diff --git a/examples/cuda/README b/examples/cuda/README
index f6163c6876..15ac9c9926 100644
--- a/examples/cuda/README
+++ b/examples/cuda/README
@@ -1,15 +1,18 @@
-These example scripts can be run with the USER-CUDA
-package, assuming you built LAMMPS with the package.
+These example scripts can be run with the USER-CUDA package, assuming
+you built LAMMPS with the package and the precision you want.
+
+Note that these benchmark problems are identical to those in the
+examples/cuda directory which use the USER-CUDA package.
 
 You can run any of the scripts as follows.  You can also reset the
 x,y,z variables in the command line to change the size of the problem.
 
-With USER-CUDA on 1 GPU:
+With the USER-CUDA package on 1 GPU:
 
 lmp_machine -c on -sf cuda < in.cuda.melt.2.5
-lmp_machine -c on -sf cuda < in.cuda.phosphate
+lmp_machine -c on -sf cuda -v x 6 -v y 6 -v z 6 < in.cuda.phosphate
 
-With USER-CUDA on 2 GPUs:
+With the USER-CUDA package on 2 GPUs:
 
 mpirun -np 2 lmp_machine -c on -sf cuda -pk cuda 2 < in.cuda.melt.2.5
 mpirun -np 2 lmp_machine -c on -sf cuda -pk cuda 2 < in.cuda.rhodo
@@ -18,4 +21,7 @@ CPU-only:
 
 lmp_machine in.cuda.melt.2.5
 mpirun -np 4 lmp_machine < in.cuda.melt.5.0
-mpirun -np 8 lmp_machine < in.cuda.rhodo
+mpirun -np 8 lmp_machine -v x 1 -v y 1 -v z 2 < in.cuda.rhodo
+
+Note that with the USER-CUDA package you must insure the number of MPI
+tasks equals the number of GPUs (both per node).
diff --git a/examples/cuda/in.cuda.melt.2.5 b/examples/cuda/in.cuda.melt.2.5
index a6d1c6e9a0..65a5ad8099 100644
--- a/examples/cuda/in.cuda.melt.2.5
+++ b/examples/cuda/in.cuda.melt.2.5
@@ -27,5 +27,5 @@ neigh_modify	delay 0 every 20 check no
 
 fix		1 all nve
 
-thermo 100
-run		5000
+thermo 		100
+run		1000
diff --git a/examples/cuda/in.cuda.melt.5.0 b/examples/cuda/in.cuda.melt.5.0
index a36202c270..e22f38520d 100644
--- a/examples/cuda/in.cuda.melt.5.0
+++ b/examples/cuda/in.cuda.melt.5.0
@@ -27,5 +27,5 @@ neigh_modify	delay 0 every 20 check no
 
 fix		1 all nve
 
-thermo 100
-run		5000
+thermo 		100
+run		1000
diff --git a/examples/cuda/in.cuda.phosphate b/examples/cuda/in.cuda.phosphate
index 19356bdcc2..7621a39fc1 100644
--- a/examples/cuda/in.cuda.phosphate
+++ b/examples/cuda/in.cuda.phosphate
@@ -24,10 +24,8 @@ kspace_style    pppm 1e-5
 
 neighbor	2.0 bin
 
-thermo 100
-
-timestep 0.001
-
-fix 1 all npt temp 400 400 0.01 iso 1000.0 1000.0 1.0
-run 1000
+thermo          100
+timestep        0.001
 
+fix 		1 all npt temp 400 400 0.01 iso 1000.0 1000.0 1.0
+run 		200
diff --git a/examples/cuda/in.cuda.rhodo b/examples/cuda/in.cuda.rhodo
index c858959bb4..dd25f27224 100644
--- a/examples/cuda/in.cuda.rhodo
+++ b/examples/cuda/in.cuda.rhodo
@@ -23,10 +23,11 @@ replicate	$x $y $z
 fix             1 all shake 0.0001 5 0 m 1.0 a 232
 fix             2 all npt temp 300.0 300.0 100.0 &
 		z 0.0 0.0 1000.0 mtk no pchain 0 tchain 1
+
 special_bonds   charmm
  
 thermo          100
 thermo_style    multi 
 timestep        2.0
 
-run		1000
+run		200