NVIDIA GeForce RTX 2070 Speed ​​Binning Exposed: TU106-400A vs TU106-400 – tech2.org

NVIDIA GeForce RTX 2070 Speed ​​Binning Exposed: TU106-400A vs TU106-400

EVGA GeForce RTX 2070 XC Ultra Games
EVGA is one of the few companies that releases multiple GeForce RTX series cards based on NVIDIA's benchmark specifications, Y at the reference price points (an important qualifier). That means you can actually buy a GeForce RTX 2070 for $ 499.99, instead of paying a premium for the NVIDIA Overclocking Founders Edition model, or any of the various overclocking SKUs of the companion board partners. NVIDIA (AIB). EVGA, for example, sells a GeForce RTX 2070 overclocked for $ 569.99. As a result, there may be more of a difference than just ready-to-use specifications. More on that in a moment.
For anyone who has not been following the launch of GeForce RTX, NVIDIA released three cards: GeForce RTX 2080 Ti, GeForce RTX 2080 and GeForce RTX 2070. All three are available in NVIDIA as models of Founders Edition, and unlike the previous generation ( Pascal), the FE models are overclocked. NVIDIA also reviewed the cooling solution, changing the blower style design to a cooler that is similar to what its AIB partners usually use.

NVIDIA also announced reference specifications at lower prices, with the presumption that its AIB partners would offer reference and overclocking models. This is how the prices are shaken:

  • GeForce RTX 2080 Ti Founders Edition: $ 1,199
  • GeForce RTX 2080 Ti Reference: $ 999
  • GeForce RTX 2080 Founders Edition: $ 799
  • GeForce RTX 2080 Reference: $ 699
  • GeForce RTX 2070 Founders Edition: $ 599
  • GeForce RTX 2070 Reference: $ 499
EVGA GeForce RTX 2070 Black Gaming
Now this is where things get interesting. Steve Burke in Gamers Nexus put in their hands two different models of GeForce RTX 2070 EVGA, and specifically these …
  • EVGA GeForce RTX 2070 XC Ultra Gaming: $ 569.99
  • EVGA GeForce RTX 2070 Black Gaming: $ 499.99

Burke removed both cards to see the GPU, and here they are a little different. The GPU under the hood on the Black Gaming card is a TU106-400-A1 and the GPU on the XC Ultra Gaming is a TU106-400A-A1. Both are Turing GPUs and are essentially the same in terms of specifications and features. However, it seems that revision "A" is a grouped part that can run faster and maintain its impulse clock for a longer period of time.

It is important to understand that not all chips are the same. That applies to all semiconductors, be it a CPU, GPU or RAM. In essence, you can buy two of the same processor model and see very different overclocking results. This is where binning comes into play. Companies like NVIDIA have the ability to test and clbadify chips according to the way they handle higher voltages and higher clock speeds.

At first glance, Burke seems to have discovered that NVIDIA is sharing its Turing chips, which in turn is selling two different versions to its board partners (at least for the GeForce RTX 2070). The TU106-400A-A1 probably costs more, so it ends up in higher priced cards with factory overclocks. Does it matter? You can, as Burke pointed out when running the two EVGA cards in 3DMark.

"The XC Ultra averages approximately 1935MHz for the entire test, while the 2070 Black experiences a clock drop from approximately 1815MHz to 1785MHz.The average frequency bounces between 1785MHz and 1800MHz for the 2070 Black.This is a difference significant for the exterior. "The frequencies of the box, and indicates the performance advantage resulting from the best-grouped GPUs," wrote Burke.

Execution in overclocking specifications compared to reference clocks in general equates to a few frames per second, depending on the game and settings. However, it is interesting that the difference between a stock clock and an overclocking card is not just the specifications, but the actual GPU. Something to keep in mind when buying a GeForce RTX card.

Source link