Matlab very slow (R2016a on Mac OSX)

2 次查看(过去 30 天)
Hi, Matlab 2016a is unreasonably slow on my iMac (3.7 GHz i7, 16GB memory). As example, just calling the help function: >> tic, help plot, toc ... Elapsed time is 87.121085 seconds.
Does anyone have any idea of what's going on, and if there is anything I can do to fix it? Thank you, Martin
  3 个评论
Martin Schain
Martin Schain 2017-3-15
Thanks for your reply John.
I am using macOS Sierra, v.10.12, and Matlab runs locally on this machine. I checked the activity monitor, the only weird thing I was able to identify was a drastic increase in %CPU used by matlab. Running a help command caused %CPU to increase (from ~.5% while idle) to ~20%. I am not sure what to make of that though. The OS does not flag Matlab to be "non-responding" during the process.
Walter Roberson
Walter Roberson 2018-1-20
Is it possible that your license is being served over the network?

请先登录,再进行评论。

回答(3 个)

Andrew Wingard
Andrew Wingard 2019-3-7
I had a similar problem on my mac, this was ultimately related to an app called Magnet I purchased through the app store. If you are running magnet, I suggest uninstalling the application and running matlab.
  2 个评论
Kristina Moore
Kristina Moore 2019-3-19
编辑:Kristina Moore 2019-3-19
WOW. I would've NEVER guessed that it was Magnet, out of all the apps I have on my Mac, but this is what fixed it for me. Thank you!
For posterity:
Although my MATLAB wasn't necessarily proccessing slow (plotting, solving systems of equations didn't take excruciatingly long as some people above have reported), but there was always a hiccup/lagginess to every little action. It was especially noticable when highlightling text in Editor to copy/paste. It was just enough to be annoying. I thought that I just had too much data in in MATLAB/too many things running on my Mac but uninstalling Magnet fixed it.
I'll report the issue to the Magnet devs--what a shame because I loved still having that capability after switching to Mac from Windows but MATLAB.
Radoya Panic
Radoya Panic 2019-12-29
I'm amazed. This was happening to me for over half a year and constantly frustrated me. Using a high quality mac and still having constant lagginess during my MATLAB use made me question the quality of my MAC. I deleted magent right away and am shocked. This worked...

请先登录,再进行评论。


Arnav Mendiratta
Arnav Mendiratta 2017-3-20
It is hard to tell the cause of this behavior without more information about your system.
Here are the plausible causes:
- Low on-system RAM
- Low Java Heap Space set in 'Preferences'
- Loading some data or executing commands in 'startup.m'. You could rule out this possibility by checking if you have any commands in your 'startup' file. To open this file, execute the following command in MATLAB Command Window:
>> edit startup
- Certain graphics card and drivers have been known to cause this issue. If it is a graphics issue, you can switch to 'softwareopengl' to resolve this. If you have figured out this is indeed a graphics card issue, the best idea would be to update your driver to latest version (or roll it back to previous version if you have the most updated one.)
- Run a benchmark by typing "bench" in the command window. It will show the performance of some standard algorithms on your computer.
If you are still facing issue, you can contact The MathWorks' Technical Support team by creating a new help ticket: http://www.mathworks.com/support/servicerequests/create.html
They will help you troubleshoot this performance issue.

Thomas Gederberg
Thomas Gederberg 2018-1-20
I have a similar problem running MATLAB 2017b on 2015 MacBook Pro (2.8 GHz Intel Core i7, 16 GB RAM). When I run benchmark, I get >> bench ans = 0.0933 0.1703 0.0464 0.1140 60.4114 63.8843
which is showing that the 2D and 3D graphics are taking an enormously long time.
  1 个评论
Thomas Gederberg
Thomas Gederberg 2018-1-20
Interesting, I rebooted my Mac and the benchmark results are much improved. >> bench ans = 0.0909 0.1384 0.0499 0.1051 1.1252 0.8111

请先登录,再进行评论。

类别

Help CenterFile Exchange 中查找有关 Graphics Performance 的更多信息

标签

Community Treasure Hunt

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

Start Hunting!

Translated by