What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

The NVIDIA GTX 980Ti Performance Review

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
When NVIDIA launched their TITAN X, everyone knew the GTX 980 Ti wouldn’t be too far behind. Based on the same GM200 core as its bigger brother, this new card is meant to lead a more affordable (yes, you read that right) lineup into the DX12 era while also preempting the upcoming launch of AMD’s Fiji architecture.

For the last two generations, the cadence of NVIDIA’s product lineup has followed a lock-step approach and this time is no different except for a few points. The last high end Ti-series card was the GTX 780 Ti, a product launched in response to AMD’s extremely competitive R9 290X. It was able to pull away from the Radeon lineup’s darling while also consuming less power and providing a substantially quieter gaming experience.

The situation this time around, almost 19 months later, is significantly different. Other than the R9 295X2, AMD has failed to launch any product refreshes that put pressure on NVIDIA’s high end Maxwell cards. As a result the GTX 980 and subsequent GM200-based TITAN X found themselves all alone, competing against the nearly two year old R9 290X. Nonetheless, the GTX 980 Ti is still being rolled out as NVIDIA keeps marching lock-step with their roadmap regardless of what AMD does.

GTX-980TI-10-12.jpg

AMD’s plans are playing a factor in how NVIDIA approaches this launch though. With all signs pointing towards an E3 announcement of Fiji and availability sometime after that, they knew the GTX 980 Ti and its customized versions would end up competing against a HBM-equipped next generation architecture in the short term.

As a result, the GTX 980 Ti will hit the $649USD price point at launch and it will come bundled with a free copy of Batman: Arkham Knight. In addition, the GTX 980 will see its price cut by $50 to $499 while the remainder of cards will maintain their current stances. This should put downwards pricing pressure on Fiji and its derivatives even before they launch.

GTX-980TI-10-1.png

What NVIDIA is promising on the performance front is significant as well with wide-ranging intergenerational improvements, particularly at 4K. While much of the UDH lead may be derived from the GTX 980 Ti’s larger memory allotment, the Maxwell architecture itself incorporates notable improvements over Kepler, particularly on the DX12 front. There’s no doubt this card has been built to excel in 4K and DX12 environments while paving the way for Pascal’s introduction.

GTX-980TI-10-3.jpg

Actually creating the GTX 980 Ti was a relatively straightforward affair. Since yields for the 28nm-based GM200 are quite good NVIDIA simply took a fully enabled die and disabled a pair of SMs. Anyone who was having flashbacks to the GTX 970 episode need not fear since tertiary rendering functions remain untouched. The only significant change is the allocation of GDDR5 per 64-bit memory controller which has been reduced from 2GB per 1GB , creating a 6GB framebuffer.

The end result of this tinkering is a 2816 core part with 176 Texture Units which are both slightly lower than the TITAN X but not by a large amount. Meanwhile, the back-end functions like ROPs (96), L2 cache (3072K) and memory bus width (384-bit) remain identical to what’s found on a fully enabled GM200. This should help with 4K gaming which requires a ton of memory and pipeline resources for adequate framerates. As with TITAN X, double precision performance still operates at 1/32 of the single precision rate.

GTX-980TI-10-83.jpg

In many ways the GTX 980 Ti is a virtual clone of NVIDIA’s TITAN X. Other than the reduced CUDA core and texture unit count, it boasts the same clock speeds, memory rates and even a 250W TDP. However those two cut SMs should result in slightly lower heat and power signatures so the GeForce Boost algorithms will have more space to play. As a result, with we could up seeing the GTX 980 Ti hitting higher average core clocks than the TITAN X. That means performance may be within spitting distance of the $999 card in situations that aren’t memory-limited. Whether or not this wish actually comes true will be discussed on the following pages.

Alongside the GTX 980 Ti launch NVIDIA is also rolling out a number of other items which we discuss in separate articles. First and foremost among these is a batch of second generation G-SYNC displays, updates to G-SYNC for more adaptability (including a Windowed Mode) and finally G-SYNC for notebooks. There are also new GameWorks developer tools for improving efficiency in high load VR environments.

The GTX 980 Ti represents an interesting step forward for NVIDIA and one which will ultimately determine how their current product stack lines up against AMD’s Fiji. While Fiji and the availability timeline for its various iterations is still very much an unknown, NVIDIA’s newest addition has already shipped in volume into the channels and there will be a broad range of custom versions being offered in a matter of days to weeks.

