Just a week ago, we published an article about NVIDIA’s new generation of Blackwell-based graphics cards. Despite a noticeable price hike, performance gains in this generation are minimal, with one notable exception: the flagship GeForce RTX 5090 significantly outperforms its predecessor in all key aspects. However, this GPU has also revealed a potential issue that could make its use in workstations running 24/7 problematic and potentially unsafe.

The newly introduced NVIDIA GeForce RTX 50 series (Blackwell architecture) brings significant changes. Notably, NVIDIA claims a doubling of integer (INT32) computation throughput per clock cycle compared to the previous Ada Lovelace architecture; this is described in the company’s whitepaper.

During the last several years, progress on the CPU performance front has seemingly stopped. Granted, last-generation CPUs are cool, silent and power-efficient. Anecdotal evidence: my new laptop (a brand new Macbook) is about as fast as the Dell ultrabook it replaced. The problem? I bought the Dell laptop some five years ago. Granted, the Dell was thicker and noisier. It’s battery never lasted longer than a few hours. But it was about as fast as the new Macbook.

Hardware acceleration of password recovery has been a hot topic for quite some time already. We were the first to adopt widely available graphic cards for this purpose and we’re proud of this. Today I’d like to share some thoughts on hardware acceleration for password recovery, its past, present, and future. I will also cover the most frequently asked questions regarding GPUs.