I love your content but mostly because lack of my knowledge I can't understand it and it gose to my watch later playlist...
@LowByteProductions
Жыл бұрын
Thank you. Please don't feel that you should wait until you understand any subject to start studying it. Feeling uncomfortable with the material is actually what you should be feeling to some degree! That is the first step to actually understanding.
@eduardpaul8413
Жыл бұрын
Another great video! I enjoy your content very much, especially the more educational themed videos.
@rfgrooteman
Жыл бұрын
Hi. Great content. Thank you so much. I just want to point out that hex 46 is not 101110 in binary. Decimal 46 is. Hex 46 is 1000110. You meant 0x2E.
@LowByteProductions
Жыл бұрын
👍
@wollwolke
Жыл бұрын
Nice video, motivated me enough to actually try and implement this in JS. For now it's just the ECB-mode, but it's rather fascinating how rather easy to code the algorithm is, considering it's security. The finite field is still a mystery to me though, oh well.
@kayakMike1000
9 ай бұрын
I find it difficult to believe you rewrote this in JS without gaining any insight into the underlying mathematics. Maybe you only manually transpiled his code into JS, but JS isn't very useful on a microcontroller.
@kilwo
Жыл бұрын
Nice video. Quick question. Could the the length be used as the iv rather than having it in the plain text. I cannot see why it couldn’t as it would affect every block the same as having it in block 1 would.
@chad3814
Жыл бұрын
I think if the length isn't encrypted, then you run into the problem of not getting an evenly distributed IV
@LowByteProductions
Жыл бұрын
Unfortunately not! Because you have to send both the code and the IV in plaintext, an attacker can alter the first block and the IV together in tandem and still come out with the same signature. And considering that a length field would only ever take at most 4 bytes, those top 12 bytes could be altered indiscriminately. When you have an implied, all zero IV, the attacker does not have that avenue.
@kayakMike1000
9 ай бұрын
I wrote code that can generate the sbox and its inverse. I think the compiled code only slightly smaller than 256 bytes
@kayakMike1000
11 ай бұрын
I tried writing a tiny RIJNDAEL implementation... its kinda hard to understand all the different steps and wierd galois math.
@LowByteProductions
11 ай бұрын
I get that. What helped me the first time around was just reading up across a number of different sources, and reading implementations in code. Sometimes seeing the same thing in a lot of different contexts can help solidify it. But yeah, galois fields are definitely tricky. Math is different beast!
@kayakMike1000
10 ай бұрын
@@LowByteProductionswell, I did get the SBOX to generate properly; turns out my affine bit matrices needed to be reversed b/c I was on a little endian system. If that makes sense... There's a magic polynomial, the bit matrix is another magical constant... I'm working on expanding the key schedule. After it's all done, I want to try a side channel attack. Maybe I could make my implementation vulnerable to a timing attack and learn how to hack myself somehow....
Пікірлер: 14