Last week we took a look at NVIDIA’s newest consumer flagship video card, the GeForce GTX 980. Based on the company’s new GM204 GPU, GTX 980 further cemented NVIDIA’s ownership of the performance crown with a combination of performance improvements, new features, and power consumption reductions. Combined with a lower price than the now-dethroned GTX 780 Ti, GTX 980 is an impressive flagship with a mix of attributes that NVIDIA hopes to entice existing 600 and 500 series owners to upgrade to.

Of course even though GTX 980 was cheaper than the outgoing GTX 780 Ti, it is still a flagship card and at $549 is priced accordingly. But as in every GeForce product lineup there is a GeForce x70 right behind it, and for GTX 980 its lower-tier, lower priced counterpart is the GeForce GTX 970. Based on the same GM204 but configured with fewer active SMMs, a slightly lower clock speed, and a lower TDP, GTX 970 fills the gap by providing a lower performance but much lower priced alternative to the flagship GTX 980. In fact at $329 it’s some 40% cheaper than GTX 980, one of the largest discounts for a second-tier GeForce card in recent memory.

For this reason GTX 970 is an interesting card on its own, if not more interesting overall than its bigger sibling. The performance decrease from the reduced clock speeds and fewer SMMs is going to be tangible, but then so is a $220 savings to the pocketbook. With GTX 980 already topping our charts, if GTX 970 can stay relatively close then it would be a very tantalizing value proposition for enthusiast gamers who want to buy in to GM204 at a lower price.

NVIDIA GPU Specification Comparison
	GTX 980	GTX 970 (CorrecteD)	GTX 780	GTX 770
CUDA Cores	2048	1664	2304	1536
Texture Units	128	104	192	128
ROPs	64	56	48	32
Core Clock	1126MHz	1050MHz	863MHz	1046MHz
Boost Clock	1216MHz	1178MHz	900Mhz	1085MHz
Memory Clock	7GHz GDDR5	7GHz GDDR5	6GHz GDDR5	7GHz GDDR5
Memory Bus Width	256-bit	256-bit	384-bit	256-bit
VRAM	4GB	4GB	3GB	2GB
FP64	1/32 FP32	1/32 FP32	1/24 FP32	1/24 FP32
TDP	165W	145W	250W	230W
GPU	GM204	GM204	GK110	GK104
Transistor Count	5.2B	5.2B	7.1B	3.5B
Manufacturing Process	TSMC 28nm	TSMC 28nm	TSMC 28nm	TSMC 28nm
Launch Date	09/18/14	09/18/14	05/23/13	05/30/13
Launch Price	$549	$329	$649	$399

Compared to GTX 980 and its full-fledged GM204 GPU, GTX 970 takes a harvested GM204 that drops 3 of the SMMs, reducing its final count to 13 SMMs or 1664 CUDA cores. It also sheds part of a ROP/L2 cache partition while retaining the 256-bit memory bus of its bigger sibling, bringing the ROP count down to 56 ROPs and the L2 cache down to 1.75MB, a configuration option new to Maxwell

Along with the reduction in SMMs clock speed is also reduced slightly for GTX 970. It ships at a base clock speed of 1050MHz and a boost clock speed of 1178MHz. This puts the theoretical performance difference between it and the GTX 980 at about 85% of the ROP performance or about 79% of the shading/texturing/geometry performance. Given that the GTX 970 is unlikely to be ROP bound with so many ROPs, the real world performance difference should much more closely track the 79% value, meaning there is still potentially a significant performance delta between the GTX 980 and GTX 970.

Elsewhere the memory configuration is unchanged from GTX 980. This means we’re looking at 4GB of GDDR5 clocked at 7GHz, all on a 256-bit bus. Compared to the GTX 770 that the GTX 970 replaces, this is a welcome and much needed upgrade from what has been the 2GB VRAM standard that NVIDIA has held to for the last two and a half years.

GTX 970’s TDP meanwhile is lower than GTX 980’s thanks to the reduced clock speeds and SMM count. The stock GTX 970 will be shipping with a TDP of just 145W, some 80W less than GTX 770’s official TDP of 225W. NVIDIA’s official designs still include 2 6-pin PCIe power sockets despite the fact that the card should technically be able to operate on just one; it is not clear at this time whether this is for overclocking purposes (150W would leave almost no power headroom) or for safety purposes since NVIDIA would be so close to going over PCIe specifications.

Like the GTX 980, NVIDIA’s target market for the GTX 970 will be owners of GTX 600/500/400 series cards and their AMD equivalents. GTX 970 is faster than GTX 770 but not immensely so, and as a result NVIDIA does not expect GTX 770 owners to want to upgrade so soon. Meanwhile GTX 670 owners and beyond are looking at 65%+ improved performance for cards at the same tier, while power consumption will remain roughly consistent from the GTX 670’s 140W GPU Boost 1.0-based power target.

Furthermore, as we mentioned in our GTX 980 review, GTX 970 has been a pure virtual (no reference card) launch, which means all of NVIDIA’s partners are launching their custom cards right out of the gate. A lot of these have been recycled or otherwise only slightly modified GTX 700/600 series designs, owing to the fact that GM204’s memory bus has been held at 256-bits and its power requirements are so low.

Meanwhile since NVIDIA did not produce reference cards, for GTX 970 reviewers are being sampled directly by NVIDIA’s partners. For our review today we will be looking at EVGA’s GeForce GTX 970 FTW ACX 2.0, the company’s highest performance GTX 970 card. Accordingly, we will be taking a look at both it’s out of the box performance and performance when reconfigured as a stock card to showcase both performance profiles.

With the discontinuation of the GTX 780 series and GTX 770, competition for the GTX 970 will be split between the GTX 760 and GTX 980 on the NVIDIA side. On the AMD side things will be even more spread out; AMD’s closest cards from a pricing perspective are the R9 280X and R9 290 priced below and above the $329 GTX 970 respectively, but as we’ll see even R9 290X is not necessarily out of the picture thanks to GM204’s strong performance.

Surprisingly even a week after the launch of the GTX 900 series, AMD has yet to officially respond to the GTX 900 series launch with any further price cuts or additional incentives beyond their existing Never Settle Forever bundle. In lieu of that some retailers have been running their own promotions; our pricing benchmark retailer Newegg has been offering 15% discounts on some of their PowerColor R9 290 series cards, meanwhile some other cards qualify for a $40 Newegg gift card (which cannot be applied retroactively to the purchase). Since the bulk of these cards don’t qualify for the price discount we’re holding our reference prices at $500 for the R9 290X and $400 for the R9 290, however the very cheapest of these PowerColor cards with the discount in play can go for as little as $450 and $340 respectively.

