Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

R2010a stable?

0 views
Skip to first unread message

Todd Welti

unread,
Mar 11, 2010, 4:24:02 PM3/11/10
to
Anyone else having stability issues with 2010a? I've crashed it 7 or 8 times in the first 2 days, versus once a week or so with 2009b.

Mark Shore

unread,
Mar 11, 2010, 8:52:02 PM3/11/10
to
"Todd Welti" <twe...@harman.com> wrote in message <hnbn1i$8vk$1...@fred.mathworks.com>...

> Anyone else having stability issues with 2010a? I've crashed it 7 or 8 times in the first 2 days, versus once a week or so with 2009b.

Nothing noticeable yet. But our respective uses may be very different.

Jan Simon

unread,
Mar 12, 2010, 5:20:32 AM3/12/10
to
Dear Todd!

> Anyone else having stability issues with 2010a? I've crashed it 7 or 8 times in the first 2 days, versus once a week or so with 2009b.

Please mention, if the crashs are reproducible, related to exhausted memory, appear in Java or Mex code, appear only when using the OpenGL renderer.

What caused the crashs in 2009b?

Kind regards, Jan

Todd Welti

unread,
Apr 1, 2010, 3:57:19 PM4/1/10
to
"Jan Simon" <matlab.T...@nMINUSsimon.de> wrote in message <hnd4hg$9tg$1...@fred.mathworks.com>...

Well, it seems ok the last couple of weeks. It must have been a fluke....

nanren888

unread,
Apr 1, 2010, 5:52:05 PM4/1/10
to
"Todd Welti" <twe...@harman.com> wrote in message <hnbn1i$8vk$1...@fred.mathworks.com>...
> Anyone else having stability issues with 2010a? I've crashed it 7 or 8 times in the first 2 days, versus once a week or so with 2009b.

Haven't really noticed it any less stable than 2009b. [XP32]

That said, the issues with out of memory & especially the java heap space are there. When memory & java heap space issues happen with heaps (sorry) of memory left it's annoying.

EBS

unread,
Apr 1, 2010, 6:00:23 PM4/1/10
to
"nanren888 " <nanren888....@hotmail.com> wrote in message <hp34i4$llc$1...@fred.mathworks.com>...

Blame Bill Gates...

Yair Altman

unread,
Apr 2, 2010, 2:26:22 AM4/2/10
to
"EBS " <ericDOT...@gmail.com> wrote in message <hp351n$sk5$1...@fred.mathworks.com>...


You're wrong. It's NOT Microsoft's fault that Matlab has a pretty poor memory management. Other applications (like your browser for example) do know how to take advantage of all the extra memory - Matlab doesn't. It's possible that Microsoft makes it harder on the programmers to take advantage of this memory, but it is not impossible.

Yair Altman
http://UndocumentedMatlab.com

0 new messages