AES 512 coin ? - coin for nvidia or i7?

[tytanick]

Hi folks, what about creting aes 512bit coin ?
Newest cpus like i7 have hardware aceleration of aes.
How about gpus, how much faster they are in aes encryption comparing to cpus ?
Meaby creating aes based alghoritm coin would be so stupid ?
Comment this ...

[tytanick]

Hmmm, am i seeing right?
are geforce cards have similar performance in aes entryption as radeon ?
http://www.purepc.pl/files/Image/artykul_zdjecia/2012/Test_Tanie_GPU/openclaes.png

[temor]

Aes is encrypting, not hashing.

[tytanick]

Aes is encrypting, not hashing.

hmm, so it cant be used in coin generation?

[ElectricMucus]

Aes is encrypting, not hashing.

hmm, so it cant be used in coin generation?

It could.
Lookup "AES hashing"

[Balthazar]

There are many cryptographic hash functions, which are based on AES and another block ciphers. The main examples are Groestl, Lesamnta and another SHA3 candidates.

P.S. I planned to use Groestl in NVC, but replaced it with usual scrypt implementation.
P.P.S. Remember that encrypting performance in AES tests limited by PCI-E and memory bus. This test results are not valid for AES hashing.

[tytanick]

There are many cryptographic hash functions, which are based on AES and another block ciphers. The main examples are Groestl, Lesamnta and another SHA3 candidates.

P.S. I planned to use Groestl in NVC, but replaced it with usual scrypt implementation.
P.P.S. Remember that encrypting performance in AES tests limited by PCI-E and memory bus. This test results are not valid for AES hashing.

So in that case, cpus will be more less the same performance as gpus?

Because it would be good coin that is gpu resistant !

[perhan007]

There are many cryptographic hash functions, which are based on AES and another block ciphers. The main examples are Groestl, Lesamnta and another SHA3 candidates.

P.S. I planned to use Groestl in NVC, but replaced it with usual scrypt implementation.
P.P.S. Remember that encrypting performance in AES tests limited by PCI-E and memory bus. This test results are not valid for AES hashing.

So in that case, cpus will be more less the same performance as gpus?

Because it would be good coin that is gpu resistant !

Yeah, good for the botnet-criminals...

[Remember remember the 5th of November]

Actually it's a paradox. If it's GPU-resistant it's CPU friendly and botnets will dominate. If it's GPU minable, large GPU farms and/or ASICs will dominate. The way I see it, Proof of Stake without Proof of Work beats this i.e PPC.

[Come-from-Beyond]

Actually it's a paradox. If it's GPU-resistant it's CPU friendly and botnets will dominate. If it's GPU minable, large GPU farms and/or ASICs will dominate. The way I see it, Proof of Stake without Proof of Work beats this i.e PPC.

U still need proof-of-work at the very beginning. Or pre-mine.

[tacotime]

Groestl's 256-bit implementation is vulnerable to collision attacks -- I think you end up with 2^64 possible hashes in the worst case.  If I remember right a lot of the AES-function utilizing secure hash functions suffer from this problem.

[Balthazar]

So in that case, cpus will be more less the same performance as gpus?

No, it isn't. I mean that this test says nothing about potential performance in AES hashing. Performance of AES crypting is limited by memory and PCI-E bus due to high amounts of transmitted data.   But for mining we don't need this, because we have midstate and 4 byte nonce field.