Odd flip timestamps

andreaswidmann · January 4, 2017, 2:34pm

Dear Mario and list,

while programming a new experiment where we want to precisely sync auditory and visual stimuli I stumbled across some confusing oddities in the flip timestamping in our lab (I hope this is not a very stupid question):

From the diagnostic PTB output (see below) I assumed that our system (Ubuntu 16.04 low-latency kernel, older nVidia Quadro NVS 290 card, nouveau driver, newest PTB) supports beamposition queries. However, the VBLTimeStamp returned by Screen('Flip‘, …) always equals the StimulusOnsetTime. On the other hand, the corresponding beampos is not -1 as I would have expected in case of missing support for beamposition queries from your explanation in the FAQ (https://github.com/Psychtoolbox-3/Psychtoolbox-3/wiki/FAQ:-Explanation-of-Flip-Timestamps) but consistently between 774 and 776.

More importantly, the returned VBLTimeStamp is consistently *later* than the corresponding FlipTimestamp (by ~400 to 450 µS). The FlipTimestamp itself has only very small deviations from a GetSecs timestamp taken after Flip (< 50 µS).

* Is this expected behavior and my mental concept is flawed (i.e., doesn’t the VBLTimeStamp reflect VBL onset and should thus be earlier than FlipTimestamp)?
* Or did I misconfigure something?
* Or is this a limitation of our admittedly somewhat older graphics card (working flawlessly otherwise)?
* And finally, to compute true StimulusOnsetTime on this system would it be safe to apply the equation to compute VBL duration used in PredictVisualOnsetForTime?

Thanks a lot! Best,
Andreas

RandR: 1024x768_100.00 (0xa9) 112.2MHz
h: width 1024 start 1096 end 1200 total 1376 skew 0
v: height 768 start 771 end 775 total 816

PTB-INFO: This is Psychtoolbox-3 for GNU/Linux X11, under GNU/Octave 64-Bit (Version 3.0.14 - Build date: Dec 22 2016).
PTB-INFO: Support status on this operating system release: Linux 4.4.0-57-lowlatency Supported.
PTB-INFO: Type 'PsychtoolboxVersion' for more detailed version information.
PTB-INFO: Most parts of the Psychtoolbox distribution are licensed to you under terms of the MIT License, with
PTB-INFO: some restrictions. See file 'License.txt' in the Psychtoolbox root folder for the exact licensing conditions.

PTB-DEBUG: Checking PCI device [Intel Corporation 4 Series Chipset DRAM Controller] with class x00060000 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 4 Series Chipset PCI Express Root Port] with class x00060400 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 4 Series Chipset HECI Controller] with class x00078000 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 4 Series Chipset Serial KT Controller] with class x00070002 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82567LM-3 Gigabit Network Connection] with class x00020000 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #4] with class x000c0300 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #5] with class x000c0300 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #6] with class x000c0300 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB2 EHCI Controller #2] with class x000c0320 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) HD Audio Controller] with class x00040300 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) PCI Express Port 1] with class x00060400 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) PCI Express Port 2] with class x00060400 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #1] with class x000c0300 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #2] with class x000c0300 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #3] with class x000c0300 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB2 EHCI Controller #1] with class x000c0320 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801 PCI Bridge] with class x00060401 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD (ICH10D) LPC Interface Controller] with classx00060100 ...
PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) SATA AHCI Controller] with class x00010601 ...
PTB-DEBUG: Checking PCI device [NVIDIA Corporation G86 [Quadro NVS 290]] with class x00030000 ...
PTB-INFO: NVIDIA Corporation - G86 [Quadro NVS 290] GPU found. Trying to establish low-level access...
PTB-DEBUG: Mapping GPU BAR address 0xf2000000 ...
PTB-DEBUG: Mapping 0x1000000 bytes...
PTB-INFO: Connected to NVidia G86 [Quadro NVS 290] GPU of NV-050 family with 2 display heads. Beamposition timestamping enabled.
PTB-INFO: Using old-style override-redirect setup path for onscreen window creation.
PTB-INFO: GLX Visual info depths is 24 bits
PTB-INFO: Running on Mesa version 11.2.0
PTB-INFO: This combo of X-Server and Mesa is considered safe for use under DRI3/Present.
PTB-INFO: Using GLEW version 2.0.0 for automatic detection of OpenGL extensions...
PTB-INFO: Using GLX_MESA_swap_control extension for control of vsync.
PTB-INFO: INTEL_swap_event support for additional swap completion correctness checks enabled.
PTB-DEBUG: PPM file magic is P6
-> Ok
# CREATOR: GIMP PNM Filter Version 1.1
PTB-DEBUG: Recognized splash image of 656 x 304 pixels, maxlevel 255. Loading...

