NVIDIA GeForce GTX 1060 review: Performance for the masses
Pascal finally has a mainstream GPU. The NVIDIA GeForce GTX 1060 joins the GeForce GTX 1070 and 1080 in punching above their price bracket and delivering previously unprecedented levels of performance. But is it really as fast as the GeForce GTX 980?
By HardwareZone Team -
Redefining mainstream performance
Finally, a truly affordable Pascal card.
This is all probably exceedingly familiar by now. With the GeForce GTX 1060, NVIDIA has completed the core of its Pascal line-up, which means it effectively has a card to appeal to three different market segments.
While the GeForce GTX 1080 is targeted at enthusiasts who want nothing but an undisputed single-GPU performance king, the GeForce GTX 1070 will appeal to gamers who want enthusiast-level performance at a slightly more palatable price. The thing is that none of these cards even remotely approach the price of the GeForce GTX Titan X, so NVIDIA has really done a great job in bringing previously ultra-enthusiast levels of performance out of the stratosphere.
In this light, the GeForce GTX 1060 is the next logical step down the ladder. Just as its bigger brothers offered unprecedented power at their respective price points, so does the GeForce GTX 1060. At US$299 for the Founders Edition (or US$249 for partner boards), this is NVIDIA’s mainstream card, and the one that will appeal to the vast majority of PC gamers who don’t need a flagship card. Just like NVIDIA's previous x60-series of graphics cards that have been the hallmark of the company's value for money option, the same is expected of the new GeForce GTX 1060.
It’s also no secret that the GeForce GTX 1060 is going head to head with AMD’s recently announced Radeon RX 480. The Founders Edition cards are pricier, but the US$249 starting price of the partner cards sets them up to be extremely attractive alternatives to the Polaris card (8GB goes for US$239, while the 4GB models go for just US$199). If we were AMD, we’d already be glancing nervously over our shoulders.
A new GPU die
The GeForce GTX 1060 is based on the new GP106 chip.
NVIDIA hasn’t exactly scrimped on its mainstream card, and that’s really heartening to see. For one, it retains all the key features of the GeForce 1070 and 1080, like the Simultaneous Multi-Projection (SMP) unit and enhanced memory compression that makes available more effective bandwidth.
We’ve already covered in detail what SMP does in our review of the GeForce GTX 1080, so you can head there to find out more. But in a nutshell, SMP’s greatest utility is found in VR applications and multi-display setups. Pascal GPUs can simultaneously create 16 different projections from a single viewpoint (it supports two viewpoints), which means it can accurately match the unique curvatures of VR displays and the monitor angles in multi-display configurations. Ultimately, the end result is better performance and a more immersive experience.
The card even supports GPU Boost 3.0 and Fast Sync – GPU Boost 3.0 enables custom per-voltage frequency offsets to maximize a card’s potential, while Fast Sync helps create a smoother and lower latency experience in less demanding titles like League of Legends.
It is also based on the same 16nm FinFET process as the more heavy hitting cards, which means it gets to benefit from the significantly improved power efficiency as well. This is a 120 watt TDP card, the same as the Maxwell-based GeForce GTX 960 that came before it, but it offers far better performance than its predecessor.
Having said that, the GeForce GTX 1060 isn’t exactly the same beast as its bigger brothers. For instance, it isn’t based on the same GP104 GPU, and features a new GP106 GPU instead. In this respect, NVIDIA is following a well-worn path – its Maxwell-based GeForce GTX 970 and 980 were similarly based on the same GM204 GPU, but the GeForce GTX 960 was based on a different GM206 GPU.
The block diagram of the GP106 die.
In a sense, this is a practical necessity. NVIDIA’s GeForce GTX x60 cards have never been in the same league as their higher-end brethren, and there’s only so far you can go in disabling certain functional units to differentiate performance.
The GP106 die is considerably smaller than the GP104’s, with an area of just 200mm² versus the 314mm² of the latter. It has just two Graphics Processing Clusters (GPCs), and a total of 10 Streaming Multiprocessors (SMs). But as on the GeForce GTX 1070 and 1080, each SM contains 128 CUDA cores, so the card has a total of 1,280 CUDA cores, about 33 percent fewer than the 1,920 cores on the GeForce GTX 1070. Each SM also comprises eight texture units, for a total of 80 on the GeForce GTX 1060.
