the list did not allow to send the attachment.
Am Sa., 26. Dez. 2020 um 17:41 Uhr schrieb Thomas Fischer <
> Thomas Fischer
> 17:36 (vor 0 Minuten)
>
> line/s in the script 68 104&105 Memory & Threads
> ----------------------------------------------------------
> 8k 4k 2666
> 4k 8k 4000
> 8k 8k 5332
>
> in case you are interested, I attach the script
>
> With best seasonal greetings, Thomas
>
>
>
> Am Sa., 26. Dez. 2020 um 17:36 Uhr schrieb Thomas Fischer <
>
[hidden email]>:
>
>> Dear Wayne and Fred,
>>
>> similar to Fred I do not have the file ImageJ.cfg. -x5120 works fine.
>> Since Fred was wondering as to the functionality of the wrapper script, I
>> tried a few replacements.
>>
>> line/s in the script 68 104&105 Memory & Threads
>> ----------------------------------------------------------
>> 8k4k 2666
>> 4k 8k 4000
>> 8k 8k 5332
>>
>> Am Fr., 25. Dez. 2020 um 23:58 Uhr schrieb Fred Damen <
[hidden email]>:
>>
>>> Greetings,
>>>
>>> Shorter version of the solution to your issue is to pass -x5120 to the
>>> imagej bash script for 5G of memory.
>>>
>>> Longer version...
>>>
>>> I use Fedora package installation of ImageJ and do not seem to have the
>>> ImageJ.cfg file that Wayne has mentioned, and my imagej bash script does
>>> not reference ImageJ.cfg, so I do not think there is any connection
>>> between these two possible ways of setting this memory option. I do not
>>> know how Edit>Options>Memory & Threads is supposed to work and have not
>>> tried to use it on Linux so I can not comment further...
>>>
>>> The code snippet that you provided below is not perfect, i.e., it defines
>>> default_mem to 4000 and in the subsequent code compares against the value
>>> 4000 and not to the value of the default_mem variable. So changing the
>>> value of of the default_mem variable is mute without correcting the
>>> subsequent code to use $default_mem instead of 4000. Although this is the
>>> answer to your inquire, you can ignore this paragraph if you use the
>>> initial suggested solution, as it renders this code irrelevant during
>>> command line parameter evaluation.
>>>
>>> It is interesting that this script uses the amount of free swap as the
>>> free_mem instead of the actual RAM...
>>>
>>> Enjoy,
>>>
>>> Fred
>>>
>>> On Fri, December 25, 2020 3:08 pm, Wayne Rasband wrote:
>>> >> On Dec 25, 2020, at 6:46 AM, Thomas Fischer <
[hidden email]>
>>> >> wrote:
>>> >>
>>> >> Hi altogether,
>>> >>
>>> >> my laptop runs under linux debian 10 and has 8 GB RAM. Edit › Options
>>> >> ›
>>> >> Memory & Threads does not allow me to increase the value above 2666MB.
>>> >
>>> > Try the Linux version of ImageJ available at
>>> >
https://imagej.nih.gov/ij/download.html. The Edit>Options>Memory &
>>> Threads
>>> > command updates the third line of the ImageJ.cfg file. For example, it
>>> > changes this line to "-Xmx4000m -cp ij.jar ij.ImageJ” when you set the
>>> > value to 4000 in the Memory & Threads dialog.
>>> >
>>> > -wayne
>>> >
>>> >> relevant lines in /usr/bin/imagej (a not so simple wrapper script used
>>> >> to
>>> >> run ImageJ) are
>>> >>
>>> >> declare -i mem
>>> >> declare -i default_mem=4000
>>> >> declare -i min_mem=16
>>> >> declare -i max_mem
>>> >> declare -i free_mem
>>> >>
>>> >> and further down
>>> >>
>>> >> elif [[ `uname` == 'Linux' ]] ; then
>>> >> if [[ `uname -m` == 'x86_64' ]] ; then
>>> >> java_path="${ij_path}/jre64/bin/java"
>>> >> max_mem=`free | awk 'NR == 2 {fmem=int($2 / 1024); if (fmem <
>>> >> 4000)
>>> >> {print fmem} else {print 4000}}'`
>>> >> free_mem=`free | awk 'NR == 3 {fmem=int($4 / 1024); if (fmem <
>>> >> 4000) {print fmem} else {print 4000}}'`
>>> >> mem=${free_mem}/3*2
>>> >> if (( $mem > $default_mem || $mem < $min_mem )) ; then
>>> >> mem=$default_mem ; fi
>>> >> else
>>> >>
>>> >> typing free:
>>> >> total used free shared buff/cache
>>> >> available
>>> >> Mem: 8064332 1579956 2659136 183288 3825240
>>> >> 5998348
>>> >> Swap: 8282108 0 8282108
>>> >>
>>> >> changing declare -i default_mem=4000 to declare -i default_mem=8000
>>> has
>>> >> no
>>> >> effect on the lines further down
>>> >>
>>> >> where is the superior position where the change to 8000 makes a
>>> >> difference
>>> >> in the file imagej?
>>> >>
>>> >> thank you, Thomas
>>> >
>>> > --
>>> > ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>>> >
>>>
>>> --
>>> ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>>>
>>