Nvidia’s upcoming graphics cards may be delayed until the summer

In May 2016, Nvidia launched the graphics architecture Pascal, which is still the company’s latest contribution to the consumer-oriented Geforce GTX series. For some time, rumors of naming the product line and architecture have replaced each other, with the GTX 1100 series and the Ampere and Turing architectures being the latest bids.

Many expected the new graphics cards to be presented during this year’s edition of the GPU Technology Conference (GTC), which ended on March 29. The focus of the presentations was instead company-oriented solutions adapted for deep learning and artificial intelligence, where Nvidia unveiled the ready-made system DGX-2 with 16 Tesla V100 and 512 GB HBM2 memory under the hood.

After talks with a source, Gamersnexus now reports that Nvidia’s next products in the Geforce GTX series will probably be launched in July at the earliest. This is based on information that the memory manufacturer SK Hynix in June will start mass production of GDDR6 memories, which is the type of memory that is said to take place in most products in the next generation.

The amount of memory will end up at 8 or 16 Gb and the capsules will use the voltage 1.35 V. They will also be more energy efficient than its predecessor, but initially also about 20 percent more expensive when it comes to manufacturing costs. It is worth mentioning, however, that Samsung, among others, has already started production of GDDR6 memories, and that SK Hynix is ​​not necessarily the only memory supplier prior to the launch.