1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 <2019> 2020 2021 2022 2023 2024 | Index | 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 <2019> 2020 2021 2022 2023 2024 |
<== Date ==> | <== Thread ==> |
---|
Subject: | New releases of areaDetector drivers ADSimDetector, ADPerkinElmer, and ADSpinnaker |
From: | Mark Rivers via Tech-talk <[email protected]> |
To: | "[email protected]" <[email protected]> |
Date: | Wed, 29 May 2019 23:13:47 +0000 |
Folks, New releases of the following areaDetector modules are now available.
**************************************************************************************** ADSimDetector R2-9 (May 29, 2019) =================== * Requires asyn R4-35 because of changes to the asynPortClient class used in simDetectorNoIOC (see below). * Changes in the simulation modes and parameters to improve utility and performance. * The Noise parameter now applies to all simulation modes, not just to Peaks mode. * The Noise computation only calls rand() when the image is reset, i.e. the data type, dimensions, etc. are changed. In normal cases it just adds the precomputed noise to each pixel starting at a random pixel. This means the noise spatial pattern is the same in each successive image, but starting at a different pixel. This makes it look quite random while being very fast. * A new Offset parameter now applies to all simulation modes. * The sineNoise and sineOffset parameters were removed, replaced by Noise and Offset. * PeakVariation was changed from longout record to ao record so it is not constrained to be an integer. * Removed the dependence of the image data on the AcquireTime. This was confusing and meant the data changed significantly when the AcquireTime changed. * Improved performance of Peaks mode by more than 10X. Previously it computed the Gaussian peak profile for each peak in the image. Now it only does the Gausssian calculation once into a scratch buffer and then copies that peak multiple times into the output image at different locations. * A new simulation mode Offset&Noise was added. This mode only does offset and noise, and is the fastest simulation mode. It can generate >700 frames/s using 1024x1024 images and asynFloat64 data type, which is >5GB/s. In asynUInt8 mode it can generate >3000 frames/s which is >3GB/s. * simDetectorNoIOC * This directory contains a pure C++ application that instantiates a simDetector, a statistics plugin, and an HDF5 file plugin outside the context of an EPICS IOC. It is intended in part to demonstrate that areaDetector drivers and plugins can be used with other control systems such as Tango. * This application has been considerably improved and simplified by using the new versions of the asynPortClient and asynParamClient classes in asyn R4-35. * iocHDF5Test * Fixed the startup scripts. **************************************************************************************** ADPerkinElmer R2-9 (29-May-2019) =============== * Fixed error reading bad pixel file. There were 2 problems: - The structures defined in Acq.h were not declared to be packed. This caused the structures to be read incorrectly from the bad pixel file. In particular the image dimensions were read wrong. Added #pragma pack(1) when including Acq.h so these structures are packed. - Because of the previous problem the driver was deliberately using the wrong fields in the structure to determine the dimensions of the image array in the file (ULY, BRX, rather than BRX, BRY). If a compiler did do structure packing then the driver would crash in loadPixelCorrectionFile because the wrong structure members were being used. * Fixed error in loadGainFile(). It was using the PE_CorrectionsDirectory instead of PE_GainFile. **************************************************************************************** ADSpinnaker (new driver for Spinnaker SDK for FLIR/Point Grey cameras) R1-0 (May 29, 2019) ---- * Initial release **************************************************************************************** Mark |