The RDNA structures with that happening as an idea sometime in early 2018. At the time, Lisa Su allocated resources for the transfer of developers to a team called Project Zen, led by Suzanne Plummer and when most of them came from the construction of the original Zen to enlarge and expand the art of creating new sculptures.
After the launch of RDNA, Plummer said they contributed part of the CPU team's experience of developing cores in the graphics team, helping with its approach and improving its frequency, performance and power. Gradually, he continues, they have taken the best they have developed internally and have tried to ensure that they apply the same development across the company.
Zen 1 has undergone a change in RDNA 1.0, the development of RDNA 2.0 will be huge
As he published on his Twitter account a few days ago his own Robert Hallock, an upgrade to be used by APUs Renoir with third-generation Vega GPUs can be up to 59% faster than the 2nd generation.
What does this have to do with RDNA 2.0? The fact that, arguably, the first signs of change the Plummer team has received has come to the APU before the introduction of a new technology for such drawings, ergo, is a revelation of what new graphics cards can be offered.
As such, you are not entirely sure how RDNA 2.0 will use the improvements in the internal circuits and transistors, and it is not known how much the component is built on, but what has been proven in most computer units and better durability. of frequency clock.
Sources also confirm card design: RX 6000, where they also confirm that they will be arriving sometime in the summer because they have to polish the details of the design and especially the efficiency of their drivers.
AMD RX 6000, Navi 21 and Navi 23: RDNA 2.0 arrives
The chips, it seems, will be Navi 21 and Navi 23, where they will pass through 500 mm2 without a problem, so with 7 nm + we should see significant improvement with regard to RDNA 1.0, not impossible to work with dedicated units Track Ray and as NVIDIA does with Tesa, the development should be very deep.
If Renoir launches and the leak is the right measuring stick, NVIDIA may face a bigger problem than perhaps at the beginning, because RDNA 1.0 as Zen 1 has been the forerunners of the biggest change we can make in just a few months.
And it seems that rumors point to a difference in performance between RDNA 1.0 and RDNA 2.0 which is very similar to the visuals between Picasso and Renoir itself. The question, which is pretty optimistic aside, is whether AMD can actually take such a huge step to work in one building, and that 59%, as Hallock said, is not an easy GAP to achieve among successive generations on the GPU.
