Intel Labs has announced two new chunks of test silicon that expand the company's definition of what it considers "job one" in terms of product development. "I've given our company a charter to make [security] job one," Justin Rattner, director of Intel labs quoted CEO Paul Otellini as saying in an interview earlier this year …
Imitation is the sincerest form of flattery
All hail the wonderful Intel innovations which are a copy of what Via delivered 5+ years ago. My 6 year old MS1000 which I use for a firewall can do that and the more recent ones can do more - RSA, RNG, etc.
Let's innovate!!! By copying from companies you bulldoze down through monopolistic practices. Viva the Intel way!!!
All hail VIA - not!
VIA is a has-been (or a "never-was"?) because they produced crappy CPU's, slow and bug-ridden chipsets, S3-derived graphics silicon that even Intel beats, and worst of all, BSOD guaranteed drivers. Good riddance.
Innovations by VIA - I can't name any, can you?
Intel is not innovating per se - hardware encryption is not exactly a new thing, they are just claiming they have the fastest AES encryption silicon. That's not copying.
if this chip will be able to get used in brute-force attacks...
Re: Brute force
Crypto that can be brute-forced (i.e. WEP) most likely does not even need specialized hardware. But if there are 2^512 possibilities, a 10x speed boost won't matter that much.
"What he was talking about is the increasing attention we're giving to security at various levels in the system, said Rattner. How can we make our products more robust in the face of attacks of all sorts — viruses, and worms, and rootkits, and all kinds of malware — as well as making them more capable of protecting secrets even in the face of attack?"
Here I was thinking that one can't fix a broken OS in silicone. The cpu doesn't really know the intent of the code it's executing. Executing a virus is not a bug if the OS allowed it in the first place, no?
One area which could use improvement is a mechanism to lock down the bootloader such that it cannot be modified - even by the OS, without user approval (requiring either physical, or network authentication). This way, even a vulnerable OS could be restored to a known state simply by rebooting.
I'm all for the RNG too!
Am I confused ...
... but isn't it incorrect to use 'pseudo' as the opposite of 'high-quality' when discussing random number generators? I thought there were such things as high-quality pseudo RNGs, and that the only RNGs that weren't 'pseudo' used a real source of randomness, like quantum effects (e.g. elecrical noise).