VillageTronic waits for blanking?

1
dear matt

1. you've got a fast machine. congrats.

2. you should update to the current release, version 2.44
http://psychtoolbox.org/download.html

3. the villagetronic card's driver has the oddity that setclut doesn't
wait for blanking when you load one element, but does wait when you load
multiple elements. there's a specific test you can run to confirm this:
WaitBlankingTest

4. you can use a call to SCREEN SetClut, loading multiple elements, in
order to get the useful side effect of waiting for blanking.

i've thought that it would be nice for PsychToolbox to detect this
automatically and deal with it, and we will think about doing this, but
the top development priority right now is to get the Windows version
cleaned up. (many people, including me, bought villagetronic cards, but
those models are no longer made, and i have no experience with their new
generation.)

best

denis

>
>Denis and David,
>
>Hi. My name's Matt and I'm a student in Herwig Baier's lab at UCSF.
>I'm gearing up to start using the psychtoolbox for doing psychophysics
>on zebrafish. I just ran the screentest on my set-up and got what's
>written below. Any light you can shed on this will be most appreciated.
>
>
>
>Thanks,
>Matt Smear
>
>
>
>****** Matt Smear's Power Mac G4/500, Mac OS 9.0.2 ******
>G4, 500 MHz, memory bus 100 MHz, 78.572 Mflop/s
>PsychToolbox 2.43, 10 July 2000, MATLAB 5.2.1.1421
>QuickTime 4.1.2
>FileSharing is on.
>Screen saver present, now disabled.
>Using UpTime-based timing, which is very accurate and reliable.
>Has serial port arbitration.
>*** Screen 0
>*************************************************************
>ATY,Rage128Ps (.Display_Rage128 version 1.0b41) in slot SLOT-A
>8 bit dacs. 1024x768 55 Hz.
>NOTE: DipPriorityAfterSetClut is true. Works around Rage128 driver bug.
>NOTE: MinimumSetClutPriority is 2. Works around Rage128 driver bug.
>pixel size 8 16 32 bits
>pages 1 1 1
>CopyWindow (ie CopyBits) 89 91 91 MB/s
>GetClut equals SetClut? == == ==
>SetClutDriverWaitsForBlanking? Yes
>At priority 7 ... No blanking
>intrpts.
>SetClut (cscSetEntries) min dur. 0.0 0.0 0.0 frames
>At priority 1 ... Blanking
>interrupts.
>SetClut (cscSetEntries) min dur. 0.0 0.0 0.0 frames
>SetClut suppresses ints. for -0.1 0.1 0.0 frames
>*** Screen 1
>*************************************************************
>MaPi_850_750 (.Display_Video_VilTro_850_750 version 1.0.1) in slot
>SLOT-C
>8 bit dacs. 800x600 100 Hz.
>(50,60,64,67,70,72,73,74,75,76,78,85,90,100,160 Hz avail)
> WARNING from DriverTest:
> Inconsistent frame rate: SetClut 137 Hz vs interrupt 100 Hz.
> SetClut does wait (at priority 7), yet SetClutDriverWaitsForBlanking
>is 0.
> Perhaps SetClut waits when loading 256 entries, but not when loading
>only 1.
> driver =
> secs: 0.2000
> windowPtrOrScreenNumber: 1
> setClutDriverAwaitsBlanking: 1
> interruptOnBlanking: 0
> setClutFraction: 0.5852
> priority: 7
> waitError 0.6 ms,noWaitError 3.0 ms,deadError 3.3 ms,badError 7.9 ms
> SetClut 137.1 Hz (7 ms) vs interrupt 100.0 Hz (10 ms)
> pixelSize 8
> SetClutDriverWaitsForBlanking 0
> SetClut period 7 ms, or 8 ms with delay 4 ms
> InterruptToSetClutEndSecs 1 2 5 6 9 19 29 38 47 57 ms
> interruptOnBlanking 0
> Please send this report to denis & david:
> web mailto:psychtoolbox@...
>
> WARNING from DriverTest:
> Inconsistent frame rate: SetClut 134 Hz vs interrupt 100 Hz.
> SetClut does wait (at priority 0.5), yet SetClutDriverWaitsForBlanking
>is 0.
> Perhaps SetClut waits when loading 256 entries, but not when loading
>only 1.
> driver =
> secs: 0.2000
> windowPtrOrScreenNumber: 1
> setClutDriverAwaitsBlanking: 1
> interruptOnBlanking: 0
> setClutFraction: 0.5858
> priority: 0.5000
> waitError 0.8 ms,noWaitError 2.9 ms,deadError 3.5 ms,badError 8.2 ms
> SetClut 134.2 Hz (7 ms) vs interrupt 100.0 Hz (10 ms)
> pixelSize 8
> SetClutDriverWaitsForBlanking 0
> SetClut period 7 ms, or 8 ms with delay 4 ms
> InterruptToSetClutEndSecs 1 2 3 6 6 6 7 7 7 8 ms
> interruptOnBlanking 0
> Please send this report to denis & david:
> web mailto:psychtoolbox@...
>
> WARNING from DriverTest:
> Inconsistent frame rate: SetClut 135 Hz vs interrupt 100 Hz.
> SetClut does wait (at priority 7), yet SetClutDriverWaitsForBlanking
>is 0.
