BabelTechReviews was at Nvidia’s GTX 1080 Austin Launch event last week, May 5-8, which included the Editor’s Press Day on Saturday. We attended a deep dive into the features and architecture of the new video card under a Non Disclosure Agreement (NDA). Earlier on Friday, Nvidia’s CEO Jensen had just fully unveiled the new DX12, fully VR-enabled, 7.1 billion transistor Pascal flagship GTX 1080 at $699 for the Founder’s Edition built by Nvidia, with a promise of partner cards starting at $599. With 8GB of vRAM, Nvidia claimed that the GTX 1080 will be significantly faster than the GTX TITAN X, the GTX 980 Ti, GTX 980 SLI, and much faster than AMD’s flagship, the Fury X. Plus, it is supposed to be able to overclock very well. These are some big claims, and today we will give our own testing results to you.
BTR received a GTX 1080 from Nvidia to take home with us, and we have put it through its stock and even overclocked paces with our updated 26-game PC benchmark suite against the GTX TITAN X and GTX 980 SLI, and also versus the Fury X, AMD’s flagship.
We are testing all of our competing cards on a clean installation of Windows 10 64-bit Home edition, using resolutions of 1920×1080, 2560×1440 and at 4K’s 3840×2160. As befits testing top video cards, we use Intel’s enthusiast Z97 platform with Core i7 4790K turboed to 4.4GHz by the BIOS, and 16GB of Kingston’s 2133MHz Predator DDR3. All games are patched to their latest versions and we are using the latest drivers for AMD’s Fury X as well as for the GTX 1080, GTX 980 SLI, and TITAN X.
Before we give you the results of our performance testing, we want to briefly cover Pascal architecture, as well as detail the specifications and features of the new GTX 1080. Since we benchmark 26 games, we have a much larger benchmark suite than any other English-speaking tech site in the world, so we are going to concentrate on performance and we will only briefly summarize the new features of the GTX 1080 and will go into further detail in future articles. But first, let’s bring our readers up to date.
The GTX 980 and the GTX 970 both launched in September 2014 as Maxwell GM204 architecture – a mid-sized chip – and both of them were faster than AMD’s flagship at the time, the R9 290X. Afterward, Nvidia launched their $1000 TITAN X as GM110, a big chip Maxwell GPU with 12GB of vRAM followed quickly by their 6GB-equipped flagship gaming card, the GTX 980 Ti for $649 which is slightly faster than the TITAN X in most games. In the meantime, AMD rebadged their 200 series lineup into 300 series, renaming the 290X into the 390X and equipping it with 8GB of slightly faster vRAM. AMD also brought out their new Fiji Flagship, the Fury X at the same $650 pricepoint as the GTX 980 Ti, but its performance generally fell short of the GeForce card. It was also touted by AMD as an Overclocker’s Dream, but Fury X did not meet expectations as they generally barely overclock.
Pascal offers large increases in performance, memory bandwidth, and power efficiency over the current Maxwell architecture. It introduces new graphics features and technologies that confirm the PC as the ultimate platform for playing AAA games and for enjoying virtual reality.
Nvidia has engineered the Pascal architecture to handle the demanding computing and gaming needs of technologies like VR. It incorporates five new technologies:
- Next-Gen GPU Architecture. Pascal is optimized for performance per watt. The GTX 1080 is 3x more power efficient than the Maxwell Architecture.
- 16nm FinFET Process. The GTX 1080 is the first gaming GPUs designed for the 16nm FinFET process, which uses smaller, faster transistors that can be packed together more densely. Its 7.2 billion transistors deliver a significant increase in performance and efficiency.
- Advanced Memory. Pascal-based GPUs are the first to use 8GB of Micron’s GDDR5X memory. The 256-bit memory interface runs at 10Gb/sec., helping to drive 1.7x higher effective memory bandwidth than that delivered by regular GDDR5.
- Superb Craftsmanship. Increases in bandwidth and power efficiency allow the GTX 1080 to run at really high clock speeds over 1700 MHz while only using 180 watts of power. New to Pascal is asynchronous compute. And new GPU Boost 3 technology supports advanced overclocking functions.
- Groundbreaking Gaming Technology. New VRWorks software features let game developers bring more immersiveness to gaming environments. And Nvidia’s Ansel technology lets gamers share their gaming experiences and explore gaming worlds in new ways.
The next generation of games will not only look better but run faster on the GeForce GTX 1080. Nvidia has developed a number of advancements for virtual reality reducing latency, improving image quality, and bringing a whole range of new content to VR.
The enhanced Nvidia VRWorks software development kit offers a higher level of “VR presence” by combining what users see, hear, and even touch with the physical behavior of the environment.