Meanwhile GTX 900 series sales have been brisk, and while the cards are still in supply not all models are available or are regularly available. At the very least everything from reference clocked cards to significantly overclocked cards are available at Newegg, so there is still a range of options. Though they are coincidentally all EVGA cards as of publication time.

Fall 2014 GPU Pricing Comparison
AMD	Price	NVIDIA
Radeon R9 295X2	$1000
	$550	GeForce GTX 980
Radeon R9 290X	$500
Radeon R9 290	$400
	$330	GeForce GTX 970
Radeon R9 280X	$280
Radeon R9 285	$250
Radeon R9 280	$220	GeForce GTX 760

 

Meet The EVGA GeForce GTX 970 FTW ACX 2.0

Our sample for today’s article is EVGA’s top-tier GTX 970 card, the EVGA GeForce GTX 970 FTW ACX 2.0, or more commonly known as simply the GTX 970 FTW. The GTX 970 FTW builds on the standard GTX 970 design by incorporating EVGA’s latest iteration of their ACX open air cooler, the ACX 2.0, and pairing that with a very sizable factory overclock.

GeForce GTX 970 Specification Comparison
	Reference	EVGA FTW
Core Clock	1050MHz	1216MHz
Boost Clock	1178MHz	1367MHz
Memory Clock	7GHz GDDR5	7GHz GDDR5
VRAM	4GB	4GB
TDP	145W	145W
Cooler	N/A	Double Slot Open Air
Length	N/A	9.5"
Warranty	N/A	3 Years
MSRP	$329	$369

Compared to the reference GTX 970 with its 1050MHz base clock and 1178MHz boost clock, the GTX 970 FTW ships with a base clock of 1216MHz and a boost clock of 1367MHz. This is a 166MHz (16%) base clock increase and 189MHz (16%) boost clock increase, making this one of the most heavily factory overclocked cards we’ve seen in some time. As we’ve discovered with GTX 980, NVIDIA’s reference clocks leave quite a bit of overclocking headroom on the table, and as a result EVGA is tapping into that headroom for their factory overclocked models. Meanwhile the memory clock with its far more limited overclocking headroom remains unchanged on the GTX 970 FTW at 7GHz.

Starting as always at the top, from a high level overview the GTX 970 FTW is a fairly typical dual fan open air cooler, which along with triple fan variants have become the de facto standard for open air cooling on mid-to-high powered video cards. The use of an open air design allows for very good temperature control and acoustics in exchange for offloading much of the cooling work onto the chassis itself.

Since this is the first GTX 970 we’ve looked at, it’s worth noting that GTX 970 in general occupies a very odd spot as far as picking the optimal cooler is concerned. High-end high-powered cards will often use blowers (or at least make them an option) for chassis compatibility and space (multi-GPU) reasons, while low-powered cards will be entirely open air because the heat load they present is relatively limited in the grand scheme of things. Then in the middle you have cards such as the GTX 970, whose 145W TDP is far less than some of these high-powered cards but still more than even an Intel Core i7-5960X.

145W of heat is still going to be enough to stress some chassis designs (especially many OEM designs) while for other chassis designs it’s only a fraction of what they can handle. What this means is that there is no ideal cooler at this level; a blower is generally not necessary, but an open air cooler is not going to be universally compatible either. EVGA as it turns out is going to be covering both possibilities – they are one of just two partners offering models with blowers – so with the exception of the open air cooled GTX 970 FTW, most of EVGA’s other GTX 970 cards have both an open air option and a blower option.

In any case, for the GTX 900 series EVGA is introducing the first of their ACX 2.0 coolers. ACX 2.0 is the latest iteration on EVGA’s open air ACX cooler, which we first saw just over a year ago on EVGA’s GTX 780 Superclocked. The original ACX cooler was an effective though none the less fairly typical twin fan cooler, delivering the kind of low load noise levels and temperatures that these cooler styles have come to represent, while EVGA went the extra mile on construction with a few consistent additions such as the use of double ball bearings on the fans.

For the ACX 2.0 cooler EVGA has focused on further optimizing their cooler design for better performance, including in ways we weren’t expecting. From a cooling performance standpoint EVGA has increased the number of fins per fan to 11 to improve airflow and simultaneously tweaked the design of the blades to make them more swept to reduce noise. However what was more surprising was that EVGA has also been redesigning the fan motors as well, and not just for noise considerations but for power considerations as well. In fact all of EVGA’s design tweaks for ACX 2.0 ultimately fold back into reducing power consumption.

The focus on power consumption for a cooler certainly sounds odd at first, but in reality it’s an excellent reflection of the state of the PC video card market and how insanely competitive NVIDIA’s partners are. Because fan power is calculated as part of the card’s overall TDP budget on NVIDIA-approved designs – 145W in the case of GTX 970 – whatever power is taken up by the fan is therefore unavailable for the GPU. When coupled with a TDP limited design – a description that fits GTX 970 to a T – this means that even shaving off a couple of watts of power from the fans can free up enough power for another boost bin or two, and so for ACX 2.0 this is exactly what EVGA has done.

EVGA: ACX 2.0 Performance at Fixed Acoustic (36dB)
	ACX 2.0	ACX
Fan Speed	2000 RPM	2300 RPM
Power	0.92W	2.05W
GPU Temp	64C	66C

 

EVGA: ACX 2.0 Performance at Fixed Acoustic (47dB)
	ACX 2.0	ACX
Fan Speed	2900 RPM	3300 RPM
Power	2.65W	6.10W
GPU Temp	58C	58C

 

The fact that we’re even talking about optimizations in the couple of watt range on a 145W card is admittedly absurd, but again it goes to show the state of the market. As EVGA argues, their low power fan design means better performance and better overclocking than competing cards without these optimizations, and in this case they are technically correct if only marginally so. Still, with a number of NVIDIA partners using similar dual fan designs, any edge EVGA can take in design and benchmark results is going to be worth it to them. And for end users? With Maxwell using 13MHz per boost bin, these power savings would translate to a consistent but highly limited 1-2% performance improvement.

