[noise] Noise HFS Kyber question

dawuud dawuud at riseup.net
Sat Feb 2 21:16:02 PST 2019

duh i just read a section from https://pq-crystals.org/kyber/data/kyber-specification.pdf
(at the top of page 17)
and now Kyber1024 seems like the obvious choice for Katzenpost.
I wonder, Why does Rhys's Noise HFS Kyber spec extension to Noise uses Kyber768?

On Sun, Feb 03, 2019 at 04:47:03AM +0000, dawuud wrote:
> Hi Peter, Rhys and the noise mailing list,
> I just got Rhys's Noise HFS Kyber extension to work on our port of the flynn golang Noise library:
> https://github.com/katzenpost/noise/tree/kyber.0
> My question is why did you choose to use Kyber768 instead of Kyber1024?
> Maybe this is a question for Peter Schwabe:
> How do these (Kyber768 and Kyber1024) compare with the safety margins of NewHope Simple?
> The reason I'm asking is because the Katzenpost (https://github.com/katzenpost)
> Noise based link layer currently uses NewHope Simple with XX in HFS mode and I'd like
> to use Kyber with parameters that will give a comparable security margin.
> Sincerely,
> David Stainton

> _______________________________________________
> Noise mailing list
> Noise at moderncrypto.org
> https://moderncrypto.org/mailman/listinfo/noise

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://moderncrypto.org/mail-archive/noise/attachments/20190203/a87e72c3/attachment.sig>

More information about the Noise mailing list