twofish keysize

[Per Tunedal Casual]

At 09:26 2004-04-19, Werner Koch wrote:
 >On Mon, 19 Apr 2004 00:19:58 +0200, Malte Gell said:
 >
 >> Maybe he expects a performance advantage when using 128 bit keys instead
 >> of 256 bit. By the way, are there estimations how much slower a 256 bit
 >> Twofish key is? I guess one will only notice it whith huge amount of
 >
 >here are encrypt/decrypt result from libgcrypt.
 >
 >           ECB             CBC             CFB             CTR
 >           --------------- --------------- --------------- ----------------
 >3DES         140ms   150ms   170ms   190ms   150ms   160ms   180ms   180ms
 >CAST5         50ms    50ms    70ms    80ms    60ms    60ms    90ms    80ms
 >BLOWFISH      90ms   100ms   110ms   120ms   110ms   110ms   130ms   130ms
 >AES           40ms    40ms    50ms    60ms    40ms    50ms    80ms    80ms
 >AES192        40ms    40ms    60ms    60ms    50ms    50ms    80ms    80ms
 >AES256        40ms    50ms    60ms    70ms    50ms    60ms    90ms    80ms
 >TWOFISH       50ms    50ms    60ms    60ms    60ms    60ms   100ms   100ms
 >TWOFISH128    50ms    50ms    60ms    60ms    60ms    60ms   100ms   100ms
 >
 >As you can see, there is no difference between twofish (256 bit) and
 >twofish 128.
 >
 >The disadvantage of a 256 bit key is that is requires double as much
 >entropy to create a session key.
 >
How large files where used in this performance test? I recently read a NIST 
evaluation: For 256-bit keys TWOFISH was slightly faster than AES on (very) 
large files.

BTW I've been told it isn't wise to encrypt files larger than a few MB 
using a block size of 64 bits. What's the limit for the block size 128 
bits, used in the new AES and TWOFISH ciphers? And what's the problem?
Per Tunedal