 8 bits) == == == LoadClut vs. GetClut? (10 bits) == == == - - - - - - - - - - - - - - - - - - - - - - - - - %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% LoadClutTest \*\*\* 's Power Mac G4/400, Mac OS 9.2.2 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* G4, 400 MHz, memory bus 100 MHz, 63.542 Mflop/s Psychtoolbox 2.55, 12 March 2005, Matlab 5.2.0.3084 FileSharing is on. Screen saver present, now disabled. \*\*\* Screen 0 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* "ATY,Bugsy\_A" (.Display\_DualHead version 1.0.1f58) in slot SLOT-C 10 bit dacs. 1152x870 75 Hz. (56,60,65,67,70,72,73,75,76,85,90,100,120,150,160,200 Hz avail) Prefs: cscGetClutBehavior, cscGetNextResolution, DipPriorityAfterSetClut, MinimumSetClutPriority 2. - - - - - - - - - - - - - - - - - - - - - - - - - 8-bit gamma table. pixelSize 8. Success! GetClut==LoadClut at 8 bits. 8-bit gamma table. pixelSize 16. Success! GetClut==LoadClut at 8 bits. 8-bit gamma table. pixelSize 32. Success! GetClut==LoadClut at 8 bits. 10-bit gamma table. pixelSize 8. Success! GetClut==LoadClut at 10 bits. 10-bit gamma table. pixelSize 16. Success! GetClut==LoadClut at 10 bits. 10-bit gamma table. pixelSize 32. Success! GetClut==LoadClut at 10 bits. \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% ClutTest \*\*\* 's Power Mac G4/400, Mac OS 9.2.2 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* G4, 400 MHz, memory bus 100 MHz, 63.203 Mflop/s Psychtoolbox 2.55, 12 March 2005, Matlab 5.2.0.3084 FileSharing is on. Screen saver present, now disabled. \*\*\* Screen 0 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* "ATY,Bugsy\_A" (.Display\_DualHead version 1.0.1f58) in slot SLOT-C 10 bit dacs. 1152x870 75 Hz. (56,60,65,67,70,72,73,75,76,85,90,100,120,150,160,200 Hz avail) Prefs: cscGetClutBehavior, cscGetNextResolution, DipPriorityAfterSetClut, MinimumSetClutPriority 2. WARNING: You're not running the latest Matlab, which is version 5.2.1. Thus we can't use Gestalt to determine whether we're running Mac OS X Classic. The graphics drivers under Classic require different settings. Please upgrade Matlab to 5.2.1 or upgrade the Psychtoolbox to 2.53, when it's released. - - - - - - - - - - - - - - - - - - - - - - - - - pixelSize 8. randomize 0. SetGamma 8 bits. GetGamma 8 bits. GetClut == predicted. pixelSize 8. randomize 1. SetGamma 8 bits. GetGamma 8 bits. GetClut == predicted. pixelSize 16. randomize 0. SetGamma 8 bits. GetGamma 8 bits. GetClut == predicted. pixelSize 16. randomize 1. SetGamma 8 bits. GetGamma 8 bits. GetClut == predicted. pixelSize 32. randomize 0. SetGamma 8 bits. GetGamma 8 bits. GetClut == predicted. pixelSize 32. randomize 1. SetGamma 8 bits. GetGamma 8 bits. GetClut == predicted. pixelSize 8. randomize 0. SetGamma 10 bits. GetGamma 10 bits. GetClut == predicted. pixelSize 8. randomize 1. SetGamma 10 bits. GetGamma 10 bits. GetClut == predicted. pixelSize 16. randomize 0. SetGamma 10 bits. GetGamma 10 bits. GetClut == predicted. pixelSize 16. randomize 1. SetGamma 10 bits. GetGamma 10 bits. GetClut == predicted. pixelSize 32. randomize 0. SetGamma 10 bits. GetGamma 10 bits. GetClut == predicted. pixelSize 32. randomize 1. SetGamma 10 bits. GetGamma 10 bits. GetClut == predicted. Screen driver supports gamma tables with precisions: 8 10 bits. \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* > --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), Denis Pelli \ wrote: > \> > \> dear steve > \> > \> yes, i worked on that. All the more modern Radeon cards have 10 bit > \> DACs, but not all the OS 9 drivers actually pass 10-bits to the > \> lookup table. (They're made by ATI.) At one point I had a long talk > \> with one of the head engineers and it's my impression that all the > \> drivers issued since then do have 10-bit support. One important > \> difference turned out to be between drivers issued by ATI (retail > \> cards) or by Apple (built-in). Their contract with apple specified > \> that upgrades to the drivers for built-ins would ONLY be changed in > \> response to requests (and presumably payment) from apple. ATI