GPU apa yang direkomendasikan untuk pemecahan kata sandi hashcat dengan kinerja terbaik?

With the release of the 40 Series NVIDIA GPU Super Update, the company's product lineup has become quite complex. In the 4070 series alone, there are four models of NVIDIA GeForce RTX available: the original 4070, 4070 Ti, and now 4070 Super and 4070 Ti Super. Understanding the differences between these cards and which models offer the best value for money in password recovery efforts is a key consideration for IT professionals.

Why use GPU to crack passwords?

In some tasks, the massively parallel computing units of today's GPUs provide far more performance than a computer's CPU, while maintaining similar power consumption. Our products support hardware acceleration technology, leveraging the available computing power of modern GPUs to accelerate password recovery attacks. Our tools support GPUs built using AMD, Intel, and NVIDIA chips, but we always recommend NVIDIA graphics cards because they offer the greatest stability and reliability in a 24×7 environment.

Our acceleration technology offloads some complex computations onto fast and scalable compute units (NVIDIA CUDA kernels), resulting in 25 to 250 times faster password recovery compared to the CPU.

NVIDIA RTX 40 Series

The first GPUs released in the Ada Lovelace-powered 40 Series were the powerful NVIDIA GeForce RTX 4090 and 4080 motherboards released in late 2022. This is followed by lower-performance models. The initial lineup included the RTX 4090, 4080, 4070 Ti, 4070, 4060 Ti (8GB and 16GB versions), and 4060.

We benchmarked some of these GPUs in  NVIDIA RTX 40 series graphics cards: faster and more efficient password recovery accelerators, and concluded that the mid-range NVIDIA GeForce RTX 4070 Ti is faster and consumes less power than its predecessor, the champion RTX 3090. Compared to the older but similarly priced RTX 3070 Ti motherboard, the new RTX 4070  offers approximately 40% improved performance and approximately 30% lower power consumption. This provides a strong basis for recommending the Ada Lovelace generation over Ampere to perform password recovery tasks.

In January 2024, NVIDIA updated its lineup, adding several models labeled "Super". The "Super" version includes the 4080 Super, 4070 Ti Super, and 4070 Super; No updates were received for low-end GPUs. NVIDIA subsequently discontinued some older models (4080 and 4070 Ti), although they were still available for sale. Meanwhile, the 4070 and 4070 Super cards remain part of the company's current product line, while the RTX 4070 has been reduced in price. The new lineup isn't easy, especially considering the discontinued models are still retailing alongside the updated cards. When selecting a GPU for a password recovery job, it is important to understand the diversity of GPU models and the performance they offer.

The number of VRAMs is largely irrelevant for password recovery. All other things being equal, there is no difference in pure attack speed between the NVIDIA GeForce RTX 4060 Ti (8GB) and GeForce RTX 4060 Ti (16GB) variants, but the latter costs more than the former. However, compared to the now-discontinued 12GB 4070 Ti, the 16GB 4070 Ti Super not only doubles the amount of VRAM, but also expands the interface width from 192 bits to 256 bits, while increasing the number of execution units. Therefore, the "Super" version of the 4070 Ti offers a 10% performance improvement in double-precision calculations.

When it comes to performance, one can benchmark a graphics card in many different ways, while expressing the results in different units. We cannot test all GPU models in our setup, so for Apple-to-Apple comparisons, we rely on data from specification sheets and data obtained from independent sources. Specification list numbers for single-precision, double-precision, semi-precision, FP16, and TFLOPS; For password recovery tasks, only the double precision metric is relevant. Comparing different models (importantly, in the same architecture) based solely on this metric allows us to predict their relative performance fairly accurately. With this information, you can evaluate their performance in certain tasks; We recommend using the numbers in the test as a basis.

Note: Additional test data can be found in the full text.

Our value for money data is based on the U.S. Suggested Retail Price only. Street prices may and do vary; Keep this in mind when purchasing the most cost-effective circuit boards. Finally, make sure you understand the impact of the circuit board's TDP to build an efficient, well-cooled, and powerful setup.

Note: The above specifications are for reference only and may vary by OEM and specific model.

Based on this information, we can draw several conclusions:

1. The "Super" update to the RTX 4070 brings significant performance improvements with only a slight increase in power consumption. Although the original RTX 4070 remains in NVIDIA's lineup, we cannot recommend it even after the price reduction.

2. The "Super" version of the RTX 4080 did not bring a significant performance improvement, but it greatly improved the price-performance ratio, mainly due to the price reduction of newer models. This, in turn, means that a non-Super version of the RTX 4080 could be considered if it is offered at the same or lower price as the new Super variant.

3. The affordable RTX 4070 Super model offers excellent value for money, but only if purchased at the suggested retail price.

Navigating NVIDIA's 40 Series Super GPU updates involves considering factors such as performance metrics, pricing changes, and TDP.By understanding these factors, IT professionals can make informed decisions about which GPU models are best suited for their specific requirements and budget.