- 2x VR Graphics Performance: VRWorks Graphics now includes a simultaneous multi-projection capability that renders natively to the unique dimensions of VR displays instead of just to traditional 2D monitors. It also renders geometry for the left and right eyes simultaneously in a single pass.
- Enveloping Audio: VRWorks Audio uses the Nvidia OptiX ray-tracing engine to trace the path of sounds across an environment in real time which better reflects the size, shape, and material of VR than previous solutions.
- Interactive Touch and Physics: Nvidia PhysX for VR detects when a hand controller interacts with a virtual object, and enables the game engine to provide a physically accurate visual and haptic response.
Nvidia has integrated these technologies into a new VR experience called the VR Funhouse.
The GTX 1080 Pascal GPU
The GTX 1080 GPU has 64 Raster Operating Units, 20 SMs of 128 Cores each totaling 2560 CUDA cores, 20 Geometry units and 160 Texture units. It uses Micron’s 256-bit GDDR5X at 10 Gbps which makes it significantly faster than GDDR5 and its 1.61GHz GPU clock has a boost of 1.73GHz or higher. We also easily managed a 2GHz clock with complete stability although we only had time to spend a few minutes with overclocking by using the new beta Precision XOC.
There are 7.2 billion transistors packed into the GTX 1080 GPU, and it uses the 16nm FinFET process which means the GTX 1080 has only a TDP of 180W thanks to a completely new GPU and a more advanced board design than the 28nm Maxwell architecture. And as befits a new architecture, Pascal uses a more advanced and efficient memory compression system
Simultaneous Multi-Projection (SMP) is a new technology that allows Pascal GPUs to render more efficiently on modern displays. SMP allows Pascal to project to 16 different viewports, each with the ability to rotate in order to better match the display. It also has the ability to project more than one viewport simultaneously. SMP is used for multiple new graphics techniques such as Perspective Surround, Lens Matched Shading, Single Pass Stereo, and MultiRes Shading. When used for virtual reality, SMP technologies can result in up to 1.5x pixel throughput, and 2x geometry throughput.
Think of a viewport as a 2D transform to scale and translation of the image plane. A viewpoint is basically a 3D transform complete with a perspective projection. Pascal allows you to efficiently render two viewpoints at the same time without submitting geometry twice.
Simultaneous Multi-projection is handled by the Pascal Polymorph Engine 4.0 as a hardware function of the GPU. It is especially designed for Virtual Reality so much more can be done with less resources as illustrated below:
Other features of Pascal – Async Compute, Dynamic Load Balancing, and Graphic Pipeline Preemption are all important upgrades over Maxwell architecture. Asynchronous Compute is used in physics, post processing and in VR, and it will be significantly faster than Maxwell in games that use it like Ashes of the Singularity, as you will see. Dynamic load balancing means the GPU can now spend less time idling and can do more work, while preemption allows the GPU to pass the graphics load to system memory quickly and without penalty. Preemption is particularly useful for VR where keeping latency low is absolutely crucial.
All of this adds up to make the Pascal GTX 1080 a speed demon by virtue of its faster clocks and improved architecture where Nvidia estimates that the GTX 1080 is approximately 1.7 times faster than the GTX 980 in gaming, and 2.7 times faster in VR.
Pascal GPUs also have support for HDR (High Dynamic Range) displays by virtue of having HDMI 2.0b and DP 1.4 built in. It even supports HDR gaming and streaming to HDR TVs which are coming soon. The GeForce GTX 1080 is DisplayPort 1.2 certified, DP 1.3/1.4 Ready, enabling support for 4K displays at 120Hz, 5K displays at 60Hz, and 8K displays at 60Hz (using two cables).
The GeForce GTX 1080 Founder’s Edition includes three DisplayPort connectors, one HDMI 2.0b connector, and one dual link DVI connector. Up to four display heads can be driven simultaneously from one card. The GTX 1080 display pipeline supports HDR gaming, as well as video encoding and decoding. New to Pascal is HDR Video ([email protected] 10/12b HEVC Decode), HDR Record/Stream ([email protected] 10b HEVC Encode), and HDR Interface Support (DP 1.4).
Something New – Fast Sync and GPU Boost 3.0
Nvidia has already brought us GSync which decouples the display from the render and is mostly useful for lower framerates so that there is no tearing nor other visual artifacting while maintaining a smooth and fluid display of frames. Fast Sync is primarily used for high frame rate situations where game frame rates far exceed the refresh rate of the monitor. It is especially useful for eSports titles that have high frame rates and require low latency and fast response times. Fast Sync delivers the low latency of Vsync Off, with no tearing.
