VBL endline beamposition queries and audio timestamping issues 2

Psychtoolbox Version: 3.0.22
MATLAB Version: 2025b
OS Name: Microsoft Windows 11 Enterprise
Version: 10.0.22631
Processor: Intel(R) Core(™) Ultra 7 165H, 3800 Mhz, 16 Core(s), 22 Logical Processor(s)
MATLAB: 9.13.0.2698988 R2022b Update 10
Speakers: 2-SoundWire Speakers
Speaker driver version: 10.0.22621.1

Hi there,

I’m reposting now that I have the latest Psychtoolbox version.

I’m trying to run the Stop Signal Task on MATLAB with new laptops and I’m getting these errors:

PTB-INFO: This is Psychtoolbox-3 for Microsoft Windows, under Matlab 64-Bit Intel (Version 3.0.22 - Build date: May 24 2025).

PTB-INFO: OS support status: Windows 11 (Version 11.0) is not supported.

PTB-INFO: For information about paid support and other commercial services, please type ‘PsychPaidSupportAndServices’.

PTB-INFO: Most parts of the Psychtoolbox distribution are licensed to you under terms of the MIT license, with some

PTB-INFO: restrictions. See file ‘License.txt’ in the Psychtoolbox root folder for the exact licensing conditions.

PTB-INFO: Psychtoolbox and its prebuilt mex files are distributed in the hope that they will be useful, but WITHOUT

PTB-INFO: ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

PTB-INFO: OpenGL-Renderer is NVIDIA Corporation :: NVIDIA RTX 1000 Ada Generation Laptop GPU/PCIe/SSE2 :: 4.6.0 NVIDIA 581.95

PTB-INFO: Screen 0 : Window 10 : VBL startline = 1200 : VBL Endline = -1

PTB-INFO: Beamposition queries unsupported or defective on this system. Using basic timestamping as fallback.

PTB-INFO: Timestamps returned by Screen(‘Flip’) will be therefore less robust and accurate.

PTB-INFO: Measured monitor refresh interval from VBLsync = 16.660302 ms [60.022922 Hz]. (50 valid samples taken, stddev=0.094264 ms.)

PTB-INFO: Reported monitor refresh interval from operating system = 16.666667 ms [60.000000 Hz].

PTB-INFO: ==============================================================================================================================

PTB-INFO: WINDOWS DWM DESKTOP COMPOSITOR IS ACTIVE. On this Windows-10 or later system, Psychtoolbox can no longer reliably detect if

PTB-INFO: this will cause trouble for timing and integrity of visual stimuli or not. You might be just fine, or you could be in trouble.

PTB-INFO: Use external measurement equipment and independent procedures to verify reliability of timing if you care about proper timing.

PTB-INFO: ==============================================================================================================================

PTB-INFO: Using PortAudio V19.7.0-devel, revision unknown

PTB-INFO: New audio device 2 with handle 0 opened as PortAudio stream:

PTB-INFO: For 1 channels Playback: Audio subsystem is MME, Audio device name is Microsoft Sound Mapper - Output

PTB-INFO: Real samplerate 22254.000000 Hz. Input latency 0.000000 msecs, Output latency 100.656062 msecs.

PTB-WARNING:PsychPortAudio(‘Stop’): Audio device with handle 0 had broken audio timestamping - and therefore timing - during this run. Don’t trust the timing!

PTB-WARNING:PsychPortAudio(‘Stop’): Audio device with handle 0 had broken audio timestamping - and therefore timing - during this run. Don’t trust the timing!

PTB-WARNING:PsychPortAudio(‘Stop’): Audio device with handle 0 had broken audio timestamping - and therefore timing - during this run. Don’t trust the timing!

PTB-WARNING:PsychPortAudio(‘Stop’): Audio device with handle 0 had broken audio timestamping - and therefore timing - during this run. Don’t trust the timing!

PTB-WARNING:PsychPortAudio(‘Stop’): Audio device with handle 0 had broken audio timestamping - and therefore timing - during this run. Don’t trust the timing!

PTB-WARNING:PsychPortAudio(‘Stop’): Audio device with handle 0 had broken audio timestamping - and therefore timing - during this run. Don’t trust the timing!

INFO: PTB’s Screen(‘Flip’, 10) command seems to have missed the requested stimulus presentation deadline

INFO: a total of 18 times out of a total of 519 flips during this session.

INFO: This number is fairly accurate (and indicative of real timing problems in your own code or your system)

INFO: if you provided requested stimulus onset times with the ‘when’ argument of Screen(‘Flip’, window [, when]);

INFO: If you called Screen(‘Flip’, window); without the ‘when’ argument, this count is more of a ‘‘mild’’ indicator

INFO: of timing behaviour than a hard reliable measurement. Large numbers may indicate problems and should at least

INFO: deserve your closer attention. Cfe. ‘help SyncTrouble’, the FAQ section at www.psychtoolbox.org and the

INFO: examples in the PDF presentation in PsychDocumentation/Psychtoolbox3-Slides.pdf for more info and timing tips.

