# Gmx anaeig problematic error "Dimensions in the eigenvector files don't match" **URL:** \ **Category:** User discussions\ **Tags:** analysis-tools\ **Created:** [April 29, 2026, 10:48am UTC](https://gromacs.bioexcel.eu/t/gmx-anaeig-problematic-error-dimensions-in-the-eigenvector-files-dont-match/13427 "2026-04-29T10:48:26Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![degroot](https://avatars.discourse-cdn.com/v4/letter/d/f4b2a3/32.png) [@degroot](https://gromacs.bioexcel.eu/u/degroot)\ **Post date:** [April 29, 2026, 10:48am UTC](https://gromacs.bioexcel.eu/t/gmx-anaeig-problematic-error-dimensions-in-the-eigenvector-files-dont-match/13427/1 "2026-04-29T10:48:26Z")
GROMACS version: 2021 - 2026 GROMACS modification: No I’m attempting a scalar product comparison of 2 sets of eigenvectors with gmx anaeig -over: gmx anaeig -v eigenvec\_pls\_sf.trr -v2 eigenvec.trr -first 1 -last 1 -over and get the following output: trr version: GMX\_trn\_file (single precision) Read non mass weighted reference structure with 80 atoms from eigenvec\_pls\_sf.trr Read non mass weighted average/minimum structure with 80 atoms from eigenvec\_pls\_sf.trr Read 19 eigenvectors (for 80 atoms) Warning: number of eigenvectors 19 does not match three times the number of atoms 80 in eigenvec\_pls\_sf.trr. Using 19 eigenvectors. Read non mass weighted reference structure with 80 atoms from eigenvec.trr Read non mass weighted average/minimum structure with 80 atoms from eigenvec.trr Read 165 eigenvectors (for 80 atoms) * * * Program: gmx anaeig, version 2026.1-dev-20260306-f41e7faea4 Source file: src/gromacs/gmxana/gmx\_anaeig.cpp (line 1297) Fatal error: Dimensions in the eigenvector files don’t match I think this should not happen because the relevant dimensions (the number of atoms) do match. It’s just that due to limited trajectory length the number of eigenvectors stored by gmx covar are different. But that shouldn’t matter for the scalar product calculation. best Bert ---
**Author:** ![Carsten](https://avatars.discourse-cdn.com/v4/letter/c/f14d63/32.png) [@Carsten](https://gromacs.bioexcel.eu/u/Carsten)\ **Post date:** [June 3, 2026, 1:11pm UTC](https://gromacs.bioexcel.eu/t/gmx-anaeig-problematic-error-dimensions-in-the-eigenvector-files-dont-match/13427/2 "2026-06-03T13:11:10Z")
Resolved in 2026.3, see [Relaxed a too strict condition in gmx anaeig (!5972) · Merge requests · GROMACS / GROMACS · GitLab](https://gitlab.com/gromacs/gromacs/-/merge_requests/5972) ---
**Author:** ![degroot](https://avatars.discourse-cdn.com/v4/letter/d/f4b2a3/32.png) [@degroot](https://gromacs.bioexcel.eu/u/degroot)\ **Post date:** [June 3, 2026, 1:17pm UTC](https://gromacs.bioexcel.eu/t/gmx-anaeig-problematic-error-dimensions-in-the-eigenvector-files-dont-match/13427/3 "2026-06-03T13:17:04Z")
cool, thanks! ---
**Author:** ![hess](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/hess/32/416_2.png) [@hess](https://gromacs.bioexcel.eu/u/hess)\ **Post date:** [June 3, 2026, 1:43pm UTC](https://gromacs.bioexcel.eu/t/gmx-anaeig-problematic-error-dimensions-in-the-eigenvector-files-dont-match/13427/4 "2026-06-03T13:43:48Z")
I fix for that got merged yesterday and will appear in 2026.3. You can apply the change yourself to make your case work. > **[Relaxed a too strict condition in gmx anaeig (!5972) · Merge requests ·...](https://gitlab.com/gromacs/gromacs/-/merge_requests/5972)** > > As described in the community forum post https://gromacs.bioexcel.eu/t/gmx-anaeig-problematic-error-dimensions-in-the-eigenvector-files-dont-match/13427 when calculating the subspace overlap for two sets of eigenvectors... ---
**Author:** ![hess](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/hess/32/416_2.png) [@hess](https://gromacs.bioexcel.eu/u/hess)\ **Post date:** [June 3, 2026, 1:45pm UTC](https://gromacs.bioexcel.eu/t/gmx-anaeig-problematic-error-dimensions-in-the-eigenvector-files-dont-match/13427/5 "2026-06-03T13:45:05Z")
Ah, what happened here, I didn’t see there answer above until after I wrote mine. Maybe I had an outdated version of this thread in my tab.