diff --git a/examples/python/in.pair_python_coulomb b/examples/python/in.pair_python_coulomb
index 87a8fa69ce..5944592473 100644
--- a/examples/python/in.pair_python_coulomb
+++ b/examples/python/in.pair_python_coulomb
@@ -3,11 +3,11 @@ atom_style	full
 
 read_data	data.spce
 
-pair_style	hybrid/overlay python 12.0 coul/long 12.0
+pair_style	hybrid/overlay coul/long 12.0 python 12.0
 kspace_style	pppm 1.0e-6
 
 pair_coeff	* * coul/long
-pair_coeff	* * python potentials.LJCutSPCE OW NULL
+pair_coeff	* * python py_pot.LJCutSPCE OW NULL
 
 pair_modify     table 0
 
diff --git a/examples/python/in.pair_python_hybrid b/examples/python/in.pair_python_hybrid
index 289de18b00..d5db729bda 100644
--- a/examples/python/in.pair_python_hybrid
+++ b/examples/python/in.pair_python_hybrid
@@ -12,7 +12,7 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 neighbor	0.3 bin
@@ -31,7 +31,7 @@ clear
 read_restart    hybrid.restart
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 fix		1 all nve
@@ -47,7 +47,7 @@ atom_style	atomic
 read_data       hybrid.data
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 neighbor	0.3 bin
diff --git a/examples/python/in.pair_python_melt b/examples/python/in.pair_python_melt
index 7b3cbf7c4d..3f775e6651 100644
--- a/examples/python/in.pair_python_melt
+++ b/examples/python/in.pair_python_melt
@@ -12,7 +12,7 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 neighbor	0.3 bin
 neigh_modify	every 20 delay 0 check no
@@ -30,7 +30,7 @@ clear
 read_restart    melt.restart
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 fix		1 all nve
 
@@ -45,7 +45,7 @@ atom_style	atomic
 read_data       melt.data
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 neighbor	0.3 bin
 neigh_modify	every 20 delay 0 check no
diff --git a/examples/python/in.pair_python_spce b/examples/python/in.pair_python_spce
index d3765ebc33..5bd9e1e23a 100644
--- a/examples/python/in.pair_python_spce
+++ b/examples/python/in.pair_python_spce
@@ -3,11 +3,11 @@ atom_style	full
 
 read_data	data.spce
 
-pair_style	hybrid/overlay python 12.0 coul/long 12.0
+pair_style	hybrid/overlay coul/long 12.0 python 12.0
 kspace_style	pppm 1.0e-6
 
 pair_coeff	* * coul/long
-pair_coeff	* * python potentials.LJCutSPCE OW NULL
+pair_coeff	* * python py_pot.LJCutSPCE OW NULL
 
 bond_style	harmonic
 angle_style	harmonic
diff --git a/examples/python/in.pair_python_table b/examples/python/in.pair_python_table
index 87dad849cc..761a6efd78 100644
--- a/examples/python/in.pair_python_table
+++ b/examples/python/in.pair_python_table
@@ -12,12 +12,12 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 # generate tabulated potential from python variant
-pair_write      1 1 10000 rsq 0.01 2.5 lj_1_1.table LJ
+pair_write      1 1 2000 rsq 0.01 2.5 lj_1_1.table LJ
 
-pair_style      table linear 10000
+pair_style      table linear 2000
 pair_coeff      1 1 lj_1_1.table LJ
 
 neighbor	0.3 bin
diff --git a/examples/python/log.4May17.pair_python_coulomb.g++.1 b/examples/python/log.4May17.pair_python_coulomb.g++.1
new file mode 100644
index 0000000000..67bb3b3496
--- /dev/null
+++ b/examples/python/log.4May17.pair_python_coulomb.g++.1
@@ -0,0 +1,178 @@
+LAMMPS (4 May 2017)
+  using 1 OpenMP thread(s) per MPI task
+units		real
+atom_style	full
+
+read_data	data.spce
+  orthogonal box = (0.02645 0.02645 0.02641) to (35.5328 35.5328 35.4736)
+  1 by 1 by 1 MPI processor grid
+  reading atoms ...
+  4500 atoms
+  scanning bonds ...
+  2 = max bonds/atom
+  scanning angles ...
+  1 = max angles/atom
+  reading bonds ...
+  3000 bonds
+  reading angles ...
+  1500 angles
+  2 = max # of 1-2 neighbors
+  1 = max # of 1-3 neighbors
+  1 = max # of 1-4 neighbors
+  2 = max # of special neighbors
+
+pair_style	hybrid/overlay coul/long 12.0 python 12.0
+kspace_style	pppm 1.0e-6
+
+pair_coeff	* * coul/long
+pair_coeff	* * python py_pot.LJCutSPCE OW NULL
+
+pair_modify     table 0
+
+bond_style	harmonic
+angle_style	harmonic
+dihedral_style	none
+improper_style	none
+
+bond_coeff	1 1000.00 1.000
+angle_coeff	1 100.0 109.47
+
+special_bonds   lj/coul 0.0 0.0 1.0
+  2 = max # of 1-2 neighbors
+  1 = max # of 1-3 neighbors
+  2 = max # of special neighbors
+
+neighbor        2.0 bin
+
+fix		1 all shake 0.0001 20 0 b 1 a 1
+  0 = # of size 2 clusters
+  0 = # of size 3 clusters
+  0 = # of size 4 clusters
+  1500 = # of frozen angles
+fix		2 all nvt temp 300.0 300.0 100.0
+
+# create combined lj/coul table for all atom types
+# generate tabulated potential from python variant
+pair_write      1 1 2000 rsq 0.1 12 spce.table OW-OW -0.8472 -0.8472
+PPPM initialization ...
+WARNING: Using polynomial approximation for long-range coulomb (../kspace.cpp:321)
+  G vector (1/distance) = 0.279652
+  grid = 40 40 40
+  stencil order = 5
+  estimated absolute RMS force accuracy = 0.000394206
+  estimated relative force accuracy = 1.18714e-06
+  using double precision FFTs
+  3d grid and FFT values/proc = 103823 64000
+Neighbor list info ...
+  update every 1 steps, delay 10 steps, check yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 14
+  ghost atom cutoff = 14
+  binsize = 7, bins = 6 6 6
+  2 neighbor lists, perpetual/occasional/extra = 2 0 0
+  (1) pair coul/long, perpetual
+      attributes: half, newton on
+      pair build: half/bin/newton
+      stencil: half/bin/3d/newton
+      bin: standard
+  (2) pair python, perpetual, skip from (1)
+      attributes: half, newton on
+      pair build: skip
+      stencil: none
+      bin: none
+pair_write      1 2 2000 rsq 0.1 12 spce.table OW-HW -0.8472  0.4236
+PPPM initialization ...
+WARNING: Using polynomial approximation for long-range coulomb (../kspace.cpp:321)
+  G vector (1/distance) = 0.279652
+  grid = 40 40 40
+  stencil order = 5
+  estimated absolute RMS force accuracy = 0.000394206
+  estimated relative force accuracy = 1.18714e-06
+  using double precision FFTs
+  3d grid and FFT values/proc = 103823 64000
+pair_write      2 2 2000 rsq 0.1 12 spce.table HW-HW  0.4236  0.4236
+PPPM initialization ...
+WARNING: Using polynomial approximation for long-range coulomb (../kspace.cpp:321)
+  G vector (1/distance) = 0.279652
+  grid = 40 40 40
+  stencil order = 5
+  estimated absolute RMS force accuracy = 0.000394206
+  estimated relative force accuracy = 1.18714e-06
+  using double precision FFTs
+  3d grid and FFT values/proc = 103823 64000
+
+# switch to tabulated potential
+pair_style      table linear 2000 pppm
+pair_coeff      1 1 spce.table OW-OW
+pair_coeff      1 2 spce.table OW-HW
+pair_coeff      2 2 spce.table HW-HW
+
+thermo 10
+run 100
+PPPM initialization ...
+WARNING: Using 12-bit tables for long-range coulomb (../kspace.cpp:321)
+  G vector (1/distance) = 0.279652
+  grid = 40 40 40
+  stencil order = 5
+  estimated absolute RMS force accuracy = 0.000394674
+  estimated relative force accuracy = 1.18855e-06
+  using double precision FFTs
+  3d grid and FFT values/proc = 103823 64000
+Neighbor list info ...
+  update every 1 steps, delay 10 steps, check yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 14
+  ghost atom cutoff = 14
+  binsize = 7, bins = 6 6 6
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair table, perpetual
+      attributes: half, newton on
+      pair build: half/bin/newton
+      stencil: half/bin/3d/newton
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 35.26 | 35.26 | 35.26 Mbytes
+Step Temp E_pair E_mol TotEng Press 
+       0            0   -100272.97            0   -100272.97   -1282.0708 
+      10    120.61568   -101350.63            0   -100272.39   -4077.5051 
+      20    136.11379   -101465.43            0   -100248.65   -5136.5677 
+      30    137.01602    -101455.3            0   -100230.46   -5347.8311 
+      40      153.424   -101582.46            0   -100210.93   -5223.1676 
+      50    167.73654   -101686.24            0   -100186.77   -4468.6687 
+      60    163.11642   -101618.16            0   -100159.99   -3291.7815 
+      70    169.64512   -101647.89            0   -100131.35    -2611.638 
+      80     182.9979   -101737.01            0   -100101.11   -2390.6293 
+      90    191.33873   -101778.71            0   -100068.24    -2239.386 
+     100     194.7458   -101775.84            0   -100034.92   -1951.9128 
+Loop time of 7.63869 on 1 procs for 100 steps with 4500 atoms
+
+Performance: 1.131 ns/day, 21.219 hours/ns, 13.091 timesteps/s
+99.6% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 5.7777     | 5.7777     | 5.7777     |   0.0 | 75.64
+Bond    | 0.00017595 | 0.00017595 | 0.00017595 |   0.0 |  0.00
+Kspace  | 1.5385     | 1.5385     | 1.5385     |   0.0 | 20.14
+Neigh   | 0.22962    | 0.22962    | 0.22962    |   0.0 |  3.01
+Comm    | 0.024123   | 0.024123   | 0.024123   |   0.0 |  0.32
+Output  | 0.00061131 | 0.00061131 | 0.00061131 |   0.0 |  0.01
+Modify  | 0.062444   | 0.062444   | 0.062444   |   0.0 |  0.82
+Other   |            | 0.005466   |            |       |  0.07
+
+Nlocal:    4500 ave 4500 max 4500 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:    21216 ave 21216 max 21216 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:    2.60177e+06 ave 2.60177e+06 max 2.60177e+06 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 2601766
+Ave neighs/atom = 578.17
+Ave special neighs/atom = 2
+Neighbor list builds = 3
+Dangerous builds = 0
+
+shell rm spce.table
+
+Total wall time: 0:00:07
diff --git a/examples/python/log.4May17.pair_python_coulomb.4 b/examples/python/log.4May17.pair_python_coulomb.g++.4
similarity index 86%
rename from examples/python/log.4May17.pair_python_coulomb.4
rename to examples/python/log.4May17.pair_python_coulomb.g++.4
index a3b13cfbaf..d5fd15ccd3 100644
--- a/examples/python/log.4May17.pair_python_coulomb.4
+++ b/examples/python/log.4May17.pair_python_coulomb.g++.4
@@ -21,11 +21,11 @@ read_data	data.spce
   1 = max # of 1-4 neighbors
   2 = max # of special neighbors
 
