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 1060 3GB Review

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,421
Location
Montreal
DX12 / 1080P: Hitman / Quantum Break

Hitman (2016)


The Hitman franchise has been around in one way or another for the better part of a decade and this latest version is arguably the best looking. Adjustable to both DX11 and DX12 APIs, it has a ton of graphics options, some of which are only available under DX12.

For our benchmark we avoid using the in-game benchmark since it doesn’t represent actual in-game situations. Instead the second mission in Paris is used. Here we walk into the mansion, mingle with the crowds and eventually end up within the fashion show area.





Quantum Break


Years from now people likely won’t be asking if a GPU can play Crysis, they’ll be asking if it was up to the task of playing Quantum Break with all settings maxed out. This game was launched as a horribly broken mess but it has evolved into an amazing looking tour de force for graphics fidelity. It also happens to be a performance killer.

Though finding an area within Quantum Break to benchmark is challenging, we finally settled upon the first level where you exit the elevator and find dozens of SWAT team members frozen in time. It combines indoor and outdoor scenery along with some of the best lighting effects we’ve ever seen.



 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,421
Location
Montreal
DX12 / 1080P: Rise of the Tomb Raider / Warhammer: Total War

Rise of the Tomb Raider


Another year and another Tomb Raider game. This time Lara’s journey continues through various beautifully rendered locales. Like Hitman, Rise of the Tomb Raider has both DX11 and DX12 API paths and incorporates a completely pointless built-in benchmark sequence.

The benchmark run we use is within the Soviet Installation level where we start in at about the midpoint, run through a warehouse with some burning its and then finish inside a fenced-in area during a snowstorm.





Warhammer: Total War


Unlike some of the latest Total War games, the hotly anticipated Warhammer title has been relatively bug free, performs well on all systems and still incorporates the level detail and graphics fidelity this series is known for. In this sequence, we use the in-game benchmarking tool to play back one of our own 40 second gameplay sessions which includes two maxed-out armies and includes all of the elements normally seen in standard gameplay. That means zooms and pans are used to pivot the camera and get a better view of the battlefield.


 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,421
Location
Montreal
Performance Consistency Over Time

Performance Consistency Over Time


Modern graphics card designs make use of several advanced hardware and software facing algorithms in an effort to hit an optimal balance between performance, acoustics, voltage, power and heat output. Traditionally this leads to maximized clock speeds within a given set of parameters. Conversely, if one of those last two metrics (those being heat and power consumption) steps into the equation in a negative manner it is quite likely that voltages and resulting core clocks will be reduced to insure the GPU remains within design specifications. We’ve seen this happen quite aggressively on some AMD cards while NVIDIA’s reference cards also tend to fluctuate their frequencies. To be clear, this is a feature by design rather than a problem in most situations.

In many cases clock speeds won’t be touched until the card in question reaches a preset temperature, whereupon the software and onboard hardware will work in tandem to carefully regulate other areas such as fan speeds and voltages to insure maximum frequency output without an overly loud fan. Since this algorithm typically doesn’t kick into full force in the first few minutes of gaming, the “true” performance of many graphics cards won’t be realized through a typical 1-3 minute benchmarking run. Hence why we use a 10-minute warm up period before all of our benchmarks.


There are two factors at play here. First and foremost, it is more than obvious that the efficient Pascal core poses absolutely problem for EVGA’ compact heatsink. On the other hand the results look like EVGA has set a hard limit of 75°C for their card and the Boost algorithms are implemented in such a way that the card doesn’t increase one iota above that point.


We can also see that fan speeds (they’re zeroed out since EVGA has implemented a profile which cuts off rotations when idle scenarios are detected) gradually increase as the heatsink tries to keep up with the core’s steady heat output. Now to be honest with you I have no doubt that RPM values that hover just over the 1000 mark will result in an extremely quiet card but is performance left on the table?


The answer to that question is somewhat multi-faceted. As you can see, the GTX 1060 3GB in both Superclocked and reference form has its frequencies step back after about two minutes of full load but after that point the values even out and don’t budge one iota.


It is really interesting to see how two BIOS’s behave on exactly the same card. Despite slight clock speed variances, there really isn’t any overall step-back in the output framerate from either of these cards.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,421
Location
Montreal
Acoustics & Power Consumption

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.


Both the stock card and EVGA’s Superclocked feature a so-called “0dB” fans which essentially shut themselves off in low load scenarios. Outside of idle situations, these cards are still extremely quiet and you’d be hard pressed to actually hear them under load.


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.


