Print

---

Hi,

Do you 'pre-read all particles into RAM"? Or do you 'copy particles to
scratch directory'?
In the first case, you might have run out of memory and your system
started swapping.
If you don't use either, 30K particles might still fit in the OS's cache
and be processed fast,
and 120K not.

> Also when trying to do convert class-001.mrc to mtz on ccpem, the program
> crashes and I cannot find in the output a reason why. Can anybody help?

I need more information.

You can also try phenix.map_to_structure_factors.

Best regards,

Takanori Nakane

> Dear all,
>
> we are using relion to do a 3D refinement of asymmetric particles with
box s
> ize 400 with a pixel size of 1.15 Å and the iterations with 30k
particles
> take 30-45 minutes, while with 120k particles they take 17 hours or more.
> Is there something that we are doing wrong? It seems like a
> disproportionate amount of time for just 3 times the number of particles.
> Does anybody have any tips to speed things up?
>
> Also when trying to do convert class-001.mrc to mtz on ccpem, the program
> crashes and I cannot find in the output a reason why. Can anybody help?
>
> Thanks,
>
> Chiara Rapisarda
>
> Post doc
> Université de Bordeaux, CNRS, INSERM
> [log in to unmask]
> Tel. +33 (5) 4000 3617
>
> Institut européen de chimie et biologie
> 2, rue Robert Escarpit
> 33607 Pessac, France
> www.iecb.u-bordeaux.fr
>
>