Moving on, while it’s not present in the initial retail units and was not available in time for our review, EVGA tells us that they are going to be implementing zero fan speed idling on the GTX 970 FTW in the near future. This idea, first popularized by Asus in their Strix line of video cards, is based around the concept of shutting down the fans entirely when the card is idling or otherwise operating at low enough temperatures that active cooling is not necessary. By cutting the fans during idle, the card is functionally turned into a passive card that makes no noise, offering even better idle noise characteristics than today’s coolers.

Cooling apparatuses that switch between active and passive cooling are not new in the PC space – your laptop has done this for years – but this is something that until very recently has never been viable on high-end video cards. With GTX 900 series cards drawing sub-10W at idle, it’s finally practical to run these cards with passive cooling at idle, and this is exactly the direction that Asus and now EVGA are going. Meanwhile since this feature is not currently enabled on the GTX 970 FTW we’re not in a position to comment on it, but if EVGA can replicate Asus’s success here then we are looking forward to trying it once the updated vBIOS becomes available.

Update 9/26: While it's still under development, EVGA has given us an early copy of their zero fan speed idle vBIOS for testing, and we can confirm it's working. Please see our Power, Temp, & Noise section for more details

Fans aside, the rest of the GTX 970 FTW’s cooling apparatus is typical of cards in this class. EVGA uses an aluminum heatsink that runs the entire 9.5 inch length of the card, and through which runs a set of 4 heatpipes. These heatpipes in turn run through an aluminum baseplate that makes contact with the GPU, which provides heat transference between the GPU and the heatpipes.

There is no active cooling of the GDDR5 RAM nor of the VRMs on this card, with both components cooled entirely by any airflow that makes it through the heatsink. For a 145W card VRMs cooled without a heatsink is not unexpected, and like the GDDR5 RAM the heat output is low enough that it should be fine as long as it receives constant airflow when under load.

Looking at the PCB itself, the PCB is strongly reminiscent of the GTX 760 reference PCB, which like EVGA’s card puts the vast majority of discrete components at the front of the card. While we don’t have an EVGA GTX 760 on hand, given the identical memory bus and lower power requirements we wouldn’t be surprised if these PCBs were nearly identical. In any case, EVGA tells us that the FTW version of their GTX 970 implements an additional VRM to allow for higher voltages, though we don’t have a non-FTW card to compare it to. Otherwise the GTX 970 uses a stretched version of this common PCB, with EVGA using the far portion of the card (beyond the PCIe sockets) for a few more discrete components. The GDDR5 VRAM itself is split between the front and back, with 4 modules on either side.

Switching over to the back side of the PCB, along with the 4 GDDR5 VRAM modules we also find EVGA’s last major feature for the GTX 970, which is triple BIOS support. Primarily intended for those users looking to overclock their cards through BIOS modifications, EVGA offers 3 BIOSes on this card that can be accessed through a switch on the back of the card. Unlike most cards with switchable BIOSes, this switch (labeled SW1) is extremely small and out of the way, so you won’t be able to find it without specifically looking for it. Even then you’ll need a small tool of some kind (e.g. #00 Phillips screwdriver) to actually manipulate the switches. Multiple BIOSes are an uncommon sight on NVIDIA cards, so this makes the GTX 970 FTW one of the few NVIDIA cards we’ve seen over the last few years with this feature.

Moving on, the overall build quality of the card is very good. Though it can’t match the overall all-in-one design of a blower, EVGA has taken care to account for both the needs of the ACX cooler and the PCB in designing this card. The PCB features a metal reinforcement plate that covers the far end and is screwed down on either side, keeping this segment of PCB from undergoing any flexing. Between that and the structural support provided by the heatsink itself, the card is as solid as any open air card could hope to be. Meanwhile the ACX 2.0 shroud is made out of a thick plastic and screwed to the heatsink at several points, so there is no flexing or bending of the shroud either.

Unlike the GTX 980, EVGA’s GTX 970 FTW is still using the older NVIDIA display I/O standard of 2x DL-DVI, 1x HDMI 2.0, and 1x DP 1.2. This card does support HDMI 2.0 since that feature is driven by the underlying GPU, but otherwise you’re looking at the same I/O options as on the GTX 600 and GTX 700 series, which means driving a G-Sync Surround setup will not be an option here.

Along with the aforementioned display I/O connections, at the top of the card you’ll find a pair of SLI connectors and a pair of 6-pin PCIe power sockets. Despite the presence of the PCIe power sockets this is still a 145W TDP card, so the use of 2 sockets provides some additional buffer room along with the card’s 110% (160W) TDP overclocking limit.

Rounding out EVGA’s package, the GTX 970 FTW also includes the company’s standard overclocking software suite. Composed of EVGA PrecisionX and EVGA OC ScannerX, these utilities provide for overclocking/monitoring and stability checking respectively.

Just launched for the GTX 900 series is PrecisionX 16, which is the first supported release of PrecisionX built in-house after the company discontinued its relationship with RivaTuner author Alexey Nicolaychuk. As a result of this behind-the-scenes change in development, PrecisionX has undergone a fairly major overhaul in replacing the RivaTuner bits with new code. None the less EVGA has done a good job of retaining all of the utility’s major features, so along with overclocking support through NVAPI it offers a suite of OSD and monitoring options. On that note, next week we will be taking an in-depth look at PrecisionX 16 along with a collection of other overclocking utilities as part of a previously scheduled round-up of overclocking utilities, so be sure to stay tuned for that.

For product support the GTX 970 FTW comes with EVGA’s standard 3 year transferable warranty, with individual 2 or 7 year extensions available for purchase upon registration. The card also qualifies for EVGA’s step-up upgrade program upon registration, so it can be upgraded for a more expensive card within 90 days for just the price difference.

Finally, EVGA has set the MSRP on the GTX 970 FTW at $369, which is $40 (12%) over the MSRP for a reference GTX 970. EVGA offers 3 tiers of factory overclocked cards, so as the highest tier card it also commands the highest price. From a hardware perspective EVGA’s factory overclock is not so high as to be unreachable on reference cards, but in this case as always, buying factory overclocked means paying more to avoid the overclocking lottery.

The Test

Quickly touching on the subject of compatibility, as readers of last week’s GTX 980 review may recall, we had initial compatibility issues with our GTX 970 FTW that prevented us from including it in our review. Since then NVIDIA has been able to isolate the issue and has put together the 334.16 drivers, which include a fix for the problem we were seeing. So we are now up and running. NVIDIA tells us that the issue only impacted certain motherboards (such as our ASRock Fatal1ty X79 Professional), and as far as we can tell that appears to be correct, as we have not seen any other reports of compatibility issues.