-pair_style	hybrid/overlay python 12.0 coul/long 12.0
+pair_style	hybrid/overlay coul/long 12.0 python 12.0
 kspace_style	pppm 1.0e-6
 
 pair_coeff	* * coul/long
-pair_coeff	* * python potentials.LJCutSPCE OW NULL
+pair_coeff	* * python py_pot.LJCutSPCE OW NULL
 
 pair_modify     table 0
 
@@ -70,16 +70,16 @@ Neighbor list info ...
   ghost atom cutoff = 14
   binsize = 7, bins = 6 6 6
   2 neighbor lists, perpetual/occasional/extra = 2 0 0
-  (1) pair python, perpetual, skip from (2)
-      attributes: half, newton on
-      pair build: skip
-      stencil: none
-      bin: none
-  (2) pair coul/long, perpetual
+  (1) pair coul/long, perpetual
       attributes: half, newton on
       pair build: half/bin/newton
       stencil: half/bin/3d/newton
       bin: standard
+  (2) pair python, perpetual, skip from (1)
+      attributes: half, newton on
+      pair build: skip
+      stencil: none
+      bin: none
 pair_write      1 2 2000 rsq 0.1 12 spce.table OW-HW -0.8472  0.4236
 PPPM initialization ...
 WARNING: Using polynomial approximation for long-range coulomb (../kspace.cpp:321)
@@ -143,22 +143,22 @@ Step Temp E_pair E_mol TotEng Press
       80     182.9979   -101737.01            0   -100101.11   -2390.6293 
       90    191.33873   -101778.71            0   -100068.24    -2239.386 
      100     194.7458   -101775.84            0   -100034.92   -1951.9128 
-Loop time of 2.38392 on 4 procs for 100 steps with 4500 atoms
+Loop time of 2.35848 on 4 procs for 100 steps with 4500 atoms
 
-Performance: 3.624 ns/day, 6.622 hours/ns, 41.948 timesteps/s
-99.1% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 3.663 ns/day, 6.551 hours/ns, 42.400 timesteps/s
+99.0% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 1.4377     | 1.5465     | 1.6848     |   7.3 | 64.87
-Bond    | 0.00010276 | 0.00012648 | 0.0001452  |   0.0 |  0.01
-Kspace  | 0.53311    | 0.66842    | 0.77484    |  10.9 | 28.04
-Neigh   | 0.066      | 0.066074   | 0.066101   |   0.0 |  2.77
-Comm    | 0.045355   | 0.048344   | 0.050747   |   1.0 |  2.03
-Output  | 0.00042391 | 0.00044996 | 0.00052667 |   0.0 |  0.02
-Modify  | 0.049891   | 0.050191   | 0.050336   |   0.1 |  2.11
-Other   |            | 0.003771   |            |       |  0.16
+Pair    | 1.5061     | 1.5612     | 1.6879     |   5.9 | 66.20
+Bond    | 9.5129e-05 | 0.00012672 | 0.00014567 |   0.0 |  0.01
+Kspace  | 0.52033    | 0.64456    | 0.69933    |   9.1 | 27.33
+Neigh   | 0.066265   | 0.066342   | 0.06644    |   0.0 |  2.81
+Comm    | 0.03394    | 0.036139   | 0.038043   |   0.8 |  1.53
+Output  | 0.00040889 | 0.00044978 | 0.00056887 |   0.0 |  0.02
+Modify  | 0.04557    | 0.045813   | 0.046082   |   0.1 |  1.94
+Other   |            | 0.003826   |            |       |  0.16
 
 Nlocal:    1125 ave 1154 max 1092 min
 Histogram: 1 0 0 0 1 0 0 1 0 1
diff --git a/examples/python/log.4May17.pair_python_hybrid.g++.1 b/examples/python/log.4May17.pair_python_hybrid.g++.1
index d9d3bf53c0..497943d4fd 100644
--- a/examples/python/log.4May17.pair_python_hybrid.g++.1
+++ b/examples/python/log.4May17.pair_python_hybrid.g++.1
@@ -1,5 +1,4 @@
 LAMMPS (4 May 2017)
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 # 3d Lennard-Jones hybrid
 
@@ -19,7 +18,7 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 neighbor	0.3 bin
@@ -59,20 +58,20 @@ Step Temp E_pair E_mol TotEng Press
      150    1.6324555   -4.7286791            0    -2.280608    5.9589514 
      200    1.6630725   -4.7750988            0   -2.2811136    5.7364886 
      250    1.6275257   -4.7224992            0    -2.281821    5.9567365 
-Loop time of 43.2436 on 1 procs for 250 steps with 4000 atoms
+Loop time of 21.9463 on 1 procs for 250 steps with 4000 atoms
 
-Performance: 2497.477 tau/day, 5.781 timesteps/s
-31.7% CPU use with 1 MPI tasks x 1 OpenMP threads
+Performance: 4921.114 tau/day, 11.391 timesteps/s
+98.6% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 42.933     | 42.933     | 42.933     |   0.0 | 99.28
-Neigh   | 0.24816    | 0.24816    | 0.24816    |   0.0 |  0.57
-Comm    | 0.027748   | 0.027748   | 0.027748   |   0.0 |  0.06
-Output  | 0.000519   | 0.000519   | 0.000519   |   0.0 |  0.00
-Modify  | 0.028028   | 0.028028   | 0.028028   |   0.0 |  0.06
-Other   |            | 0.005912   |            |       |  0.01
+Pair    | 21.819     | 21.819     | 21.819     |   0.0 | 99.42
+Neigh   | 0.094718   | 0.094718   | 0.094718   |   0.0 |  0.43
+Comm    | 0.01407    | 0.01407    | 0.01407    |   0.0 |  0.06
+Output  | 0.00024915 | 0.00024915 | 0.00024915 |   0.0 |  0.00
+Modify  | 0.015002   | 0.015002   | 0.015002   |   0.0 |  0.07
+Other   |            | 0.003232   |            |       |  0.01
 
 Nlocal:    4000 ave 4000 max 4000 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
@@ -90,7 +89,6 @@ write_data      hybrid.data
 write_restart   hybrid.restart
 
 clear
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 
 read_restart    hybrid.restart
@@ -99,7 +97,7 @@ read_restart    hybrid.restart
   4000 atoms
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 fix		1 all nve
@@ -136,20 +134,20 @@ Step Temp E_pair E_mol TotEng Press
      400    1.6540555   -4.7622999            0    -2.281837    5.8200413 
      450    1.6264734   -4.7200865            0   -2.2809863    5.9546991 
      500    1.6366891   -4.7350979            0   -2.2806781    5.9369284 
-Loop time of 46.2882 on 1 procs for 250 steps with 4000 atoms
+Loop time of 21.9098 on 1 procs for 250 steps with 4000 atoms
 