By far the most enticing element of this launch is the GTX 980 Ti’s price. $649 certainly isn’t inexpensive in the grand scheme of things but it is certainly more affordable than the GTX 780 Ti’s initial price of $699. This should get high performance, DX12 compatible, 4K capable cards into the hands of far more gamers and tide NVIDIA’s lineup over until Pascal gets rolled out early next year. It just remains to be seen how well the GTX 980 Ti lines up against the TITAN X; it could be VERY close.

GTX-980TI-10-2.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
A Closer Look at the GTX 980 Ti

A Closer Look at the GTX 980 Ti


GTX-980TI-10-4.jpg

Retail packages from board partners will vary from one to another but reviewers received their reference samples in a sleek box with “The World’s Most Advanced GPU” emblazoned on it. It’s hard to deny that statement considering the advantage NVIDIA currently has over their competitors.

GTX-980TI-10-5.jpg

The GTX 980 Ti’s exterior design can be traced back to the GTX 780 days since NVIDIA’s reference board design really hasn’t changed much since then. It still looks great though. In this case the card uses the same high quality anodized aluminum shroud as NVIDIA’s TITAN X, making it 10.5” long and more than capable of fitting within certain SFF chassis.

GTX-980TI-10-6.jpg

Since the GTX 98- Ti utilizes the same GM200 cores as the TITAN X while boasting the capability to run at higher clock speeds, NVIDIA decided to remain with the 8+6 input power connector layout. This should allow the GTX 980 Ti to still hit its target 250W TDP while still allowing a good amount of room for overclocking.

GTX-980TI-10-8.jpg

Flipping the card over we see a naked PCB without any additional memory modules or a secondary aluminum cover. Supposedly, the elimination of a backplate cooler on NVIDIA’s latest cards improves SLI temperatures but we’d beg to differ. Nearly all modern ATX motherboards with SLI support have at the very least one additional PCI-E slot separating the two primary graphics slots. As a result, any multi-card setup receives more than enough fresh air without any airflow restriction. Let’s call a spade a spade: the lack of a backplate saves a few bucks, nothing more.

GTX-980TI-10-10.jpg

Like the TITAN X, taking off the shroud reveals and extensive three-stage internal vapor chamber heatsink design that consists of two primary aluminum fin arrays alongside a full-coverage plate that covers the memory and several VRM components. There are also channels to direct and accelerate the fan’s airflow so it can be most effective in dissipating any built-up heat.

Since cooler designs like the EVGA ACX 2.0, Gigabyte Windforce 3X and ASUS DirectCU II are rated for well above 300W of thermal output, expect all of those to make a comeback one the GTX 980 Ti. More importantly, NVIDIA’s thermal management and advanced chip design mean they don’t have to utilize water cooling to insure cool temperatures.

GTX-980TI-10-11.jpg

The card itself makes use of an advanced 6+2 phase PWM with insulated chokes to reduce whine at high framerates. Interestingly enough, there’s room for another power input connector on the PCB which perhaps points to it pulling double duty as a Tesla / Quadro workstation board as well. Once again, this mirrors what we found on the TITAN X.

GTX-980TI-10-7.jpg

Consisting of a trio of DisplayPort outputs, a lone HDMI 2.0 port and a DVI output, the GTX 980 Ti’s backplate has been carried over from the GTX 980. There’s really no need to change the layout here.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
Test System & Setup

Main Test System

Processor: Intel i7 4930K @ 4.7GHz
Memory: G.Skill Trident 16GB @ 2133MHz 10-10-12-29-1T
Motherboard: ASUS P9X79-E WS
Cooling: NH-U14S
SSD: 2x Kingston HyperX 3K 480GB
Power Supply: Corsair AX1200
Monitor: Dell U2713HM (1440P) / ASUS PQ321Q (4K)
OS: Windows 8.1 Professional


Drivers:
AMD 15.5 Beta
NVIDIA 352.90


*Notes:

- All games tested have been patched to their latest version

- The OS has had all the latest hotfixes and updates installed

- All scores you see are the averages after 2 benchmark runs

All IQ settings were adjusted in-game and all GPU control panels were set to use application settings


The Methodology of Frame Testing, Distilled


How do you benchmark an onscreen experience? That question has plagued graphics card evaluations for years. While framerates give an accurate measurement of raw performance , there’s a lot more going on behind the scenes which a basic frames per second measurement by FRAPS or a similar application just can’t show. A good example of this is how “stuttering” can occur but may not be picked up by typical min/max/average benchmarking.

