Testing

Test Rig

  • CPU: Intel Core i7 3770K @ 4.2GHz
  • Motherboard: Asus Maximus V Formula
  • RAM: Corsair Vengeance 1600MHz 9-9-9-24 16GB
  • Storage: Crucial C300 128GB
  • PSU: SuperFlower Golden king 550W 80+ Platinum
  • CASE: Coolermaster Cosmos 2

Latest driver was downloaded from the manufacturer’s website.

 

Methodology

For graphics card testing, here at XSReviews our aim is to provide a set of clean, reliable benchmarks. For each test, the test was repeated three times and then an average taken or where results were inconsistent, tests were continued until two concordant results were reached.

The benchmarks consist of a mixture of games and synthetic benchmarks; game testing is carried out through a specifically designed benchmark suite for that particular title or by using a time demo and the utility FRAPS to record the frame rates. The following software was used in the testing of the graphics cards:

  • 3Dmark 11
  • 3Dmark Vantage
  • Furmark – for temperature testing
  • Unique Heaven 3.0
  • DIRT 3
  • Call of Juarez (1080P, High, MSAA X4)
  • Resident Evil 5 Bench (Max, 1080P)
  • Streetfighter(Max, 1080P)
  • FFXIV(Max, 1080P)
 

3Dmark 11

3Dmark11 is the latest benchmark released from industry standard firm Futuremark. It is being touted as the de-facto way to test DirectX 11 enabled graphics cards and their capabilities of processing the latest 3D features found in modern games.

[easychart type=”horizbar” title=”3Dmark11 Score” groupnames=”6950, 7850, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”Score” group1values=”5369″ group2values=”6735″ group3values=”4717″ group4values=”9112″ group5values=”8950″ group6values=”9683″ ]

Futuremark describes the 3Dmark score as “an overall measure of your system’s 3D gaming capabilities, based on comprehensive real-time 3D graphics and processor tests. By comparing your score with those submitted by millions of other gamers you can see how your gaming rig performs, making it easier to choose the most effective upgrades or finding other ways to optimize your system.”
The GTX680 AMP! is in a league of its own, showing off its factory overclock.

 

3Dmark Vantage

3DMark Vantage is the industry standard performance benchmark for DirectX 10 gaming PCs. It includes two graphics tests, two CPU tests and six feature tests. Trusted by gamers worldwide to give accurate and unbiased results, 3DMark Vantage is the best way to consistently and reliably test DirectX 10 under game-like loads. PhysX has been disabled for this test to make it fair against the AMD cards.

[easychart type=”horizbar” title=”3Dmark Vantage ” groupnames=”6950, 7870, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”Score” group1values=”21446″ group2values=”26131″ group3values=”20312″ group4values=”32246″ group5values=”32416″ group6values=”33483″ ]

In Vantage again we see very close scores, this GTX680 AMP! is far above any other card tested to date.

 

Unigine Heaven

“Heaven Benchmark is a DirectX 11 GPU benchmark based on advanced Unigine™ engine from Unigine Corp. It reveals the enchanting magic of floating islands with a tiny village hidden in the cloudy skies. Interactive mode provides emerging experience of exploring the intricate world of steampunk.”

[easychart type=”horizbar” title=”Unigine Heaven 3″ groupnames=”6950, 7870, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”FPS” group1values=”50.2″ group2values=”66.7″ group3values=”33.6″ group4values=”85.2″ group5values=”84.4″ group6values=”90.8″ ]

The settings used are:  Shaders – High, Tessellation – Normal, AA – Off, Anistropy – 4, 1280 X 1024 Full screen.

Five FPS may not seem huge at this speed but if the settings were higherthose 5 FPS become much more important, especially in the 25 – 30 FPS range.

 

DIRT 3

Dirt 3 (stylised DiRT 3) is a rallying video game and the third in the Dirt series of the Colin McRae Rally series, developed and published by Codemasters. Featuring DX11, it’s a great test for the latest graphics cards.

[easychart type=”horizbar” title=”Dirt 3″ groupnames=”6950, 7870, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”Ultra, High, Low” group1values=”41.09,149.46,160.73″ group2values=”80.26,187.11,211.71″ group3values=”60.29,143.22,168.12″ group4values=”110.81,213.67,232.41″ group5values=”110.96,212.65,234.12″ group6values=”118.35,214.71,235.53″ ]

As expected, the GTX680 AMP! has a lead here, it’s close but we’re starting to see a CPU bottle neck here.

 

Call of Juarez (1080P, High, MSAA X4)

[easychart type=”horizbar” title=”Call of Juarez” groupnames=”6950, 7870, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”FPS” group1values=”80.1″ group2values=”103.7″ group3values=”74.7″ group4values=”131.41″ group5values=”134.3″ group6values=”142.9″ ]

11 FPS higher here, that’s an impressive increase over a normal GTX680.

 

Resident Evil 5 Bench (Max, 1080P)

[easychart type=”horizbar” title=”Resident Evil 5 Bench” groupnames=”6950, 7870, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”FPS” group1values=”123.3″ group2values=”156.6″ group3values=”104.8″ group4values=”168″ group5values=”171.5″ group6values=”178.4″ ]

Again, we see a 10 FPS increase.

 

Streetfighter(Max, 1080P)

[easychart type=”horizbar” title=”Streetfighter” groupnames=”6950, 7870, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”FPS” group1values=”143.59″ group2values=”181.01″ group3values=”175.1″ group4values=”274.23″ group5values=”273.51″ group6values=”289.27″ ]

Street fighter sees some really high rates.

 

FFXIV(Max, 1080P)

[easychart type=”horizbar” title=”FFXIV” groupnames=”6950, 7870, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”Score” group1values=”3977″ group2values=”49989″ group3values=”4338″ group4values=”6952″ group5values=”6974″ group6values=”7279″ ]

The GTX680 AMP! Leads by over 300 points, excellent!

 

Overclocking and Thermals

As usual we performed temperature testing, a couple of seconds into running Furmark on Extreme settings I found the GPU to be getting quite hot, the temperatures were safe but still higher than I had hoped to achieve.

After some research I found adding washers to the heatsink screws helped by adding extra pressure between the GPU and cooler, temperatures dropped to just 75C, a much improved temperature.

[easychart type=”horizbar” title=”Temperature” groupnames=”6950, 7870, 560 Ti, GTX 680, GTX 670 AMP!, GTX 680 AMP!” valuenames=”Idle, Load” group1values=”47,78″ group2values=”41,67″ group3values=”33,76″ group4values=”42,84″ group5values=”33,65″ group6values=”39,75″ ]

After talking to Zotac, it seems this won’t be an issue with retail cards and more likely that this was caused because it’s a pre-release review sample that’s been passed around a few times to different reviewers.

At stock settings the card idles at 39C, the load temps, as noted above reached 75C, 10C more than the GTX670 AMP! And importantly 9C lower than a reference card. At idle the card was almost silent, atleast inaudible to my ears, under load the fan did begin to speed up but was still quite fairly quiet and again, much quieter than a reference GTX680. In actual game play you find yourself very rarely hearing anything from the card, even without a side panel and sitting right next to you.

I installed MSI AfterBurner and set the power target to 132%. With Heaven bench running the card set itself to 1176MHz.  We reached around about 1225MHz and that seemed to be the limit of this card.

About The Author
William Hemmens