Moving on, for the purposes of our testing we will be looking at both the GTX 970 FTW in its shipping configuration and in a reference clocked configuration. EVGA has given us the reference GTX 970 vBIOS to flash to this card (taking advantage of the triple BIOS feature), allowing us to turn it into a standard GTX 970 for that part of our testing.

CPU:	Intel Core i7-4960X @ 4.2GHz
Motherboard:	ASRock Fatal1ty X79 Professional
Power Supply:	Corsair AX1200i
Hard Disk:	Samsung SSD 840 EVO (750GB)
Memory:	G.Skill RipjawZ DDR3-1866 4 x 8GB (9-10-9-26)
Case:	NZXT Phantom 630 Windowed Edition
Monitor:	Asus PQ321
Video Cards:	AMD Radeon R9 290X AMD Radeon R9 290 AMD Radeon R9 280X AMD Radeon HD 7970 AMD Radeon HD 6970 EVGA GeForce GTX 970 FTW ACX 2.0 NVIDIA GeForce GTX 980 NVIDIA GeForce GTX 780 Ti NVIDIA GeForce GTX 780 NVIDIA GeForce GTX 770 NVIDIA GeForce GTX 680 NVIDIA GeForce GTX 670 NVIDIA GeForce GTX 580
Video Drivers:	NVIDIA Release 344.07 Beta NVIDIA Release 344.16 Beta AMD Catalyst 14.300.1005 Beta
OS:	Windows 8.1U1 Pro

 

Metro: Last Light

As always, kicking off our look at performance is 4A Games’ latest entry in their Metro series of subterranean shooters, Metro: Last Light. The original Metro: 2033 was a graphically punishing game for its time and Metro: Last Light is in its own right too. On the other hand it scales well with resolution and quality settings, so it’s still playable on lower end hardware.

It seems fitting that we start with a game where the GTX 970 and R9 290XU start out tied. Even at the R9 290XU’s strongest hand – 4K – the GTX 970 is at parity and that remains for 1440p as well. Only at 1080p does the GTX 970 even begin to trail the R9 290XU.

This game ends up being a very good summary of what we’re going to see. The GTX 970 and R9 290XU trade blows from game-to-game, but in the end the two are a tie, just as we see here. Which for NVIDIA is a great outcome, as it means they’re tying a card that's nearly 50% more expensive.

Meanwhile if we push on the gas pedal a bit harder with the GTX 970 FTW, we see the EVGA card pull ahead of the R9 290XU and stock GTX 970 by 8%. This is a bit above average overall for the GTX 970 FTW, but it does conveniently highlight the fact that even if AMD officially pushed the clock speeds of R9 290XU a bit more, they’d still end up with GTX 970 right next to them.

Company of Heroes 2

Our second benchmark in our benchmark suite is Relic Games’ Company of Heroes 2, the developer’s World War II Eastern Front themed RTS. For Company of Heroes 2 Relic was kind enough to put together a very strenuous built-in benchmark that was captured from one of the most demanding, snow-bound maps in the game, giving us a great look at CoH2’s performance at its worst. Consequently if a card can do well here then it should have no trouble throughout the rest of the game.

Company of Heroes 2 was one of the only games that the GTX 980 didn’t take a significant lead in, and consequently it’s one of the few games that GTX 970 will lose by a significant margin. With the exception of 1080p the stock GTX 970 can’t keep pace with the R9 290, let alone the R9 290XU. Overall NVIDIA’s second-tier card will trail the AMD flagship by 12%, which in the grand scheme of things is still going to be much narrower than the price difference.

In this case Company of Heroes 2 seems especially fond of shader performance. So the GTX 970 drops off by a bit more than in most other games, coming in at just 85% of the performance of GTX 980.

With AMD already doing well on minimum framerates against the GTX 980, against the GTX 970 NVIDIA ends up further behind. When push comes to shove in this game’s hardest scenes, the GTX 970 buckles a bit more than the competition.

Bioshock Infinite

Bioshock Infinite is Irrational Games’ latest entry in the Bioshock franchise. Though it’s based on Unreal Engine 3 – making it our obligatory UE3 game – Irrational had added a number of effects that make the game rather GPU-intensive on its highest settings. As an added bonus it includes a built-in benchmark composed of several scenes, a rarity for UE3 engine games, so we can easily get a good representation of what Bioshock’s performance is like.

Bioshock Infinite immediately turns the tables on AMD, and outside of a near-tie at 4K the GTX 970 takes a significant lead here. At both 1440p and 1080p the GTX 970 clears the R9 290XU by 21%, and even against the GTX 780 it enjoys quite a lead.

Battlefield 4

Our latest addition to our benchmark suite and our current major multiplayer action game of our benchmark suite is Battlefield 4, DICE’s 2013 multiplayer military shooter. After a rocky start, Battlefield 4 has finally reached a point where it’s stable enough for benchmark use, giving us the ability to profile one of the most popular and strenuous shooters out there. As these benchmarks are from single player mode, based on our experiences our rule of thumb here is that multiplayer framerates will dip to half our single player framerates, which means a card needs to be able to average at least 60fps if it’s to be able to hold up in multiplayer.

In Battlefield 4 resolution makes all the difference. AMD’s 4K advantage is in full force here, while that solid lead errodes and the GTX 970 catches up for 1440p and 1080p..

Looking at just the NVIDIA lineup for a second, while not even GTX 980 was able to cross 60fps at 1440p, it does prove that its 17% performance advantage counts for something by being able to push framerates in the high 50s, all while GTX 970 can’t even crack 50fps. EVGA’s FTW overclock will get you there though, and for that matter it can even cross the 60fps mark at 4K.

Crysis 3

Still one of our most punishing benchmarks, Crysis 3 needs no introduction. With Crysis 3, Crytek has gone back to trying to kill computers and still holds “most punishing shooter” title in our benchmark suite. Only in a handful of setups can we even run Crysis 3 at its highest (Very High) settings, and that’s still without AA. Crysis 1 was an excellent template for the kind of performance required to drive games for the next few years, and Crysis 3 looks to be much the same for 2014.

Crysis 3 is another game where the outcome between the R9 290XU and GTX 970 depends on the resolution. At 4K Low the GTX 970 trails the R9 290XU by 10%, only for the two to get within a frame of each other at 1440p High. Past that it’s all NVIDIA at 1080p. This once again neatly illustrates that AMD still holds a general resolution scaling advantage over NVIDIA and the Maxwell 2 architecture. Though since we’re looking at a $329 card that’s cheaper than any 4K monitor, that’s not an advantage that’s going to be of much value in the real world.

