Featured

Nvidia GTX 750 Ti 2GB “Maxwell” Graphics Card Review

Testing Procedure

Methods

To test the performance of video cards at eTeknix we run a variety of tests at four different resolutions, where supported and appropriate, across a suite of games and benchmarks. We run each of these benchmarks three times to take an average and use the latest WHQL-certified video card drivers from AMD and Nvidia to test with.

To attain Furmark load noise levels we hold a decibel meter approximately 3 inches away from the card after running Furmark for five minutes. To attain 3DMark Load noise we do the same in the last stage of a 3DMark 11 benchmark run. To achieve idle noise we allow the system to stay idle at the desktop for 5 minutes before taking a noise reading.

To achieve Furmark load temperatures we take the maximum temperature of the GPU core during Furmark load as recorded by CPUID HW Monitor. To achieve idle temperatures we take the minimum recorded GPU core temperature in CPUID HW Monitor after desktop-idling for 5 minutes. We also do the same in 3DMark 11 to simulate a more realistic load. We then convert these temperatures into Delta temperatures – that is = actual recorded temperature minus the ambient temperature of the room. Note we do not alter the default fan profile in any way, all graphics cards are left to run at their stock settings.

To measure power consumption we take the maximum power consumption at idle on the desktop over a 5 minute period, and for load we take the maximum stable power consumption during Furmark load. We also do the same during a 3DMark 11 run, taking the maximum stable power consumption from the last combined test only. To address recent developments in thermal throttling (and Furmark throttling more specifically) we take our measurements within the first 30 seconds to a minute on cards that dramatically clock down when hot (e.g. the R9 290X) as this causes there power consumption to appear much lower than it actually can be.

To formulate our performance metric we use a weighted average method and each benchmark or test gets a weight of 1 out of a total calculation with 9 weights (meaning all resolutions and variations of the same test are only ever one weight). Those 9 weights are 1 for each test we do excluding Battlefield 4 (so 6 games and 3 benchmarks) which is omitted as we do not have results for all the graphics cards in that game. We then calculate the relative score for each card in each test (by dividing the cumulative frame rate/score of each card by the average cumulative frame rate/score from all the graphics cards in group). We then take those 9 relative scores and average those for each card, before dividing those averages by the average of the card being reviewed in the test so that the value of the card being reviewed is 1 and every other card’s performance metric is in relation to it. The end result are numbers that vary around 1 with 0.5 being half the performance of the tested card and 2 being double the performance of the tested card. We then convert those numbers into percentages to make them more readable so 1 becomes 100% and 0.5 becomes 50%.

Test System:

Motherboard – Asus Rampage IV Extreme X79 LGA 2011 (chipset fan disabled)
Processor – Intel Core i7 3960X at stock clock speeds of 3.3GHz with Turbo Mode disabled.
RAM – 16GB (2 X 8GB) Corsair Vengeance Pro Series 1866MHz at 9-10-9-27
CPU Cooler – Corsair H100i with Quiet Fan Profile
Power Supply – Corsair HX1050W
Main Storage Drive – Kingston HyperX 240GB SSD over SATA III interface
Chassis – Lian Li T60 Test Bench
Displays – Dell U2711 Ultra Sharp for 2560 by 1440 and we use the LG IPS 234 & LG IPS 224 with the Dell U2711 Ultra Sharp to run 5760 by 1080.
Operating System – Windows 7 Ultimate 64 Bit

We would like to thank Asus, Corsair, Kingston, Lian Li and all our other partners who supplied us with test equipment and hardware. Their generosity makes our testing possible and without them we wouldn’t be able to produce the reviews we do, so thank you!

Games Used