CONFOCALMICROSCOPY Archives

August 2003

CONFOCALMICROSCOPY@LISTS.UMN.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Karl Garsha <[log in to unmask]>
Reply To:
Confocal Microscopy List <[log in to unmask]>
Date:
Thu, 14 Aug 2003 09:57:59 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (119 lines)
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hello Juan-Carlos,
The 400 MB problem is a known issue with the newest build of the Leica
software as ported to Windows XP-running XP in NT emulation mode won't
help either. Changing the paging file configuration does't have an
impact.  Defragging the hard drive doesn't seem to help either.
Apparently the legacy code base for the confocal software uses some sort
of unconventional virtual memory swap space on the hard disk rather than
making use of available system memory or the default swap space on
windows systems.  It is a low-level problem hard-coded into the software
and there probably won't be an easy fix until the low level software
internals are updated.  Even if you get the 400MB warning by trying to
aquire a dataset that should approach this meager limit, the software
will fail to allow opening or aquiring of datasets in some instances.
When this happens you will have to restart the software in order to
work.  The only solution I'm presently aware of is to reinstall the
entire system (operating system and all) from a bit-for-bit disk image
of a clean factory install. This isn't the same as a clean install of
the operating system and software--some tweak done at the factory needs
to be replicated and service didn't have the details of this hack when
we tried to move to XP.  Any data kept on the system drive(s) will be
destroyed in this process, so make sure you have everything backed up.
Depending on how often the software is used, you may have to do this
frequently. You'll need to get the disk image disks from Leica and
you'll need Norton Ghost, or some similar software to image your own
drive after the reformat/image install.  The WinXP port of the Leica
software also has issues with the service diagnostics functionality, and
there are probably some other issues I'm not aware of.
    Apparently somebody doing the system installs at the factory knows
what the workaround is, but nobody else at Leica can seem to find out if
there is a fix to the problem in the os config or file system short of
complete wiping of the system and reinstall of a ghosted drive image--at
least this was the case a couple of months ago.
    The LCS software will work on WindowsNT, so we are running
WindowsNT.  WinNT probably won't jive with your newer computer hardware
though, but you may be able to get an old used 'Leica Certified' machine
from Leica.  The old machine with WindowsNT shouldn't give you problems
with large datasets.  Be advised, however, that Microsoft will no longer
be supporting WindowsNT  and the Leica-modified service pack required to
run the confocal instrument control software leaves gaping security
holes in the system and you will get hacked unless you are properly
firewalled or offline.
    In short, the software internals need to be rewritten to catch up
with improvements in computer hardware and trends in multi-dimensional
imaging.  It can take 7-10 minutes to open a 325MB dataset on a dual
processor machine with a gigabyte of ram and dual 60gb harddrives in
RAID 0 configuration.  This has something to do with the software
creating an image database somewhere--probably in this mysterious
virtual memory space--each time an experiment with a large dataset is
opened.  Saving takes equally long (much longer on a stock Leica NT
box), which is silly.  I'm not sure that the physical memory is used at
all aside from holding the program, instrument settings and the
operating system.  Images are written to the /Windows/Temp directory in
raw form as they are aquired, later, when the images are saved, they are
converted into a series of tiff files and the instrument parameter
settings are written to text inside the directory of your choosing.
    Good luck, and please let me know if you find an alternative
solution to the nefarious 400MB problem.
    Regards,
    Karl

Juan-Carlos Sarria wrote:

>Search the CONFOCAL archive at
>http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
>Hi, confocalists!
>
>We have an AOBS SP2 confocal microscope.
>all works pretty well but we have a little memory problem.
>Our computer have 2GB of RAM, but our LEICA software always tell at startup:
>
>"System performance optimization failed.
>Maximum data-size is about 400 MB"
>
>The offline software, tells the same (in other 2GB computer).
>
>We are starting living cell following and we are starting gigabite imaging...
>
>We have almost all tried to change the memory assignment without finding the good way.
>Maybe it could be trivial for one of you?....
>
>If somebody know how.... Help!
>
>Thanks a lot!
>
>Floyd
>
>____________________________________________________________
>
>Sarria J.-C. Floyd
>Swiss Federal Institute of Technology (EPFL) - Lausanne
>School of Life Sciences - The Neurosciences Institute
>EPFL/SV/Cellular Neurobiology Lab (LNC)/ Image Processing & Analysis
>AI-Ecublens, CH-1015 Lausanne, Switzerland
>
>Phone : +41 (21) 693 96 29
>Mobile: +41 (78) 830 26 00
>Email: [log in to unmask]
>URL: http://sv.epfl.ch
>
>

--
Karl Garsha
Light Microscopy Specialist
Imaging Technology Group
Beckman Institute for Advanced Science and Technology
University of Illinois at Urbana-Champaign
405 North Mathews Avenue
Urbana, IL 61801
Office: B650J
Phone: 217.244.6292
Fax: 217.244.6219
Mobile: 217.390.1874
www.itg.uiuc.edu

ATOM RSS1 RSS2