Crysis: Warhead

Up next is our legacy title for 2014, Crysis: Warhead. The stand-alone expansion to 2007’s Crysis, at over 5 years old Crysis: Warhead can still beat most systems down. Crysis was intended to be future-looking as far as performance and visual quality goes, and it has clearly achieved that. We’ve only finally reached the point where single-GPU cards have come out that can hit 60fps at 1920 with 4xAA, never mind 2560 and beyond.

With the GTX 980 already falling just short of beating the R9 290XU in Crysis: Warhead, this is the only other instance where the GTX 970 isn’t a completely competitive card. Even the EVGA FTW overclock can’t help it catch up to the R9 290, let alone the R9 290XU. Crysis simply calls for more shader power than what the GTX 970 can realistically deliver.

The story is much the same with minimum framerates. Although GTX 970 doesn’t fare any worse relatively speaking, it also doesn’t get to close the gap.

Total War: Rome 2

The second strategy game in our benchmark suite, Total War: Rome 2 is the latest game in the Total War franchise. Total War games have traditionally been a mix of CPU and GPU bottlenecks, so it takes a good system on both ends of the equation to do well here. In this case the game comes with a built-in benchmark that plays out over a forested area with a large number of units, definitely stressing the GPU in particular.

For this game in particular we’ve also gone and turned down the shadows to medium. Rome’s shadows are extremely CPU intensive (as opposed to GPU intensive), so this keeps us from CPU bottlenecking nearly as easily.

Yet again we have a situation where the winner and loser is effectively decided by the resolution in use. GTX 970 will trail at 4K, only to take a slight lead at 1440p. As we’ve stated before 1440p and 1080p are going to be the sweet spots for GTX 970 based on its $329 price tag, so for GTX 970 this means it’s winning or tying in the resolutions where it matters the most.

Of course if you factor in the FTW overclock then the point becomes moot. With the GTX 970 and R9 290XU tracking so close together, the higher clock speeds mean that a GTX 970 card can push ahead of R9 290XU on average.

Thief

Our latest addition to our benchmark suite is Eidos Monreal’s stealth action game, Thief. Set amidst a Victorian-era fantasy environment, Thief is an Unreal Engine 3 based title which makes use of a number of supplementary Direct3D 11 effects, including tessellation and advanced lighting. Adding further quality to the game on its highest settings is support for SSAA, which can eliminate most forms of aliasing while bringing even the most powerful video cards to their knees.

Until we hit 1080p, the GTX 970 once again runs neck-and-neck with the R9 290XU. Otherwise GTX 970 trails its full-fledged sibling by about 15%, reiterating the GTX 980’s rather consistent performance advantage.

Meanwhile GTX 970 does fare a bit better in minimum framerates. It’s not by much, but it ever so slightly remains ahead in the lowest performance situations.

GRID 2

The final game in our benchmark suite is also our racing entry, Codemasters’ GRID 2. Codemasters continues to set the bar for graphical fidelity in racing games, and with GRID 2 they’ve gone back to racing on the pavement, bringing to life cities and highways alike. Based on their in-house EGO engine, GRID 2 includes a DirectCompute based advanced lighting system in its highest quality settings, which incurs a significant performance penalty but does a good job of emulating more realistic lighting within the game world.

Our final game once again sees the GTX 970 start out trailing the R9 290XU, only to start pulling ahead once the resolution drops. At 1440p it’s practically a tie, and at 1080p it becomes a clear victory for the GTX 970.

Meanwhile it’s interesting to note that at an average performance gap of 10%, this is the game with the smallest performance difference between the GTX 980 and GTX 970. Compared to ROP throughput and memory bandwidth, shader and texture throughput isn’t being tested here by as much, which helps to negate some of the GTX 970’s innate disadvantage.

Synthetics

As always we’ll also take a quick look at synthetic performance. These tests mainly serve as a canary for finding important architectural and configuration changes.

Tessellation performance has scaled very closely with the change in SMMs and clock speeds, just as we would expect here.

Texel throughput has also taken a hit in accordance with the loss of SMMs and clock speed. Based on gaming performance the GTX 970 doesn’t appear to be too badly handicapped here, but it definitely doesn’t have much in the way of texel throughput to spare.

Pixel throughput on the other hand ends up being extremely odd and not at all what we were expecting. The GTX 970 takes an incredible dive here, with its pixel fillrate dropping by 26%. At a high level this test is bounded by memory bandwidth and ROP throughput, and both of these factors should be identical between GTX 980 and GTX 970. Instead we see GTX 970 lose more performance than should theoretically be possible, as the 26% drop is more than the accumulated difference between the clock speed and SMM differences.

At this point we’re still trying to figure out exactly what’s going on. We have no other evidence that there’s a difference in ROP throughput or memory bandwidth between the GTX 980 and GTX 970 so it is not clear to us where the difference lies. One possibility is that this is somehow bottlenecked at the Raster Engine level – where each of the four engines accounts for 25% of the work – but the pigeonhole principle means that NVIDIA can’t disable a GPC since at least 1 SMM must be active in each GPC partition. This matter will require further research.

Compute

With GTX 980 NVIDIA surprised us with their stunning turnaround in in GPU compute performance, which saw them capable of reaching the top in many compute benchmarks they couldn’t before. GTX 970 meanwhile should benefit from these architectural and driver improvements, though since compute is nearly analogous to shader performance this is also a case where the performance difference between the GTX 970 and GTX 980 stands to be among its widest.

As always we’ll start with LuxMark2.0, the official benchmark of SmallLuxGPU 2.0. SmallLuxGPU is an OpenCL accelerated ray tracer that is part of the larger LuxRender suite. Ray tracing has become a stronghold for GPUs in recent years as ray tracing maps well to GPU pipelines, allowing artists to render scenes much more quickly than with CPUs alone

Thanks to GTX 980 taking the top spot here, GTX 970 still maintains a small lead over R9 290XU. So even with the GTX 970's weaker performance, it can still manage to outperform AMD's flagship in this case.

For our second set of compute benchmarks we have CompuBench 1.5, the successor to CLBenchmark. We’re not due for a benchmark suite refresh until the end of the year, however as CLBenchmark does not know what to make of GTX 980 and is rather old overall, we’ve upgraded to CompBench 1.5 for this review.