> Perhaps SetClut waits when loading 32 entries, but not when loading
>only 1.
> driver =
> secs: 0.2000
> windowPtrOrScreenNumber: 1
> setClutDriverAwaitsBlanking: 1
> interruptOnBlanking: 0
> setClutFraction: 0.5977
> priority: 7
> waitError 0.5 ms,noWaitError 3.2 ms,deadError 3.4 ms,badError 7.9 ms
> SetClut 134.8 Hz (7 ms) vs interrupt 100.2 Hz (10 ms)
> pixelSize 16
> SetClutDriverWaitsForBlanking 0
> SetClut period 7 ms, or 8 ms with delay 4 ms
> InterruptToSetClutEndSecs 1 2 3 4 4 4 4 5 7 7 ms
> interruptOnBlanking 0
> Please send this report to denis & david:
> web mailto:psychtoolbox@...
>
> WARNING from DriverTest:
> Inconsistent frame rate: SetClut 133 Hz vs interrupt 100 Hz.
> SetClut does wait (at priority 0.5), yet SetClutDriverWaitsForBlanking
>is 0.
> Perhaps SetClut waits when loading 32 entries, but not when loading
>only 1.
> driver =
> secs: 0.2000
> windowPtrOrScreenNumber: 1
> setClutDriverAwaitsBlanking: 1
> interruptOnBlanking: 0
> setClutFraction: 0.6284
> priority: 0.5000
> waitError 0.7 ms,noWaitError 3.0 ms,deadError 3.5 ms,badError 8.3 ms
> SetClut 133.2 Hz (8 ms) vs interrupt 100.1 Hz (10 ms)
> pixelSize 16
> SetClutDriverWaitsForBlanking 0
> SetClut period 8 ms, or 8 ms with delay 4 ms
> InterruptToSetClutEndSecs 3 3 4 4 4 4 4 7 8 9 ms
> interruptOnBlanking 0
> Please send this report to denis & david:
> web mailto:psychtoolbox@...
>
> WARNING from DriverTest:
> Inconsistent frame rate: SetClut 132 Hz vs interrupt 100 Hz.
> SetClut does wait (at priority 7), yet SetClutDriverWaitsForBlanking
>is 0.
> Perhaps SetClut waits when loading 256 entries, but not when loading
>only 1.
> driver =
> secs: 0.2000
> windowPtrOrScreenNumber: 1
> setClutDriverAwaitsBlanking: 1
> interruptOnBlanking: 1
> setClutFraction: 0.5647
> priority: 7
> waitError 0.7 ms,noWaitError 3.1 ms,deadError 3.6 ms,badError 8.3 ms
> SetClut 132.1 Hz (8 ms) vs interrupt 100.1 Hz (10 ms)
> pixelSize 32
> SetClutDriverWaitsForBlanking 0
> SetClut period 8 ms, or 8 ms with delay 4 ms
> InterruptToSetClutEndSecs 2 2 3 3 3 3 4 7 8 9 ms
> interruptOnBlanking 1
> Please send this report to denis & david:
> web mailto:psychtoolbox@...
>
> WARNING from DriverTest:
> Inconsistent frame rate: SetClut 134 Hz vs interrupt 100 Hz.
> SetClut does wait (at priority 0.5), yet SetClutDriverWaitsForBlanking
>is 0.
> Perhaps SetClut waits when loading 256 entries, but not when loading
>only 1.
> driver =
> secs: 0.2000
> windowPtrOrScreenNumber: 1
> setClutDriverAwaitsBlanking: 1
> interruptOnBlanking: 0
> setClutFraction: 0.5856
> priority: 0.5000
> waitError 0.9 ms,noWaitError 2.8 ms,deadError 3.4 ms,badError 8.4 ms
> SetClut 134.4 Hz (7 ms) vs interrupt 100.1 Hz (10 ms)
> pixelSize 32
> SetClutDriverWaitsForBlanking 0
> SetClut period 7 ms, or 8 ms with delay 4 ms
> InterruptToSetClutEndSecs 1 2 2 3 4 4 4 4 6 7 ms
> interruptOnBlanking 0
> Please send this report to denis & david:
> web mailto:psychtoolbox@...
>
>pixel size 8 16 32 bits
>pages 34 17 8
>CopyWindow (ie CopyBits) 61 62 62 MB/s
>GetClut equals SetClut? == == ==
>SetClutDriverWaitsForBlanking? No
>At priority 7 ... No blanking
>intrpts.
>SetClut (cscSetEntries) min dur. 0.6 0.6 0.6 frames
>*** SetClutDriverAwaitsBlanking? Yes
>At priority 0.5 ... Blanking
>interrupts.
>SetClut (cscSetEntries) min dur. 0.6 0.6 0.6 frames
>SetClut suppresses ints. for 0.0 0.0 0.0 frames
>*** SetClutDriverAwaitsBlanking? Yes
>**************************************************************************
>
>
2
>> at least for your VillageTronic, it appears that SetClut never
>> waits, no matter how much of the clut you load. i cannot think of any
>> work around for that. the psychtoolbox will still allow you to wait for
>> blanking by using of the blanking interrupt.
>
>Rats. Let me make sure I understand the implications of this -- does it mean
>that if I try to make a movie that involves animating the LUT I _can't_
>synchronize LUT changes with screen refreshes, but I _can_ sync frame
>advances
>with the refresh.
>
>Is my situation better or worse than this?


dear matt

it's better. SetClut is immediate. You can use WaitVBL to synch to
refresh. So a loop like the one below should work fine, loading a new
clut on each frame.

for i=1:100
screen(w,'waitblanking');
screen(w,'setclut',clut(i));
end

best

denis