<p dir="ltr">Hi Ralf,</p>
<p dir="ltr">I must admit, I'll puzzled by this behaviour. Most of what you're describing makes sense. The reason it's possible to generate and process very large track files is that most commands typically operate on the track file as a stream: each track is processed in turn, and only a handful of tracks need to be loaded at any one time. MRView on the other hand will need to load the whole file at once to display the tracks all together, and will in fact require even more RAM than the track file itself for this (to store the color of each point as well as its position). So this the reason MRView will crash out on large track files.</p>
<p dir="ltr">However, this doesn't explain why you would end up with corrupt track files once you get close to the RAM limit. I've never seen this behaviour personally, and I have no idea what could cause this. Given that you're operating on a virtual machine, I can only speculate that it might be related to the virtual machine's handling of external devices (I assume /mnt/hgfs refers to either a network drive or an external USB drive?). Maybe you could try to run the same command but store the results on the virtual machine's own virtual drive (assuming you have enough storage allocated)? If that works as expected, that would at least provide a clue as where the problem might be (hopefully not MRtrix...).</p>
<p dir="ltr">Sorry I can't be more help...<br>
Cheers,</p>
<p dir="ltr">Donald</p>
<p dir="ltr">--<br>
Dr J-Donald Tournier (PhD)</p>
<p dir="ltr">Senior Lecturer, Biomedical Engineering<br>
Division of Imaging Sciences & Biomedical Engineering<br>
King's College London</p>
<p dir="ltr">A: Department of Perinatal Imaging & Health, 1st Floor South Wing, St Thomas' Hospital, London. SE1 7EH<br>
T: +44 (0)20 7188 7118 ext 53613<br>
W: <a href="http://www.kcl.ac.uk/medicine/research/divisions/imaging/departments/biomedengineering">http://www.kcl.ac.uk/medicine/research/divisions/imaging/departments/biomedengineering</a><br>
</p>
<div class="gmail_quot<blockquote class=" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">Hi Donald,</p>
<p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px"><br>
</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">thanks for the answer.</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
At first Host and Guest system are 64bit and using NTFS and ext4.</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">The VMWare uses 4GB RAM.</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
<br></p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">4GB RAM:</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px"><br></p>
<p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
700,000 works 2.1 GB</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">800,000 didn't work 2,33 GB not viewable, and doesn't work for TDI</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
<br></p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">I raised the RAM to around 6GB:</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
<br></p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">850,000 works 2,48</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
1,000,000 works 2,91</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">1,500,000 works 4,37 not viewable, but works for TDI</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
2,000,000 works 5,83 not viewable, but works for TDI</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px"><br></p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
<br></p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">What I don't understand is. You dont need RAM for creating the file. But you need a lot RAM for viewing the file. If you watch the RAM load with MRView it fits, and if load a file with the 6GB RAM and its to much it goes near 100% and the terminal kills the process. Thats fine with me. But with 4GB RAM .. the files, that overloads the RAM by viewing...these files doesnt show the RAM load up. Instead it doesnt even start loading the RAM, it breaks with (<b>is not in MDS format (unrecognised magic number).</b></p>
<p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">Also, if I try to calculate TDI the same error message is thrown.</p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
<br></p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">So far, I have no limits for the 6GB RAM and the TDI processing. Viewing breaks between 1,000,000 between 1,500,000.</p>
<p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px"><br></p><p style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">Its fine with me, I can raise the RAM or playing with the SWAP file. But I'm curious about it. Maybe there is an easy answer or it helps somebody else. </p>
<p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px"><br></p><p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
**********************************************************************</p>
<p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px"><br></p><p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
<br>
