# .trr file utilities

**URL:** <https://gromacs.bioexcel.eu/t/trr-file-utilities/10579>\
**Category:** User discussions\
**Tags:** mdrun\
**Created:** [November 18, 2024, 7:59pm UTC](https://gromacs.bioexcel.eu/t/trr-file-utilities/10579 "2024-11-18T19:59:02Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![fernandoovb](https://avatars.discourse-cdn.com/v4/letter/f/e9c0ed/32.png) [@fernandoovb](https://gromacs.bioexcel.eu/u/fernandoovb)\
**Post date:** [November 18, 2024, 7:59pm UTC](https://gromacs.bioexcel.eu/t/trr-file-utilities/10579/1 "2024-11-18T19:59:02Z")

</div>

GROMACS version: 2022  
GROMACS modification: No

Greetings!  
I’m a new user to gromacs, I would like to learn more about the software. Right now, I’m asking for help with my membrane simulations.  
I noticed that bulk membrane simulations, rather than proteins, create a .trr file that is much larger than .xtc file.  
It is my understanding that .trr is a trajectory file, and so is the .xtc file. So the question is, why is .trr file created? How can I block its creation? Is there useful informations I can extract from .trr file that I can’t get from the .xtc file?

Thank you for your attention, and sorry for my english.  
fernandoovb

---

<div class="post-metadata">

**Author:** ![MagnusL](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/magnusl/32/2380_2.png) [@MagnusL](https://gromacs.bioexcel.eu/u/MagnusL)\
**Post date:** [November 19, 2024, 7:42am UTC](https://gromacs.bioexcel.eu/t/trr-file-utilities/10579/2 "2024-11-19T07:42:30Z")

</div>

There is some more information here: [File formats - GROMACS 2024.4 documentation](https://manual.gromacs.org/current/reference-manual/file-formats.html#trajectory-files). But you are right that .trr files are larger than .xtc files. There are two reasons for this, one is that .trr files may contain velocities and forces (depending on your simulation settings) and they are also uncompressed. .xtc files, on the other hand, use a lossy compression with the precision set by `compressed-x-precision` (the default value 1000 means 0.001 nm).

There is no difference in output type depending on if you simulate membranes or proteins, but perhaps there are input files from different sources that use different settings.

The types of output is specified in the mdp file and documented at [Molecular dynamics parameters (.mdp options) - GROMACS 2024.4 documentation](https://manual.gromacs.org/current/user-guide/mdp-options.html#output-control). Unless you know that you need uncompressed coordinate output (or forces or velocities) I would recommend:

```auto
nstxout = 0
nstvout = 0
nstfout = 0

```

The interval of compressed coordinates can vary a lot from one use-case to another. I would argue that values in the range of 5,000-500,000 would cover most situations (also depending on the `dt` setting). But some people want to write with even fewer (than 5,000) steps in between.