-Performance: 2333.206 tau/day, 5.401 timesteps/s
-31.7% CPU use with 1 MPI tasks x 1 OpenMP threads
+Performance: 4929.303 tau/day, 11.410 timesteps/s
+98.8% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 45.662     | 45.662     | 45.662     |   0.0 | 98.65
-Neigh   | 0.55234    | 0.55234    | 0.55234    |   0.0 |  1.19
-Comm    | 0.035614   | 0.035614   | 0.035614   |   0.0 |  0.08
-Output  | 0.000544   | 0.000544   | 0.000544   |   0.0 |  0.00
-Modify  | 0.029269   | 0.029269   | 0.029269   |   0.0 |  0.06
-Other   |            | 0.008735   |            |       |  0.02
+Pair    | 21.68      | 21.68      | 21.68      |   0.0 | 98.95
+Neigh   | 0.19625    | 0.19625    | 0.19625    |   0.0 |  0.90
+Comm    | 0.014877   | 0.014877   | 0.014877   |   0.0 |  0.07
+Output  | 0.00027227 | 0.00027227 | 0.00027227 |   0.0 |  0.00
+Modify  | 0.013663   | 0.013663   | 0.013663   |   0.0 |  0.06
+Other   |            | 0.004371   |            |       |  0.02
 
 Nlocal:    4000 ave 4000 max 4000 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
@@ -164,7 +162,6 @@ Neighbor list builds = 25
 Dangerous builds = 25
 
 clear
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 
 units		lj
@@ -179,7 +176,7 @@ read_data       hybrid.data
   4000 velocities
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 neighbor	0.3 bin
@@ -219,20 +216,20 @@ Step Temp E_pair E_mol TotEng Press
      150    1.6537193   -4.7627023            0   -2.2827434    5.8177704 
      200    1.6258731   -4.7205017            0   -2.2823017     5.952511 
      250    1.6370862   -4.7373176            0   -2.2823022     5.925807 
-Loop time of 46.4094 on 1 procs for 250 steps with 4000 atoms
+Loop time of 22.091 on 1 procs for 250 steps with 4000 atoms
 
-Performance: 2327.115 tau/day, 5.387 timesteps/s
-31.6% CPU use with 1 MPI tasks x 1 OpenMP threads
+Performance: 4888.868 tau/day, 11.317 timesteps/s
+98.8% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 46.066     | 46.066     | 46.066     |   0.0 | 99.26
-Neigh   | 0.27099    | 0.27099    | 0.27099    |   0.0 |  0.58
-Comm    | 0.033778   | 0.033778   | 0.033778   |   0.0 |  0.07
-Output  | 0.000507   | 0.000507   | 0.000507   |   0.0 |  0.00
-Modify  | 0.030938   | 0.030938   | 0.030938   |   0.0 |  0.07
-Other   |            | 0.006695   |            |       |  0.01
+Pair    | 21.966     | 21.966     | 21.966     |   0.0 | 99.43
+Neigh   | 0.094647   | 0.094647   | 0.094647   |   0.0 |  0.43
+Comm    | 0.013071   | 0.013071   | 0.013071   |   0.0 |  0.06
+Output  | 0.00027871 | 0.00027871 | 0.00027871 |   0.0 |  0.00
+Modify  | 0.013882   | 0.013882   | 0.013882   |   0.0 |  0.06
+Other   |            | 0.003102   |            |       |  0.01
 
 Nlocal:    4000 ave 4000 max 4000 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
@@ -247,4 +244,4 @@ Neighbor list builds = 12
 Dangerous builds not checked
 
 shell rm hybrid.data hybrid.restart
-Total wall time: 0:02:20
+Total wall time: 0:01:07
diff --git a/examples/python/log.4May17.pair_python_hybrid.g++.4 b/examples/python/log.4May17.pair_python_hybrid.g++.4
index e45514a803..0331f34a5e 100644
--- a/examples/python/log.4May17.pair_python_hybrid.g++.4
+++ b/examples/python/log.4May17.pair_python_hybrid.g++.4
@@ -1,5 +1,4 @@
 LAMMPS (4 May 2017)
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 # 3d Lennard-Jones hybrid
 
@@ -19,7 +18,7 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 neighbor	0.3 bin
@@ -59,20 +58,20 @@ Step Temp E_pair E_mol TotEng Press
      150    1.6596605   -4.7699432            0   -2.2810749    5.7830138 
      200    1.6371874   -4.7365462            0   -2.2813789    5.9246674 
      250    1.6323462   -4.7292021            0   -2.2812949    5.9762238 
-Loop time of 11.1422 on 4 procs for 250 steps with 4000 atoms
+Loop time of 6.01723 on 4 procs for 250 steps with 4000 atoms
 
-Performance: 9692.888 tau/day, 22.437 timesteps/s
-35.1% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 17948.472 tau/day, 41.547 timesteps/s
+98.2% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 10.448     | 10.772     | 10.937     |   5.9 | 96.67
-Neigh   | 0.062061   | 0.062949   | 0.06439    |   0.4 |  0.56
-Comm    | 0.12929    | 0.29444    | 0.61802    |  35.8 |  2.64
-Output  | 0.000301   | 0.000684   | 0.001824   |   0.0 |  0.01
-Modify  | 0.009803   | 0.0098622  | 0.010014   |   0.1 |  0.09
-Other   |            | 0.002618   |            |       |  0.02
+Pair    | 5.1507     | 5.4989     | 5.9629     |  13.1 | 91.39
+Neigh   | 0.024123   | 0.024877   | 0.025959   |   0.5 |  0.41
+Comm    | 0.02258    | 0.48785    | 0.83691    |  44.1 |  8.11
+Output  | 0.00039768 | 0.00045246 | 0.00052929 |   0.0 |  0.01
+Modify  | 0.0036325  | 0.0037773  | 0.0038905  |   0.2 |  0.06
+Other   |            | 0.001357   |            |       |  0.02
 
 Nlocal:    1000 ave 1010 max 982 min
 Histogram: 1 0 0 0 0 0 1 0 0 2
@@ -90,7 +89,6 @@ write_data      hybrid.data
 write_restart   hybrid.restart
 
 clear
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 
 read_restart    hybrid.restart
@@ -99,7 +97,7 @@ read_restart    hybrid.restart
   4000 atoms
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 fix		1 all nve
@@ -136,20 +134,20 @@ Step Temp E_pair E_mol TotEng Press
      400    1.6388136   -4.7387093            0   -2.2811035    5.9331084 
      450    1.6431295   -4.7452215            0   -2.2811435    5.8929898 
      500     1.643316   -4.7454222            0   -2.2810644    5.8454817 
-Loop time of 11.287 on 4 procs for 250 steps with 4000 atoms
+Loop time of 6.09991 on 4 procs for 250 steps with 4000 atoms
 
-Performance: 9568.520 tau/day, 22.149 timesteps/s
-34.9% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 17705.179 tau/day, 40.984 timesteps/s
+98.2% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 10.274     | 10.76      | 11.02      |   8.8 | 95.33
-Neigh   | 0.12639    | 0.1291     | 0.13056    |   0.5 |  1.14
-Comm    | 0.12094    | 0.38226    | 0.87078    |  46.7 |  3.39
-Output  | 0.000297   | 0.0006965  | 0.001867   |   0.0 |  0.01
-Modify  | 0.010445   | 0.010638   | 0.011054   |   0.2 |  0.09
-Other   |            | 0.003901   |            |       |  0.03
+Pair    | 5.2315     | 5.5179     | 6.0183     |  13.7 | 90.46
+Neigh   | 0.049134   | 0.051424   | 0.053837   |   0.8 |  0.84
+Comm    | 0.021671   | 0.52455    | 0.8132     |  44.5 |  8.60
+Output  | 0.00019336 | 0.00026017 | 0.00032115 |   0.0 |  0.00
+Modify  | 0.0036032  | 0.0036635  | 0.0038021  |   0.1 |  0.06
+Other   |            | 0.002068   |            |       |  0.03
 
 Nlocal:    1000 ave 1012 max 983 min
 Histogram: 1 0 0 0 0 0 2 0 0 1
@@ -164,7 +162,6 @@ Neighbor list builds = 25
 Dangerous builds = 25
 
 clear
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 
 units		lj
@@ -179,7 +176,7 @@ read_data       hybrid.data
   4000 velocities
 
 pair_style	hybrid lj/cut 2.5 python 2.5
-pair_coeff	* * python potentials.LJCutMelt lj NULL
+pair_coeff	* * python py_pot.LJCutMelt lj NULL
 pair_coeff      * 2 lj/cut 1.0 1.0
 
 neighbor	0.3 bin
@@ -219,20 +216,20 @@ Step Temp E_pair E_mol TotEng Press
      150    1.6384234   -4.7389689            0   -2.2819482    5.9315273 
      200    1.6428814   -4.7460743            0   -2.2823683    5.8888228 
      250    1.6432631   -4.7466603            0   -2.2823818    5.8398819 
-Loop time of 11.1573 on 4 procs for 250 steps with 4000 atoms
+Loop time of 6.04476 on 4 procs for 250 steps with 4000 atoms
 