OpenGL-Vendor / renderer / version are: nouveau - Gallium 0.4 on NV86 - 3.0 Mesa 11.2.0

OpenGL-Extensions are: GL_ARB_multisample GL_EXT_abgr GL_EXT_bgra GL_EXT_blend_color GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_copy_texture GL_EXT_polygon_offset GL_EXT_subtexture GL_EXT_texture_object GL_EXT_vertex_array GL_EXT_compiled_vertex_array GL_EXT_texture GL_EXT_texture3D GL_IBM_rasterpos_clipGL_ARB_point_parameters GL_EXT_draw_range_elements GL_EXT_packed_pixels GL_EXT_point_parameters GL_EXT_rescale_normal GL_EXT_separate_specular_color GL_EXT_texture_edge_clamp GL_SGIS_generate_mipmap GL_SGIS_texture_border_clamp GL_SGIS_texture_edge_clamp GL_SGIS_texture_lod GL_ARB_framebuffer_sRGB GL_ARB_multitexture GL_EXT_framebuffer_sRGB GL_IBM_multimode_draw_arrays GL_IBM_texture_mirrored_repeat GL_ARB_texture_cube_map GL_ARB_texture_env_add GL_ARB_transpose_matrix GL_EXT_blend_func_separate GL_EXT_fog_coord GL_EXT_multi_draw_arrays GL_EXT_secondary_color GL_EXT_texture_env_add GL_EXT_texture_filter_anisotropic GL_EXT_texture_lod_bias GL_INGR_blend_func_separate GL_NV_blend_square GL_NV_light_max_exponent GL_NV_texgen_reflection GL_NV_texture_env_combine4 GL_S3_s3tc GL_SUN_multi_draw_arrays GL_ARB_texture_border_clamp GL_ARB_texture_compression GL_EXT_framebuffer_object GL_EXT_texture_compression_s3tc GL_EXT_texture_env_combine GL_EXT_texture_env_dot3 GL_MESA_window_pos GL_NV_packed_depth_stencil GL_NV_texture_rectangleGL_ARB_depth_texture GL_ARB_occlusion_query GL_ARB_shadow GL_ARB_texture_env_combine GL_ARB_texture_env_crossbar GL_ARB_texture_env_dot3 GL_ARB_texture_mirrored_repeat GL_ARB_window_pos GL_EXT_stencil_two_side GL_EXT_texture_cube_map GL_NV_depth_clamp GL_NV_fog_distance GL_APPLE_packed_pixels GL_APPLE_vertex_array_object GL_ARB_draw_buffers GL_ARB_fragment_program GL_ARB_fragment_shader GL_ARB_shader_objects GL_ARB_vertex_program GL_ARB_vertex_shader GL_ATI_draw_buffers GL_ATI_texture_env_combine3 GL_ATI_texture_float GL_EXT_depth_bounds_test GL_EXT_shadow_funcs GL_EXT_stencil_wrap GL_MESA_pack_invert GL_NV_primitive_restart GL_ARB_depth_clamp GL_ARB_fragment_program_shadow GL_ARB_half_float_pixel GL_ARB_occlusion_query2 GL_ARB_point_sprite GL_ARB_shading_language_100 GL_ARB_sync GL_ARB_texture_non_power_of_two GL_ARB_vertex_buffer_object GL_ATI_blend_equation_separate GL_EXT_blend_equation_separate GL_OES_read_format GL_ARB_color_buffer_float GL_ARB_pixel_buffer_object GL_ARB_texture_compression_rgtc GL_ARB_texture_float GL_ARB_texture_rectangle GL_ATI_texture_compression_3dc GL_EXT_packed_float GL_EXT_pixel_buffer_object GL_EXT_texture_compression_dxt1 GL_EXT_texture_compression_rgtc GL_EXT_texture_mirror_clamp GL_EXT_texture_rectangle GL_EXT_texture_sRGB GL_EXT_texture_shared_exponent GL_ARB_framebuffer_object GL_EXT_framebuffer_blit GL_EXT_framebuffer_multisample GL_EXT_packed_depth_stencil GL_ARB_vertex_array_object GL_ATI_separate_stencil GL_ATI_texture_mirror_once GL_EXT_draw_buffers2 GL_EXT_draw_instanced GL_EXT_gpu_program_parameters GL_EXT_texture_array GL_EXT_texture_compression_latc GL_EXT_texture_integer GL_EXT_texture_sRGB_decode GL_EXT_timer_query GL_OES_EGL_image GL_AMD_performance_monitor GL_ARB_copy_buffer GL_ARB_depth_buffer_float GL_ARB_draw_instanced GL_ARB_half_float_vertex GL_ARB_instanced_arrays GL_ARB_map_buffer_range GL_ARB_texture_rg GL_ARB_texture_swizzle GL_ARB_vertex_array_bgra GL_EXT_texture_swizzle GL_EXT_vertex_array_bgra GL_NV_conditional_render GL_AMD_conservative_depth GL_ARB_ES2_compatibility GL_ARB_blend_func_extended GL_ARB_debug_output GL_ARB_draw_elements_base_vertex GL_ARB_explicit_attrib_location GL_ARB_fragment_coord_conventions GL_ARB_provoking_vertex GL_ARB_sampler_objects GL_ARB_seamless_cube_map GL_ARB_shader_texture_lod GL_ARB_texture_multisample GL_ARB_texture_rgb10_a2ui GL_ARB_uniform_buffer_object GL_ARB_vertex_type_2_10_10_10_rev GL_EXT_provoking_vertex GL_EXT_texture_snorm GL_MESA_texture_signed_rgba GL_NV_texture_barrier GL_ARB_get_program_binary GL_ARB_robustness GL_ARB_separate_shader_objects GL_ARB_shader_bit_encoding GL_ARB_timer_query GL_NV_vdpau_interop GL_ANGLE_texture_compression_dxt3 GL_ANGLE_texture_compression_dxt5 GL_ARB_base_instance GL_ARB_compressed_texture_pixel_storage GL_ARB_conservative_depth GL_ARB_internalformat_query GL_ARB_map_buffer_alignment GL_ARB_shading_language_420pack GL_ARB_shading_language_packing GL_ARB_texture_storage GL_EXT_framebuffer_multisample_blit_scaled GL_EXT_transform_feedback GL_AMD_shader_trinary_minmax GL_ARB_ES3_compatibility GL_ARB_arrays_of_arrays GL_ARB_clear_buffer_object GL_ARB_copy_image GL_ARB_explicit_uniform_location GL_ARB_invalidate_subdata GL_ARB_program_interface_query GL_ARB_stencil_texturing GL_ARB_texture_query_levels GL_ARB_texture_storage_multisample GL_ARB_texture_view GL_ARB_vertex_attrib_binding GL_KHR_debug GL_ARB_buffer_storage GL_ARB_clear_texture GL_ARB_multi_bind GL_ARB_texture_mirror_clamp_to_edge GL_ARB_texture_stencil8 GL_ARB_vertex_type_10f_11f_11f_rev GL_EXT_shader_integer_mix GL_ARB_clip_control GL_ARB_conditional_render_inverted GL_ARB_derivative_control GL_ARB_get_texture_sub_image GL_ARB_pipeline_statistics_query GL_ARB_shader_texture_image_samples GL_ARB_texture_barrier GL_EXT_polygon_offset_clamp GL_KHR_context_flush_control