Fast Sync is available in the Pascal Control Panel, but we haven’t had a chance to play with it yet, and we will give you a full report when we can evaluate it properly. It looks like a great solution to have low latency with no tearing at high framerates.
SLI and new SLI Bridges
Pascal requires more bandwidth than Maxwell, and the old SLI bridges are not sufficient so Nvidia has designed new ones.
Nvidia has indicated that they will continue to allow for more than two video cards to be linked in SLI, but that they are concentrating on only supporting two cards.
Nvidia has improved their GPU boost from version 2.0 to 3.0. Maxwell uses a fixed frequency offset but Pascal uses a per voltage point frequency offset for more flexibility for overclocking. EVGA’s Precision XOC beta now includes an overclocking scanner that automatically determines the frequency offset curve according to the theoretical maximum GPU clocks. It determines this for each voltage point by running a built in benchmark, allowing users to generate a custom frequency offset curve that can even be fine-tuned manually later.
We did not spend much time with the new Precision XOC, but will devote a future article to overclocking the GTX 1080, and will be able to give you a complete report on the complexities of GPU Boost 3.0. Instead, we manually clocked our GTX 1080 to just over 2,000MHz (+190MHz offset to the core and +150MHz to the memory) to test scaling in a few games and we got absolute stability without really trying.
Nvidia also announced Ansel, a powerful game capture tool that allows gamers to explore, capture, and share the artistry of gaming in radical new ways.
With Ansel, gamers can compose the gameplay shots by pointing the camera in any direction and from any vantage point within a gaming world allowed by the devs. They can capture screenshots at up to 32x screen resolution, and then zoom into a chosen location without losing fidelity. With photo-filters, they can add many different effects in real time before taking their shot, or process them afterward. They can even capture 360-degree stereo photospheres for viewing with a VR headset or with Google Cardboard.
Ansel will be available in upcoming releases and by game patches such as Tom Clancy’s The Division, The Witness, Lawbreakers, The Witcher 3, Paragon, No Man’s Sky, Obduction, Fortnite and Unreal Tournament.
The Founder’s Edition of the GTX 1080
The GeForce GTX 1080 Founders Edition is a graphics card designed and built by Nvidia that will sell for $699, $100 more than the base models of the partner cards. It was crafted with premium materials and components, including a die cast aluminum body and low profile backplate. It is ideal for situations where air needs to be exhausted out of a case, and perfect for SLI compared with open designs which do not do well in small form factor cases.
Compared with Maxwell reference versions, the power supply has been upgraded from 4 phases to a 5 phase dualFET design and tuned for bandwidth, phase balancing and quiet acoustics. Power efficiency has increased by approximately 6% compared to the GTX 980, and peak to peak voltage noise was reduced from 209mV to 120mV for improved overclocking.
The GTX 1080 uses a vapor chamber cooling, a first for a Nvidia 180W graphics card. It features a new low profile backplate with a removable section to allow better airflow between multiple graphics cards in adjacent SLI configuration.
It comes in a big and rather heavy box. If you remove the top of the box, it makes a nice display for the card.
The GeForce GTX 1080 “Founders Edition” will be available on May 27 for $699. It will be available from ASUS, Colorful, EVGA, Gainward, GALAX, Gigabyte, Innovision 3D, MSI, NVIDIA. Palit, PNY and Zotac. Custom boards from partners will vary by region, and pricing is expected to start at $599. The Founders Edition of the GTX 1080 are premium built by Nvidia reference versions that will be available for purchase for the entire life of the line instead of only at launch.
There is so much more that we simply cannot cover, including MultiRes Shading which can now be applied to desktop gaming. This applies more GPU power toward the center of the screen resulting in more detail and higher resolution where gamers tend to focus their attention.
Pascal also brings VRWorks Audio using the Nvidia OptiX ray-tracing engine to trace the path of sounds across an environment in real time which better reflects the size, shape, and material of VR than previous solutions. Much more can be said about VR, but we are concentrating on PC gaming and especially performance. However, let’s look at the Pascal GTX 1080 specifications first.
Here are the specifications for the GTX 1080:
How does the GTX 1080 compare with its rival, AMD’s Fury X?
It doesn’t. The GTX 1080 is simply in a class above the Fury X and even well above the GTX 980, GTX 980 Ti and above the TITAN X. How far? We are going to look at the performance of 26 games to compare the GTX 1080 with the TITAN X, with the GTX 980 Ti and with the GTX 980 SLI, and versus the Fury X.
However, before we do performance testing, let’s take a closer look at the GTX 1080 and check out overclocking and noise.