NVIDIA GeForce GTX 650 Ti Boost - The New & Improved Mid-Range Kepler

Meet the NVIDIA GeForce GTX 650 Ti Boost

Last October, NVIDIA released the GeForce GTX 650 Ti, a mid-range, Kepler-based graphics card aimed at providing high-quality, Full HD, DX11 gaming at an affordable price. While we were generally quite impressed with the card, we were disappointed that NVIDIA's GPU Boost technology, a standout feature found on other higher-end Kepler-based graphics cards that dynamically adjusts GPU clock speed according to the operating environment of the graphics card, was missing.

Today, NVIDIA is releasing the new and improved GeForce GTX 650 Ti Boost, an updated version of the original GTX 650 Ti with a higher 980MHz core clock speed, 6008MHz DDR memory speed, 2GB VRAM and most notably, full support for GPU Boost.

Still using the same mid-range GK106 GPU core, the GTX 650 Ti Boost actually has a lot more in common with the GeForce GTX 660 than the GTX 650 Ti. Both have the same number of transistors, the same 2GB GDDR5 VRAM, the same port layout (two DVI, 1 HDMI, 1 DisplayPort), the same core and memory clock speeds, the same 192-bit memory interface and the same 144.2GB/s memory bandwidth. Both share the same reference design cooler as well. In fact, the only difference between the two is that the GTX 660 has 192 more CUDA Cores, 16 more Texture Mapping Units, and 8 more ROPs. 

With an SRP of US$169, the GeForce GTX 650 Ti Boost neatly fills the mid-range gap currently existing between the US$129 GeForce GTX 650 Ti and the US$199 GeForce GTX 660.

Rounding out NVIDIA's mid-range lineup, a 1GB version of the GTX 650 Ti Boost will also be available for US$149 (which will directly compete with AMD's recently released US$149 Radeon HD 7790).

Along with the US$109 GeForce GTX 650, that gives gamers no less than five options to choose from between the US$100-200 range.

Local pricing for the NVIDIA GeForce GTX 650 Ti Boost is expected to be around S$250-300. We'll update you more on the local price aspects as soon as we take a look at some custom edition cards.

[hwzcompare]

[products=381612,336116,326408,380323,380336,380340]

[width=175]

[caption=NVIDIA GeForce GTX 650 Ti Boost and competitive SKUs compared]

[showprices=0]

[/hwzcompare]

Test Setup

For those who've been following our reviews, take note that we've replaced our old test rig with a brand new one for 2013. Here are the new specs we'll be running with:

• Intel Core i7-3960X (3.3GHz)
• ASUS P9X79 Pro (Intel X79 chipset) Motherboard
• 4 x 2GB DDR3-1600 G.Skill Ripjaws Memory
• Seagate 7200.10 200GB SATA hard drive (OS)
• Western Digital Caviar Black 7200 RPM 1TB SATA hard drive (Benchmarks + Games)
• Windows 7 Ultimate 64-bit

Here's the list of cards we'll be testing and the drivers used.

• NVIDIA GeForce GTX 650 Ti Boost 2GB GDDR5 (NVIDIA ForceWare 314.21 Beta)
• NVIDIA GeForce GTX 650 Ti 1GB GDDR5 (NVIDIA ForceWare 310.90)
• NVIDIA GeForce GTX 660 2GB GDDR5 (NVIDIA ForceWare 314.14)
• ASUS Radeon HD 7790 DirectCU2 1GB GDDR5 (AMD Catalyst 13.3 Beta)
• AMD Radeon HD 7770 1GB GDDR5 (AMD Catalyst 13.2 Beta)
• AMD Radeon HD 7850 2GB GDDR5 (AMD Catalyst 13.2 Beta) 

N.B.: The ForceWare 314.21 Beta drivers we were supplied with only worked with the GTX 650 Ti Boost, and as such, it may hold a slight advantage over the other NVIDIA cards using marginally older drivers. 

Benchmarks

We've also updated our benchmark list, adding Unigine's new Valley benchmark, as well as 3DMark 2013 and the big one, Crysis 3! We've also updated Unigine from version 3.0 to 4.0.

• Futuremark 3DMark 11
• Futuremark 3DMark (2013)
• Unigine 4.0 "Heaven"
• Unigine "Valley" 1.0
• Battlefield: Bad Company 2
• Crysis Warhead
• Crysis 2
• Dirt 3
• Assassin's Creed 3
• Call of Duty: Black Ops 2
• Hitman: Absolution
• Far Cry 3
• Crysis 3

3DMark 11 Results

As always, we start with 3DMark 11, our trusty synthetic benchmark designed to test a GPU's performance at various aspects of DirectX 11 such as tessellation and DirectCompute. 

The GTX 650 Ti Boost immediately showed a huge improvement over the original GTX 650 Ti, outscoring it by about 20% on the Performance Preset and by 30% on Extreme. At both presets, it trailed the GTX 660 by about 16%. 

Compared to the ASUS Radeon HD 7790, surprisingly it trailed by about 5% at the Performance Preset, but was able to outscore it by about 13% at Extreme. Impressively it also outscored the higher-end Radeon HD 7850 by 6% on Extreme.