In the cases where the GTX 980 does well, so does the GTX 970. In the cases where the GTX 980 wasn’t fast enough to top the charts, the GTX 970 will be similarly close behind. Overall compared to AMD’s lineup we see the whole gamut, from a tie between the GTX 970 and R9 290XU to victories for either card.

Our 3rd compute benchmark is Sony Vegas Pro 12, an OpenGL and OpenCL video editing and authoring package. Vegas can use GPUs in a few different ways, the primary uses being to accelerate the video effects and compositing process itself, and in the video encoding step. With video encoding being increasingly offloaded to dedicated DSPs these days we’re focusing on the editing and compositing process, rendering to a low CPU overhead format (XDCAM EX). This specific test comes from Sony, and measures how long it takes to render a video.

As expected, GTX 970 sheds a bit of performance here. AMD still holds a lead here overall, and against GTX 970 that lead is a little bit larger.

Moving on, our 4th compute benchmark is FAHBench, the official Folding @ Home benchmark. Folding @ Home is the popular Stanford-backed research and distributed computing initiative that has work distributed to millions of volunteer computers over the internet, each of which is responsible for a tiny slice of a protein folding simulation. FAHBench can test both single precision and double precision floating point performance, with single precision being the most useful metric for most consumer cards due to their low double precision performance. Each precision has two modes, explicit and implicit, the difference being whether water atoms are included in the simulation, which adds quite a bit of work and overhead. This is another OpenCL test, utilizing the OpenCL path for FAHCore 17.

With the GTX 980 holding such a commanding lead here, even with the GTX 970’s lower performance it still is more than enough to easily beat any other card in single precision Folding @ Home workloads. Only in double precision with NVIDIA’s anemic 1:32 ratio does GTX 970 falter.

Wrapping things up, our final compute benchmark is an in-house project developed by our very own Dr. Ian Cutress. SystemCompute is our first C++ AMP benchmark, utilizing Microsoft’s simple C++ extensions to allow the easy use of GPU computing in C++ programs. SystemCompute in turn is a collection of benchmarks for several different fundamental compute algorithms, with the final score represented in points. DirectCompute is the compute backend for C++ AMP on Windows, so this forms our other DirectCompute test.

Recently this has been a stronger benchmark for AMD cards than NVIDIA cards, and consequently GTX 970 doesn’t enjoy quite the lead it sees elsewhere. Though not too far behind R9 280X and even R9 290, like GTX 980 it can’t crunch numbers quite fast enough to keep up with R9 290XU.

Power, Temperature, & Noise

As always, last but not least is our look at power, temperature, and noise. Next to price and performance of course, these are some of the most important aspects of a GPU, due in large part to the impact of noise. All things considered, a loud card is undesirable unless there’s a sufficiently good reason – or sufficiently good performance – to ignore the noise.

Having already seen GM204 in action with the GTX 980 series, we have a pretty solid foundation to base expectations on for the GTX 970. GTX 970 should offer reduced power consumption at reduced performance, and with EVGA’s open air cooler this card will never hit 80C, so if it is ever throttled it will be entirely by TDP and not by temperatures as was the case with GTX 980.

GeForce GTX 900 Series Voltages
EVGA GTX 970 Boost Voltage	EVGA GTX 970 Idle Voltage	GTX 980 Boost Voltage
1.218v	0.856v	1.225v

We’ll start as always with voltages, which in this case are in-line with what we have already seen with GTX 980. At 1.218v for the card’s maximum boost bin, the GTX 970 is only a step below the GTX 980 in voltages. We sometimes see these cards binned to operate at lower voltages for power saving purposes, though with GM204’s already low power consumption I’m not sure how necessary that is from a product development standpoint.

Next up, let’s take a look at average clock speeds.

GeForce GTX 900 Series Average Clockspeeds
	EVGA GTX 970 FTW	EVGA GTX 970 (Stock)	GTX 980
Max Boost Clock	1418MHz	1252MHz	1252MHz
Metro: LL	1337MHz	1222MHz	1192MHz
CoH2	1265MHz	1143MHz	1177MHz
Bioshock	1370MHz	1221MHz	1201MHz
Battlefield 4	1373MHz	1227MHz	1227MHz
Crysis 3	1354MHz	1202MHz	1227MHz
TW: Rome 2	1332MHz	1177MHz	1161MHz
Thief	1329MHz	1202MHz	1190MHz
GRID 2	1340MHz	1202MHz	1151MHz
Furmark	912MHz	910MHz	923MHz

Despite not being temperature limited, what we can see right away is that regardless of the clock speed settings it uses, the GTX 970 FTW is TDP limited under all scenarios. At no point in time are we able to maintain the card’s top boost bin, and instead the card spends its time fluctuating between the boost bins it can hold while maintaining power consumption of 145W. The actual drop off from the maximum boost bin depends heavily on the game; some games average clock speeds close to the maximum, while others have to pull way back.

Overall the GTX 970 FTW is a bit more TDP limited in its factory overclocked FTW configuration than its reference clocked configuration, as higher clock speeds draw greater power levels even though voltages are constant. Otherwise we find that in the FTW configuration the card is on average clocked 138MHz higher than the reference configuration, which isn’t quite as high as the 166-189MHz difference in their official specifications. In other words, the FTW’s performance advantage is equivalent to a 140MHz overclock.

Moving on to power and starting with idle power consumption, the GTX 970 FTW does not fare quite as well as the GTX 980. Despite having fewer SMMs lit up, EVGA’s card draws a consistent 3W more at the wall. In lieu of other data it’s difficult to determine exactly whether this is due to chip-to-chip variation or EVGA’s board, but the latter explanation is more plausible at this moment.

Load power consumption for GTX 970 follows the same path as the GTX 980, almost down to the watt in this case. The GTX 970 FTW actually sees slightly higher power consumption at the wall, while reducing it to reference clocks brings down the power consumption to a flat 300W. Since we are underclocking a factory overclocked card I’m not going to read into the underclocked results too much, but for the FTW we’re looking at power consumption similar to GTX 980 but not quite the performance to match. This is likely thanks to the GTX 980 operating a bit under its TDP due to thermal throttling.

Still, GTX 970 FTW remains as one of the lower power consuming cards of anything in our lineup. Similarly performing cards such as the GTX 780 and Radeon R9 290XU (which is virtually tied on performance here) clearly draw more power, significantly so in the case of the latter.