Before we go on, a basic explanation of FRAPS’ frames per second benchmarking method is important. FRAPS determines FPS rates by simply logging and averaging out how many frames are rendered within a single second. The average framerate measurement is taken by dividing the total number of rendered frames by the length of the benchmark being run. For example, if a 60 second sequence is used and the GPU renders 4,000 frames over the course of that time, the average result will be 66.67FPS. The minimum and maximum values meanwhile are simply two data points representing single second intervals which took the longest and shortest amount of time to render. Combining these values together gives an accurate, albeit very narrow snapshot of graphics subsystem performance and it isn’t quite representative of what you’ll actually see on the screen.

FCAT on the other hand has the capability to log onscreen average framerates for each second of a benchmark sequence, resulting in the “FPS over time” graphs. It does this by simply logging the reported framerate result once per second. However, in real world applications, a single second is actually a long period of time, meaning the human eye can pick up on onscreen deviations much quicker than this method can actually report them. So what can actually happens within each second of time? A whole lot since each second of gameplay time can consist of dozens or even hundreds (if your graphics card is fast enough) of frames. This brings us to frame time testing and where the Frame Time Analysis Tool gets factored into this equation.

Frame times simply represent the length of time (in milliseconds) it takes the graphics card to render and display each individual frame. Measuring the interval between frames allows for a detailed millisecond by millisecond evaluation of frame times rather than averaging things out over a full second. The larger the amount of time, the longer each frame takes to render. This detailed reporting just isn’t possible with standard benchmark methods.

We are now using FCAT for ALL benchmark results, other than 4K.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
Performance Consistency & Temperatures Over Time

Performance Consistency & Temperatures Over Time


While the GTX 980 Ti may have less CUDA cores and texture units than the TITAN X, these cut components also give it a bit more headroom for higher clock speeds before the core hits its preset TDP. NVIDIA does list Base and Boost clocks that are perfectly aligned with a fully enabled GM200 but that shouldn’t stop it from reaching better speeds provided temperatures remain low enough.

GTX-980TI-10-80.jpg

The first test doesn’t really give much hope for higher clock speeds since core temperatures perfectly align with the TITAN X. With that being said, this could simply be a matter of the card maximizing its temperature envelope in an effort to boost frequencies.

GTX-980TI-10-79.jpg

As we can see, that’s exactly what NVIDIA’s Boost algorithm is doing here; with its cut down core the GTX 980 Ti actually has more room to increase engine speeds before it hits max TDP. Meanwhile the TITAN X needed to actually reduce frequencies as temperatures hit 82°C and the Boost algorithms cut things back to compensate for higher power consumption.

GTX-980TI-10-78.jpg

With about 5% less CUDA cores and Texture Units yet roughly 7% higher clock speeds, the GTX 980 Ti is actually able to just eke ahead of its more expensive sibling in this test. Some of this performance difference could also be due to slightly tighter timings on the Ti’s memory interface but that’s just supposition on our part.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
Thermal Imaging, Acoustics & Power Consumption

Thermal Imaging


GTX-980TI-10-100.jpg
GTX-980TI-10-101.jpg

Even though the GTX 980 Ti doesn’t build up a ton of heat, it would have been nice to see the reference cooler come equipped with a secondary backplate heatsink. We’ll have to wait for board partners to address that limitation.


Acoustical Testing


What you see below are the baseline idle dB(A) results attained for a relatively quiet open-case system (specs are in the Methodology section) sans GPU along with the attained results for each individual card in idle and load scenarios. The meter we use has been calibrated and is placed at seated ear-level exactly 12” away from the GPU’s fan. For the load scenarios, Hitman Absolution is used in order to generate a constant load on the GPU(s) over the course of 15 minutes.

GTX-980TI-10-81.jpg

If you are concerned about silence, then any blower-style heatsink will be a poor fit. However, the GXT 980 Ti is remarkably quiet for a card with a 250W TDP.


System Power Consumption


For this test we hooked up our power supply to a UPM power meter that will log the power consumption of the whole system twice every second. In order to stress the GPU as much as possible we used 15 minutes of Unigine Valley running on a loop while letting the card sit at a stable Windows desktop for 15 minutes to determine the peak idle power consumption.

