Commits · 83284ce6e721a63e1e2c851595c5db6b349db040 · tsoc / openmm

10 Apr, 2013 1 commit
- Added volatile keyword to fix crashes on some GPUs · 83284ce6
  Peter Eastman authored Apr 10, 2013
  
  83284ce6
22 Mar, 2013 1 commit
- Merged 5.1Optimizations branch back to trunk · 93c467b2
  Peter Eastman authored Mar 22, 2013
  
  93c467b2
14 Dec, 2012 1 commit

When converting to fixed point, multiply by 0x100000000 instead of 0xFFFFFFFF.... · 18355094

Peter Eastman authored Dec 14, 2012

When converting to fixed point, multiply by 0x100000000 instead of 0xFFFFFFFF.  This should be (very very slightly) more accurate, since its reciprocal can be exactly represented in floating point.

18355094

16 Nov, 2012 1 commit
- Restrict the number of threads in GK kernels based on available shared memory · b8cba79f
  Peter Eastman authored Nov 16, 2012
  
  b8cba79f
01 Nov, 2012 1 commit
- Fixed lots of bugs found by mixed/double precision test cases · 490cc912
  Peter Eastman authored Nov 01, 2012
  
  490cc912
28 Sep, 2012 1 commit
- Renamed cuda2 platform to cuda · 58b094ce
  Peter Eastman authored Sep 28, 2012
  
  58b094ce
25 Sep, 2012 1 commit
- Minor cleanup to simplify code and eliminate unnecessary memory access · 1cb85ba3
  Peter Eastman authored Sep 25, 2012
  
  1cb85ba3
22 Sep, 2012 1 commit
- Fixed bugs in AMOEBA GK · 3c130a24
  Peter Eastman authored Sep 22, 2012
  
  3c130a24
20 Sep, 2012 1 commit
- Implemented surface area term for GK · 04fe2449
  Peter Eastman authored Sep 20, 2012
  
  04fe2449
17 Sep, 2012 1 commit
- Bug fixes to AmoebaGeneralizedKirkwoodForce · 6c0082a5
  Peter Eastman authored Sep 17, 2012
  
  6c0082a5
14 Sep, 2012 1 commit
- Continuing to convert AmoebaGeneralizedKirkwoodForce to new CUDA platform · b587b396
  Peter Eastman authored Sep 14, 2012
  
  b587b396
13 Sep, 2012 1 commit
- Began converting AmoebaGeneralizedKirkwoodForce to new CUDA platform · 222378c6
  Peter Eastman authored Sep 13, 2012
  
  222378c6