The story is much the same with FurMark. Since this a purely TDP limited scenario the GTX 970 FTW configuration draws 284W at the wall regardless of the clock speed settings used. In this worst case scenario power consumption is 10W less than the GTX 980 at the wall and 71W less than AMD’s R9 290. Meanwhile since the GTX 970 is TDP limited under all gaming workloads, in this case this is not a bad proxy for overall power consumption while gaming.

While idle power consumption didn’t quite top the charts, idle temperatures do. At 28C the GTX 970 FTW is barely above room temperature.

Owing to its open air cooler, low TDP, and TDP throttling, the load temperature for the GTX 970 FTW tops out at 63C in all scenarios. Whether it’s gaming or FurMark, 100% TDP means the card finds its cooling/temperature equilibrium at nearly 20C less than its thermal throttling point.

In this case since this is one of the only open air coolers in this lineup – the only other card was the R9 280X – this means the GTX 970 FTW tops the charts thanks to its design. We’ve already covered the pros and cons of open air coolers many times before, and when it comes to temperatures this is a definite pro.

If there is an Achilles heel for the GTX 970 FTW then it is idle noise levels. EVGA’s ACX 2.0 cooler just doesn’t idle very well; even at its minimum fan speed, it still pushes enough air to push the sound meter to 42.7 dB. At this noise level the GTX 970 FTW is not in terrible shape, but near-silent at idle it is not. As it currently stands this card is not going to be a great choice for users looking for a system that can idle with little noise, especially compared to some of the 38dB cards we’ve seen over the last year.

This I suspect is why EVGA is looking into changing their card over to using passive cooling at idle. If EVGA can successfully pull off that modification and turn off the fans when idling, then they will resolve the card’s one true weakness and go from the bottom of this chart to the top.

Truth be told, even load noise levels aren’t all that impressive here. 47.5dB is still near the top of our charts, but the GTX 980 of all things ended up being ever so slightly quieter. Which perhaps more than anything else goes to show just what that $550 price tag is paying for when it comes to cooling.

Without more GTX 970 cards it’s hard to place EVGA’s card among its true competition. But given what we’ve seen previous 150W cards do in the past, it’s safe to say that EVGA is running a bit loud at both idle and load for the amount of heat they have to remove. I suspect the ACX 2.0 cooler used here is a bit oversized for a 145W card, which is similarly reflected in the low temperature results. Given the kind of power they have at hand, EVGA could likely do better with a more conservative fan curve.

Looking at the broader picture, while EVGA may not do well against other open air coolers, they are going to do well against most other blowers here. Even the GTX 670 is louder, with a 2.5dB difference in EVGA’s favor.

When it comes to FurMark the GTX 970 FTW goes to the top of our charts, but it remains within a stone’s throw of the GTX 980.

Overall the GTX 970 FTW is a bit of a mixed bag. EVGA’s aggressive fan curve and ACX 2.0 cooler produce some amazing GPU temperatures, even for a 145W card. On the other hand EVGA’s noise levels are simply so-so for an open air cooler, with EVGA benefitting from most of the competition in our charts being louder blowers.

To be clear it’s still a solid performance and 47.5 dB for the second fastest single-GPU card on the planet is incredibly quiet. I don’t even have to throw in an R9 290XU comparison to make that point, though that card is AMD’s closest competitor from a gaming performance standpoint. Still, I only wonder how well other open air coolers could do in this situation given what we’ve seen from other cards in the past. If nothing else, perhaps we’ve been spoiled by the GTX 980 and its top-tier blower.

Update 9/26: Though EVGA missed the initial cut-off for this review, later in the day they turned in an beta/engineering version of their upcoming vBIOS with the zero fan speed idle changes worked in, along with a new fan curve. We can confirm that the zero fan speed function is working in this BIOS, rendering the card silent at idle.

Overall EVGA is keeping the fans off until about 63C, at which point the fan controller starts ramping up the fans until it finally supplies enough power for them to kick in. Even under a max TDP load (FurMark) the fans are up and running by 67C.

At the same time EVGA’s new fan curve has adjusted their equilibrium point from 63C to 73C. At max TDP load our card now tops out at 73C, which corresponds to a fan duty cycle of 29%, or 42dB. This new curve places the fan load noise levels at less than the previous curve’s idle levels, and now makes this one of the quietest midrange enthusiast cards we have ever reviewed.

EVGA ACX 2.0 Cooler Performance
	Shipping BIOS	Engineering BIOS
Idle Fan Speed	30%	0%
Load Fan Speed	38%	29%
Idle Temperature	28C	33C
Load Temperature	63C	73C
Idle Noise	42.7dB	36.3dB
Load Noise	47.5dB	42.0dB

For the time being this BIOS is still under development (and is subject to change), with EVGA telling us that their goal is to release it to users in the “coming weeks.” Since this BIOS is still in development and not available to the public we will not be changing our official results, but it certainly warrants a proper second look once EVGA ships this BIOS.

Overclocking

With GTX 980 we saw first-hand how GM204 had very significant overclocking headroom. Even without the ability to meaningful overvolt on NVIDIA cards, we were able to push our base GPU clock speed up from 1126MHz to 1377MHz, or in terms of the maximum boost bin, from 1265MHz to 1515MHz. Consequently with GTX 970 shipping at lower clock speeds, we have very lofty expectations here.

But running counter to that will be TDP. As we have already seen, GTX 970 is TDP limited right out of the gate, so even if our card has more clock speed headroom, its 110% TDP limit doesn’t leave much more in the way of power headroom. Furthermore as this is already a factory overclocked card, there’s no guarantee that EVGA has left us much overclocking headroom to play with in the first place.

EVGA GeForce GTX 970 FTW Overclocking
	FTW	Overclocked
Core Clock	1216MHz	1241MHz
Boost Clock	1367MHz	1392MHz
Max Boost Clock	1418MHz	1455MHz
Memory Clock	7GHz	7.8GHz
Max Voltage	1.218v	1.243v

And in fact our results show they haven’t. We aren’t able to get even another 50MHz out of our GPU before errors start setting in; 25MHz is all we will get, which pushes our base GPU clock speed from 1216MHz to 1241MHz, and our maximum boost clock from 1418MHz to 1455MHz. Overall this is a weaker overclock than GTX 980, though not immensely so.

Meanwhile memory overclocking was just as fruitful as it was on GTX 980, with our card being able to handle up to 7.8GHz on its GDDR5 memory. As we saw with GTX 980 we’re nearly as memory bandwidth bottlenecked as we are GPU bottlenecked, but we will take what performance we can get.

