Hi all,
Just reinforcing that issue:
If you compile mitk in shared, DO LINK all other dependencies of mitk in
shared as well. Namely VTK and ITK in shared. Not doing it would create
problems with static objects and other undesirable behaviours (for instance
not to update the Levelwindow or rendering manager not working properly,
memory leaks, etc)
Juan Antonio Moya
Scientific developer
Computational Imaging Lab, Pompeu Fabra University
Networking Research Center on Bioengineering, Biomaterials and Nanomedicine
(CIBER-BBN)
Information & Communication Technologies Department - D. 313
Pompeu Fabra University, Pg. Circumval·lació 8
08003 Barcelona - Spain
Phone +34 935 421350
-----Mensaje original-----
De: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] En nombre de Seitel
Mathias
Enviado el: viernes, 22 de febrero de 2008 11:08
Para: Klaus Drechsler; [email protected]
Asunto: Re: [mitk-users] Bug in mitkImageMapper2D
Hello Klaus,
We could reproduce this behavior now, and apparently it is due to the wrong
version of ITK being used.
A few days back, there was a major change in the way the MITK core library
and the Qmitk library are generated. Both can be compiled as shared libs now
(DLLs on windows). This significantly reduces overall linking times.
However, to make use of this feature, it is necessary to also use
dynamically build versions of both ITK and VTK (you can check this via CMake
or by looking for dll files in your ITK/VTK binary folders).
To reproduce the bug, we tried to use the "wrong" ITK version (statically
build) which produced the same results as you reported.
Thanks for pointing out the memory leakage problem. I'll make note of this
in our bug tracker.
Regards,
Mathias
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Klaus
Drechsler
Sent: Thursday, February 21, 2008 11:29 AM
To: [email protected]
Subject: Re: [mitk-users] Bug in mitkImageMapper2D
Hi Mathias,
> I tried this out on my machine (basically with the same system
> configuration), but so far couldn't reproduce the bug. What dataset are
> you using? What extent and dimensions does it have? Is it a
> time-resolved dataset?
I tried it with two datasets in different file formats and I also tried a
debug and a release toolkit-chain (qt,vt,it,mitk).
You can try it with the followig public dataset, that I found on the
internet:
http://pubimage.hcuge.ch:8080/DATA/RATIB1.zip (~250 MB)
> To get more clarity about what is happening here, you could set a
> breakpoint at BaseData::SetGeometry() and have a look at the call
> stack... Hopefully this should reveal why the Geometry changes, and how
> this is linked to the RegionGrowing functionality. It would be great if
> you could try this out and give us a feedback!
I will try it out, but it may take some time.
BTW: I also noticed that there seems to be memory leaks somewhere.
When I open and close the above dataset a few times then the memory usage
info in the sample app and windows´ task manager shows an increasing memory
usage:
after start: 71 MB
after loading: 230 MB
after closing: 93 MB
after loading: 240 MB
after closing: 102 MB
...
.
.
(I dont know, if it is a problem with MITK, VTK or ITK, but should be fixed)
Regards Klaus
--
Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten
Browser-Versionen downloaden: http://www.gmx.net/de/go/browser
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
mitk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mitk-users
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
mitk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mitk-users
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
mitk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mitk-users