-Performance: 9679.760 tau/day, 22.407 timesteps/s
-35.0% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 17866.705 tau/day, 41.358 timesteps/s
+98.3% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 10.166     | 10.713     | 10.932     |   9.7 | 96.01
-Neigh   | 0.060687   | 0.062175   | 0.063163   |   0.4 |  0.56
-Comm    | 0.14931    | 0.36938    | 0.91686    |  52.5 |  3.31
-Output  | 0.00036    | 0.00058175 | 0.001228   |   0.0 |  0.01
-Modify  | 0.009918   | 0.010237   | 0.010388   |   0.2 |  0.09
-Other   |            | 0.002356   |            |       |  0.02
+Pair    | 5.2589     | 5.5841     | 5.9788     |  11.1 | 92.38
+Neigh   | 0.023942   | 0.024705   | 0.025509   |   0.4 |  0.41
+Comm    | 0.034946   | 0.43056    | 0.75671    |  40.0 |  7.12
+Output  | 0.00022149 | 0.00029725 | 0.0003593  |   0.0 |  0.00
+Modify  | 0.003613   | 0.0037647  | 0.003829   |   0.1 |  0.06
+Other   |            | 0.001313   |            |       |  0.02
 
 Nlocal:    1000 ave 1013 max 989 min
 Histogram: 1 0 0 1 0 1 0 0 0 1
@@ -247,4 +244,4 @@ Neighbor list builds = 12
 Dangerous builds not checked
 
 shell rm hybrid.data hybrid.restart
-Total wall time: 0:00:35
+Total wall time: 0:00:18
diff --git a/examples/python/log.4May17.pair_python_melt.g++.1 b/examples/python/log.4May17.pair_python_melt.g++.1
index 3459dd4f87..fcb902fb3c 100644
--- a/examples/python/log.4May17.pair_python_melt.g++.1
+++ b/examples/python/log.4May17.pair_python_melt.g++.1
@@ -1,5 +1,4 @@
 LAMMPS (4 May 2017)
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 # 3d Lennard-Jones melt
 
@@ -19,7 +18,7 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 neighbor	0.3 bin
 neigh_modify	every 20 delay 0 check no
@@ -48,20 +47,20 @@ Step Temp E_pair E_mol TotEng Press
      150    1.6324555   -4.7286791            0    -2.280608    5.9589514 
      200    1.6630725   -4.7750988            0   -2.2811136    5.7364886 
      250    1.6275257   -4.7224992            0    -2.281821    5.9567365 
-Loop time of 24.2466 on 1 procs for 250 steps with 4000 atoms
+Loop time of 21.6481 on 1 procs for 250 steps with 4000 atoms
 
-Performance: 4454.233 tau/day, 10.311 timesteps/s
-59.9% CPU use with 1 MPI tasks x 1 OpenMP threads
+Performance: 4988.899 tau/day, 11.548 timesteps/s
+98.5% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 24.079     | 24.079     | 24.079     |   0.0 | 99.31
-Neigh   | 0.13174    | 0.13174    | 0.13174    |   0.0 |  0.54
-Comm    | 0.016789   | 0.016789   | 0.016789   |   0.0 |  0.07
-Output  | 0.000271   | 0.000271   | 0.000271   |   0.0 |  0.00
-Modify  | 0.015073   | 0.015073   | 0.015073   |   0.0 |  0.06
-Other   |            | 0.003428   |            |       |  0.01
+Pair    | 21.529     | 21.529     | 21.529     |   0.0 | 99.45
+Neigh   | 0.08819    | 0.08819    | 0.08819    |   0.0 |  0.41
+Comm    | 0.013276   | 0.013276   | 0.013276   |   0.0 |  0.06
+Output  | 0.00025654 | 0.00025654 | 0.00025654 |   0.0 |  0.00
+Modify  | 0.014466   | 0.014466   | 0.014466   |   0.0 |  0.07
+Other   |            | 0.003143   |            |       |  0.01
 
 Nlocal:    4000 ave 4000 max 4000 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
@@ -79,7 +78,6 @@ write_data      melt.data
 write_restart   melt.restart
 
 clear
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 
 read_restart    melt.restart
@@ -88,7 +86,7 @@ read_restart    melt.restart
   4000 atoms
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 fix		1 all nve
 
@@ -114,20 +112,20 @@ Step Temp E_pair E_mol TotEng Press
      400    1.6540555   -4.7622999            0    -2.281837    5.8200413 
      450    1.6264734   -4.7200865            0   -2.2809863    5.9546991 
      500    1.6366891   -4.7350979            0   -2.2806781    5.9369284 
-Loop time of 24.3239 on 1 procs for 250 steps with 4000 atoms
+Loop time of 21.9592 on 1 procs for 250 steps with 4000 atoms
 
-Performance: 4440.069 tau/day, 10.278 timesteps/s
-60.0% CPU use with 1 MPI tasks x 1 OpenMP threads
+Performance: 4918.203 tau/day, 11.385 timesteps/s
+98.5% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 24.017     | 24.017     | 24.017     |   0.0 | 98.74
-Neigh   | 0.26927    | 0.26927    | 0.26927    |   0.0 |  1.11
-Comm    | 0.018113   | 0.018113   | 0.018113   |   0.0 |  0.07
-Output  | 0.000254   | 0.000254   | 0.000254   |   0.0 |  0.00
-Modify  | 0.015259   | 0.015259   | 0.015259   |   0.0 |  0.06
-Other   |            | 0.004524   |            |       |  0.02
+Pair    | 21.74      | 21.74      | 21.74      |   0.0 | 99.00
+Neigh   | 0.18588    | 0.18588    | 0.18588    |   0.0 |  0.85
+Comm    | 0.01476    | 0.01476    | 0.01476    |   0.0 |  0.07
+Output  | 0.00022244 | 0.00022244 | 0.00022244 |   0.0 |  0.00
+Modify  | 0.01356    | 0.01356    | 0.01356    |   0.0 |  0.06
+Other   |            | 0.004382   |            |       |  0.02
 
 Nlocal:    4000 ave 4000 max 4000 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
@@ -142,7 +140,6 @@ Neighbor list builds = 25
 Dangerous builds = 25
 
 clear
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 
 units		lj
@@ -157,7 +154,7 @@ read_data       melt.data
   4000 velocities
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 neighbor	0.3 bin
 neigh_modify	every 20 delay 0 check no
@@ -186,20 +183,20 @@ Step Temp E_pair E_mol TotEng Press
      150    1.6537193   -4.7627023            0   -2.2827434    5.8177704 
      200    1.6258731   -4.7205017            0   -2.2823017     5.952511 
      250    1.6370862   -4.7373176            0   -2.2823022     5.925807 
-Loop time of 22.9051 on 1 procs for 250 steps with 4000 atoms
+Loop time of 21.8255 on 1 procs for 250 steps with 4000 atoms
 
-Performance: 4715.116 tau/day, 10.915 timesteps/s
-60.1% CPU use with 1 MPI tasks x 1 OpenMP threads
+Performance: 4948.331 tau/day, 11.454 timesteps/s
+98.6% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 22.752     | 22.752     | 22.752     |   0.0 | 99.33
-Neigh   | 0.12254    | 0.12254    | 0.12254    |   0.0 |  0.53
-Comm    | 0.013385   | 0.013385   | 0.013385   |   0.0 |  0.06
-Output  | 0.000254   | 0.000254   | 0.000254   |   0.0 |  0.00
-Modify  | 0.014159   | 0.014159   | 0.014159   |   0.0 |  0.06
-Other   |            | 0.002851   |            |       |  0.01
+Pair    | 21.707     | 21.707     | 21.707     |   0.0 | 99.46
+Neigh   | 0.088455   | 0.088455   | 0.088455   |   0.0 |  0.41
+Comm    | 0.01311    | 0.01311    | 0.01311    |   0.0 |  0.06
+Output  | 0.00025082 | 0.00025082 | 0.00025082 |   0.0 |  0.00
+Modify  | 0.013836   | 0.013836   | 0.013836   |   0.0 |  0.06
+Other   |            | 0.003096   |            |       |  0.01
 
 Nlocal:    4000 ave 4000 max 4000 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
@@ -214,4 +211,4 @@ Neighbor list builds = 12
 Dangerous builds not checked
 
 shell rm melt.data melt.restart
-Total wall time: 0:01:13
+Total wall time: 0:01:07
diff --git a/examples/python/log.4May17.pair_python_melt.g++.4 b/examples/python/log.4May17.pair_python_melt.g++.4
index 7e4ba25acf..d504efc4ac 100644
--- a/examples/python/log.4May17.pair_python_melt.g++.4
+++ b/examples/python/log.4May17.pair_python_melt.g++.4
@@ -1,5 +1,4 @@
 LAMMPS (4 May 2017)
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 # 3d Lennard-Jones melt
 
@@ -19,7 +18,7 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 neighbor	0.3 bin
 neigh_modify	every 20 delay 0 check no
@@ -48,20 +47,20 @@ Step Temp E_pair E_mol TotEng Press
      150    1.6596605   -4.7699432            0   -2.2810749    5.7830138 
      200    1.6371874   -4.7365462            0   -2.2813789    5.9246674 
      250    1.6323462   -4.7292021            0   -2.2812949    5.9762238 
-Loop time of 12.7083 on 4 procs for 250 steps with 4000 atoms
+Loop time of 5.83903 on 4 procs for 250 steps with 4000 atoms
 