As you’d expect from such a mild overclock, the performance increase is very limited. Our overclocked GTX 970 FTW does close on GTX 980 even more, but even with this full overclock it won’t overcome the 3 SMM deficit.

Overall in all likelihood the GTX 970 FTW benefits more from the 10% increase in TDP than it does the clock speed increase. GTX 970 – and GM204 in general – clearly desires to be fed with more voltage and more power overall than what any NVIDIA approved card is going to see.

Power consumption and noise tick up, but only slightly. The limited 10% TDP increase means that the amount of power the card can draw and dissipate as heat only increases slightly. You aren’t getting much more performance, but you also aren’t getting much more noise.

Final Words

While it was clear early on in our testing that the GeForce GTX 980 would be the new single-GPU champion, even after finishing testing of NVIDIA’s new flagship we weren’t sure what to expect out of the GTX 970. On the one hand NVIDIA had driven a very significant gap between the GTX 970 and the GTX 980 on theoretical performance – on paper the GTX 970 should be around 80% the performance of the GTX 980 – and on the other hand the GTX 980 was in practice some 16% faster than AMD’s flagship R9 290X. Immediately you can see the potential for even a slower GM204 card to be a threat to AMD, but it’s only an assumption until you have the data in hand.

It’s only fitting then that with the GTX 970 now up and running we find ourselves in a virtual tie between NVIDIA and AMD. Despite not even being NVIDIA’s flagship GM204 card, the GTX 970 is still fast enough to race the R9 290X to a dead heat – at 1440p the GTX 970 averages just 1% faster than the R9 290X. Only at 4K can AMD’s flagship pull ahead, and even then the situation becomes reversed entirely in NVIDIA’s favor at 1080p. As such the R9 290X still has a niche of its own, but considering the fact that GTX 970 is a $329 card I don’t seriously expect it to be used for 4K gaming, so the 1440p and 1080p comparisons are going to be the most appropriate comparisons here.

As is the case with AMD/NVIDIA ties in the past, this is an anything-but-equal sort of situation, but it strongly sells the idea of just how fast the GTX 970 is and just how dangerous it is to AMD. Ultimately what you are going to find is that the GTX 970 scores some big leads in some games only to fall well behind the R9 290X in others, and in other games still the two cards are tied through and through. In the end for pure performance neither card is superior, and in this case that’s a huge victory for NVIDIA.

With the GTX 970 NVIDIA only ties the R9 290X, but in the process they do so while consuming nearly 90W less power, generating far less noise, and most importantly delivering all of this at a two-thirds the cost. GTX 980 gave NVIDIA a well-earned lead over AMD, but it’s the one-two punch of GTX 980 and GTX 970 together that so solidly cement NVIDIA’s position as the top GPU manufacturer. It’s one thing for the $499 R9 290X to lose to NVIDIA’s $549 flagship, but to be outright tied by NVIDIA’s second tier card is a slap in the face that AMD won’t soon forget.

There’s not much more that can be said at this point other than that as of this moment the high-end performance landscape is entirely in NVIDIA’s favor. They have undercut AMD with better hardware at a lower price, leaving AMD in a very tenuous position. AMD would have to cut R9 290X’s priceby nearly $200 to be performance competitive, and even then they can’t come close to matching NVIDIA’s big edge in power consumption. To that end it’s a lot like the GTX 670 launch, but even in that case NVIDIA’s overall hardware and pricing advantage wasn’t quite as immense as it is today.

At anything over $300 there are only two single-GPU cards to consider: GTX 980 and GTX 970. Nothing else matters. For much of the last year NVIDIA has been more than performance competitive but not price competitive with AMD. So I am very happy to see NVIDIA finally reverse that trend and to do so in such a big way.

Moving on, in NVIDIA’s lineup the GTX 970 is a potent performer, but NVIDIA has left themselves a big enough gap that it doesn’t completely undercut their new flagship. GTX 980 remains 15% faster than GTX 970, and that’s no small difference. For $220 more GTX 980 is certainly not the value option, but then this is an NVIDIA flagship we’re talking about, and NVIDIA has always charged a premium there. Instead if you can’t afford the high price of the GTX 980 or simply don’t want to pay that premium, the GTX 970 is an excellent alternative. By pricing the card at $329 NVIDIA has done a great job of making a significant fraction of GM204’s performance available at a better price, and for this reason GTX 970 stands a very good chance of being the value champion for this generation.

Finally let’s talk about EVGA’s GeForce GTX 970 FTW ACX 2.0. EVGA has clearly put a lot of thought into their card and there is a good reason they remain one of NVIDIA’s closest partners. As the only GTX 970 we’ve looked at thus far we’re really only able to look at it on a pass-fail basis, but on that basis it’s a clear pass. The ACX 2.0 cooler is incredibly powerful when paired with the 145W GTX 970 (almost a bit too much) and EVGA continues to deliver some great features and software through items such as their triple BIOS capability and their software suite.

At the same time the FTW in particular is a solid value proposition, though how solid will depend on what you compare it to. Compared to GTX 980 it’s going to fall short by 7% or so despite the overclock, but in the process it essentially cuts the GTX 980’s lead in half. Given the $220 price difference between the reference GTX 970 and GTX 980, that $40 FTW premium is a great alternative. On the other hand this is in the end a factory overclocked card, and it’s entirely likely that most reference clocked GTX 970s could achieve similar clock speeds without paying the $40 premium. As is usually the case with factory overclocked cards, with the GTX 970 FTW you are paying for the peace of mind that comes from a sure thing and the support behind it, as opposed to playing the overclocking lottery.

With that said, I feel like EVGA does walk away from this launch with one vulnerability exposed, and that is the ACX 2.0 cooler. EVGA’s amazing cooling performance is undercut by their middle of the road noise performance, which although is still very good in light of GTX 970’s overall gaming performance, it is not as good as what we have seen other open air coolers do in the past. It’s by no means a deal breaker – especially given all of EVGA’s other advantages – but given the kind of quiet cooling possibilities that a 145W GPU should enable, EVGA is not exploiting it as well as they could.

And while this configuration isn’t going to be optimal for stock users, I don’t doubt for a second that EVGA’s overclocking community will have a field day with this one. With this much cooling headroom to work with the ACX 2.0 cooler is going hold up very well for users who want to overvolt on air.