NVIDIA Responds to GTX 970 3.5GB Memory Issue

As I'm working a GTX 970 review I am reminded of some things that have been uncovered in the Maxwell generation.  In this case there is a performance issue under extremely high memory usage.  When addressing more than 3.5GB of memory the GTX 970 will run slower than a GTX 980 under the same conditions.

Having read the explanation over at PCPer I'm in full agreement with Josh.

A few days ago when we were presented with evidence of the 970 not fully utilizing all 4 GB of memory, I theorized that it had to do with the reduction of SMM units. It makes sense from an efficiency standpoint to perhaps "hard code" memory addresses for each SMM. The thought behind that would be that 4 GB of memory is a huge amount of a video card, and the potential performance gains of a more flexible system would be pretty minimal.

I believe that the memory controller is working as intended and not a bug. When designing a large GPU, there will invariably be compromises made. From all indications NVIDIA decided to save time, die size, and power by simplifying the memory controller and crossbar setup. These things have a direct impact on time to market and power efficiency.  NVIDIA probably figured that a couple percentage of performance lost was outweighed by the added complexity, power consumption, and engineering resources that it would have taken to gain those few percentage points back. ~ Josh Walrath PCPer

It makes sense that cutting down the GPU would have an impact on how the framebuffer is being used and might explain "why" chips are cut down the way they are.  It might also explain why most Nvidia graphics cards only came with 3GB/6GB of memory framebuffer. 

Either way the performance problems are only under extreme conditions and while the GTX 970 might be a good price match most gamers the conditions that cause the performance issues are likely best suited for GTX 980 cards anyway.

Related Web URL: http://www.pcper.com/news/Graphics-Cards/NVIDIA-Re...