In addition, the GP106 die features six 32-bit memory controllers, which gives the card a 192-bit memory bus width. Eight ROPs are also tied to each memory controller, so you’ll find 48 ROPs altogether on board. Memory bandwidth is much improved over the GeForce GTX 960 as well – with 6GB of GDDR5 memory and an effective memory clock of 8,000MHz, the card has up to 192GB/s of memory bandwidth, up from 112GB/s on its Maxwell equivalent.
All that aside, the defining feature of the Pascal cards have been their high clock speeds, and the GeForce GTX 1060 is no exception. The card boasts a base clock of 1,506MHz and boost clock of 1,708MHz. As it turns out, the GeForce GTX 1060 actually has the same base clock as the GeForce GTX 1070, but pairs that with a higher boost clock (the 1070’s boost clock sits at 1,683MHz).
Finally, here's a screenshot of the card's specifications to sum everything up:
A quick snapshot of the card's specifications.
A closer look at the GeForce GTX 1060 Founders Edition
The GeForce GTX 1060 Founders Edition sports a similar diecast aluminum body as its bigger brothers.
As with the GeForce GTX 1070 and 1080, NVIDIA has announced a Founders Edition version of the GeForce GTX 1060. These Founders Edition GeForce GTX 1060 cards and custom variants from add-in board partners are now being made available together, but the difference is that this Founders Edition card will only be available to buy from NVIDIA’s website. In comparison, you could easily buy a Founders Edition version of the GeForce GTX 1070 or 1080 from NVIDIA’s partners.
The odd Founders Edition premium applies here as well. The GeForce GTX 1060 Founders Edition is going for US$299, but the custom cards from NVIDIA’s partners will start at just US$249.
With that said, the same faceted design and blower fan on the earlier Founders Edition cards has carried over to the GeForce GTX 1060 as well. This means a very high-quality diecast aluminum body – in no way does this feel like a card intended for more budget conscious gamers.
Still, the design has been slightly changed from the GP104-based cards, and you don’t get the luxury of a full-length metal backplate.
The card doesn't have a full-length metal backplate like the GeForce GTX 1070 and 1080.
That aside, NVIDIA has still made certain improvements over the reference design of the GeForce GTX 960. The heatsink is now over 50 percent larger, which means increased surface area and much more efficient heat dissipation (when combined with two copper heatpipes at the base). The card also features a low-impedance power delivery network and custom voltage regulators for smoother power delivery, much as on the GeForce GTX 1070 and 1080 Founders Edition cards.
At around 250mm long, the card is also a bit shorter than its bigger brothers, which measure around 267mm in length. Because of the blower-style fan, heat is exhausted out the back of the case and not inside, so the 1060 should be a fairly good fit for small form factor systems where airflow isn’t the best.
In the way of display connectors, the card has the fairly standard selection of three DisplayPort 1.4 ports, one HDMI 2.0b port, and one dual-link DVI connector. These are all the latest display standards, so the GeForce GTX 1060 is fully-equipped to support things like HDR content. And with a TDP of just 120 watts, the card needs just a 6-pin PCIe connector for power.
Here's a look at the display connectors on the rear.
While the GeForce GTX 1070 and 1080 cards come with a single 8-pin PCIe connector, the GeForce GTX 1060 has just a lone 6-pin connector.
There’s just one other thing to note about the card – it doesn’t support SLI. There is literally no SLI connector, so NVIDIA is really putting its foot down about withholding multi-GPU support for the GeForce GTX 1060. This also ties in with NVIDIA’s efforts to roll back support for more extreme multi-GPU configurations. As things stand, even the GeForce GTX 1070 and 1080 support only two-way SLI.
This move actually makes sense. NVIDIA has to do a lot of work to optimize various SLI configurations for multiple titles, and the payoff is considerably little seeing as the number of folks who go beyond two-way SLI are few and far between. Furthermore, it doesn’t really make a lot of sense to go for two GeForce GTX 1060s, seeing as that would be more expensive than the US$449 GeForce GTX 1070 Founders Edition and might not even perform as well.
