Speed of Davies attack

I have skimmed through "Pairs and Triplets of DES S-Boxes" by Davies
and Murphy, and "An Improvement of Davies' Attack on DES" by Biham and
Biryukov, but I'm no expert in cryptanalysis, I'm still learning. So,
if anyone could tell me in simple language, this Biham-Biryukov
improvement od Davies' attack, how faster is it than brute force? Can
I ask something like "if breaking DES by brute force took 1 year, how
much would inproved Davies take"? Can efficiency of cryptanalytic
attacks be discussed in such a terms?

--
"Now the storm has passed over me
I'm left to drift on a dead calm sea
And watch her forever through the cracks in the beams
Nailed across the doorways of the bedrooms of my dreams"
.

Relevant Pages

  • Re: Any truth to rumor that NSA had Public Key Crypto first?
    ... the best differential attack against full single DES requires 2^47 ... takes about 2^60 steps and uses ...2^60 chosen plaintexts". ...
    (sci.crypt)
  • Re: Some questions on DES Encryption...
    ... If all you have to go on is a single plaintext-ciphertext pair, ... DES uses a 56-bit key, ... > I think you're thinking of linear or differential cryptanalysis here. ... world discovered this attack. ...
    (SecProg)
  • Re: FUD about CGD and GBDE
    ... >When DES was designed, nobody knew that differential attacks existed. ... *one* attack significantly better than brute force was found: ... slightly-weakened version of Skipjack, an NSA cipher. ... note that 256-bit AES is approved for Top Secret traffic. ...
    (freebsd-hackers)
  • FW: Some questions on DES Encryption...
    ... The DES algorithm is public record. ... If you're researching crypto, ... Even 56 bit DES is strong enough that an attacker will almost always attack ...
    (SecProg)
  • Re: What is the accepted technical definition of the word break?
    ... >> attack the algorithm in practice. ... Nevertheless TripleDES is not ... sufficient to break three-key triple DES. ... of single DES operations and assumes the other operations to be much faster, ...
    (sci.crypt)