This particular version of the GTX 1060 may have half the memory of its bigger brother and utilizes a core with a single SM disabled but that doesn’t necessarily mean it consumes significantly less power. At least on our sample we are looking at an 8W savings and that’s well within the margin of error so I'm guessing we received a slightly more power hungry core this time around. Meanwhile the Superclocked does require a bit more current but that’s to be expected due to its higher clock speeds.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,421
Location
Montreal
Overclocking Results

Overclocking Results


Overclocking the GTX 1060, regardless of whether it’s a 3GB or 6GB card, has proven to be a challenge in the past and this time was no different with one exception: the GDDR5 on this card had the most headroom of any 1060 to date. Will that translate to other 3GB versions? I have absolutely no idea but the lower amount of load on the memory controller could be a contributing factor to slightly higher memory OC headroom. One other thing to note is that I didn’t experience one iota of difference between the Reference and SC BIOS on this card in terms of achievable clock speeds.


Core clocks on this particular sample topped out at a consistent 2037MHz which is just shy of what the Founders Edition hit a few months ago. All things considered that’s not all that much of an improvement over the 1800MHz to 1900MHz this card achieved without any overclock whatsoever. As you can imagine, that speed was capped by the voltage which, even with EVGA’s excellent Precision software, works on a nebulous sliding scale of 0 to 100. Memory on the other hand did hit an impressive speed of just over 9Gbps.

With such a limited core overclock, one other thing becomes apparent as we move towards the in-game results: these cards aren’t bottlenecked by their memory speed. Despite a 1GHz GDDR5 overclock the amount of additional performance that can be squeezed out of the GTX 1060 3GB is minimal at best unless you get one that allows for substantially higher engine clocks.


 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,421
Location
Montreal
Conclusion

Conclusion


I initially went into this review with some preset expectations about the GTX 1060 3GB and its relative positioning within the current $150 to $250 GPU landscape. I know I’m beginning to sound like a broken record but this slim market segment holds a ton of possibilities, so much so that potential buyers can quickly become overwhelmed with options. Do they save a few bucks and go with a GTX 1050Ti or spend a bit more and grab something like the RX 470 or GTX 1060 3GB? For the most part the GTX 1060 3GB does indeed give a good accounting of itself but there are a few caveats which may cause you to look elsewhere.

NVIDIA’s target audience for this card are people who want more performance than a GTX 1050 Ti can provide but don’t necessarily want to fork out the extra $50 demanded by the GTX 1060 6GB. As a result it rests in an odd no-man’s-land and acts as a true bridge product between two very different market segments. The problem with this is pretty straightforward: to achieve a lower price NVIDIA needed to cut something and in this case it was a single SM along with the GDDR5 memory allotment being marched over to the sacrificial altar. And right now the GTX 1060’s 3GB of memory may not be a massive hindrance to in-game framerates but it may provide a distinctive barrier in the not too distant future.


Throughout the benchmarking process one of my assumptions was that provided a game didn’t require over 3GB of memory, this card would run relatively close to its 6GB sibling regardless of the cut-down core. That actually didn’t happen quite as often as I expected even though our sample’s Boost frequencies mirrored the GTX 1060 6GB Founders Edition being used as a direct comparison. There were some noteworthy and very large discrepancies.

Only two games in this review (Doom and Gears of War) required more than 3GB at certain points during testing. In those two games NVIDIA’s budget-friendly GTX 1060 got massacred and in one case was even outmatched by the GTX 1050 Ti. Throughout testing and even removing those outlier situations the less expensive card was consistently 10% to 15%behind and that's something which just can’t be overlooked.

The framerate shortfall manifests itself to a greater extent in DX12 with Doom and Gears of War in particular being a standout examples of what the future may hold. As DX12 matures and developers begin bringing its bottleneck-mitigating features to bear, the GTX 1060 3GB could flourish but it could flounder as well. If you are looking for a bit more future-proofing and want a card that can easily jump from 1080P to 1440P, look towards the GTX 1060 6GB or RX 480. They both represent $50 of money well spent in this case.

One of the GTX 1060 3GB primary successes is the fact that, much like the GTX 1050 Ti, it competes by way of avoidance. It costs $20 more than the average price of AMD’s RX 470 and completely justifies that premium more often that not. For gamers who intend to remain at 1080P for the time being, it costs much less than the GTX 1060 6GB or RX 480 8GB and that’s money which can be directed towards a better CPU, SSD or cooling setup.

Then there’s EVGA Superclocked version and while I don’t typically recommend pre-overclocked cards due to their inherently terrible value, this one is different. For that minor $10 investment you get a bit more performance and a better heatsink design without sacrificing the compact form factor. If you are looking at a GTX 1060 3GB, this one is a no-brainer.

From an overall value standpoint I’m on the fence about the GTX 1060 3GB. On one hand, right now it provides an excellent combination of performance, efficiency and price to folks who want a timely upgrade for their old systems. Unfortunately the persistent question will always be how long that 3GB will remain enough to keep this particular card’s head above water.
 
Last edited:

Latest posts

Twitter

Top