Test Setup
The detailed specifications of our current graphics card testbed system are as follow:-
- Intel Core i7-5960X
- ASUS X99-Pro (Intel X99 chipset) motherboard
- 2 x 4GB Corsair Vengeance LPX DDR4-2133 (Auto timings: CAS 15-15-15-36)
- Samsung SSD 840 Pro 256GB SATA 6Gbps solid state drive (OS + benchmark + games)
- Western Digital Caviar Black 1TB SATA 6Gbps hard drive (general storage)
- Windows 10 Pro 64-bit
- Intel INF 10.1.1.14
NVIDIA supplied us with beta driver version 368.64 for testing, and that’s what we used for both benchmarking and overclocking. For our comparison cards, we naturally included the other Pascal cards, in addition to the GeForce GTX 960 and 980 (since NVIDIA says the two are supposed to be on par). We also threw in the Radeon RX 480, the direct competitor from AMD, and the Radeon R9 Fury, to see how it compares with one of the top performers from the red camp.
The full line-up of graphics cards and their driver versions are listed below:
- NVIDIA GeForce GTX 1060 (ForceWare 368.64)
- NVIDIA GeForce GTX 1070 (ForceWare 368.19)
- NVIDIA GeForce GTX 1080 (ForceWare 368.19)
- MSI GeForce GTX 980 Gaming 4G (ForceWare 365.19)
- ASUS Strix GeForce GTX 960 2GB (ForceWare 368.81)
- AMD Radeon RX 480 8GB (Crimson Edition 16.6.2)
- ASUS Strix Radeon R9 Fury (Crimson Edition 16.3.1)
[hwzcompare]
[products=562776,556120,554644]
[width=200]
[caption=NVIDIA's Pascal cards compared]
[showprices=1]
[/hwzcompare]
[hwzcompare]
[products=479126,491673,560282,515698]
[width=200]
[caption=Test cards compared]
[showprices=1]
[/hwzcompare]
Benchmarks
Our performance benchmarks now include a handful of the latest titles. Two benchmark games, Hitman and Ashes of the Singularity, also take advantage of DirectX 12, so we were able to get a look at how the GeForce GTX 1060 performed using Microsoft’s latest gaming API. We’ve also included the new 3DMark DirectX 12 benchmark, Time Spy, to provide more insight into the card’s ability to take advantage of DirectX 12 features.
Here’s the list of the benchmarks we used:
- Futuremark 3DMark (2013) with Time Spy
- Middle-earth: Shadow of Mordor
- Crysis 3
- Tom Clancy’s The Division
- Ashes of the Singularity
- Hitman
We used the Fire Strike Extreme test in 3DMark (2013) for our power and temperature tests.
Futuremark 3DMark (2013)
While NVIDIA says that the GeForce GTX 1060 is supposed to deliver the same kind of performance as the GeForce GTX 980, that wasn’t quite the case here. It was about 10 percent behind the GeForce GTX 980 in Fire Strike, although the two were virtually on par when looking at the results for the DirectX 12 Time Spy benchmark. NVIDIA has made great improvements in DirectX 12 performance in Pascal, and that’s evident in the results here as well. Another factor to consider is the factory overclocks that the MSI GeForce GTX 980 comes with – NVIDIA was likely talking about the slower reference speeds when it made those claims, so some discrepancies are to be expected.
The GeForce GTX 1060 also edges out the AMD Radeon RX 480, its direct competitor. It was around 10 percent faster in Fire Strike Extreme, a decent amount considering that only US$10 separates the price of both cards (when comparing against the price of partner boards). It also inched ahead of the RX 480 in Time Spy, again showing that NVIDIA has truly caught up with AMD in taking advantage of the performance benefits offered by DirectX 12.
And if you’re wondering what separates the GeForce GTX 1060 from the 1070, the answer is a whopping 35 percent performance differential. The GeForce GTX 1060 isn’t anywhere in the same league as the GP104-based cards, and it caters to a separate, more budget conscious audience. With that said, the gains over the GeForce GTX 960 are significant – in Fire Strike, it is a staggering 57 percent faster, which amounts to quite a hefty upgrade in just one generation.
