A few months have passed since we saw how NVIDIA managed to reach the top spot of the most valuable companies in the world, all thanks to the creation of a specific type of hardware for a technology that, for some reason or another, we all know that. The importance he had Nvidia in the artificial intelligence sector is extremely large, so its main competitors did not even have the opportunity to enter this market, which they also intend to dominate next year with their next GPU new generation for AI, B300.
Hardware for AI is one of the biggest implementations we've seen in recent years. Many people and companies have been looking for a way to obtain a series of specific components that allow them to get the most out of the power it is capable of. to offer. this technology. This is where Nvidia stands out, because they have not only focused on developing components for large servers, but also on their traditional graphics cards as well as the material has AI The consumer-oriented approach they have developed has enabled everyone to take full advantage of the potential the company offers.
NVIDIA reportedly managed to improve the performance of its GPUs for AI by 50%
One of the goals that practically all brands that manufacture components such as processors or graphics cards have had in recent years is to try to offer the greatest possible capacity to use artificial intelligence, either through a NPU, or directly thanks to the capacity they have. For example, GPUs calculate millions of operations per second. In this regard, being a company that has dominated the graphics card sector for many years, NVIDIA has played a great role in the development of these components, but apparently things are only getting better for the brand.
We know various information about the new B300 that the company plans to launch next year, some GPU designed for servers offering extremely high power, up to a 50% higher to what the generation offered B200. All this has a cost, as we know well, and that is that one of the leaks that we have experienced in addition to this increase in power is also the increase in consumption, which would be around 200 W higher than that of the previous generation, a total of 1,400 W.
But compared to the fact that this significantly increases performance, the increase in consumption is actually not that high. One of the main reasons why they will be able to outperform the previous generation is the use of 12-Hi HBM3E memory stacks, which will provide 288 GB of memory with 8 TB/s of bandwidth. On the other hand, they could also adopt the 800G ConnectX-8 NIC that they developed themselves and which has twice the bandwidth of the current 400G ConnectX-7, in addition to offering 48 PCIe lanes versus the 32 of their predecessor.
