[gmx-users] no CUDA-capable device detected

[SebastianWaltz]

Dear gromacs user,

I have some problems getting a simulation running on a tesla T10 with
CUDA driver version 3.20 running.
My .mdp file looks like:

;
title                    = ttt
cpp                    =  /lib/cpp
include               = -I../top
constraints         =  none
;define                =  -DPOSRES            ; for possition restraints
integrator          =  md-vv            ; leap-frog integrator
emtol               =  100.0             ; max-force to converge
emstep              =  0.005            ; initial step size
dt                  =  0.002            ; ps !
nsteps              =  500000              ; total 2 ps
nstcomm             =  5000            ; frequency for center of mass
motion removal

nstxout             =  500            ; frequency for writting the
trajectory
nstvout             =  500            ; frequency for writting the velocity
nstfout             =  500            ; frequency to write forces to
output trajectory
nstlog              =  5000            ; frequency to write the log file
nstenergy           =  5000
nstxtcout           =  500            ; frequency to write energies to
energy file

nstlist             =  5            ; Frequency to update the neighbor list
ns_type             =  grid            ; Make a grid in the box and only
check atoms in neighboring grid cells when constructing a new neighbor
rlist               =  1.2            ; cut-off distance for the
short-range neighbor list

coulombtype         =  PME            ; Fast Particle-Mesh Ewald
electrostatics
vdwtype            =  cut-off
rcoulomb            =  1.2            ; cut-off distance for the coulomb
field
rvdw                =  1.2            ; cut-off distance for the vdw field
fourierspacing      =  0.12            ; The maximum grid spacing for
the FFT grid
pme_order           =  4            ; Interpolation order for PME
optimize_fft        =  yes
pbc                =  xyz

Tcoupl              =  andersen
tc-grps             =  System
tau_t               =  0.1 
ref_t               =  300

energygrps          =  System

Pcoupl              =  no
tau_p               =  1
compressibility     =  4.1e-5
ref_p               =  1.0

gen_vel             =  yes
gen_temp            =  300
gen_seed            =  171113

I'm starting the simulation using the comand:

mdrun-gpu -device "OpenMM:Platform=CUDA,DeviceID=0,Memtest=15,Force-device=yes" -deffnm GPU_test.tpr

When I get the warning that OpenMM only works with the Andersen
thermostat and so on, and stops with:

Error getting CUDA device no CUDA-capable device detected.

Every help is very welcome

Sebastian