If we look at the GPU Boost chart, you can see that the 650 Ti Boost achieved a maximum boosted core clock speed of 1071MHz, an impressive 91MHz higher than its default clock speed.

3DMark (2013) Results

3DMark (2013)'s Fire Strike benchmark is Futuremark's most ambitious and technical benchmark to date utilizing a grueling two-part test with extreme levels of tessellation and volumetric illumination, as well as complex smoke simulation using compute shaders and dynamic particle illumination, while Fire Strike Extreme ramps up the difficulty with more tessellation, more particle effects and more taxing DirectCompute calculations.

The GTX 650 Ti Boost was again impressive, vastly outscoring the GTX 650 Ti while trailing the GTX 660 by only 11% on both Fire Strike and Fire Strike Extreme.

Against AMD, the 650 Ti Boost achieved similar scores to the HD 7790 on Fire Strike, but managed to outscore it at the more intensive Fire Strike Extreme (obviously due to the the 1GB graphics memory on the HD 7790). However, it lost to the HD 7850 on both tests. 

Unigine 4.0 "Heaven" Results

Unigine's "Heaven" benchmark tests GPUs with extreme tessellation, dynamic sky with volumetric clouds, real-time global illumination, and screen-space ambient occlusion among others.

The GTX 650 Ti Boost surprised us here by beating the ASUS Radeon HD 7790 by as much as 20% in almost every setting and also outscored the HD 7850 by a few percent on all settings. In fact, it was much closer in performance to the GTX 660, trailing it by just a few percent with AA turned off. Even with 8x AA, the difference between the GTX 650 Ti Boost and GTX 660 was only a few FPS apart.

Unigine "Valley" 1.0 Results

Valley is a new benchmark from Unigine that renders a highly detailed seamless landscape covering a total area of 64,000 square kilometers. It also utilizes dynamic sky effects, volumetric clouds, sun shafts, DOF, and ambient occlusion.

We saw a similar result to our Heaven benchmark, with the GTX 650 Ti Boost outscoring the ASUS Radeon HD 7790 by as much as 20% and the HD 7850 by as much as 8%, only slightly trailing the GTX 660 by less than 1%.

Crysis Warhead Results

For our first gaming benchmark we'll be using DX10 game Crysis Warhead. While it's getting a bit old, it does feature a fairly heavy amount of tessellation, which is quite taxing on a GPU.

The GTX 650 Ti Boost continued to outperform the ASUS Radeon HD 7790, averaging between 15-20% higher FPS across all settings. Against the GTX 660, the 650 Ti Boost trailed by about 5-8%.

Battlefield: Bad Company 2 Results

On Battlefield: Bad Company 2, the 650 Ti Boost again performed only slightly under the GTX 660, while outscoring the HD 7790 by around 6-10%.

Crysis 2 Results

Crysis 2's Ultra setting puts GPUs through their paces with a grueling test of tessellation, Parallax Occlusion Mapping, water rendering, and particle motion blur and shadowing.

The GTX 650 Ti Boost wasn't quite as impressive here, although it still managed to beat the HD 7790 by about 2-4% across all resolutions. However it trailed both the HD 7850 and the GTX 660 by about 15%.

Dirt 3 Results

The GTX 650 Ti Boost performed better at Dirt 3, trailing the GTX 660 by about 3% while beating the HD 7790 by around 5-6% and the HD 7850 by around 1-2%.

Assassin's Creed 3 Results

On Assassin's Creed 3, the GTX 650 Ti Boost again performed well, outscoring the HD 7790 by about 20% and trailing behind the GTX 660 by just 1 FPS across all resolutions.

Call of Duty: Black Ops 2 Results

Another strong performance for the GTX 650 Ti Boost saw it matching the GTX 660 in every resolution. It's possible that the GTX 650 Ti Boost benefited from slightly newer drivers, however, it was impressive nonetheless. 

Against the HD 7790, the GTX 650 Ti Boost continued to outscore it by between 15% up to a massive 30% at the 2560 x 1600 resolution.

Far Cry 3 Results

Moving onto some truly challenging games, first up, Far Cry 3, which is running on the Dunia 2 Engine, a modified version of Crytek's CryEngine. As such, this benchmark features heavy tessellation, high amounts of volumetric lighting, and deferred radiance transfer volumes (global illumination).

The GTX 650 Ti Boost performed fairly well on all settings and resolutions, scoring about halfway between the HD 7790 and GTX 660 - although all three cards dropped to unplayable framerates with MSAA set to 8x.

Of course, these mid-range cards are really designed for use at Full HD resolution with post-processed anti-aliasing options like FXAA, not MSAA.

Hitman: Absolution Results

Hitman: Absolution is another challenging benchmark, running on IO Interactive's proprietary Glacier2 engine, a very hardware intensive engine which is able to render up to 1200 NPCs at the same time. The benchmark tool used to test this game is extremely intensive and features a massive crowd of NPCs, Reflective Shadow Mapping (RSM), Direct Compute accelerated Bokeh Depth of Field, extreme tessellation and Ambient Occlusion.

