nVidia GTX285 Bottleneck Investigation

Introduction

This article is another installment in the ongoing investigation of bottlenecking in video cards, with previous installments having been done on a 8800 Ultra and a GTX260+.

To quickly recap on the basic premise: nVidia hardware allows separate core, shader and memory clocks. Therefore, underclocking one of these values while leaving the rest at stock allows you to find out what part of the GPU is the biggest bottleneck in any given situation.

There’s a limit to how far the core and shader clocks can separate from each other and in the case of the GTX285, it’s a drop of 1476 MHz to 1296 MHz, or about 12%. This is smaller than the 8800 Ultra’s possible drop of 19%, but is higher than the GTX260+’s 7% limit. It’s interesting how these cards differ with this maximum value.

I’ll run the card at stock speed followed by underclocking each clock by 12%, leaving the rest at stock to see which has the most performance impact. Along with three new games being used, Far Cry 2 will also be retested with a better benchmark.

Stock clocks are as follows: core 648, shader 1476, memory 1242.

The underclocked values and corresponding colors are as follows: core 570, shader 1296, memory 1093, each with an underclock of approximately 12%.