Middle-earth: Shadow of Mordor
A similar pattern was observed in Middle-earth: Shadow of Mordor. The GeForce GTX 1060 absolutely outclassed the GeForce GTX 960, which performed quite abysmally even at a resolution of 2,560 x 1,600 pixels. Granted, our 960 was the 2GB version, and it’s likely that the 4GB version would fare somewhat better, but that doesn’t discount the fact that NVIDIA’s mainstream card now offers a lot more compared to just a generation ago.
The GeForce GTX 1060 also managed to beat the Radeon RX 480 once again. At a 1600p resolution, it was around 9 percent faster.
However, it once again fell behind the factory overclocked MSI GeForce GTX 980 Gaming 4G. To its credit, it didn’t lag behind by much, and it even eked out a lead at the highest resolution, likely thanks to its 6GB of GDDR5 memory, compared to the latter card’s 4GB.
Although neither card managed playable frame rates at that resolution, the result goes to show that the GeForce GTX 1060 might nevertheless be better equipped to deal with bandwidth hungry applications like VR.
Crysis 3
The GeForce GTX 1060 generally did quite well against the Radeon RX 480 in Crysis 3. Although it fell behind in the least demanding benchmark, it eventually overtook the latter as we ramped up the resolution and graphics settings. At the most demanding settings, NVIDIA’s card was actually a good 19 percent faster than AMD’s.
But when it came to the GeForce GTX 980, the GeForce GTX 1060 didn’t quite measure up here. The factory overclocks aside, the GeForce GTX 980 was still 34 percent faster than the 1060 at a 1200p resolution and with anti-aliasing turned off. The gap narrowed considerably as the settings grew more taxing, and it looks like the GeForce GTX 1060 requires more trying conditions to really stand up to the 980. For instance, just turning on 8X MSAA at a 1200p resolution shrinks the gap to around 16 percent.
Tom Clancy’s The Division
In Tom Clancy’s The Division, the GeForce GTX 1060 and Radeon RX 480 took turns one-upping each other. Ultimately, it came down to the resolution – the GeForce GTX 1060 did better at the higher 1600p resolution, while the RX 480 came out ahead at 1200p. Strangely enough, the RX 480 is technically the card with higher memory bandwidth (and more memory in our RX 480 card), so it would look like that isn’t the main limiting factor in The Division, and other things like better shader performance are helping the 1060 at 1600p.
This was also another game where the GeForce GTX 1060 lagged behind the 980. At 1600p and Ultra settings, it was around 16 percent slower than the 980. But all things considered, the GeForce GTX 1060 didn’t actually do too badly. It was still a huge step ahead of its predecessor, with nearly double the performance at the most demanding settings. It was also capable of delivering playable frame rates at 1600p, which is more than could be said for mainstream cards from just a year ago.
Ashes of the Singularity
Note: We were unable to get Ashes of the Singularity to run in DirectX 12 mode and Crazy settings for the GeForce GTX 960 and 1060, which is why those figures are missing.
We’ll focus on what DirectX 12 performance figures we managed to obtain here, seeing as the previous games were all DirectX 11 titles. As it turns out, the Radeon RX 480 turned out a better performance in DirectX 12 mode, even though it was slower than the GeForce GTX 1060 in DirectX 11 mode. At 1600p and High settings, the Radeon RX 480 was around 7 percent faster.
The card also didn’t show much improvement when moving from DirectX 11 to 12. Even the GeForce GTX 1070 and 1080 had comparatively moderate improvements compared to AMD cards like the Radeon R9 Fury, so it would look like AMD’s implementation of asynchronous compute capabilities is still favored in AotS, despite any official statements to the contrary. At 1080p and High settings, the GeForce GTX 1080 enjoyed a 19 percent boost in DirectX 12, whereas the Radeon RX 480 netted a 33 percent increase.
Hitman
Note: As with Ashes of the Singularity, Hitman couldn’t run in DirectX 12 with all the graphics options maxed out on the GeForce GTX 1060.
As in AotS, there was virtually no difference in performance when moving from DirectX 11 to 12 for the GeForce GTX 1060. From the numbers we managed to obtained, it looks like the newest Pascal card doesn’t quite have the computing heft to reap a significant performance boost, even if it does support the new features in DirectX 12. That aside, performance was still very respectable for a US$299 card, and it could even handle the game at 1600p and Ultra settings.