</p><p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px"><br></p><p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
Hi Ralf,</p><p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">Difficult to know what's going on without a bit more information, but my guess is things start going wrong when you hit the 32-bit threshold. I'm not sure why it wouldn't at least stop tracking if that were the case though, I thought I'd put in some checks in case this type of problem occurred... In any event, the fact that you can generate a decent number of tracks successfully strongly suggests that the problem is related to file size in some way, and a quick calculation suggests that 500,000 tracks would probably take up of the order of 2-4Gb. You can probably check whether your file grows as expected beyond 2 or 4Gb: if it stops at that point, that would be the problem...</p>
<p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">Assuming that's the case, the problem would be either that you're running a 32 version of MRtrix, that your virtual machine is itself 32 bit, or that the filesystem you're writing to doesn't support large file sizes (e.g. the FAT filesystem as used in DOS-formatted hard drives). Since it does look like you're writing to an external hard drive, I would have a good look at how it's formatted... If it's a FAT partition, you might want to consider reformatting as a NTFS partition.</p>
<p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">Hope this helps,<br>Donald.</p><p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">
<br>--<br>Dr J-Donald Tournier (PhD)</p><p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">Senior Lecturer, Biomedical Engineering<br>Division of Imaging Sciences & Biomedical Engineering<br>
King's College London</p><p dir="ltr" style="font-size:medium;margin-bottom:0px;font-family:Times;margin-top:0px">A: Department of Perinatal Imaging & Health, 1st Floor South Wing, St Thomas' Hospital, London. SE1 7EH<br>
T: +44 (0)20 7188 7118 ext 53613<br>W: <a href="https://exmail.med.uni-magdeburg.de/owa/redir.aspx?C=6L5TSjZ7GEqdCag4y93XoyXbV5Kb5NAIAHJqEQfU-6Bpo1SU6NsXe3_85akimfXLSlXNiSAwvZI.&URL=http%3a%2f%2fwww.kcl.ac.uk%2fmedicine%2fresearch%2fdivisions%2fimaging%2fdepartments%2fbiomedengineering" target="_blank">http://www.kcl.ac.uk/medicine/research/divisions/imaging/departments/biomedengineering</a><br>
</p><div class="gmail_quote" style="font-size:medium;font-family:Times">On 13 Jan 2014 13:53, "Ralf" <<a href="https://exmail.med.uni-magdeburg.de/owa/redir.aspx?C=6L5TSjZ7GEqdCag4y93XoyXbV5Kb5NAIAHJqEQfU-6Bpo1SU6NsXe3_85akimfXLSlXNiSAwvZI.&URL=mailto%3aluetzken%40med.ovgu.de" target="_blank">luetzken@med.ovgu.de</a>> wrote:<br type="attribution">
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div>Hello,</div><div><br></div><div>did somebody encounter the error?</div>
<div><br></div><div>I can load whole brain tracks with 500,000 but 1,000,000 and above doesnt work. Im not sure where the threshold is.</div><div><br></div><div>Terminal:</div><div><br></div><div>generated the track:</div>
<div><br></div><div>30111 405207 generated, </div><div>30111 2019750 generated,</div><div>1500000 selected [100%]</div><div>[1]+ Done </div><div><br></div><div>real<span style="white-space:pre-wrap"> </span>68m21.159s</div>
<div>user<span style="white-space:pre-wrap"> </span>61m49.668s</div><div>sys<span style="white-space:pre-wrap"> </span>7m42.241s</div><div><br></div><div>ubuntu@ubuntu-virtual-machine:/mnt/hgfs/HR_DWI/ho33_2404_1.4_B1500_ElectroVectorScheme/dicomdaten_nr3538$ mrview anat_ho33_2404.mif &[1] 4237<br>
</div><div><br></div><div>Then I try loading the Track with the GUI:</div><div><br></div><div>ubuntu@ubuntu-virtual-machine:/mnt/hgfs/HR_DWI/ho33_2404_1.4_B1500_ElectroVectorScheme/dicomdaten_nr3538$ mrview: file "/mnt/hgfs/HR_DWI/ho33_2404_1.4_B1500_ElectroVectorScheme/dicomdaten_nr3538/whole_brain_1500000t.tck"<b> is not in MDS format (unrecognised magic number)</b></div>
<div><b><br></b></div><div>No further details about the error are given.<br></div><div><br></div><div>Hope somebody can help,</div><div><br></div><div>Ralf Lützkendorf</div></div><br>_______________________________________________<br>
Mrtrix-discussion mailing list<br><a href="https://exmail.med.uni-magdeburg.de/owa/redir.aspx?C=6L5TSjZ7GEqdCag4y93XoyXbV5Kb5NAIAHJqEQfU-6Bpo1SU6NsXe3_85akimfXLSlXNiSAwvZI.&URL=mailto%3aMrtrix-discussion%40www.nitrc.org" target="_blank">Mrtrix-discussion@www.nitrc.org</a><br>
<a href="https://exmail.med.uni-magdeburg.de/owa/redir.aspx?C=6L5TSjZ7GEqdCag4y93XoyXbV5Kb5NAIAHJqEQfU-6Bpo1SU6NsXe3_85akimfXLSlXNiSAwvZI.&URL=http%3a%2f%2fwww.nitrc.org%2fmailman%2flistinfo%2fmrtrix-discussion" target="_blank">http://www.nitrc.org/mailman/listinfo/mrtrix-discussion</a></blockquote>
</div></div>
<br>_______________________________________________<br>
Mrtrix-discussion mailing list<br>
<a href="mailto:Mrtrix-discussion@www.nitrc.org">Mrtrix-discussion@www.nitrc.org</a><br>
<a href="http://www.nitrc.org/mailman/listinfo/mrtrix-discussion" target="_blank">http://www.nitrc.org/mailman/listinfo/mrtrix-discussion</a><br>
<br></div>