Nvidia’s GTX 680 arrives! “Faster, Smoother, Richer” – is it enough to take the Performance Crown?
Nvidia has finally released its long-awaited GeForce GTX 680 based on its brand new 28nm Kepler DX11.1 architecture today. This new GPU – Graphics Processing Unit – a term originally originated by Nvidia is a continuation of their strategy since G80 launched over six years ago to create a General Purpose Processor – co-equal with the CPU – that also renders amazing graphics.
The GTX 680 is the culmination of five years of Nvidia’s efforts with their new DX11.1 Kepler architecture, their flagship GPU and the replacement for the GTX 580 which launched at the end of 2010. Nvidia realizes that they are 2-1/2 months later than their rival AMD graphics with 28nm DX11.1 video cards. It appears that Nvidia waited to guarantee that there would be enough chips to supply a big demand when it finally launches today as yields on the new 28nm chips are less than ideal.
In fact, AMD has already launched much of its HD 7000 DX11.1 series with new premium pricing for their HD 7970 at $550, down to their HD 7750 for $129. So today, NVIDIA has launched its new Kepler GeForce lineup with its GTX 680 flagship The GTX 480 comes with a MSRP of $499. So we need to answer the question: How does it compare to AMD’s top single-GPU video card – the HD 7970?
To properly bring you this review, we purchased a PowerColor HD 7970 from NewEgg for $550 (above) and put it through its paces this week with the very latest WHQL drivers – Catalyst 12-2, released March 7. This driver set should be quite mature as AMD has had ten weeks to fine tune their HD 7970 drivers. In contrast, the release GTX 680 GeForce 300.99 drivers that we are using are still beta and should leave some room for improvement by Nvidia’s driver team in the months to come.
So you will see us pit our PowerColor reference design HD 7970 (above right) – at stock as well as overclocked – against the stock and overclocked GTX 680 (above left) using 23 modern games and 3 synthetic benchmarks mostly using 1920×1200 and 2560×1600 resolutions. We shall also compare Nvidia’s 3-panel Surround working now off of a single GTX 680 versus the HD 7970 in Eyefinity at 5760×1080 resolution. Lastly, we also bench 3D Vision 2 and PhysX, ‘on’ versus ‘off’ at the very popular 1920×1080 resolution.
We are also comparing the performance of our reference dual-GPU video cards, HD 6990 and the GTX 590 (above) as well as the very fastest video cards of AMD’s and Nvidia’s last 40nm generation, the HD 6970 and the GTX 580 (below).
What’s New with Kepler’s GTX 680?
One of the best three minute promotional videos of a new launch was (accidentally) posted by Nvidia yesterday. It is well worth watching and will give you a great introduction to the GTX 680:
Nvidia’s marketing buzzwords for the GTX 680 launch are, “Faster. Smoother. Richer.” The GTX 680 is designed for extreme efficiency and high performance.
The GTX 680’s Kepler architecture is now SMX-based with 1336 CUDA cores. It promises better geometry and texture processing than Fermi thanks to its improved instruction throughput and redesign. In addition, Nvidia brings “GPU Boost”, a dynamic way to boost clocks speeds and maximize performance for each game.
New kinds of anti-aliasing – FXAA and TXAA – are now said to compete with MSAA in terms of IQ while not sacrificing performance. And there is a new “Adaptive VSync” that is supposed to reduce tearing and stuttering associated with regular VSync. Great hardware needs great software to support it and Nvidia is also a software company.
For the first time, it is now possible to play games spanning 3 displays in Surround or in 3D Surround off of a single GeForce GPU, the GTX 680. And the card can also drive a fourth accessory display. PhysX has also been improved.
How does the GTX 680 compare with its rival, AMD’s HD 7970?
This evaluation attempts to analyze and compare GTX 680 and HD 7970 performance and hopefully we can announce a performance winner. We will also look at the details to see what the new Nvidia Kepler GPU brings to the table. We also believe we have a good handle on how AMD is going to respond to Nvidia’s GTX 680 launch and we will share our analysis and insights with you.
The competing cards
The GTX 680, GTX 580, the GTX 590, the HD 7970, the HD 6990, the HD 6970 are the top cards from Nvidia and AMD of this and the last generation and we will see where they sit in relation to each other. The HD 6990 and the GTX 590 are dual-GPU cards and are considered the fastest production video cards currently. And it is important to see how much performance increase the GTX 680 has brought over the GTX 580.
Since we do not want any chance of our CPU “bottlenecking” our graphics, we are testing all of our Graphics cards by using our Intel Core i7-920 at 4.20 GHz (using the 21x multiplier in turbo mode), 6 GB Kingston DDR3 and a Gigabyte X58 full 16x + 16x PCIe CrossFire/SLI motherboard. This Core i7-920 although more than three years old easily competes with a mildly overclocked Core i7-2700K and is sufficient to differentiate high end video cards at high resolution and high detail settings. Before we do performance testing, let’s take a look at the GTX 680 and quickly recap its new Kepler DX11.1 architecture and features.