-Performance: 8498.384 tau/day, 19.672 timesteps/s
-31.5% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 18496.226 tau/day, 42.815 timesteps/s
+98.2% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 11.491     | 11.96      | 12.464     |  10.0 | 94.11
-Neigh   | 0.065058   | 0.065956   | 0.067066   |   0.3 |  0.52
-Comm    | 0.16288    | 0.66706    | 1.1373     |  42.2 |  5.25
-Output  | 0.000416   | 0.00085025 | 0.002121   |   0.0 |  0.01
-Modify  | 0.010849   | 0.011123   | 0.011321   |   0.2 |  0.09
-Other   |            | 0.003005   |            |       |  0.02
+Pair    | 5.152      | 5.5209     | 5.7679     |   9.6 | 94.55
+Neigh   | 0.022809   | 0.023364   | 0.023891   |   0.3 |  0.40
+Comm    | 0.041927   | 0.28952    | 0.65893    |  42.2 |  4.96
+Output  | 0.0002389  | 0.00024772 | 0.00026727 |   0.0 |  0.00
+Modify  | 0.0036368  | 0.0036796  | 0.0037563  |   0.1 |  0.06
+Other   |            | 0.001328   |            |       |  0.02
 
 Nlocal:    1000 ave 1010 max 982 min
 Histogram: 1 0 0 0 0 0 1 0 0 2
@@ -79,7 +78,6 @@ write_data      melt.data
 write_restart   melt.restart
 
 clear
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 
 read_restart    melt.restart
@@ -88,7 +86,7 @@ read_restart    melt.restart
   4000 atoms
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 fix		1 all nve
 
@@ -114,20 +112,20 @@ Step Temp E_pair E_mol TotEng Press
      400    1.6388136   -4.7387093            0   -2.2811035    5.9331084 
      450    1.6431295   -4.7452215            0   -2.2811435    5.8929898 
      500     1.643316   -4.7454222            0   -2.2810644    5.8454817 
-Loop time of 12.6852 on 4 procs for 250 steps with 4000 atoms
+Loop time of 5.85683 on 4 procs for 250 steps with 4000 atoms
 
-Performance: 8513.855 tau/day, 19.708 timesteps/s
-31.6% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 18440.001 tau/day, 42.685 timesteps/s
+98.4% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 11.653     | 11.92      | 12.472     |   9.6 | 93.97
-Neigh   | 0.13284    | 0.13556    | 0.13729    |   0.5 |  1.07
-Comm    | 0.051389   | 0.60884    | 0.88175    |  43.0 |  4.80
-Output  | 0.000362   | 0.0046985  | 0.008143   |   5.1 |  0.04
-Modify  | 0.011007   | 0.011344   | 0.011857   |   0.3 |  0.09
-Other   |            | 0.004278   |            |       |  0.03
+Pair    | 5.2483     | 5.5095     | 5.7744     |   8.0 | 94.07
+Neigh   | 0.047228   | 0.047998   | 0.049293   |   0.4 |  0.82
+Comm    | 0.027134   | 0.29341    | 0.55554    |  34.6 |  5.01
+Output  | 0.00020003 | 0.00021219 | 0.0002358  |   0.0 |  0.00
+Modify  | 0.0035472  | 0.0036988  | 0.0038681  |   0.2 |  0.06
+Other   |            | 0.001984   |            |       |  0.03
 
 Nlocal:    1000 ave 1012 max 983 min
 Histogram: 1 0 0 0 0 0 2 0 0 1
@@ -142,7 +140,6 @@ Neighbor list builds = 25
 Dangerous builds = 25
 
 clear
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 
 units		lj
@@ -157,7 +154,7 @@ read_data       melt.data
   4000 velocities
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 neighbor	0.3 bin
 neigh_modify	every 20 delay 0 check no
@@ -186,20 +183,20 @@ Step Temp E_pair E_mol TotEng Press
      150    1.6384234   -4.7389689            0   -2.2819482    5.9315273 
      200    1.6428814   -4.7460743            0   -2.2823683    5.8888228 
      250    1.6432631   -4.7466603            0   -2.2823818    5.8398819 
-Loop time of 12.5324 on 4 procs for 250 steps with 4000 atoms
+Loop time of 5.86684 on 4 procs for 250 steps with 4000 atoms
 
-Performance: 8617.631 tau/day, 19.948 timesteps/s
-31.6% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 18408.545 tau/day, 42.612 timesteps/s
+98.4% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 11.648     | 11.918     | 12.387     |   8.3 | 95.10
-Neigh   | 0.064038   | 0.06537    | 0.065914   |   0.3 |  0.52
-Comm    | 0.065189   | 0.53362    | 0.80384    |  39.4 |  4.26
-Output  | 0.000346   | 0.0007525  | 0.001938   |   0.0 |  0.01
-Modify  | 0.011255   | 0.01155    | 0.011852   |   0.2 |  0.09
-Other   |            | 0.002751   |            |       |  0.02
+Pair    | 5.3207     | 5.5695     | 5.8071     |   7.6 | 94.93
+Neigh   | 0.023073   | 0.023405   | 0.023834   |   0.2 |  0.40
+Comm    | 0.030558   | 0.2686     | 0.51789    |  34.7 |  4.58
+Output  | 0.00028825 | 0.00036758 | 0.00042987 |   0.0 |  0.01
+Modify  | 0.0034878  | 0.0036733  | 0.0039375  |   0.3 |  0.06
+Other   |            | 0.001259   |            |       |  0.02
 
 Nlocal:    1000 ave 1013 max 989 min
 Histogram: 1 0 0 1 0 1 0 0 0 1
@@ -214,4 +211,4 @@ Neighbor list builds = 12
 Dangerous builds not checked
 
 shell rm melt.data melt.restart
-Total wall time: 0:00:39
+Total wall time: 0:00:18
diff --git a/examples/python/log.4May17.pair_python_spce.g++.1 b/examples/python/log.4May17.pair_python_spce.g++.1
index 4e429b9cdd..1a535df0e9 100644
--- a/examples/python/log.4May17.pair_python_spce.g++.1
+++ b/examples/python/log.4May17.pair_python_spce.g++.1
@@ -1,5 +1,4 @@
 LAMMPS (4 May 2017)
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 units		real
 atom_style	full
@@ -22,11 +21,11 @@ read_data	data.spce
   1 = max # of 1-4 neighbors
   2 = max # of special neighbors
 
-pair_style	hybrid/overlay python 12.0 coul/long 12.0
+pair_style	hybrid/overlay coul/long 12.0 python 12.0
 kspace_style	pppm 1.0e-6
 
 pair_coeff	* * coul/long
-pair_coeff	* * python potentials.LJCutSPCE OW NULL
+pair_coeff	* * python py_pot.LJCutSPCE OW NULL
 
 bond_style	harmonic
 angle_style	harmonic
@@ -68,16 +67,16 @@ Neighbor list info ...
   ghost atom cutoff = 14
   binsize = 7, bins = 6 6 6
   2 neighbor lists, perpetual/occasional/extra = 2 0 0
-  (1) pair python, perpetual, skip from (2)
-      attributes: half, newton on
-      pair build: skip
-      stencil: none
-      bin: none
-  (2) pair coul/long, perpetual
+  (1) pair coul/long, perpetual
       attributes: half, newton on
       pair build: half/bin/newton
       stencil: half/bin/3d/newton
       bin: standard
+  (2) pair python, perpetual, skip from (1)
+      attributes: half, newton on
+      pair build: skip
+      stencil: none
+      bin: none
 Per MPI rank memory allocation (min/avg/max) = 41.05 | 41.05 | 41.05 Mbytes
 Step Temp E_pair E_mol TotEng Press 
        0            0   -16692.369            0   -16692.369    -1289.222 
@@ -91,33 +90,33 @@ Step Temp E_pair E_mol TotEng Press
       80    182.94811   -18155.978            0   -16520.523   -2393.3156 
       90    191.29902   -18197.887            0   -16487.779   -2242.7104 
      100    194.70949   -18195.021            0   -16454.425   -1955.2916 
-Loop time of 63.3145 on 1 procs for 100 steps with 4500 atoms
+Loop time of 23.0818 on 1 procs for 100 steps with 4500 atoms
 
-Performance: 0.136 ns/day, 175.874 hours/ns, 1.579 timesteps/s
-86.0% CPU use with 1 MPI tasks x 1 OpenMP threads
+Performance: 0.374 ns/day, 64.116 hours/ns, 4.332 timesteps/s
+98.9% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 61.815     | 61.815     | 61.815     |   0.0 | 97.63
-Bond    | 0.000132   | 0.000132   | 0.000132   |   0.0 |  0.00
-Kspace  | 1.2226     | 1.2226     | 1.2226     |   0.0 |  1.93
-Neigh   | 0.21684    | 0.21684    | 0.21684    |   0.0 |  0.34
-Comm    | 0.015175   | 0.015175   | 0.015175   |   0.0 |  0.02
-Output  | 0.000405   | 0.000405   | 0.000405   |   0.0 |  0.00
-Modify  | 0.040088   | 0.040088   | 0.040088   |   0.0 |  0.06
-Other   |            | 0.003896   |            |       |  0.01
+Pair    | 21.186     | 21.186     | 21.186     |   0.0 | 91.79
+Bond    | 0.00022054 | 0.00022054 | 0.00022054 |   0.0 |  0.00
+Kspace  | 1.5442     | 1.5442     | 1.5442     |   0.0 |  6.69
+Neigh   | 0.25672    | 0.25672    | 0.25672    |   0.0 |  1.11
+Comm    | 0.023787   | 0.023787   | 0.023787   |   0.0 |  0.10
+Output  | 0.00060248 | 0.00060248 | 0.00060248 |   0.0 |  0.00
+Modify  | 0.064809   | 0.064809   | 0.064809   |   0.0 |  0.28
+Other   |            | 0.005301   |            |       |  0.02
 
