Download
Randon number generation & testing. The best suite of statistical tests for fast PRNGs anywhere. Multithreaded for speed, command line tools for automation, no upper limit on data size. Also, a variety of C++ pseudo-random number generators with well designed interfaces aimed at practical uses, not just research.
Features
- A convenient & powerful interface to RNG algorithms
- A variety of fast high quality RNG algorithms
- Fast & effective statistical tests for RNGs
License
Public DomainFollow Practically Random
Other Useful Business Software
Enterprise-grade ITSM, for every business
Freshservice is an intuitive, AI-powered platform that helps IT, operations, and business teams deliver exceptional service without the usual complexity. Automate repetitive tasks, resolve issues faster, and provide seamless support across the organization. From managing incidents and assets to driving smarter decisions, Freshservice makes it easy to stay efficient and scale with confidence.
Rate This Project
Login To Rate This Project
User Reviews
-
Proved to be essential for PRVHASH development.
-
Helped me see patterns I could not see. Fun to use, smart, well-written, concise.
-
Best program I've ever found to test RNG's on Windows (and anywhere else), by far. Never got the (apparently next best) TestU01 to work on Windows, but even if I had, it would now likely be eighty-sixed anyway. My only wish for Pract Rand would be to allow the -tlshow switch to be automatically repeatable to avoid having to copy the switch multiple times. For example, to repeat results every 4 hours there might be a -tlshowRep 4h.
-
I really like the capabilities which PractRand offers. So far I have been using testing suite called TestU01. PractRand offers number of unique features which TestU01 does not have. This includes: * testing virtually unlimited lenght of random data. You will be limited only by runtime. I have tested some generators till 128TB of data. On the other hand, TestU01's most extensive BigCrush test can consume only few hunders GB of data. This can be too little to find deviation from randomness. * applying number of automatic filters. For example, it's testing on the fly the randomness of low bits of provided data. This happening in parallel to the testing of full data. * it's very fast, in single - threaded mode operating at rate ~4GiB/min. The processing speed can be increased by using multi-threaded mode. * I was able to detect problems with generators which has passed TestU01 testing.