Why does MATLAB 7.0 (R14) start up very slowly, and is slower in execution than MATLAB 6.5 (R13)?

52 views (last 30 days)
On my Windows machine with both MATLAB 7.0 (R14) and MATLAB 6.5 (R13) installed, MATLAB 7.0 is much slower than MATLAB 6.5, and takes an excessively long time to launch.

Accepted Answer

MathWorks Support Team
MathWorks Support Team on 21 Dec 2018
Two varieties of performance concerns have been reported in MATLAB 7.0 (R14). The first behavior involves MATLAB 7.0 (R14) taking a very long time to launch, particularly in the phase where "Initializing" is displayed next to the MATLAB Start button. However, once MATLAB 7.0 (R14) is launched, performance is acceptable. If this describes the behavior of your system, see solution 1 below. If MATLAB continues to exhibit poor performance after launching, and throughout the entire MATLAB session, see solution 2.
Note: The first problem has also been reported for MATLAB 7.2 (R2006a). Solution 1 applies to R2006a as well.
Solution 1:
The extended startup times exhibited by MATLAB 7.0 (R14) are due to the presence of other applications which use FLEXnet for license management, as MATLAB processes all entries in the LM_LICENSE_PATH environment variable during the "Initialization" phase, even if it is the first entry on the path. This behavior may arise even if you have an Individual License. To reduce the effect of this, try starting MATLAB with the -c option, which tells MATLAB which license.dat file to use directly, instead of searching the path. This is called from the Windows command line in the following format:
matlab -c $MATLABROOT\bin\win32\license.dat
where $MATLABROOT is the MATLAB root directory as returned by entering
matlabroot
at the MATLAB Command Prompt.
If this is successful, your Windows shortcuts to MATLAB can be modified in the same way. To modify a Windows shortcut to start MATLAB with the same "-c" option, perform the following:
1. Right-click on a shortcut to MATLAB and select "Properties"
2. In the "Target" field, enter the following:
$MATLABROOT\bin\win32\MATLAB.exe -c $MATLABROOT\bin\win32\license.dat
In Windows 2000, it may require quotes:
"$MATLABROOT\bin\win32\MATLAB.exe" "-c" "$MATLABROOT\bin\win32\license.dat"
(i.e., THREE different quotes) for this option to work correctly.
3. Click "Okay".
If you have many folders on your path whose contents do not change often, consider copying them to the MATLAB toolbox folder. This will allow MATLAB to cache these folders along with the rest of the toolbox folders, thus considerably reducing the initialization time for the MATLAB path. Note that MATLAB will not cache any folders other than those in the toolbox folder.
Solution 2a:
There is an issue with MATLAB 7.0 (R14) and the graphics libraries supplied with certain graphics cards, which do not use memory efficiently, and cause MATLAB 7.0 (R14) to use virtual memory more frequently than would be expected. In this case, disabling the hardware graphics acceleration features of the card may improve the performance of MATLAB 7.0 (R14). To do so, follow the steps below.
1. Close MATLAB.
2. Right-click on the Windows desktop, and select "Properties".
3. In the Display Properties window that opens, select the "Settings" tab, and then click on the "Advanced" button.
4. In the window which opens, select the "Troubleshoot" tab.
5. Set the "Hardware acceleration" slider all the way to the left, to "None".
6. Click OK in this window, and then click OK again in the Display Properties window.
7. Restart MATLAB.
Solution 2b:
In some instances, customers have reported that the anti-virus application on their machine, which was working in the background, was slowing down MATLAB. For these customers, disabling the active scan option for at least the MATLAB directories improved performance substantially.
If neither of these solutions resolved your performance concerns, or if you have a concern with the performance of a specific feature, contact Technical Support:

More Answers (0)

Categories

Find more on Startup and Shutdown in Help Center and File Exchange

Products

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!