Hi Weiguang,
Note that the "prevmostboundonly_" reduced snapshot files contain not only the most bound particles of the current subhalos, but in addition all particles that have ever been a most bound particle in the past (i.e. where most bound in previous group catalogues).
All the most bound particles in a given group/subhalo are again stored together, in the sequence of the FOF halos, and nested inside them, their subhalos.
There are special fields GroupLenPrevMostBnd and SubhaloLenPrevMostBnd in the group catalogue that tell how many particles with this most-bound property are stored for every group/subhalo. With this it is possible to selectively get the particles associated with a given group/subhalo (you need to compute the offset from the beginning yourself however). For every subhalo, there should be at least 1 particle, but if there are several, the first one is always the one with the corresponding SubhaloIDMostbound of the current group/subhalo catalogue.
Regards,
Volker
> On 6. Jan 2022, at 21:17, Weiguang Cui <cuiweiguang_at_gmail.com> wrote:
>
> Dear Volker,
>
> Thank you for this reply. Now I understand. What about the outputs which only have the most bound particles instead of a full snapshot? Are the IDs from the "prevmostboundonly_" can be directly used as the IDs in the group catalog?
> I suppose so as they are "saved in group order", but just want to get confirmation from you.
>
> Best,
> Weiguang
>
> -------------------------------------------
> https://weiguangcui.github.io/
>
>
> On Thu, Jan 6, 2022 at 4:12 PM Volker Springel <vspringel_at_mpa-garching.mpg.de> wrote:
>
> Dear Weiguang,
>
> Yes, the IDs of group particles are not saved any more in the group catalogues, as this information is redundant - the particle ID block in the snapshot files contains the *identical* information, because the particles are stored in group order by default (except that at the end of the snapshot ID block, also particle IDs not contained in any group are listed).
>
> The gadget4 code still contains the function fof_subfind_prepare_ID_list() necessary to prepare the IDs such that they can also be added to the group catalogue itself. In the public Arepo code, this option is still there (with the FOF_STOREIDS switch), and it could be reenabled for Gadget4, if desired, by again adding the lines enabled by this switch. However, as I said, it was deliberately removed in Gadget4 as being redundant.
>
> Best,
> Volker
>
> > On 5. Jan 2022, at 14:28, Weiguang Cui <cuiweiguang_at_gmail.com> wrote:
> >
> > Dear all,
> >
> > It seems the particle IDs for the FoF groups are not saved in the output FoF catalog...
> > ```
> > [_at_login8a groups_301]$ h5dump -g IDs fof_subhalo_tab_301.0.hdf5
> > HDF5 "fof_subhalo_tab_301.0.hdf5" {
> > GROUP "IDs" {
> > }
> > }
> > ```
> > Is this because of any configure or parameter options that I forgot to set for outputting this information? Or the `GroupOffsetType` and `GroupLenType` can be directly applied to the snapshot files?
> > Thank you!
> >
> > Best,
> > Weiguang
> >
> > -------------------------------------------
> > https://weiguangcui.github.io/
> >
> > -----------------------------------------------------------
> >
> > If you wish to unsubscribe from this mailing, send mail to
> > minimalist_at_MPA-Garching.MPG.de with a subject of: unsubscribe gadget-list
> > A web-archive of this mailing list is available here:
> > http://www.mpa-garching.mpg.de/gadget/gadget-list
>
>
>
>
> -----------------------------------------------------------
>
> If you wish to unsubscribe from this mailing, send mail to
> minimalist_at_MPA-Garching.MPG.de with a subject of: unsubscribe gadget-list
> A web-archive of this mailing list is available here:
> http://www.mpa-garching.mpg.de/gadget/gadget-list
>
> -----------------------------------------------------------
>
> If you wish to unsubscribe from this mailing, send mail to
> minimalist_at_MPA-Garching.MPG.de with a subject of: unsubscribe gadget-list
> A web-archive of this mailing list is available here:
> http://www.mpa-garching.mpg.de/gadget/gadget-list
Received on 2022-01-06 22:05:23