Micromanager acquisition speed improvement?

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Micromanager acquisition speed improvement?

Sam Dundon
I recently (< 1 month) configured our spinning disk confocal to run using Micromanager. One individual needs to take acquisitions with 6 frames/second. This was possible in our older software on the same computer and the same equipment, but in Micromanager the max I can get at the moment is 3 (100 ms exposures, but the rate does not increase if we drop it to 10 so I think this is caused by delays in the software communications, not exposure time). So far I have tried:

-Leaving all shutters open during acquisition
-Not automatically saving the image, but giving it the max RAM and then saving after acquisition is completed
-Stripping out all equipment from the configuration that is not 100% necessary for these particular acquisitions (this made the biggest difference)

By doing these things, I got it up to 3 frames/sec from 1-2 frames/sec. My suspicion is that there isn't much else we can do with the current equipment and computer, but does anyone have any other suggestions I could try?
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Micromanager acquisition speed improvement?

Kyle Douglass
Hi Sam,

What camera are you using? Some sCMOS cameras have a setting where they "preclear" the sensor between frames to reduce noise. This has the effect of reducing the frame rate because a dead time is introduced between frames. For this reason exposure time is not strictly the inverse frame rate.

I would suggest checking your camera settings in the MM device properties browser for a setting along these lines.

Cheers,
Kyle

Dr. Kyle M. Douglass
Post-doctoral Researcher
EPFL - The Laboratory of Experimental Biophysics
http://leb.epfl.ch/
http://kmdouglass.github.io

________________________________________
From: Sam Dundon [[hidden email]]
Sent: Friday, June 9, 2017 7:43 PM
To: [hidden email]
Subject: [micro-manager-general] Micromanager acquisition speed improvement?

I recently (< 1 month) configured our spinning disk confocal to run using
Micromanager. One individual needs to take acquisitions with 6
frames/second. This was possible in our older software on the same computer
and the same equipment, but in Micromanager the max I can get at the moment
is 3 (100 ms exposures, but the rate does not increase if we drop it to 10
so I think this is caused by delays in the software communications, not
exposure time). So far I have tried:

-Leaving all shutters open during acquisition
-Not automatically saving the image, but giving it the max RAM and then
saving after acquisition is completed
-Stripping out all equipment from the configuration that is not 100%
necessary for these particular acquisitions (this made the biggest
difference)

By doing these things, I got it up to 3 frames/sec from 1-2 frames/sec. My
suspicion is that there isn't much else we can do with the current equipment
and computer, but does anyone have any other suggestions I could try?



--
View this message in context: http://micro-manager.3463995.n2.nabble.com/Micromanager-acquisition-speed-improvement-tp7588515.html
Sent from the Micro-Manager mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Micromanager acquisition speed improvement?

Sam Dundon
We have an Andor Laser Combiner PCU 100 series ER-PCUV-101 driving a Prior NanoScan Z NZ100 for Z control and controlling the halogen lamp shutter. I am now remembering that this *also* communicates with the camera. As currently set up, I am controlling the camera via communication with the computer, not via the Andor box. Maybe it is possible to trigger the camera with this through Micromanager to speed things up? The camera is an Andor iXON EMCCD from 2009. Model info on the side says DU-897E-C00-#BV-500

Other hardware includes the
-Spinning disk: Yokogawa CSUX1-A1N-E
-Prior XY Stage controller H30XYE/B
-Sutter Lambda 10-B controller for the emission filter wheel after the spinning disk.
-The microscope body is an Olympus IX71. Almost nothing on this is automated, so there was no sacrifice to removing it from the configuration to improve speed.
-488 and 561 nm LS 20 mW OBIS lasers from Coherent (recently purchased).

The Sutter controller also communicates with the Andor PCU, but at the moment this only controls the transmitted light shutter via TTL. Filter wheel position is communicated directly to the Sutter box from the computer.

For now, we are doing a Z-series time lapse with a single laser. In the best case scenario we would be able to do dual-color as well.

Thank you,
Sam
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Micromanager acquisition speed improvement?

MarshallC
Sam Dundon wrote
We have an Andor Laser Combiner PCU 100 series ER-PCUV-101 driving a Prior NanoScan Z NZ100 for Z control and controlling the halogen lamp shutter. I am now remembering that this *also* communicates with the camera. As currently set up, I am controlling the camera via communication with the computer, not via the Andor box. Maybe it is possible to trigger the camera with this through Micromanager to speed things up? The camera is an Andor iXON EMCCD from 2009. Model info on the side says DU-897E-C00-#BV-500
A couple things to check:
-Are you operating in frame transfer mode?
-What is the pixel readout rate?
-What is the vertical clock speed?
-Is the EM amplifier on?

