[OpenSPIM] Suboptimal acquisition speed

[Feinstein, Timothy N]

Thanks Pete, no, after consulting with Hamamatsu engineers I realized that "lighsheet mode" is exclusively for instruments that use a scanned Bessel beam or Airy beam.  Standard cylinder-lens light sheets have to use standard acquisition modes.  In this case I keep the stage still during my acquisitions so the camera can run at its theoretical max speed.  

After exploring some a bit it looks like our FLASH4 v2 only gives its theoretical max FPS in 'streaming mode', which means it dumps data straight from the camera into memory and then saves it out of the circular buffer when it's done.  I can set that up in the proprietary Hamamatsu software (HCImage Live) but I cannot figure out how to do it in Micro-Manager 1.4.3/openSPIM.  Other people have pulled it off in the past by writing beanshell scripts.  

BTW, earlier I promised to report on the Space Mouse.  It works fantastic with the Picard stage.  Using it speeds up our acquisitions dramatically relative to moving the stage with sliders in the openSPIM plugin.  My only complaint is that the rotation is far too coarse; no matter what the settings I cannot get it to rotate less than 20 degrees at a time.  Picard recommended that I buy a kit to upgrade from the O-ring to a toothed belt to get more rotational precision.  I will post a note if that helps.  

All the best, 


Tim

Timothy Feinstein, Ph.D.
Research Scientist
University of Pittsburgh Department of Developmental Biology


-----Original Message-----
From: Peter Gabriel Pitrone [mailto:pitrone at mpi-cbg.de] 
Sent: Tuesday, March 22, 2016 1:35 PM
To: Feinstein, Timothy N
Cc: openspim at openspim.org
Subject: Re: [OpenSPIM] Suboptimal acquisition speed

Hello Tim,

Two questions need answering first:

1.) Are you in "Lightsheet mode"?!

This would "half" your frame rate, as the camera would read in only one direction (top to bottom) instead of bi-directional (from the center to the top and bottom).

And

2.) Assuming you are using the Picard stage: What is your motor dwell time set to?

As a standard setting it is set to 10 ms as to settle the sample from swinging, which could be your issue if you already set your exposure to 10 ms...

I hope that helps explain the problems you could be facing, if not good luck finding what else it could be.


Regards,
Pete

----- Original Message -----
From: "Feinstein, Timothy N" <tnf8 at pitt.edu>
To: openspim at openspim.org
Sent: Monday, March 21, 2016 7:34:02 PM
Subject: [OpenSPIM] Suboptimal acquisition speed

Hello all, 



Does anyone have experience troubleshooting frame rate problems? We have a Hamamatsu Flash 4.0 v2 that is theoretically capable of 100 FPS with a CameraLink card, but we get about half that using Micro-Manager/openSPIM. I have spent some time with Hamamatsu engineers and as far as we can tell the hardware works fine; we can reach expected speeds with Hamamatsu’s diagnostic software (XCAP). Rather it seems like MM is inefficiently communicating with the API, possibly not ‘expecting’ or ‘asking for’ frames at the same interval that they are being delivered from the camera. I really do not want to pay huge premium for Metamorph unless we absolutely have to, especially without knowing whether that will help. 



Thanks for any suggestions. 


Best, 




Tim 



Timothy Feinstein, Ph.D. 

Research Scientist 

University of Pittsburgh Department of Developmental Biology 



_______________________________________________
OpenSPIM mailing list
OpenSPIM at openspim.org
http://openspim.org/mailman/listinfo/openspim