 Nlocal:    4500 ave 4500 max 4500 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
 Nghost:    21216 ave 21216 max 21216 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
-Neighs:    1.44594e+06 ave 1.44594e+06 max 1.44594e+06 min
+Neighs:    2.60176e+06 ave 2.60176e+06 max 2.60176e+06 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
 
-Total # of neighbors = 1445935
-Ave neighs/atom = 321.319
+Total # of neighbors = 2601762
+Ave neighs/atom = 578.169
 Ave special neighs/atom = 2
 Neighbor list builds = 3
 Dangerous builds = 0
-Total wall time: 0:01:05
+Total wall time: 0:00:23
diff --git a/examples/python/log.4May17.pair_python_spce.g++.4 b/examples/python/log.4May17.pair_python_spce.g++.4
index 15c11cd376..c277663287 100644
--- a/examples/python/log.4May17.pair_python_spce.g++.4
+++ b/examples/python/log.4May17.pair_python_spce.g++.4
@@ -1,5 +1,4 @@
 LAMMPS (4 May 2017)
-OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
   using 1 OpenMP thread(s) per MPI task
 units		real
 atom_style	full
@@ -22,11 +21,11 @@ read_data	data.spce
   1 = max # of 1-4 neighbors
   2 = max # of special neighbors
 
-pair_style	hybrid/overlay python 12.0 coul/long 12.0
+pair_style	hybrid/overlay coul/long 12.0 python 12.0
 kspace_style	pppm 1.0e-6
 
 pair_coeff	* * coul/long
-pair_coeff	* * python potentials.LJCutSPCE OW NULL
+pair_coeff	* * python py_pot.LJCutSPCE OW NULL
 
 bond_style	harmonic
 angle_style	harmonic
@@ -68,16 +67,16 @@ Neighbor list info ...
   ghost atom cutoff = 14
   binsize = 7, bins = 6 6 6
   2 neighbor lists, perpetual/occasional/extra = 2 0 0
-  (1) pair python, perpetual, skip from (2)
-      attributes: half, newton on
-      pair build: skip
-      stencil: none
-      bin: none
-  (2) pair coul/long, perpetual
+  (1) pair coul/long, perpetual
       attributes: half, newton on
       pair build: half/bin/newton
       stencil: half/bin/3d/newton
       bin: standard
+  (2) pair python, perpetual, skip from (1)
+      attributes: half, newton on
+      pair build: skip
+      stencil: none
+      bin: none
 Per MPI rank memory allocation (min/avg/max) = 14.59 | 14.59 | 14.59 Mbytes
 Step Temp E_pair E_mol TotEng Press 
        0            0   -16692.369            0   -16692.369    -1289.222 
@@ -91,33 +90,33 @@ Step Temp E_pair E_mol TotEng Press
       80    182.94811   -18155.978            0   -16520.523   -2393.3156 
       90    191.29902   -18197.887            0   -16487.779   -2242.7104 
      100    194.70949   -18195.021            0   -16454.425   -1955.2916 
-Loop time of 29.6024 on 4 procs for 100 steps with 4500 atoms
+Loop time of 6.588 on 4 procs for 100 steps with 4500 atoms
 
-Performance: 0.292 ns/day, 82.229 hours/ns, 3.378 timesteps/s
-52.5% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 1.311 ns/day, 18.300 hours/ns, 15.179 timesteps/s
+98.4% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 26.631     | 27.507     | 28.649     |  14.3 | 92.92
-Bond    | 0.00021    | 0.00022675 | 0.000248   |   0.0 |  0.00
-Kspace  | 0.72315    | 1.8708     | 2.7365     |  54.7 |  6.32
-Neigh   | 0.10667    | 0.1067     | 0.10674    |   0.0 |  0.36
-Comm    | 0.045357   | 0.054035   | 0.064607   |   3.6 |  0.18
-Output  | 0.000424   | 0.00086625 | 0.002189   |   0.0 |  0.00
-Modify  | 0.056602   | 0.056667   | 0.056763   |   0.0 |  0.19
-Other   |            | 0.006337   |            |       |  0.02
+Pair    | 5.3756     | 5.5417     | 5.8745     |   8.3 | 84.12
+Bond    | 0.0001049  | 0.00013965 | 0.0001812  |   0.0 |  0.00
+Kspace  | 0.54765    | 0.87786    | 1.042      |  20.8 | 13.33
+Neigh   | 0.072695   | 0.072884   | 0.072973   |   0.0 |  1.11
+Comm    | 0.04138    | 0.043576   | 0.045475   |   0.7 |  0.66
+Output  | 0.00041032 | 0.00043947 | 0.00052142 |   0.0 |  0.01
+Modify  | 0.047381   | 0.047567   | 0.047745   |   0.1 |  0.72
+Other   |            | 0.003845   |            |       |  0.06
 
 Nlocal:    1125 ave 1154 max 1092 min
 Histogram: 1 0 0 0 1 0 0 1 0 1
 Nghost:    12256.2 ave 12296 max 12213 min
 Histogram: 1 0 1 0 0 0 0 0 1 1
-Neighs:    361484 ave 376583 max 347969 min
+Neighs:    650440 ave 678828 max 626375 min
 Histogram: 1 0 0 0 2 0 0 0 0 1
 
-Total # of neighbors = 1445935
-Ave neighs/atom = 321.319
+Total # of neighbors = 2601762
+Ave neighs/atom = 578.169
 Ave special neighs/atom = 2
 Neighbor list builds = 3
 Dangerous builds = 0
-Total wall time: 0:00:30
+Total wall time: 0:00:06
diff --git a/examples/python/log.4May17.pair_python_table.g++.1 b/examples/python/log.4May17.pair_python_table.g++.1
index 21d42693ba..bd3e86501b 100644
--- a/examples/python/log.4May17.pair_python_table.g++.1
+++ b/examples/python/log.4May17.pair_python_table.g++.1
@@ -18,10 +18,10 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 # generate tabulated potential from python variant
-pair_write      1 1 10000 rsq 0.01 2.5 lj_1_1.table LJ
+pair_write      1 1 2000 rsq 0.01 2.5 lj_1_1.table LJ
 Neighbor list info ...
   update every 1 steps, delay 10 steps, check yes
   max neighbors/atom: 2000, page size: 100000
@@ -35,9 +35,9 @@ Neighbor list info ...
       stencil: half/bin/3d/newton
       bin: standard
 
-pair_style      table linear 10000
+pair_style      table linear 2000
 pair_coeff      1 1 lj_1_1.table LJ
-WARNING: 1 of 1000 force values in table are inconsistent with -dE/dr.
+WARNING: 2 of 2000 force values in table are inconsistent with -dE/dr.
   Should only be flagged at inflection points (../pair_table.cpp:476)
 
 neighbor	0.3 bin
@@ -61,36 +61,36 @@ Neighbor list info ...
       bin: standard
 Per MPI rank memory allocation (min/avg/max) = 3.184 | 3.184 | 3.184 Mbytes
 Step Temp E_pair E_mol TotEng Press 
-       0            3   -6.7733675            0   -2.2744925   -3.7033435 
-      50    1.6758875   -4.7951764            0   -2.2819736    5.6705794 
-     100    1.6458266   -4.7488945            0   -2.2807717    5.8696895 
-     150    1.6324439   -4.7283321            0   -2.2802784    5.9594952 
-     200    1.6630547   -4.7746809            0   -2.2807225    5.7372657 
-     250    1.6278968   -4.7226363            0   -2.2814016    5.9559236 
-Loop time of 1.0498 on 1 procs for 250 steps with 4000 atoms
+       0            3   -6.7733629            0   -2.2744879   -3.7032813 
+      50    1.6758731   -4.7953067            0   -2.2821255    5.6706553 
+     100    1.6458118   -4.7490281            0   -2.2809276    5.8697466 
+     150     1.632425   -4.7284533            0   -2.2804279    5.9595684 
+     200    1.6631578   -4.7749889            0   -2.2808759    5.7365839 
+     250    1.6277062   -4.7224727            0   -2.2815238    5.9572913 
+Loop time of 0.996652 on 1 procs for 250 steps with 4000 atoms
 