PTB-DEBUG: Linux X11 backend with FOSS drivers - Enabling locked pixeltoken-write + flush workaround for XLib thread-safety.
PTB-DEBUG: Interrogating Low-level renderer capabilities for onscreen window with handle 10:
Indicator variables: FBO's 1, ATI_texture_float 1, ARB_texture_float 1, Vendor nouveau, Renderer Gallium 0.4 on NV86.
Indicator variables: maxcolorattachments = 8, maxrectangletexturesize = 8192, maxnativealuinstructions = 16384.
GPU supports non-power-of-two textures.
Hardware supports signed normalized textures of 16 bpc integer format.
Basic framebuffer objects with rectangle texture rendertargets supported --> RGBA8 rendertargets with blending.
Framebuffer objects support fast blitting between each other.
Framebuffer objects support anti-aliasing via multisampling.
Framebuffer objects support single-pass multisample resolve blits and image rescaling.
Hardware supports floating point textures of 16bpc and 32bpc float format.
Assuming NV30 core or later...
Assuming NV40 core or later (maxcolattachments=8): Hardware supports floating point blending and filtering on 16bpc float format.
Hardware also supports floating point framebuffers of 16bpc and 32bpc float format.
Hardware supports full 32 bit floating point precision shading.
Assuming G80 core or later (maxtexsize=8192): Hardware supports full floating point blending and filtering on 16bpc and 32bpc float format.
Assuming hardware supports native OpenGL primitive smoothing (points, lines).
PTB-Info: Running on 'The X.Org Foundation' XServer, Vendor release 11804000.
OML_sync_control indicators: glXGetSyncValuesOML=0x7f6c1bcd4c40 , glXWaitForMscOML=0x7f6c1bcd4d80, glXWaitForSbcOML=0x7f6c1bcd4e50, glXSwapBuffersMscOML=0x7f6c1bcd4cb0
OML_sync_control indicators: glxewIsSupported() says 1.
System supports OpenML OML_sync_control extension for high-precision scheduled swaps and timestamping.
OpenML OML_sync_control extension enabled for all scheduled swaps.
PTB-DEBUG: Interrogation done.

RandR: 1024x768_100.00 (0xa9) 112.2MHz
h: width 1024 start 1096 end 1200 total 1376 skew 0
v: height 768 start 771 end 775 total 816
RandR: 1024x768_100.00 (0xa9) 112.2MHz
h: width 1024 start 1096 end 1200 total 1376 skew 0
v: height 768 start 771 end 775 total 816
PTB-DEBUG: glClear splash image top-left reference pixel: 255 255 255
PTB-INFO: Threshold Settings for successfull video refresh calibration are: maxStdDev = 0.200000 msecs,maxDeviation = 10.000000 %, minSamples = 50, maxDuration = 5.000000 secs.
PTB-DEBUG: Screen 1 [head 1]: vbl_start = 812 vbl_end = 44 vactive = 816.
PTB-INFO: Screen 1 [head 1]: Applying beamposition corrective offsets: vblbias = 44, vbltotal = 816.

