I'm testing ECDH key agreement using different libs - Crypto++ 5.5.2, openssl 0.9.8g (www.openssl.org) and Bouncy Castle Crypto APIs C# Release 1.3 (http://www.bouncycastle.org/csharp/).
When I use standard curves based on GF(p), for example secp521r1, I recive identical agreed values generated by each library (with the same [private key/other party's public key] pair of course). But if I trying curves based on GF(2^n), for example sect571r1, I've got difference between resulting value generated by Crypto++ and values generated by openssl & Bouncy Castle Crypto APIs (these libs returns same result). Also Crypto++ works much more slowly with these curves. I think careful testing of ECDH class with test vectors or something is required. With best regards, Andrei. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the "Crypto++ Users" Google Group. To unsubscribe, send an email to [EMAIL PROTECTED] More information about Crypto++ and this group is available at http://www.cryptopp.com. -~----------~----~----~----~------~----~------~--~---
