**Press release **: https://web.archive.org/web/20231220075442/https://www.prnewswire.com/news-releases/peerplays-rng-functionality-completes-gli-testing-301096484.html
Peerplays RNG is a decentralized, software-driven random number generator. This unique approach provides tamper-resistance, fairness, transparency, and verifiability. The algorithm uses a combination of ripemd160 and SHA256 cryptographic functions coupled with Peerplays blockchain's blocks to generate the random numbers.
The Peerplays blockchain acts as a source of unique, random seeds to the random number generator functions. Blockchain by its nature provides high entropy in each block due to the fact that the transactions and computations performed in each block is entirely arbitrary as it acts as a high source of entropy. A public blockchain with a large number of transactions, trades, and geo-distributed mining or block producer nodes acts as a truly resilient noice provider. In addition, the open nature of blockchain provides the scope for anyone to dynamically monitor the output in the background in an interoperable manner.
Random Number Generator Algorithm
Peerplays blockchain exposes extensive APIs to build decentralized Apps to businesses and hobbyist developers alike. We are a public blockchain providing inbuilt functionalities and interfaces for sports, eSports, igaming products, and lottery products and secure messaging capacities. This helps chains to see a high number of transactions which are supported by a dozen and ever-growing number of block producers who acts as miners and signs the blocks. Every 3 second a new block with unique transactions are generated. These blocks provide a never-ending, unique source of noise.
Random number is generated from secret hashes of previous and current blocks, which are combined into a single data stream and encoded using ripemd160 algorithm and fed into a random number generator as a seed. This random number is then exposed to the consumer over a user-friendly API. As we have multiple nodes acting as noise and random number generators, we can provide high throughput.
By design we expect the random number requests to be of internet scale. This derives from the fact that the underlying blockchain supports 1000s of transactions per second. Our unique distributed load balancing model due to a large number of nodes can readily support extremely large throughput. Unlike centralized systems we can scale all the while ensuring security and performance.
Implementation details
Our blockchain derives from the robust Graphene platform. The platform boasts its roots to the LMAX-disruptor which provides extremely large transactional capacity. The tooling is C++ and BOOST libraries making it deploy on wide range of OSes and hardware. The stack also provides end to end programming libraries written in popular languages like Javascript and Python. All the source code is shared in Open source MIT license enabling ease of adaption to various business models and the same time being completely transparent.