With that said, performance was a mixed bag when the GeForce GTX 1060 was compared against other cards. At 1080p and Ultra settings, the GeForce GTX 980 was still 17 percent faster. The Radeon RX 480 also had a particularly strong showing, where it edged out even the latter card and was around 23 percent quicker than the 1060.
And given that the RX 480 is outperforming the GeForce GTX 980 even at the higher resolutions, and given the especially poor performance of the GeForce GTX 960 with just 2GB of memory, it looks like the Polaris’ card 256GB/s of memory bandwidth is serving it well. Hitman appears to be more demanding in this aspect of performance, and that would explain why the GeForce GTX 1060 lagged behind the RX 480, which it has shown that it can beat.
Temperature and Power Consumption
There’s nothing much to remark on when it comes to the temperature and power consumption figures. With a TDP of just 120 watts, the GeForce GTX 1060 expectedly ran cooler and consumed less power than its more powerful counterparts.
It’s worth noting that its TDP is actually lower than the Radeon RX 480’s 150 watt TDP, and the latter card also ran hotter at 70 degrees Celsius.
Overclocking
We managed to overclock the GeForce GTX 1060 to a high of 1,756MHz (or 120MHz above its base clock), which translates to a boost clock of 1,948MHz. With that said, the card still managed to boost even higher to around 2,126MHz in 3DMark, so that amounts to quite a significant speed jump.
This resulted in around a 7 percent improvement in Fire Strike, which was on par with the gains the Radeon RX 480 produced when we overclocked it. The biggest gain was actually seen in Fire Strike Ultra, where there was close to a 10 percent improvement.
As it turns out, the proportional performance increases from overclocking are also similar to what we observed with the GeForce GTX 1070 and 1080, and in that respect, the card doesn’t fare too shabbily at all.
Not quite a GeForce GTX 980, but a decent budget performer nonetheless
The GeForce GTX 1060 completes the core of NVIDIA's Pascal line-up.
The GeForce GTX 1060 is really just the logical conclusion to what NVIDIA has been doing with the rest of its Pascal series. The company effectively introduced new levels of performance at all-time-low price points, and it’s safe to say that its new cards are winners through and through.
In fact, the marketing pitch for the GeForce GTX 1060 closely parallels what NVIDIA did with the GeForce GTX 1070. Just as NVIDIA said the latter card would offer similar performance to the GeForce GTX Titan X, it trotted out the GeForce GTX 980 as a new yardstick for the GeForce GTX 1060.
The Titan X was the undisputed king of the last generation of cards, which meant it was also woefully out of the reach of many a buyer. Similarly, the GeForce GTX 980 reigned as the single-GPU king for a while, and at that point in time, the level of performance it offered was unthinkable in a mainstream card that a far larger swath of the market would be able to afford.
With the GeForce GTX 1060, a slice of that sort of performance is now available to the masses. We say a slice, because despite NVIDIA’s claims, its newest Pascal card doesn’t quite measure up to the GeForce GTX 980. It comes close, but it’s unfortunately not really there yet.
The GeForce GTX 1060 doesn't quite measure up to the GeForce GTX 980, but it is still a huge upgrade over what NVIDIA was offering on its mainstream card just a generation ago.
Having said that, we’re not going to be judging the card based on whether or not it lives up to NVIDIA’s claims. Taken on its own, the GeForce GTX 1060 delivers performance at a price that was just a dream only a year ago. In the way of competitors, the most obvious candidate is the AMD Radeon RX 480, and the two cards trade blows throughout our benchmarks.
But if we were to pick a winner, we’d have to give an edge to the AMD card because of the lead it takes in some of the newer games and the performance boost it nets in DirectX 12 mode. It also supports CrossFire, and while we thought that it wasn’t that wise of an investment, it’s still nice to see that the option is open. And to cap it all off, it’s also cheaper.
That’s not to say that the GeForce GTX 1060 is a bad card. It’s impressive what NVIDIA has done with the card, and all at the same 120 watt TDP as the vastly weaker GeForce GTX 960. Maybe it doesn't entirely out muscle the competition, but it is a huge step forward nonetheless.
Our articles may contain affiliate links. If you buy through these links, we may earn a small commission.