-Performance: 102877.190 tau/day, 238.142 timesteps/s
-99.7% CPU use with 1 MPI tasks x 1 OpenMP threads
+Performance: 108362.785 tau/day, 250.840 timesteps/s
+99.8% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 0.93242    | 0.93242    | 0.93242    |   0.0 | 88.82
-Neigh   | 0.088495   | 0.088495   | 0.088495   |   0.0 |  8.43
-Comm    | 0.012153   | 0.012153   | 0.012153   |   0.0 |  1.16
-Output  | 0.00013924 | 0.00013924 | 0.00013924 |   0.0 |  0.01
-Modify  | 0.013729   | 0.013729   | 0.013729   |   0.0 |  1.31
-Other   |            | 0.002855   |            |       |  0.27
+Pair    | 0.87999    | 0.87999    | 0.87999    |   0.0 | 88.29
+Neigh   | 0.087921   | 0.087921   | 0.087921   |   0.0 |  8.82
+Comm    | 0.012098   | 0.012098   | 0.012098   |   0.0 |  1.21
+Output  | 0.00013614 | 0.00013614 | 0.00013614 |   0.0 |  0.01
+Modify  | 0.01363    | 0.01363    | 0.01363    |   0.0 |  1.37
+Other   |            | 0.002882   |            |       |  0.29
 
 Nlocal:    4000 ave 4000 max 4000 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
-Nghost:    5504 ave 5504 max 5504 min
+Nghost:    5500 ave 5500 max 5500 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
-Neighs:    151497 ave 151497 max 151497 min
+Neighs:    151496 ave 151496 max 151496 min
 Histogram: 1 0 0 0 0 0 0 0 0 0
 
-Total # of neighbors = 151497
-Ave neighs/atom = 37.8743
+Total # of neighbors = 151496
+Ave neighs/atom = 37.874
 Neighbor list builds = 12
 Dangerous builds not checked
 
diff --git a/examples/python/log.4May17.pair_python_table.g++.4 b/examples/python/log.4May17.pair_python_table.g++.4
index a64ebd6631..8dfe82097e 100644
--- a/examples/python/log.4May17.pair_python_table.g++.4
+++ b/examples/python/log.4May17.pair_python_table.g++.4
@@ -18,10 +18,10 @@ mass		* 1.0
 velocity	all create 3.0 87287
 
 pair_style	python 2.5
-pair_coeff	* * potentials.LJCutMelt lj
+pair_coeff	* * py_pot.LJCutMelt lj
 
 # generate tabulated potential from python variant
-pair_write      1 1 10000 rsq 0.01 2.5 lj_1_1.table LJ
+pair_write      1 1 2000 rsq 0.01 2.5 lj_1_1.table LJ
 Neighbor list info ...
   update every 1 steps, delay 10 steps, check yes
   max neighbors/atom: 2000, page size: 100000
@@ -35,9 +35,9 @@ Neighbor list info ...
       stencil: half/bin/3d/newton
       bin: standard
 
-pair_style      table linear 10000
+pair_style      table linear 2000
 pair_coeff      1 1 lj_1_1.table LJ
-WARNING: 1 of 10000 force values in table are inconsistent with -dE/dr.
+WARNING: 2 of 2000 force values in table are inconsistent with -dE/dr.
   Should only be flagged at inflection points (../pair_table.cpp:476)
 
 neighbor	0.3 bin
@@ -61,36 +61,36 @@ Neighbor list info ...
       bin: standard
 Per MPI rank memory allocation (min/avg/max) = 2.69 | 2.69 | 2.69 Mbytes
 Step Temp E_pair E_mol TotEng Press 
-       0            3   -6.7733675            0   -2.2744925   -3.7033435 
-      50    1.6754092    -4.794723            0   -2.2822376    5.6616601 
-     100    1.6503295   -4.7559815            0   -2.2811061    5.8051261 
-     150    1.6596603   -4.7699379            0   -2.2810699    5.7830168 
-     200    1.6371948   -4.7365549            0   -2.2813766    5.9245585 
-     250    1.6321199   -4.7288017            0   -2.2812339    5.9776124 
-Loop time of 0.313548 on 4 procs for 250 steps with 4000 atoms
+       0            3   -6.7733629            0   -2.2744879   -3.7032813 
+      50     1.675395   -4.7945736            0   -2.2821094    5.6620623 
+     100    1.6503067   -4.7558145            0   -2.2809733    5.8055967 
+     150    1.6595852   -4.7697199            0   -2.2809644    5.7837898 
+     200    1.6371471   -4.7363942            0   -2.2812874     5.924977 
+     250    1.6315623   -4.7278268            0   -2.2810951    5.9807196 
+Loop time of 0.300176 on 4 procs for 250 steps with 4000 atoms
 
-Performance: 344444.576 tau/day, 797.325 timesteps/s
-99.1% CPU use with 4 MPI tasks x 1 OpenMP threads
+Performance: 359789.395 tau/day, 832.846 timesteps/s
+99.3% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 0.24963    | 0.25443    | 0.2632     |   1.1 | 81.15
-Neigh   | 0.023249   | 0.023735   | 0.024497   |   0.3 |  7.57
-Comm    | 0.020689   | 0.030402   | 0.035249   |   3.4 |  9.70
-Output  | 0.00020766 | 0.00021476 | 0.00023031 |   0.0 |  0.07
-Modify  | 0.0034959  | 0.0035564  | 0.0036762  |   0.1 |  1.13
-Other   |            | 0.001206   |            |       |  0.38
+Pair    | 0.23104    | 0.23876    | 0.2451     |   1.2 | 79.54
+Neigh   | 0.022763   | 0.023687   | 0.024305   |   0.4 |  7.89
+Comm    | 0.025416   | 0.032499   | 0.041304   |   3.7 | 10.83
+Output  | 0.00015378 | 0.00016057 | 0.00017667 |   0.0 |  0.05
+Modify  | 0.0035894  | 0.0036637  | 0.0037456  |   0.1 |  1.22
+Other   |            | 0.001409   |            |       |  0.47
 
-Nlocal:    1000 ave 1010 max 982 min
+Nlocal:    1000 ave 1010 max 981 min
 Histogram: 1 0 0 0 0 0 1 0 0 2
-Nghost:    2703.75 ave 2713 max 2689 min
-Histogram: 1 0 0 0 0 0 0 2 0 1
-Neighs:    37915.5 ave 39231 max 36202 min
+Nghost:    2703 ave 2715 max 2688 min
 Histogram: 1 0 0 0 0 1 1 0 0 1
+Neighs:    37915.2 ave 39191 max 36151 min
+Histogram: 1 0 0 0 0 1 0 1 0 1
 
-Total # of neighbors = 151662
-Ave neighs/atom = 37.9155
+Total # of neighbors = 151661
+Ave neighs/atom = 37.9153
 Neighbor list builds = 12
 Dangerous builds not checked
 
diff --git a/examples/python/py_pot.py b/examples/python/py_pot.py
index dbce8cd445..5699bd082c 100644
--- a/examples/python/py_pot.py
+++ b/examples/python/py_pot.py
@@ -1,73 +1,65 @@
 from __future__ import print_function
 
-class LJCutMelt(object):
-
+class LAMMPSPairPotential(object):
     def __init__(self):
         self.pmap=dict()
-        # set coeffs: eps, sig, 48*eps*sig**12, 24*eps*sig**6,
-        #                        4*eps*sig**12,  4*eps*sig**6
-        self.coeff = {'lj'  : {'lj'  : (1.0,1.0,48.0,24.0,4.0,4.0),
-                               'NULL': (0.0,1.0, 0.0, 0.0,0.0,0.0)},
-                      'NULL': {'lj'  : (0.0,1.0, 0.0, 0.0,0.0,0.0),
-                               'NULL': (0.0,1.0, 0.0, 0.0,0.0,0.0)}}
+        self.units='lj'
+    def map_coeff(self,name,ltype):
+        self.pmap[ltype]=name
+    def check_units(self,units):
+        if (units != self.units):
+           raise Exception("Conflicting units: %s vs. %s" % (self.units,units))
 
-    def map_coeff(self,name,type):
-        if name in self.coeff:
-           self.pmap[type] = name
-        else:
-           raise Exception("cannot match atom type %s" % name)
+class LJCutMelt(LAMMPSPairPotential):
+    def __init__(self):
+        super(LJCutMelt,self).__init__()
+        # set coeffs: 48*eps*sig**12, 24*eps*sig**6,
+        #              4*eps*sig**12,  4*eps*sig**6
+        self.units = 'lj'
+        self.coeff = {'lj'  : {'lj'  : (48.0,24.0,4.0,4.0)}}
 
     def compute_force(self,rsq,itype,jtype):
         coeff = self.coeff[self.pmap[itype]][self.pmap[jtype]]
         r2inv  = 1.0/rsq
         r6inv  = r2inv*r2inv*r2inv
-        lj1 = coeff[2]
-        lj2 = coeff[3]
-        return (r6inv * (lj1*r6inv - lj2))
+        lj1 = coeff[0]
+        lj2 = coeff[1]
+        return (r6inv * (lj1*r6inv - lj2))*r2inv
 
     def compute_energy(self,rsq,itype,jtype):
         coeff = self.coeff[self.pmap[itype]][self.pmap[jtype]]
         r2inv  = 1.0/rsq
         r6inv  = r2inv*r2inv*r2inv
-        lj3 = coeff[4]
-        lj4 = coeff[5]
+        lj3 = coeff[2]
+        lj4 = coeff[3]
         return (r6inv * (lj3*r6inv - lj4))
 
