@MaartenBodewes-modelection Mostly, I am trying to figure out why people suck at implementing AES. I have an asynchronous implementation that fits nicely into a super-scaler architecture, which is everything since a PowerPC601 as it's how we now do division units. There's so much lore in cryptography and I wanted to sort through some of it for AES.
When you have someone who understand something, it's not lore; however, I just don't hang out with cryptographers.
I find that cryptographers see hardware as an inconvenience and not implementation realities.
@bdegnan There is at least some focus on side channels in the latest NIST competitions. But cryptographers of the type that create algorithms need to be mostly mathematicians. That's a rather abstract science, so it is no surprise that many won't be into hardware.
@MaartenBodewes-modelection It varies. I consider myself primarily a cryptomathematician, but I have co-authored electrical engineering papers on energy efficient cryptography and also have a patent on handling information leakage from sidechannels coming out in the next couple of weeks.
@DanielS Thank goodness there are exceptions to the rule. I find myself to be a pretty good dev for somebody that does cryptography - although my math skills suck so hard I'm not presenting myself as a "cryptographer" - I'm more at home at the protocol / API level :)
Huh, I first tried to avoid "rule" in the sentence before that and now I'm using it anyway. Well, y'know what I mean.
From this answer I am able to understand that if CM field discriminant for a particular curve is small then it provide us a fast endomorphism which in turn allow rho method to speed up by $\sqrt{\frac{1}{3}}$.
What I cannot understand is:
How one calculate this CM discriminant? Is this same as $...
@DanielS I'd like to see what you did in that patent. Once it's out, send me a link.
I was thinking of making AES128 out of 7400 parts just so students can try a side-channel attack. One of the things with FPGAs is that the target routing is heuristic. I know that on a certain paper, it was rerouted until they got results. Almost no one does power attacks on actual POTS hardware.
The Side Channel
The Side Channel