GTX-980TI-10-82.jpg

Due to its higher average clock speeds we half expected the GTX 980 Ti’s power consumption envelope to be aligned with the TITAN X’s but that didn’t quite happen. This may be a case of sample to sample variance but our Ti proved to be quite efficient when under load.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
1440P: AC:Unity / Battlefield 4

Assassin’s Creed: Unity


GTX-980TI-10-54.jpg

GTX-980TI-10-30.jpg


Battlefield 4


<iframe width="640" height="360" src="//www.youtube.com/embed/y9nwvLwltqk?rel=0" frameborder="0" allowfullscreen></iframe>​

In this sequence, we use the Singapore level which combines three of the game’s major elements: a decayed urban environment, a water-inundated city and finally a forested area. We chose not to include multiplayer results simply due to their randomness injecting results that make apples to apples comparisons impossible.

GTX-980TI-10-55.jpg

GTX-980TI-10-31.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
1440P: Dragon Age: Inquisition / Dying Light

Dragon Age: Inquisition


<iframe width="640" height="360" src="https://www.youtube.com/embed/z7wRSmle-DY" frameborder="0" allowfullscreen></iframe>

Dragon Age: Inquisition is one of the most popular games around due to its engaging gameplay and open-world style. In our benchmark sequence we run through two typical areas: a busy town and through an outdoor environment.

GTX-980TI-10-56.jpg

GTX-980TI-10-32.jpg



Dying Light


<iframe width="640" height="360" src="https://www.youtube.com/embed/MHc6Vq-1ins" frameborder="0" allowfullscreen></iframe>​

Dying Light is a relatively late addition to our benchmarking process but with good reason: it required multiple patches to optimize performance. While one of the patches handicapped viewing distance, this is still one of the most demanding games available.

GTX-980TI-10-57.jpg

GTX-980TI-10-33.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
1440P: Far Cry 4 / Grand Theft Auto V

Far Cry 4


<iframe width="640" height="360" src="https://www.youtube.com/embed/sC7-_Q1cSro" frameborder="0" allowfullscreen></iframe>​

The latest game in Ubisoft’s Far Cry series takes up where the others left off by boasting some of the most impressive visuals we’ve seen. In order to emulate typical gameplay we run through the game’s main village, head out through an open area and then transition to the lower areas via a zipline.

GTX-980TI-10-58.jpg

GTX-980TI-10-34.jpg


Grand Theft Auto V


GTX-980TI-10-59.jpg

GTX-980TI-10-35.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
1440P: Hitman Absolution / Middle Earth: Shadow of Mordor

Hitman Absolution


<iframe width="560" height="315" src="http://www.youtube.com/embed/8UXx0gbkUl0?rel=0" frameborder="0" allowfullscreen></iframe>​

Hitman is arguably one of the most popular FPS (first person “sneaking”) franchises around and this time around Agent 47 goes rogue so mayhem soon follows. Our benchmark sequence is taken from the beginning of the Terminus level which is one of the most graphically-intensive areas of the entire game. It features an environment virtually bathed in rain and puddles making for numerous reflections and complicated lighting effects.

GTX-980TI-10-60.jpg

GTX-980TI-10-36.jpg


Middle Earth: Shadow of Mordor


With its high resolution textures and several other visual tweaks, Shadow of Mordor’s open world is also one of the most detailed around. This means it puts massive load on graphics cards and should help point towards which GPUs will excel at next generation titles.

GTX-980TI-10-61.jpg

GTX-980TI-10-37.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
1440P: Thief / Tomb Raider

Thief


<iframe width="640" height="360" src="//www.youtube.com/embed/p-a-8mr00rY?rel=0" frameborder="0" allowfullscreen></iframe>​

When it was released, Thief was arguably one of the most anticipated games around. From a graphics standpoint, it is something of a tour de force. Not only does it look great but the engine combines several advanced lighting and shading techniques that are among the best we’ve seen. One of the most demanding sections is actually within the first level where you must scale rooftops amidst a thunder storm. The rain and lightning flashes add to the graphics load, though the lightning flashes occur randomly so you will likely see interspersed dips in the charts below due to this.

GTX-980TI-10-62.jpg

GTX-980TI-10-38.jpg


Tomb Raider


<iframe width="560" height="315" src="http://www.youtube.com/embed/okFRgtsbPWE" frameborder="0" allowfullscreen></iframe>​

Tomb Raider is one of the most iconic brands in PC gaming and this iteration brings Lara Croft back in DX11 glory. This happens to not only be one of the most popular games around but it is also one of the best looking by using the entire bag of DX11 tricks to properly deliver an atmospheric gaming experience.

In this run-through we use a section of the Shanty Town level. While it may not represent the caves, tunnels and tombs of many other levels, it is one of the most demanding sequences in Tomb Raider.[/I

GTX-980TI-10-63.jpg

GTX-980TI-10-39.jpg
 

Latest posts

Top