PTB-DEBUG: Output of all acquired samples of calibration run follows:
PTB-DEBUG: Sample 0: 0.000000
PTB-DEBUG: Sample 1: 0.010005
PTB-DEBUG: Sample 2: 0.010005
PTB-DEBUG: Sample 3: 0.010006
PTB-DEBUG: Sample 4: 0.010005
PTB-DEBUG: Sample 5: 0.010005
PTB-DEBUG: Sample 6: 0.010005
PTB-DEBUG: Sample 7: 0.010005
PTB-DEBUG: Sample 8: 0.010006
PTB-DEBUG: Sample 9: 0.010005
PTB-DEBUG: Sample 10: 0.010005
PTB-DEBUG: Sample 11: 0.010005
PTB-DEBUG: Sample 12: 0.010005
PTB-DEBUG: Sample 13: 0.010005
PTB-DEBUG: Sample 14: 0.010005
PTB-DEBUG: Sample 15: 0.010006
PTB-DEBUG: Sample 16: 0.010005
PTB-DEBUG: Sample 17: 0.010005
PTB-DEBUG: Sample 18: 0.010005
PTB-DEBUG: Sample 19: 0.010005
PTB-DEBUG: Sample 20: 0.010007
PTB-DEBUG: Sample 21: 0.010005
PTB-DEBUG: Sample 22: 0.010005
PTB-DEBUG: Sample 23: 0.010005
PTB-DEBUG: Sample 24: 0.010005
PTB-DEBUG: Sample 25: 0.010005
PTB-DEBUG: Sample 26: 0.010006
PTB-DEBUG: Sample 27: 0.010005
PTB-DEBUG: Sample 28: 0.010005
PTB-DEBUG: Sample 29: 0.010005
PTB-DEBUG: Sample 30: 0.010005
PTB-DEBUG: Sample 31: 0.010005
PTB-DEBUG: Sample 32: 0.010005
PTB-DEBUG: Sample 33: 0.010005
PTB-DEBUG: Sample 34: 0.010005
PTB-DEBUG: Sample 35: 0.010005
PTB-DEBUG: Sample 36: 0.010005
PTB-DEBUG: Sample 37: 0.010005
PTB-DEBUG: Sample 38: 0.010005
PTB-DEBUG: Sample 39: 0.010005
PTB-DEBUG: Sample 40: 0.010005
PTB-DEBUG: Sample 41: 0.010005
PTB-DEBUG: Sample 42: 0.010005
PTB-DEBUG: Sample 43: 0.010006
PTB-DEBUG: Sample 44: 0.010005
PTB-DEBUG: Sample 45: 0.010006
PTB-DEBUG: Sample 46: 0.010005
PTB-DEBUG: Sample 47: 0.010005
PTB-DEBUG: Sample 48: 0.010005
PTB-DEBUG: Sample 49: 0.010006
PTB-DEBUG: Sample 50: 0.010005
PTB-DEBUG: End of calibration data for this run...

PTB-INFO: OpenGL-Renderer is nouveau :: Gallium 0.4 on NV86 :: 3.0 Mesa 11.2.0
PTB-INFO: VBL startline = 768 , VBL Endline = 815
PTB-INFO: Measured monitor refresh interval from beamposition = 10.005163 ms [99.948400 Hz].
PTB-INFO: Will try to use OS-Builtin OpenML sync control support for accurate Flip timestamping.
PTB-INFO: Measured monitor refresh interval from VBLsync = 10.005183 ms [99.948195 Hz]. (50 valid samples taken, stddev=0.000490 ms.)
PTB-INFO: Reported monitor refresh interval from operating system = 10.002901 ms [99.971001 Hz].
PTB-INFO: Small deviations between reported values are normal and no reason to worry.
PTB-INFO: Support for fast OffscreenWindows enabled.

andreaswidmann · January 4, 2017, 5:58pm

Dear Mario,

thanks a lot for the detailed response and fixing my outdated mental concept! I will add your explanation to the FAQ page, ok?

Sorry, your explanation triggered two additional questions:

* At the end of PredictVisualOnsetForTime() the VBL duration is calculated and added to the estimated vbl time (if beamposition queries are supported). Wouldn’t that systematically overestimate expected stimulus onset time if VBLTimestamp already reflects real stimulus onset on OpenML systems? (the observed offset was the reason to start looking at the timestamps in more detail)

* And if I want to send a TTL trigger with the visual stimulus onset optimally I wait until VBLTimestamp if it is later than FlipTimestamp/GetSecs() on OpenML systems? Actually, wasn’t aware of that yet.

And yes, we verify audio-video sync with oscilloscope and photodiode.