Any help would be greatly appreciated.

Thanks,
Bree

Comment from a user on my last post:
Hm. Seems like there are 2 problems, graphics and audio. Regarding the graphics problem, is this the latest NVIDIA driver available for your system? Because, very recently, I had vsync problems with a desktop system, where Windows had auto-installed an outdated version 576.88, which was no good. My problem was fixed once I updated to 591.44 (and shortly after to 591.74).
My answer: I have 581.95. I can now see online there’s a 582.16 version. I’ll try that.

Ok, now that you are on a supported version of PTB.

Sound:

The sound issue suggests something wrong with the PsychPortAudio('Open', ...) call in your script. By default, PsychPortAudio would choose a sound device backend that provides relatively low latency and high timing precision, unless you specified the optional 3rd parameter reqlatencyclass as 0, which requests high latency + no timing precision mode. The parameter should be left out, or at default value [] or a value of 1 or greater to select high precision. The other way you could get this is if you specified the optional 1st parameter deviceid to force selection of a specific combination of sound device and sound backend, enforcing choice of the MME backend, which is always high latency and no timing precision.

E.g., BasicSoundOutputDemo.m uses this MME choice via reqlatencyclass=0, but all other sound demos use the default [] (== 1) or higher for better timing precision and lower latency. Normally this will use the Windows WASAPI backend.

Visual VBL endline stuff and broken / disabled beamposition timestamping

That is unusual that those don’t work on a regular NVidia single-gpu machine. Do a clear all then run our standard timing test PerceptualVBLSyncTest or VBLSyncTest to see if those work. If so, then it is something in your script forcing proper visual timing off.

Otherwise, I have the suspicion that your machine is a dual-graphics NVidia Optimus Laptop with Intel iGPU + NVidia dGPU? A Google search of “NVIDIA RTX 1000 Ada Generation Laptop” suggests this. What specific brand and machine model is this? Provide a link to a product description if possible.

In that case you might have a problem, as most dual-graphics laptops under MS-Windows have often unfixable broken visual timing. See help HybridGraphics, and our Hardware requirements on our webpage, which strongly advise against such laptops if one wants to use them with MS-Windows.

Hi Mario,

Thanks for your help with this.

I have attached the order details for this laptop, hopefully that has useful information.

We had these laptops custom built to suit Psychtoolbox, but perhaps something was missed. It looks like there’s only one graphics card on the order summary form, but Task Manager > Performance is showing two: GPU 0 = Intel(R) Arc ™ Pro Graphics, GPU 1 = NVIDIA RTX 1000 Ada Generation Laptop GPU

That’s funny, because this laptop seems to be custom built for the exact opposite.

Btw. you haven’t run our test scripts yet, as I asked you earlier, to rule out problems with your script.

Yes, it is a dual-gpu “Optimus” hybrid-graphics laptop, the kind I strongly advise against as a troublemaker, especially on MS-Windows, in the “System Requirements” section of our website. Also, NVidia gpu’s are somewhat advised against on our website, unless you never intend to run the machine under Linux (where AMD and Intel generally work excellent, whereas NVidia is only ok’ish/manageable with known restrictions and more setup hassle).

Anyway, maybe you get lucky and this is salvageable. According to a lot of googling and some AI searches, some Dell Precision workstations with Intel iGPU’s + NVidia dGPU’s do have something called a “Direct Graphics Controller Direct Output Mode”, which can be enabled in the Laptops BIOS/UEFI firmware setup. It allows to completely disable and bypass the Intel iGPU and effectively turn the laptop into a single NVidia dGPU machine, using suitable graphics multiplexers (see “help HybridGraphics” in Psychtoolbox for explanation). Higher power consumption, but also higher performance and hopefully trouble-free presentation timing.

Prompting Google search’s AI if your “Dell mobile precision workstation 5490” supports this, gives a confident “yes”, but references the Dell manual for a different Dell Precision 5750 laptop model, so this could be a case of the AI being confidently wrong, just hallucinating the answer it thinks I want to read. So lets see. The following instructions for a Dell 5750 may work for your laptop:

Adding to that, this links gives the response of the AI to one of my. prompts. In case Direct Output Mode is not possible on your machine, there is the Optimus configuration shown in that conversation as section “Windows level settings”. Not likely though that it will fix your problem, as that is just NVidia standard Optimus, and what MS-Windows already selected in auto-mode by itself - or Psychtoolbox would not have reported the NVidia gpu being used as OpenGL renderer.

Another good way solving it would be “Advanced Optimus” (NVIDIA Advanced Optimus Overview | NVIDIA), involving automated runtime mux switching, but AI prompting suggests that that is not available on your machine, only basic Optimus.