I have an iXon 888 Ultra on my microscope, so not exactly the same, but I believe they're similar enough to make this relevant.  Depending on how these parameters are set it will greatly effect the frame rate.  Do you have Solis?  One place to start might be to open Solis, activate optacquire (I think that's what it's called) in fast kinetics mode (again, can't remember the exact name, but I think that's correct) and make a note of the settings.  Then you can launch micromanager and enter those settings to see if that gets you to your desired frame rate.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Micromanager acquisition speed improvement?

Sam Dundon
-Are you operating in frame transfer mode?
Yes

-What is the pixel readout rate?
Readout Mode is set at 10.000 MHz, the reported Readout Time is 29.0892.

-What is the vertical clock speed?
0.5 μs. I have messed around with this and with the Vertical Clock Voltage according to the information on this page: https://micro-manager.org/wiki/Andor. Going down to 0.3 μs I just get smeared images.

-Is the EM amplifier on?
Yes

We do not have Solis, we were running this on Andor IQ2 previously. I looked through all the viewable settings (possibly our camera is too old to have OptAcquire settings? I see nothing about this) and applied these settings in Micromanager. The only difference from the settings I was already using was the Vertical Readout was set to 1.7 μs instead of 0.5 μs. Obviously this did not speed up acquisition in Micromanager. The trigger control was set to "Free run" which I think must be the equivalent of "Software" in Micromanager.

I have been doing all my testing in Micromanager 2.0 beta because it reports the real time framerate. I have not been able to find this information in the 1.4 release, though it is possible I am missing it. With the Andor IQ2 software we can achieve 6 fps, with Micromanager the best I've been able to achieve is 3 fps.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: -|EXT|- Re: Micromanager acquisition speed improvement?

AndorThirdParty
Hi Sam,

I made a quick attempt at recreating your issue with an old PCI iXon 888 I have access to, but with your settings I get framerates of around 8 fps.

I have a few questions before I investigate further.

Are you able to recreate this issue with only the camera attached, or do some of the other devices have to be included in the configuration?

Could you check if you have the same problem when using internal triggering on the camera? Just in case the software triggers are somehow being delayed by communication with the other devices.

In your time lapse acquisition, what is the interval between acquisitions?

Just a note, I can also see the framerate displayed on the image acquisition window in Micromanager 1.4 (in the bottom right for Live Mode, and the bottom middle for MDA).

Thanks,
Jack Gaston
Software Engineer
-----Original Message-----
From: Sam Dundon [mailto:[hidden email]]
Sent: 13 June 2017 18:18
To: [hidden email]
Subject: -|EXT|- Re: [micro-manager-general] Micromanager acquisition speed improvement?

-Are you operating in frame transfer mode?
Yes

-What is the pixel readout rate?
Readout Mode is set at 10.000 MHz, the reported Readout Time is 29.0892.

-What is the vertical clock speed?
0.5 μs. I have messed around with this and with the Vertical Clock Voltage according to the information on this page:
https://micro-manager.org/wiki/Andor. Going down to 0.3 μs I just get smeared images.

-Is the EM amplifier on?
Yes

We do not have Solis, we were running this on Andor IQ2 previously. I looked through all the viewable settings (possibly our camera is too old to have OptAcquire settings? I see nothing about this) and applied these settings in Micromanager. The only difference from the settings I was already using was the Vertical Readout was set to 1.7 μs instead of 0.5 μs. Obviously this did not speed up acquisition in Micromanager. The trigger control was set to "Free run" which I think must be the equivalent of "Software" in Micromanager.

I have been doing all my testing in Micromanager 2.0 beta because it reports the real time framerate. I have not been able to find this information in the 1.4 release, though it is possible I am missing it. With the Andor IQ2 software we can achieve 6 fps, with Micromanager the best I've been able to achieve is 3 fps.



--
View this message in context: http://micro-manager.3463995.n2.nabble.com/Micromanager-acquisition-speed-improvement-tp7588515p7588536.html
Sent from the Micro-Manager mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


+++Scanned for Viruses by ForcePoint+++


___________________________________________________________________________This e-mail is confidential and is for the addressee only.  Please refer to www.oxinst.com/email-statement for regulatory information.
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: -|EXT|- Re: Micromanager acquisition speed improvement?

Sam Dundon
With only the camera attached, the framerate fluctuates between 2-4. Adding in the Andor Laser Combiner brings this up to 3-5, presumably because of TTL triggering through this box. Adding in the spinning disk doesn't make a difference.

I do not notice a speed difference between internal vs. software triggering.

For the experiments we need a fast framerate for, the time point interval is 1 s. I have changed this to 1 min to see if that makes any difference and it did not.

Also, all of this is running on an older 32-bit XP system. Possibly this is part of the problem? It is interesting that 1.4 does display the "data" framerate for others. For me, the MDA window reports the current time since acquisition start, position, channel, and z-slice. In between time points it displays the time to next image. I still do not see the framerate reported (again, maybe a 32-bit issue?).

This is the configuration I have been using for testing (which does not include our clean up filter or the microscope body), in case any useful information can be gleaned: Framerate.cfg
Loading...