Hitman's benchmark is extremely taxing, especially at 8x MSAA, simply due to the sheer number of models that need to be anti-aliased. Having said that, the GTX 650 Ti Boost put in a solid performance, slightly trailing the GTX 660 at most resolutions, and even outscoring it at 2560 x 1600 resolution with AA turned off.

Crysis 3 Results

Our newest gaming benchmark is the much anticipated sequel to Crysis 2 and, much like its predecessors, continues to be an absolute beast of a game, utilizing CryEngine 3 to the max with extreme amounts of tessellation, per-pixel per-object motion blur, Bokeh Depth of Field, displacement mapping on small terrain, particle and volumetric lighting and fog shadows, improved dynamic cloth and vegetation, dynamic caustics and diffuse shadows.

Crysis 3 is really not designed for mid-range graphics cards (especially at the maximum settings that we test at), although surprisingly, the GTX 650 Ti Boost performed quite well with AA turned off, scoring almost exactly the same as the GTX 660. While it was only able to achieve 30 FPS at Ultra settings with the resolution turned down to 1680 x 1050, turning down a few options at 1920 x 1200 saw frame rates climb above 30 FPS.

Overclocking Results

For our overclocking efforts, we were able to raise the GTX 650 Ti Boost core clock speeds to 1150MHz with memory raised to 6600MHz DDR. This resulted in a respectable 10% gain on 3DMark 11's Performance preset and 14% on the Extreme preset. 

AMD's Radeon HD 7850 remains the best overclocker, trailing the GTX 650 Ti Boost at Extreme at the reference clocks, but surpassing it at max overclocked settings, gaining a massive 25% performance boost in the process.

Temperature Results

Similar to our performance benchmark results, the GTX 650 Ti Boost's temperature performance was much closer to that of the GTX 660.

Power Consumption Results

Power consumption was also much closer to the GTX 660 than the GTX 650 Ti, at just 5 Watts lower at load than the GTX 660. This was expected due to the way NVIDIA's GPU Boost always fully utilizes the available power threshold of the card.

Price-Performance Comparison

When it comes to mid-range graphics cards, the key metric most gamers look for is its price-to-performance ratio, in other words, how much bang do you get for your buck?

For this comparison, we calculated the average FPS of each card, based on its sum total score in each of our gaming benchmarks at 1920 x 1200 resolution, with both AA off and set to maximum where appropriate (as such 3DMark scores are excluded for obvious reasons that it doesn't deal with FPS scores). We then divided that score by the card's current retail price in US dollars to derive exactly what 1 US dollar will get you in terms of FPS for each card. We chose this resolution as the vast majority of gamers (especially at this price-point) play at this resolution. Note that we've used an ASUS Radeon HD 7790 as we didn't managed to get any reference equivalent card.

At US$149, the GeForce GTX 650 Ti Boost provides exceptional value, almost 10% better than the next best value card, the original GTX 650 Ti.

If you're trying to decide between the GTX 650 Ti Boost and GTX 660, as you can see, both have very similar average FPS, however, the 660 GTX does perform slightly better at higher resolutions, and also has more overclocking potential (not shown in the above chart). But then again, these mid-range cards weren't really designed for gaming beyond standard full HD resolutions. If you look at the results beyond 1920 x 1200 pixels resolution, they are mostly below the playable 30 FPS mark.

Having said that, if you only intend to play at full HD resolution, the GTX 650 Ti Boost should be sufficient and it's not worth paying the extra US$30 for the GTX 660. However, if you're interested in overclocking your card or playing at higher resolutions, you might want to consider gunning for something better than the midrange card and go for at least a GeForce GTX 660 Ti (not compared in this article).

Official local pricing for the GeForce GTX 650 Ti has not been released yet, however, with the GTX 660 currently retailing for between S$340-S$380, anything under S$330 will provide fantastic value while providing a similar level of performance to the GeForce GTX 660.

Conclusion

NVIDIA seems to be concentrating on the mid-range market segment, with no less than five cards priced between $US100-200 segment. Out of all of them, the GeForce GTX 650 Ti Boost is undeniably the best buy, offering fantastic value with performance very close to that of the GTX 660, and surpassing both AMD's HD 7790 and HD 7850 on most tests.

It is also the cheapest card to offer 2GB of GDDR5 VRAM, which gives it better future-proofing value than the 1GB cards, and could also be important if you intend to upgrade to a higher resolution monitor or dual-monitor setup in the future.

If we have any complaints to make about the GTX 650 Ti Boost, it's that its overclocking potential seems rather average - possibly because it is already performing at close to maximum potential. After all, if the GK106 die was really of better quality, it would have been binned for a GTX 660 card. So from that perspective and considering that it's a mid-range card, there's really not much to complain about. Still, it will be interesting to see how much extra performance NVIDIA's add on partners can squeeze out of their custom OC cards when they're released.

Additionally, users who liked the original GTX 650 Ti for its small, compact design may be turned off by the rather large size of the GTX 650 Ti Boost. The good news is that many NVIDIA partners releasing their versions will sport a more compact design. So stay tuned for more of those comparisons soon.

Our articles may contain affiliate links. If you buy through these links, we may earn a small commission.