Blazing Fast Password Recovery With New ATI Cards

An anonymous reader writes "ElcomSoft accelerates the recovery of Wi-Fi passwords and password-protected iPhone and iPod backups by using ATI video cards. The support of ATI Radeon 5000 series video accelerators allows ElcomSoft to perform password recovery up to 20 times faster compared to Intel top of the line quad-core CPUs, and up to two times faster compared to enterprise-level NVIDIA Tesla solutions. Benchmarks performed by ElcomSoft demonstrate that ATI Radeon HD5970 accelerated password recovery works up to 20 times faster than Core i7-960, Intel's current top of the line CPU unit."

Comments

Stop with the advertising

[ShadowRangerRIT]

This isn't really about GPUs, it's an advert for ElcomSoft products. The whole summary is in marketing-speak for crying out loud.

Re:Stop with the advertising

[Lord Ender]

As an IT security guy, I found this to be informative, actually. When analyzing the security of a system or organization, I need to know not just what is theoretically possible, but what can be done with already-existing software and hardware.

This article gives me some idea as to what attacks are currently practical (and for what key lengths).

When research or engineering achievements come from the commercial (rather than academic) sector, it isn't really reasonable to expect an academic tone. They're tooting their own horn, but they are doing it about something important.

GPUs

[Thyamine]

This isn't the first story about how crazy fast GPUs are for crunching. I know very little about that level of hardware, but why aren't we incorporating these types of things into CPUs? Is the coding/assembly so different that it doesn't translate? Do they only do certain kinds of processing really well (it is a GPU after all), so it couldn't handle other more 'mundane' OS needs?

Re:GPUs

[ShadowRangerRIT]

That's not really the same thing. The Intel 80 core prototype was still a CPU at heart, they just made improvements to communication. GPUs are quite different. GPUs are designed as primarily floating point processors (though newer ones can do low precision integer math with similar efficiency), but more importantly, they are vector processors with virtually no support for conditional statements and optimized for sequential access to memory instead of random access. They're halfway between dedicated circuitry and a general purpose CPU; what they can do, they do *very* well, and they can generalize a little, but tasks they weren't designed for need to be rewritten to accommodate their quirks, and eventually reach a point of diminishing returns. Integrating GPUs into the CPU will allow more programs to use it (and possibly speed processing and enable new scenarios where the CPU and GPU need to communicate frequently), but for run of the mill computing tasks, the relatively inflexible design of GPUs is a problem.