Thank you! Best regards
Andreas

> these results look all perfect. I guess that FAQ is simply not quite up to date. It was written sometime before 2010, before we had the improved timestamping on Linux + FOSS drivers. It describes the beamposition timestamping used on Windows, OSX with AMD and NVidia cards + PsychtoolboxKernelDriver, and Linux with proprietary NVidia driver.
>
> On Linux with open-source drivers we use the OS builtin timestamping, which implements the OpenML OML_sync_control extension, and is generally superior to PTB's mechanism. On Intel, AMD and NVidia graphics cards, as well as the RaspberryPi and a couple of other embedded gpu's, the current algorithm is essentially a substantially improved version of PTB's mechanism, but other implementations would be possible as long as they satisfy the OML_sync_control spec to a sufficient degree.
>
> With OpenML timestamping, the returned timestamps are always visual stimulus onset timestamps, ie. the 2nd return argument of Flip, the time when the first new top-left stimulus pixel gets send out to the display. VBLTimestamp is set == stimulus onset timestamp, so the rule VBLTimestamp <= StimulusOnsetTimestamp holds. In theory we could calculate and return the VBLTimestamp from StimulusOnsetTimestamp, but in practice i can not think of any situation where this would provide a benefit.
>
> FlipTimestamp, which is not much more than a GetSecs taken at end of Flip, can and often will be earlier than StimulusOnsetTimestamp, as flip processing usually already completes while the display is still in VBLANK, preparing for display of the new stimulus frame, that's why StimulusOnsetTimestamp == VBLTimestamp > FlipTimestamp.
>
> PTB would fall back to the older beamposition timestamping if it would detect a problem with the OpenML timestamping. We have something like up to 4-fold redundancy on Linux with open-source drivers when it comes to visual timing and also extra mechanisms to detect and correct/compensate for trouble compared to other OS'es. The Screen Preference setting VBLTimestampingmode 0 (or 1?) would force you back on beamposition timestamping.
>
> -> Wrt. PredictVisualOnsetForTime(), yes that's what you should use to predict visual onset, and feed into PsychPortAudio as target time for good audio-video sync. In fact, i changed PsychPortAudioTimingTest.m - our example of doing and testing audio-video sync - in the latest Beta update to demonstrate that approach via PredictVisualOnsetForTime().
>
> -> One thing to keep in mind is that you should still test audio-video sync independently once for your setup before running data collection, as there is no way to detect purely in software if the audio timing works correctly. Also there can still be an unavoidable fixed offset between requested and true audio onset/offset, dependent on audio hardware. However, that offset is fixed, can be eliminated by calibration and was not ever observed by myself anywhere to be larger than ~1 msec. And Linux gets tested for proper audio-timing by myself with a Datapixx (via PsychPortAudioTimingTest and PsychPortAudioDatapixxTimingtest) for each kernel release at least on some typical onboard sound chips like in your machine and so far never failed in over 6 years of testing.
>
> Happy testing,
> -mario
>
>
> while programming a new experiment where we want to precisely sync auditory and visual stimuli I stumbled across some confusing oddities in the flip timestamping in our lab (I hope this is not a very stupid question):
>
> From the diagnostic PTB output (see below) I assumed that our system (Ubuntu 16.04 low-latency kernel, older nVidia Quadro NVS 290 card, nouveau driver, newest PTB) supports beamposition queries. However, the VBLTimeStamp returned by Screen('Flip‘, …) always equals the StimulusOnsetTime. On the other hand, the corresponding beampos is not -1 as I would have expected in case of missing support for beamposition queries from your explanation in the FAQ (https://github.com/Psychtoolbox-3/Psychtoolbox-3/wiki/FAQ:-Explanation-of-Flip-Timestamps) but consistently between 774 and 776.
>
> More importantly, the returned VBLTimeStamp is consistently *later* than the corresponding FlipTimestamp (by ~400 to 450 µS). The FlipTimestamp itself has only very small deviations from a GetSecs timestamp taken after Flip (< 50 µS).
>
> * Is this expected behavior and my mental concept is flawed (i.e., doesn’t the VBLTimeStamp reflect VBL onset and should thus be earlier than FlipTimestamp)?
> * Or did I misconfigure something?
> * Or is this a limitation of our admittedly somewhat older graphics card (working flawlessly otherwise)?
> * And finally, to compute true StimulusOnsetTime on this system would it be safe to apply the equation to compute VBL duration used in PredictVisualOnsetForTime?
>
> Thanks a lot! Best,
