ZOTAC NVIDIA GeForce GTX Titan Review
Last week NVIDIA announced the GTX Titan, a card which has its roots in a high cost supercomputer GPU design brought to the desktop. It is an answer to what AMD has been able to achieve with their dual GPU setups, and an attempt to solidify their position at the top of the single GPU set-ups of the current generation.
Today we take the GTX Titan through a selection of real world gaming tests which include Crysis 3, Aliens: Colonial Marines, Far Cry 3, Assassins Creed 3, Hitman: Absolution, StarCraft 2: Heart of the Swarm, Battlefield 3, Skyrim: Dragonborn, DOTA 2, SWTOR, Mass Effect 3, F1 2012 and Borderlands 2. Mix that with the latest 3DMark, Heaven Benchmark, GPU computing, Media playback and all the usual power/thermal and overclocking tests... add frame latency, competing multi and single GPU setups and we have quite a lot to get through today.
So what exactly is the GTX Titan?
GTX Titan is based on the GK110 GPU, the same family as the GTX 600 series which uses the GK104 for the 660/670/680/690 parts. GK110 was designed to offer massive amounts of computing power originally and as a result has significantly higher specifications than those desktop parts, somewhat similar in fact to 2x GK104 combined.
We will get into the exact specifications on the next page but before that, a few notes on some of the features that make GTX Titan unique at this time.
GPU Boost 2.0:
GPU Boost is present on cards like the GTX 680 and is able to adjust the speed of that product depending on the operating conditions. Simply, that means increasing speed to maximise performance as long as the GPU remained under 170w power draw.
With GTX Titan and GPU Boost 2.0 NVIDIA have moved to thermal based readings and will boost the speed of the card to the highest possible level as long as the temperature remains below 80°C (with a power level not exceeding 265w). Clock speed and voltage are tweaked to maintain this as is fan speed with NVIDIA giving focus to low noise operation.
As some added bonuses the user also has control of the temperature target in software and can say, for example, that they wish the target to be 85°C rather than 80°C. Then for those who like overclocking with water, the added bonus is that the lower temperatures this affords will mean consistently higher clocks due to being under 80°C.
With the GTX 600 series NVIDIA introduced Adaptive V-Sync, a feature which allowed them to improve the gaming experience of those who found the standard high FPS with tearing or limited FPS with improved visuals to be a choice they didn't like. Now with GTX Titan NVIDIA are introducing Display Overclocking.
In normal circumstances our monitors, with V-Sync enabled, limit framerate to 60fps (based on 60Hz screens). That is regardless of the performance the GPU could offer. With GTX Titan we can attempt to "overclock" the screen, pushing the refresh rate and therefore framerate higher. NOTE: This is not supported by all monitors.
Double Precision Computing:
GTX Titan is the first of NVIDIAs consumer products to ship with full performance double precision compute potential. Previous products had limited support, such as the GTX 680 offering 8 double precision CUDA Cores. This is a feature which can be enabled/disabled in software as for gaming Double Precision is not required/advantageous, especially when we consider that the GPU clock speed is reduced when double precision is enabled.
So those are the three key features of the GTX Titan. Here is a quick run-through of the GTX 600 series features it also supports.
To skip forward to page 2 of the review click here.
One of the key design principles when designing Kepler was "Faster" and to enable this NVIDIA have developed what they call an "Extreme SM architecture" that contains (per GPU) 1536 CUDA core which offer improved instruction throughput, texturing and geometry processing over previous GPUS.
The GPU has its roots in GTX 480 which first used a parallel geometry pipeline and compute architecture (inc fast context switching). In Kepler, or the GTX 680 variants, we have various hardware blocks which perform specific tasks with the GPC being the master high level hardware block with dedicated resource for rasterization, shading, texturing and compute.
Within the GPC we have the latest generation of Streaming Multiprocessors (named SMX) which have their performance enhanced over Fermi based equivalents while running on less power (NOTE: See how this affects power consumption and why later in the review). SMX are the components which power our games using their CUDA cores to handle pixel/vertex/geometry shading and physics/compute operations. Texture units deal with the filtering, load/store units fetch and save information and the PolyMorph Engine looks after vertex fetch, tessellation, viewport transform, attribute setup and stream output. Also included are Special Function Units (SFUs) for transcendental and graphics interpolation instructions.
Connecting the core to its memory is a new memory interface. NVIDIA have reworked the memory subsystem to maximise the memory clock possible and therefore the bandwidth with rated speed of 6Gbps quoted. Elsewhere we have a new hardware based H.264 encoder (NVENC) which takes the video processing load away from the CUDA cores, enhancing performance and reducing power consumption.
Another area that comes under the "Faster" design ethic is NVIDIAs new GPU Boost technology. The basic idea behind this is that the GTX 690/680/670 has real time hardware monitoring that varies the clock speed of the core in small increments depending on load. This is completely automatic and not dependant on software and is similar to the speed step technology we have seen in desktop CPUs for some time, though to a much finer level of available speeds... less than 10MHz increments. The core speed doesn't just scale down when at idle or low load though, when required it also "boosts" up with the default speed rated as 1020MHz (This is an average, often peaking higher while remaining within the defined power range for the card). This has a major (positive) impact on power use, overclocking, temperatures and essentially the day to day workings of our card so we recommend visiting Page 17 of this review for detailed information and examples of how it works.
The significant changes brought to market by NVIDIA with the GTX 6x0 and its R300 drivers continue with three key technologies. First up is FXAA in all games, we have seen FXAA (smoother edges, smaller performance impact than MSAA) used in select games before now however from today we can enable the technology in the driver and that will be covered later in this review. Next up is surround gaming from one card. With the GTX 690/680/670 NVIDIA now allow us to connect three displays to one card/GPU and power resolutions such as 5760x1080 and while doing so we can also enable 3D Vision. As an added extra NVIDIA also allow the use of an "accessory" display, a fourth screen which can run at the same time as our three gaming screens for use on software such as chat, mail and browsing applications.
Next up is Adaptive VSync which looks to solve issues commonly associated with traditional V-Sync on/off setups. On past products users generally had to choose between V-Sync "on" which impacted framerates or V-Sync "off" which gave them the full speed of the card but with the introduction of image tearing. Adaptive VSync disables VSync when framerates are below 60fps, reducing stutters and when they return to over 60fps it is enabled again to reduce tearing.
And finally we have TXAA... a new mode of anti-aliasing which we will cover later in the article...