quickly > \> added 10 bit support for all the retail cards, but i don't think that > \> this happened for all the drivers for the built-in cards. > \> > \> thus, i suggest you make sure you've got the latest driver, and then, > \> if your test indicates absence of 10-bit support, believe the result > \> and move on. > \> > \> > \> all my notes about this are here: > \> [http://psychtoolbox.org/mac.html#drivers](http://psychtoolbox.org/mac.html#drivers) > \> > \> i'd welcome any new information you think we should add to that page. > \> i strongly suggest running ScreenTest for a thorough characterization > \> of the card. > \> > \> best > \> > \> denis > \> > \> > \> > \> On Jul 10, 2006, at 3:23 PM, sengelweb wrote: > \> > \> \> > \> \> Hi, > \> \> > \> \> Does anyone have experience using the Radeon 9200 in OS 9? I'm > \> \> trying to get it to work on > \> \> an old mac using the lastest os 9 version of psych. toolbox, and I > \> \> cant get the supposed 10- > \> \> bit DACs to function-- TwoBitFlickerTest fails, and when I measure > \> \> successive gray levels > \> \> (same nunber, between 0-1023, sent to all 3 DACs) with a good light > \> \> meter I only see > \> \> changes in luminance every 4 levels, as if its ignoring the last > \> \> two bits of the values I give it. > \> \> > \> \> Anyone work through this before? > \> \> > \> \> Thanks, > \> \> > \> \> Steve > \> \> > \> \> . > \> \> > \> \> > \> ---
**Author:** ![mpqrowe](https://avatars.discourse-cdn.com/v4/letter/m/df788c/32.png) [@mpqrowe](https://psychtoolbox.discourse.group/u/mpqrowe)\ **Post date:** [July 17, 2006, 8:09pm UTC](https://psychtoolbox.discourse.group/t/radeon-9200-and-mac-os-9-ptb/1662/4 "2006-07-17T20:09:50Z")
Hi, I also have a Radeon 9200 Mac Edition running in a Mac (G3) under OS 9. Until last week I thought I was getting 10-bits per channel, but when I tested it, I discovered that my monitor also was only bumping up once every four changes in level (as reported by Steve: [http://groups.yahoo.com/group/psychtoolbox/message/4718)](http://groups.yahoo.com/group/psychtoolbox/message/4718)). However... I just discovered that that's only true (for me anyways) when the monitor is connected to the DVI port. I have other calibration and timing issues, but I've just verified ten bit resolution if I use the VGA connector. If you haven't already tried that, you might want to. I'm running PsychToolbox 2.55 and Matlab 5.2.1.1421. -- Mickey P. Rowe ([mrowe@...](mailto:mrowe@...)) ---
**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:** [July 18, 2006, 9:11am UTC](https://psychtoolbox.discourse.group/t/radeon-9200-and-mac-os-9-ptb/1662/5 "2006-07-18T09:11:14Z")
Hi Mickey The DVI specification only natively supports 8bit resolution per colour channel. If you have a graphics card with more than 8bit DACs, you have to use the analogue output to get the extra bits (i.e. use the built-in HD15 output, or a DVI-to-HD15 adapter). If you are connecting a standard LCD monitor to the graphics card (using the DVI or the analogue HD15 output), you may only be able to measure 6bit or 8bit resolution out of the display. Steve --- In [psychtoolbox@yahoogroups.com](mailto:psychtoolbox@yahoogroups.com), "Mickey Rowe;893-2446" \ wrote: > \> > \> > \> Hi, > \> > \> I also have a Radeon 9200 Mac Edition running in a Mac (G3) under OS > \> 9. Until last week I thought I was getting 10-bits per channel, but > \> when I tested it, I discovered that my monitor also was only bumping > \> up once every four changes in level (as reported by Steve: > \> [http://groups.yahoo.com/group/psychtoolbox/message/4718)](http://groups.yahoo.com/group/psychtoolbox/message/4718)). However... > \> I just discovered that that's only true (for me anyways) when the > \> monitor is connected to the DVI port. I have other calibration and > \> timing issues, but I've just verified ten bit resolution if I use the > \> VGA connector. > \> > \> If you haven't already tried that, you might want to. I'm running > \> PsychToolbox 2.55 and Matlab 5.2.1.1421. > \> > \> -- > \> Mickey P. Rowe (mrowe@...) > \>