> Andreas
>
>
> RandR: 1024x768_100.00 (0xa9) 112.2MHz
> h: width 1024 start 1096 end 1200 total 1376 skew 0
> v: height 768 start 771 end 775 total 816
>
>
> PTB-INFO: This is Psychtoolbox-3 for GNU/Linux X11, under GNU/Octave 64-Bit (Version 3.0.14 - Build date: Dec 22 2016).
> PTB-INFO: Support status on this operating system release: Linux 4.4.0-57-lowlatency Supported.
> PTB-INFO: Type 'PsychtoolboxVersion' for more detailed version information.
> PTB-INFO: Most parts of the Psychtoolbox distribution are licensed to you under terms of the MIT License, with
> PTB-INFO: some restrictions. See file 'License.txt' in the Psychtoolbox root folder for the exact licensing conditions.
>
> PTB-DEBUG: Checking PCI device [Intel Corporation 4 Series Chipset DRAM Controller] with class x00060000 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 4 Series Chipset PCI Express Root Port] with class x00060400 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 4 Series Chipset HECI Controller] with class x00078000 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 4 Series Chipset Serial KT Controller] with class x00070002 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82567LM-3 Gigabit Network Connection] with class x00020000 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #4] with class x000c0300 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #5] with class x000c0300 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #6] with class x000c0300 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB2 EHCI Controller #2] with class x000c0320 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) HD Audio Controller] with class x00040300 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) PCI Express Port 1] with class x00060400 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) PCI Express Port 2] with class x00060400 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #1] with class x000c0300 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #2] with class x000c0300 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB UHCI Controller #3] with class x000c0300 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) USB2 EHCI Controller #1] with class x000c0320 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801 PCI Bridge] with class x00060401 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD (ICH10D) LPC Interface Controller] with classx00060100 ...
> PTB-DEBUG: Checking PCI device [Intel Corporation 82801JD/DO (ICH10 Family) SATA AHCI Controller] with class x00010601 ...
> PTB-DEBUG: Checking PCI device [NVIDIA Corporation G86 [Quadro NVS 290]] with class x00030000 ...
> PTB-INFO: NVIDIA Corporation - G86 [Quadro NVS 290] GPU found. Trying to establish low-level access...
> PTB-DEBUG: Mapping GPU BAR address 0xf2000000 ...
> PTB-DEBUG: Mapping 0x1000000 bytes...
> PTB-INFO: Connected to NVidia G86 [Quadro NVS 290] GPU of NV-050 family with 2 display heads. Beamposition timestamping enabled.
> PTB-INFO: Using old-style override-redirect setup path for onscreen window creation.
> PTB-INFO: GLX Visual info depths is 24 bits
> PTB-INFO: Running on Mesa version 11.2.0
> PTB-INFO: This combo of X-Server and Mesa is considered safe for use under DRI3/Present.
> PTB-INFO: Using GLEW version 2.0.0 for automatic detection of OpenGL extensions...
> PTB-INFO: Using GLX_MESA_swap_control extension for control of vsync.
> PTB-INFO: INTEL_swap_event support for additional swap completion correctness checks enabled.
> PTB-DEBUG: PPM file magic is P6
> -> Ok
> # CREATOR: GIMP PNM Filter Version 1.1
> PTB-DEBUG: Recognized splash image of 656 x 304 pixels, maxlevel 255. Loading...
>
>
> OpenGL-Vendor / renderer / version are: nouveau - Gallium 0.4 on NV86 - 3.0 Mesa 11.2.0
>
>
> OpenGL-Extensions are: GL_ARB_multisample GL_EXT_abgr GL_EXT_bgra GL_EXT_blend_color GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_copy_texture GL_EXT_polygon_offset GL_EXT_subtexture GL_EXT_texture_object GL_EXT_vertex_array GL_EXT_compiled_vertex_array GL_EXT_texture GL_EXT_texture3D GL_IBM_rasterpos_clipGL_ARB_point_parameters GL_EXT_draw_range_elements GL_EXT_packed_pixels GL_EXT_point_parameters GL_EXT_rescale_normal GL_EXT_separate_specular_color GL_EXT_texture_edge_clamp GL_SGIS_generate_mipmap GL_SGIS_texture_border_clamp GL_SGIS_texture_edge_clamp GL_SGIS_texture_lod GL_ARB_framebuffer_sRGB GL_ARB_multitexture GL_EXT_framebuffer_sRGB GL_IBM_multimode_draw_arrays GL_IBM_texture_mirrored_repeat GL_ARB_texture_cube_map GL_ARB_texture_env_add GL_ARB_transpose_matrix GL_EXT_blend_func_separate GL_EXT_fog_coord GL_EXT_multi_draw_arrays GL_EXT_secondary_color GL_EXT_texture_env_add GL_EXT_texture_filter_anisotropic GL_EXT_texture_lod_bias GL_INGR_blend_func_separate GL_NV_blend_square GL_NV_light_max_exponent GL_NV_texgen_reflection GL_NV_texture_env_combine4 