-class LJCutSPCE(object):
 
+class LJCutSPCE(LAMMPSPairPotential):
     def __init__(self):
-        self.pmap=dict()
-        # SPCE oxygen in real units
+        super(LJCutSPCE,self).__init__()
+        self.units='real'
+        # SPCE oxygen LJ parameters in real units
         eps=0.15535
         sig=3.166
-
-        # set coeffs: eps, sig, 48*eps*sig**12, 24*eps*sig**6,
-        #                        4*eps*sig**12,  4*eps*sig**6
-        self.coeff = {'OW'  : {'OW'  : (1.0,1.0,
-                                48.0*eps*sig**12,24.0*eps*sig**6,
-                                 4.0*eps*sig**12, 4.0*eps*sig**6),
-                               'NULL': (0.0,1.0, 0.0, 0.0,0.0,0.0)},
-                      'NULL': {'OW'  : (0.0,1.0, 0.0, 0.0,0.0,0.0),
-                               'NULL': (0.0,1.0, 0.0, 0.0,0.0,0.0)}}
-
-    def map_coeff(self,name,type):
-        if name in self.coeff:
-           self.pmap[type] = name
-        else:
-           raise Exception("cannot match atom type %s" % name)
+        self.coeff = {'OW'  : {'OW'  : (48.0*eps*sig**12,24.0*eps*sig**6,
+                                         4.0*eps*sig**12, 4.0*eps*sig**6),
+                               'HW'  : (0.0,0.0, 0.0,0.0)},
+                      'HW'  : {'OW'  : (0.0,0.0, 0.0,0.0),
+                               'HW'  : (0.0,0.0, 0.0,0.0)}}
 
     def compute_force(self,rsq,itype,jtype):
         coeff = self.coeff[self.pmap[itype]][self.pmap[jtype]]
         r2inv  = 1.0/rsq
         r6inv  = r2inv*r2inv*r2inv
-        lj1 = coeff[2]
-        lj2 = coeff[3]
-        return (r6inv * (lj1*r6inv - lj2))
+        lj1 = coeff[0]
+        lj2 = coeff[1]
+        return (r6inv * (lj1*r6inv - lj2))*r2inv
 
     def compute_energy(self,rsq,itype,jtype):
         coeff = self.coeff[self.pmap[itype]][self.pmap[jtype]]
         r2inv  = 1.0/rsq
         r6inv  = r2inv*r2inv*r2inv
-        lj3 = coeff[4]
-        lj4 = coeff[5]
+        lj3 = coeff[2]
+        lj4 = coeff[3]
         return (r6inv * (lj3*r6inv - lj4))
diff --git a/src/PYTHON/pair_python.cpp b/src/PYTHON/pair_python.cpp
index 1d80145c58..384aa5a94b 100644
--- a/src/PYTHON/pair_python.cpp
+++ b/src/PYTHON/pair_python.cpp
@@ -24,6 +24,7 @@
 #include "comm.h"
 #include "force.h"
 #include "memory.h"
+#include "update.h"
 #include "neigh_list.h"
 #include "python.h"
 #include "error.h"
@@ -41,9 +42,10 @@ PairPython::PairPython(LAMMPS *lmp) : Pair(lmp) {
   one_coeff = 1;
   reinitflag = 0;
 
-  python->init();
-
   py_potential = NULL;
+  skip_types = NULL;
+
+  python->init();
 
   // add current directory to PYTHONPATH
   PyObject * py_path = PySys_GetObject((char *)"path");
@@ -60,7 +62,8 @@ PairPython::PairPython(LAMMPS *lmp) : Pair(lmp) {
 
 PairPython::~PairPython()
 {
-  if(py_potential) Py_DECREF((PyObject*) py_potential);
+  if (py_potential) Py_DECREF((PyObject*) py_potential);
+  delete[] skip_types;
 
   if (allocated) {
     memory->destroy(setflag);
@@ -160,6 +163,9 @@ void PairPython::compute(int eflag, int vflag)
       rsq = delx*delx + dely*dely + delz*delz;
       jtype = type[j];
 
+      // with hybrid/overlay we might get called for skipped types
+      if (skip_types[itype] || skip_types[jtype]) continue;
+
       py_jtype = PY_INT_FROM_LONG(jtype);
       PyTuple_SetItem(py_compute_args,2,py_jtype);
 
@@ -173,7 +179,7 @@ void PairPython::compute(int eflag, int vflag)
           PyGILState_Release(gstate);
           error->all(FLERR,"Calling 'compute_force' function failed");
         }
-        fpair = factor_lj*PyFloat_AsDouble(py_value)/rsq;
+        fpair = factor_lj*PyFloat_AsDouble(py_value);
 
         f[i][0] += delx*fpair;
         f[i][1] += dely*fpair;
@@ -299,6 +305,39 @@ void PairPython::coeff(int narg, char **arg)
 
   py_potential = (void *) py_pair_instance;
 
+  PyObject *py_check_units = PyObject_GetAttrString(py_pair_instance,"check_units");
+  if (!py_check_units) {
+    PyErr_Print();
+    PyErr_Clear();
+    PyGILState_Release(gstate);
+    error->all(FLERR,"Could not find 'check_units' method'");
+  }
+  if (!PyCallable_Check(py_check_units)) {
+    PyErr_Print();
+    PyErr_Clear();
+    PyGILState_Release(gstate);
+    error->all(FLERR,"Python 'check_units' is not callable");
+  }
+  PyObject *py_units_args = PyTuple_New(1);
+  if (!py_units_args) {
+    PyErr_Print();
+    PyErr_Clear();
+    PyGILState_Release(gstate);
+    error->all(FLERR,"Could not create tuple for 'check_units' function arguments");
+  }
+
+  PyObject *py_name = PY_STRING_FROM_STRING(update->unit_style);
+  PyTuple_SetItem(py_units_args,0,py_name);
+  PyObject *py_value = PyObject_CallObject(py_check_units,py_units_args);
+  if (!py_value) {
+    PyErr_Print();
+    PyErr_Clear();
+    PyGILState_Release(gstate);
+    error->all(FLERR,"Calling 'check_units' function failed");
+  }
+  Py_DECREF(py_units_args);
+
+
   PyObject *py_map_coeff = PyObject_GetAttrString(py_pair_instance,"map_coeff");
   if (!py_map_coeff) {
     PyErr_Print();
@@ -321,9 +360,15 @@ void PairPython::coeff(int narg, char **arg)
     error->all(FLERR,"Could not create tuple for 'map_coeff' function arguments");
   }
 
-  PyObject *py_type, *py_name, *py_value;
+  delete[] skip_types;
+  skip_types = new int[ntypes+1];
+  skip_types[0] = 1;
   for (int i = 1; i <= ntypes ; i++) {
-    py_type = PY_INT_FROM_LONG(i);
+    if (strcmp(arg[2+i],"NULL") == 0) {
+      skip_types[i] = 1;
+      continue;
+    } else skip_types[i] = 0;
+    PyObject *py_type = PY_INT_FROM_LONG(i);
     py_name = PY_STRING_FROM_STRING(arg[2+i]);
     PyTuple_SetItem(py_map_args,0,py_name);
     PyTuple_SetItem(py_map_args,1,py_type);
@@ -336,10 +381,8 @@ void PairPython::coeff(int narg, char **arg)
     }
 
     for (int j = i; j <= ntypes ; j++) {
-      if (strcmp(arg[2+i],"NULL") != 0) {
-        setflag[i][j] = 1;
-        cutsq[i][j] = cut_global*cut_global;
-      }
+      setflag[i][j] = 1;
+      cutsq[i][j] = cut_global*cut_global;
     }
   }
   Py_DECREF(py_map_args);
@@ -359,6 +402,11 @@ double PairPython::single(int i, int j, int itype, int jtype, double rsq,
                          double factor_coul, double factor_lj,
                          double &fforce)
 {
+  // with hybrid/overlay we might get called for skipped types
+  if (skip_types[itype] || skip_types[jtype]) {
+    fforce = 0.0;
+    return 0.0;
+  }
 
   // prepare access to compute_force and compute_energy functions
 
@@ -417,7 +465,7 @@ double PairPython::single(int i, int j, int itype, int jtype, double rsq,
     PyGILState_Release(gstate);
     error->all(FLERR,"Calling 'compute_force' function failed");
   }
-  fforce = factor_lj*PyFloat_AsDouble(py_value)/rsq;
+  fforce = factor_lj*PyFloat_AsDouble(py_value);
 
   py_value = PyObject_CallObject(py_compute_energy,py_compute_args);
   if (!py_value) {
diff --git a/src/PYTHON/pair_python.h b/src/PYTHON/pair_python.h
index 3c9c34fbdf..440b39e482 100644
--- a/src/PYTHON/pair_python.h
+++ b/src/PYTHON/pair_python.h
@@ -47,6 +47,7 @@ class PairPython : public Pair {
  protected:
   double cut_global;
   void * py_potential;
+  int  * skip_types;
 
   virtual void allocate();
 };