This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
>From: William C Klein <address@hidden> >Organization: Valparaiso >Keywords: 200204151650.g3FGoPa24623 McIDAS MCGUI Bill, >A few more questions and comments: > >1. What is the message that we get as we start McIDAS: > "using slow processing for Z-32 images" This is an informational message to let you know that when running X with depts of greater than 8 bits, your display processes will be a little slower. This used to be noticable on slow machines, but is not on current hardware. >2. By changing resolution in the $ mcidas config, we are unable to > retrieve datasets I'm sorry, but I am not sure what modification you made that is causing your problems. Do you mean that things stopped working when you changed the number of lines and/or elements of the display? This should have no effect on who the data is being gotten from or the definitions of the datasets your are trying to access. >3. Simon has commented on a script that would run to automagically > set up datasets for users instead of them poking around and > setting up these by themselves. Does this exist at this point? > Have I confused you? ;) In fact, the user 'mcidas' can setup defaults that will be usable by all McIDAS users. When 'mcidas' does the following: <login> cd data cp DATALOC.BAT LOCDATA.BAT <edit LOCDATA.BAT and setup where various datasets are to be found> cd ~/workdata batch.k LOCDATA.BAT Then a file named ADDESITE.TXT should be created in ~mcidas/data. Now, each user account that will be running McIDAS should have two environment variables setup for ADDE dataset location information: MCTABLE_READ - tells McIDAS what files to read when trying to find out which server to contact to get data from a dataset MCTABLE_WRITE - tells McIDAS what file to write when the user runs a DATALOC command to setup pointing at datasets MCTABLE_READ for a user that is not 'mcidas' should read: MCTABLE_READ="/home/user/mcidas/data/MCTABLE.TXT;/home/mcidas/data/ADDESITE.TXT" This tells the user's McIDAS session to first look in the user's own ~/mcidas/data/MCTABLE.TXT file first when trying to resolve the name of a server for a particular dataset and, if it doesn't find the information to then look in ~mcidas/data/ADDESITE.TXT. Since ~mcidas/data/ADDESITE.TXT is created by 'mcidas' during McIDAS installation/configuration, the user should be ready to go with dataset location information when s/he brings up her/his McIDAS session. >4. Currently, the DMGRID is 'active'. I'm not able to get the NCEP > forcast grids though. Something you said in a previous email makes me ask the following: Are you not getting the GRID data, or are you not decoding the GRID data, or is your impression of there not being any GRID data a result of the MCTUI interface not having an action that allows you to easily look at the GRID data? >You had commented on needing space in > order to do this. We are sitting on 35G drive with 7G used and 9M > of DS3 bandwidth to use. Will this suffice? You have plenty of disk space and bandwidth to process grids. On the assumption that you are not decoding model data into McIDAS GRID files, let me review what is needed in order for this to work: 1) you have to request the model data in your LDM ~ldm/etc/ldmd.conf file. The datastream that contains the model data is HRS. 2) you have to put an action in your LDM ~ldm/etc/pqact.conf file that sends the model data that you are receiving to a McIDAS-XCD process so that it can be decoded. The entry in pqact will look like: HRS .* PIPE xcd_run HRS 3) GRID decoding in McIDAS has to be turned on. Your comment above says that this is already the case 4) you would have to setup McIDAS file REDIRECTions for the GRID files that will be created. This should have been done in the step where you create a local copy of EXAMPLE.NAM in LOCAL.NAM and then edited LOCAL.NAM to set the locations for the various data files. After creating LOCAL.NAM, you would have had to make the REDIRECTions active in your 'mcidas' account as follows: <login as 'mcidas'> cd ~mcidas/workdata redirect.k REST LOCAL.NAM 5) To see the model data that is being decoded, you would have to have setup the RTGRIDS dataset. This would have been done when you made the local copy of DSSERVE.BAT in LSSERVE.BAT and then edited LSSERVE.BAT. You would make the dataset definitions active in the 'mcidas' account as follows: <login as 'mcidas'> cd ~mcidas/workdata te.k XCDDATA \"/directory_where_you_want_GRID_data_decoded batch.k LSSERVE.BAT So, our job now is to figure out exactly what in the above list is not working. Tom