Our tests show that GRTensorII 1.79 works as expected with Maple 11. Install as with Maple 10, 9.5, 9, 8.

Our tests show that GRTensorII 1.79 works as expected with Maple 10. Install as with Maple 9.5, 9, 8.

We recommend that you use the classic worksheet interface and that you do not use the current Java worksheet interface supplied with Maple 9.x.

Demonstrations by Dumitru N. Vulcanov now on line.

An interactive on line database is now open at grdb.org

The GRTensorII site is now registered as grtensor.org

For something new at GRTensorII (that isn't quite ready yet) go to LiveTranslate.

The latest Release of GRTensorII is 1.79 for Maple 6, 7, 8, 9 and 9.5. (If you use a Microsoft operating system, make sure your maple.ini file is in the users subdirectory with Maple 7, 8 and 9.)

With Maple 8 and 9, for the current release of GRTensorII, you need to load the libraries needed for your calculation. In previous versions of Maple they autoload. A typical initialization file is here. Simply download this file and follow the instructions in it. If you have developed your own libraries, load them in the same way.

Note: Previous to maple 6, maple used 17 bits to represent the internal length of an object on 32 bit machines. As of maple 6, maple now uses 26 bits instead. For most cases, this means that maple 6 is now likely to run out of memory (both physical and virtual) on any machine before it encounters an 'object too large' message.

  This Release offers the following changes since Release 1.70:

  • Converted name of Newman-Penrose Lambda to NPLambda
  • Changed the way indices are handled in diff() within grdef()
  • Fixed error in Levi-Cevita symbol calculated for a basis
  • Two new simplification routines.
  • qload() of a previously loaded spacetime is converted internally to grmetric()
  • Complete clearing of definitions in grundefine()
  • Fix allowing raise/lower of indices for tensors with more than 5 indices
  • Fix automatic determination of symmetries for grdef() objects whose indices have been raised/lowered
  • Improved grtransform(): following the suggestions of Edward Huff.
  • Warnings attached to autoAlias(): to avoid possible sources of confusion
  • Fix indexing problem pointed out by Kostas Kokkotas and Daniel Finley.
  • Ability to enter the basis directly in terms of differentials.
  • GRJunction isn't here yet. You have to go to our old site.