GL_S3_s3tc GL_SUN_multi_draw_arrays GL_ARB_texture_border_clamp GL_ARB_texture_compression GL_EXT_framebuffer_object GL_EXT_texture_compression_s3tc GL_EXT_texture_env_combine GL_EXT_texture_env_dot3 GL_MESA_window_pos GL_NV_packed_depth_stencil GL_NV_texture_rectangleGL_ARB_depth_texture GL_ARB_occlusion_query GL_ARB_shadow GL_ARB_texture_env_combine GL_ARB_texture_env_crossbar GL_ARB_texture_env_dot3 GL_ARB_texture_mirrored_repeat GL_ARB_window_pos GL_EXT_stencil_two_side GL_EXT_texture_cube_map GL_NV_depth_clamp GL_NV_fog_distance GL_APPLE_packed_pixels GL_APPLE_vertex_array_object GL_ARB_draw_buffers GL_ARB_fragment_program GL_ARB_fragment_shader GL_ARB_shader_objects GL_ARB_vertex_program GL_ARB_vertex_shader GL_ATI_draw_buffers GL_ATI_texture_env_combine3 GL_ATI_texture_float GL_EXT_depth_bounds_test GL_EXT_shadow_funcs GL_EXT_stencil_wrap GL_MESA_pack_invert GL_NV_primitive_restart GL_ARB_depth_clamp GL_ARB_fragment_program_shadow GL_ARB_half_float_pixel GL_ARB_occlusion_query2 GL_ARB_point_sprite GL_ARB_shading_language_100 GL_ARB_sync GL_ARB_texture_non_power_of_two GL_ARB_vertex_buffer_object GL_ATI_blend_equation_separate GL_EXT_blend_equation_separate GL_OES_read_format GL_ARB_color_buffer_float GL_ARB_pixel_buffer_object GL_ARB_texture_compression_rgtc GL_ARB_texture_float GL_ARB_texture_rectangle GL_ATI_texture_compression_3dc GL_EXT_packed_float GL_EXT_pixel_buffer_object GL_EXT_texture_compression_dxt1 GL_EXT_texture_compression_rgtc GL_EXT_texture_mirror_clamp GL_EXT_texture_rectangle GL_EXT_texture_sRGB GL_EXT_texture_shared_exponent GL_ARB_framebuffer_object GL_EXT_framebuffer_blit GL_EXT_framebuffer_multisample GL_EXT_packed_depth_stencil GL_ARB_vertex_array_object GL_ATI_separate_stencil GL_ATI_texture_mirror_once GL_EXT_draw_buffers2 GL_EXT_draw_instanced GL_EXT_gpu_program_parameters GL_EXT_texture_array GL_EXT_texture_compression_latc GL_EXT_texture_integer GL_EXT_texture_sRGB_decode GL_EXT_timer_query GL_OES_EGL_image GL_AMD_performance_monitor GL_ARB_copy_buffer GL_ARB_depth_buffer_float GL_ARB_draw_instanced GL_ARB_half_float_vertex GL_ARB_instanced_arrays GL_ARB_map_buffer_range GL_ARB_texture_rg GL_ARB_texture_swizzle GL_ARB_vertex_array_bgra GL_EXT_texture_swizzle GL_EXT_vertex_array_bgra GL_NV_conditional_render GL_AMD_conservative_depth GL_ARB_ES2_compatibility GL_ARB_blend_func_extended GL_ARB_debug_output GL_ARB_draw_elements_base_vertex GL_ARB_explicit_attrib_location GL_ARB_fragment_coord_conventions GL_ARB_provoking_vertex GL_ARB_sampler_objects GL_ARB_seamless_cube_map GL_ARB_shader_texture_lod GL_ARB_texture_multisample GL_ARB_texture_rgb10_a2ui GL_ARB_uniform_buffer_object GL_ARB_vertex_type_2_10_10_10_rev GL_EXT_provoking_vertex GL_EXT_texture_snorm GL_MESA_texture_signed_rgba GL_NV_texture_barrier GL_ARB_get_program_binary GL_ARB_robustness GL_ARB_separate_shader_objects GL_ARB_shader_bit_encoding GL_ARB_timer_query GL_NV_vdpau_interop GL_ANGLE_texture_compression_dxt3 GL_ANGLE_texture_compression_dxt5 GL_ARB_base_instance GL_ARB_compressed_texture_pixel_storage GL_ARB_conservative_depth GL_ARB_internalformat_query GL_ARB_map_buffer_alignment GL_ARB_shading_language_420pack GL_ARB_shading_language_packing GL_ARB_texture_storage GL_EXT_framebuffer_multisample_blit_scaled GL_EXT_transform_feedback GL_AMD_shader_trinary_minmax GL_ARB_ES3_compatibility GL_ARB_arrays_of_arrays GL_ARB_clear_buffer_object GL_ARB_copy_image GL_ARB_explicit_uniform_location GL_ARB_invalidate_subdata GL_ARB_program_interface_query GL_ARB_stencil_texturing GL_ARB_texture_query_levels GL_ARB_texture_storage_multisample GL_ARB_texture_view GL_ARB_vertex_attrib_binding GL_KHR_debug GL_ARB_buffer_storage GL_ARB_clear_texture GL_ARB_multi_bind GL_ARB_texture_mirror_clamp_to_edge GL_ARB_texture_stencil8 GL_ARB_vertex_type_10f_11f_11f_rev GL_EXT_shader_integer_mix GL_ARB_clip_control GL_ARB_conditional_render_inverted GL_ARB_derivative_control GL_ARB_get_texture_sub_image GL_ARB_pipeline_statistics_query GL_ARB_shader_texture_image_samples GL_ARB_texture_barrier GL_EXT_polygon_offset_clamp GL_KHR_context_flush_control
>
> PTB-DEBUG: Linux X11 backend with FOSS drivers - Enabling locked pixeltoken-write + flush workaround for XLib thread-safety.
> PTB-DEBUG: Interrogating Low-level renderer capabilities for onscreen window with handle 10:
> Indicator variables: FBO's 1, ATI_texture_float 1, ARB_texture_float 1, Vendor nouveau, Renderer Gallium 0.4 on NV86.
> Indicator variables: maxcolorattachments = 8, maxrectangletexturesize = 8192, maxnativealuinstructions = 16384.
> GPU supports non-power-of-two textures.
> Hardware supports signed normalized textures of 16 bpc integer format.
> Basic framebuffer objects with rectangle texture rendertargets supported --> RGBA8 rendertargets with blending.
> Framebuffer objects support fast blitting between each other.
