Hello,
http://www.optitrack.com/static/documen ... 0Guide.pdf
I've found in this OptiHub guide (end of page 4) that cameras V100/V100R2 have a time delay between the trigger event of the OptiHub synchronisation and the start of the exposure of the cameras.
However, this document seems a bit outdated since V100 do not exist anymore and V100R2 have been renamed Flex3.
I would like to know if such a delay also exists on Flex13 and Prime13 cameras and if yes do you know its value (or at least how to compute it such as it is explained in this OptiHub guide)
Thanks,
Elie
Time lag between trigger event and start of exposure
-
steven.andrews
- NaturalPoint Employee
- Posts: 720
- Joined: Mon Jan 19, 2015 11:52 am
Re: Time lag between trigger event and start of exposure
Hello Elie,
Thank you for reaching out to us regarding your questions
For the Flex13, the delay from the external signal to the start of the camera exposure is 480us.
For all Prime cameras, the delay (in micro-seconds) is given by:
delay_us = [1000000/frame_rate]/2 - exposure_us/2
I hope you find this information to be useful.
Best regards,
Steven
--
Steven Andrews
OptiTrack | Customer Support Engineer
Thank you for reaching out to us regarding your questions
For the Flex13, the delay from the external signal to the start of the camera exposure is 480us.
For all Prime cameras, the delay (in micro-seconds) is given by:
delay_us = [1000000/frame_rate]/2 - exposure_us/2
I hope you find this information to be useful.
Best regards,
Steven
--
Steven Andrews
OptiTrack | Customer Support Engineer
Re: Time lag between trigger event and start of exposure
Thanks a lot. That helped me.
But I have another question (I did not start a new topic as it is a bit specific):
I try to limit the latency of my system, here is its description:
My main computer is sending a usb pulse at each display refresh to the External sync in of the optihub.
The optitrack (running with Motive 1.5.0 on another computer) is thus synchronised with my display refresh (120Hz) and captures an image at each frame (using Flex 13). Marker data is then send by UDP and received by the main computer to be treated and displayed.
In my program, I measured the time between when the pulse is sent and the moment I receive the UDP packet with the optitrack data. I found a value around 12.5ms (that is pretty big).
I assume that sending the pulse is quasi immediate as it is only electronically stuff. The optitrack data is captured 0.48ms after this pulse (you gave me that), during 1.5ms (exposure time). UDP broadcast should also be extremely fast and cannot explain for so many delay.
Are you aware of any time delay between the exposure time and the moment the data are streamed (with Flex 13 + Motive 1.5.0) ?
I suspect that some process (picture analysis, usb communication, streamed data buffering ?) inside the cameras or Motive takes some time.
Do you have any info about this ?
If yes (or no), do you think that this problem would be solved by using Prime 13 instead of Flex 13 ?
Elie
But I have another question (I did not start a new topic as it is a bit specific):
I try to limit the latency of my system, here is its description:
My main computer is sending a usb pulse at each display refresh to the External sync in of the optihub.
The optitrack (running with Motive 1.5.0 on another computer) is thus synchronised with my display refresh (120Hz) and captures an image at each frame (using Flex 13). Marker data is then send by UDP and received by the main computer to be treated and displayed.
In my program, I measured the time between when the pulse is sent and the moment I receive the UDP packet with the optitrack data. I found a value around 12.5ms (that is pretty big).
I assume that sending the pulse is quasi immediate as it is only electronically stuff. The optitrack data is captured 0.48ms after this pulse (you gave me that), during 1.5ms (exposure time). UDP broadcast should also be extremely fast and cannot explain for so many delay.
Are you aware of any time delay between the exposure time and the moment the data are streamed (with Flex 13 + Motive 1.5.0) ?
I suspect that some process (picture analysis, usb communication, streamed data buffering ?) inside the cameras or Motive takes some time.
Do you have any info about this ?
If yes (or no), do you think that this problem would be solved by using Prime 13 instead of Flex 13 ?
Elie
-
steven.andrews
- NaturalPoint Employee
- Posts: 720
- Joined: Mon Jan 19, 2015 11:52 am
Re: Time lag between trigger event and start of exposure
Hi Elie,
Based on the information I can find online, the Flex 3's are reported to have a latency of 10ms and the Prime 13's 4.2ms. While the delay between trigger and exposure is one thing, I believe these numbers reflect the time it takes for data transfer. On top of this, if you are tracking realtime 3D data, Motive requires an amount of time to process the 2D data.
http://optitrack.com/hardware/compare/
This might help to explain the latency that is greater than the value you requested. And yes, it does seem that using Prime cameras would provide you faster results.
Cheers,
Steven
--
Steven Andrews
OptiTrack | Customer Support Engineer
Cheers,
Steven
Based on the information I can find online, the Flex 3's are reported to have a latency of 10ms and the Prime 13's 4.2ms. While the delay between trigger and exposure is one thing, I believe these numbers reflect the time it takes for data transfer. On top of this, if you are tracking realtime 3D data, Motive requires an amount of time to process the 2D data.
http://optitrack.com/hardware/compare/
This might help to explain the latency that is greater than the value you requested. And yes, it does seem that using Prime cameras would provide you faster results.
Cheers,
Steven
--
Steven Andrews
OptiTrack | Customer Support Engineer
Cheers,
Steven