> Framebuffer objects support anti-aliasing via multisampling.
> Framebuffer objects support single-pass multisample resolve blits and image rescaling.
> Hardware supports floating point textures of 16bpc and 32bpc float format.
> Assuming NV30 core or later...
> Assuming NV40 core or later (maxcolattachments=8): Hardware supports floating point blending and filtering on 16bpc float format.
> Hardware also supports floating point framebuffers of 16bpc and 32bpc float format.
> Hardware supports full 32 bit floating point precision shading.
> Assuming G80 core or later (maxtexsize=8192): Hardware supports full floating point blending and filtering on 16bpc and 32bpc float format.
> Assuming hardware supports native OpenGL primitive smoothing (points, lines).
> PTB-Info: Running on 'The X.Org Foundation' XServer, Vendor release 11804000.
> OML_sync_control indicators: glXGetSyncValuesOML=0x7f6c1bcd4c40 , glXWaitForMscOML=0x7f6c1bcd4d80, glXWaitForSbcOML=0x7f6c1bcd4e50, glXSwapBuffersMscOML=0x7f6c1bcd4cb0
> OML_sync_control indicators: glxewIsSupported() says 1.
> System supports OpenML OML_sync_control extension for high-precision scheduled swaps and timestamping.
> OpenML OML_sync_control extension enabled for all scheduled swaps.
> PTB-DEBUG: Interrogation done.
>
> RandR: 1024x768_100.00 (0xa9) 112.2MHz
> h: width 1024 start 1096 end 1200 total 1376 skew 0
> v: height 768 start 771 end 775 total 816
> RandR: 1024x768_100.00 (0xa9) 112.2MHz
> h: width 1024 start 1096 end 1200 total 1376 skew 0
> v: height 768 start 771 end 775 total 816
> PTB-DEBUG: glClear splash image top-left reference pixel: 255 255 255
> PTB-INFO: Threshold Settings for successfull video refresh calibration are: maxStdDev = 0.200000 msecs,maxDeviation = 10.000000 %, minSamples = 50, maxDuration = 5.000000 secs.
> PTB-DEBUG: Screen 1 [head 1]: vbl_start = 812 vbl_end = 44 vactive = 816.
> PTB-INFO: Screen 1 [head 1]: Applying beamposition corrective offsets: vblbias = 44, vbltotal = 816.
>
>
> PTB-DEBUG: Output of all acquired samples of calibration run follows:
> PTB-DEBUG: Sample 0: 0.000000
> PTB-DEBUG: Sample 1: 0.010005
> PTB-DEBUG: Sample 2: 0.010005
> PTB-DEBUG: Sample 3: 0.010006
> PTB-DEBUG: Sample 4: 0.010005
> PTB-DEBUG: Sample 5: 0.010005
> PTB-DEBUG: Sample 6: 0.010005
> PTB-DEBUG: Sample 7: 0.010005
> PTB-DEBUG: Sample 8: 0.010006
> PTB-DEBUG: Sample 9: 0.010005
> PTB-DEBUG: Sample 10: 0.010005
> PTB-DEBUG: Sample 11: 0.010005
> PTB-DEBUG: Sample 12: 0.010005
> PTB-DEBUG: Sample 13: 0.010005
> PTB-DEBUG: Sample 14: 0.010005
> PTB-DEBUG: Sample 15: 0.010006
> PTB-DEBUG: Sample 16: 0.010005
> PTB-DEBUG: Sample 17: 0.010005
> PTB-DEBUG: Sample 18: 0.010005
> PTB-DEBUG: Sample 19: 0.010005
> PTB-DEBUG: Sample 20: 0.010007
> PTB-DEBUG: Sample 21: 0.010005
> PTB-DEBUG: Sample 22: 0.010005
> PTB-DEBUG: Sample 23: 0.010005
> PTB-DEBUG: Sample 24: 0.010005
> PTB-DEBUG: Sample 25: 0.010005
> PTB-DEBUG: Sample 26: 0.010006
> PTB-DEBUG: Sample 27: 0.010005
> PTB-DEBUG: Sample 28: 0.010005
> PTB-DEBUG: Sample 29: 0.010005
> PTB-DEBUG: Sample 30: 0.010005
> PTB-DEBUG: Sample 31: 0.010005
> PTB-DEBUG: Sample 32: 0.010005
> PTB-DEBUG: Sample 33: 0.010005
> PTB-DEBUG: Sample 34: 0.010005
> PTB-DEBUG: Sample 35: 0.010005
> PTB-DEBUG: Sample 36: 0.010005
> PTB-DEBUG: Sample 37: 0.010005
> PTB-DEBUG: Sample 38: 0.010005
> PTB-DEBUG: Sample 39: 0.010005
> PTB-DEBUG: Sample 40: 0.010005
> PTB-DEBUG: Sample 41: 0.010005
> PTB-DEBUG: Sample 42: 0.010005
> PTB-DEBUG: Sample 43: 0.010006
> PTB-DEBUG: Sample 44: 0.010005
> PTB-DEBUG: Sample 45: 0.010006
> PTB-DEBUG: Sample 46: 0.010005
> PTB-DEBUG: Sample 47: 0.010005
> PTB-DEBUG: Sample 48: 0.010005
> PTB-DEBUG: Sample 49: 0.010006
> PTB-DEBUG: Sample 50: 0.010005
> PTB-DEBUG: End of calibration data for this run...
>
>
>
> PTB-INFO: OpenGL-Renderer is nouveau :: Gallium 0.4 on NV86 :: 3.0 Mesa 11.2.0
> PTB-INFO: VBL startline = 768 , VBL Endline = 815
> PTB-INFO: Measured monitor refresh interval from beamposition = 10.005163 ms [99.948400 Hz].
> PTB-INFO: Will try to use OS-Builtin OpenML sync control support for accurate Flip timestamping.
> PTB-INFO: Measured monitor refresh interval from VBLsync = 10.005183 ms [99.948195 Hz]. (50 valid samples taken, stddev=0.000490 ms.)
> PTB-INFO: Reported monitor refresh interval from operating system = 10.002901 ms [99.971001 Hz].
> PTB-INFO: Small deviations between reported values are normal and no reason to worry.
> PTB-INFO: Support for fast OffscreenWindows enabled.
>
>
>
>

Topic		Replies	Views
flip timing jitters, missing beampos lines	1	595	June 25, 2007
Flip?	1	1996	December 23, 2008
Timing issue	0	165	July 22, 2008
"didn't use pageflipping for flip"	2	799	July 14, 2013
An indescribable problem, seemingly also a hardware issue Programming Help windows	8	149	March 10, 2024

Odd flip timestamps

Related Topics