The NVIDIA GeForce GTX 980 Review: Maxwell Mark 2
by Ryan Smith on September 18, 2014 10:30 PM ESTOverclocking GTX 980
One of GTX 750 Ti’s more remarkable features was its overclocking headroom. GM107 could overclock so well that upon initial release, NVIDIA did not program in enough overclocking headroom in their drivers to allow for many GTX 750 Ti cards to be overclocked to their true limits. This is a legacy we would be glad to see repeated for GTX 980, and is a legacy we are going to put to the test.
As with NVIDIA’s Kepler cards, NVIDIA’s Maxwell cards are subject to NVIDIA’s stringent power and voltage limitations. Overvolting is limited to NVIDIA’s built in overvoltage function, which isn’t so much a voltage control as it is the ability to unlock 1-2 more boost bins and their associated voltages. Meanwhile TDP controls are limited to whatever value NVIDIA believes is safe for that model card, which can vary depending on its GPU and its power delivery design.
For GTX 980 we have a 125% TDP limit, meanwhile we are able to overvolt by 1 boost bin to 1265MHz, which utilizes a voltage of 1.25v.
| GeForce GTX 980 Overclocking | ||||
| Stock | Overclocked | |||
| Core Clock | 1126MHz | 1377MHz | ||
| Boost Clock | 1216MHz | 1466MHz | ||
| Max Boost Clock | 1265MHz | 1515MHz | ||
| Memory Clock | 7GHz | 7.8GHz | ||
| Max Voltage | 1.25v | 1.25v | ||
GTX 980 does not let us down, and like its lower end Maxwell 1 based counterpart the GTX 980 turns in an overclocking performance just short of absurd. Even without real voltage controls we were able to push another 250MHz (22%) out of our GM204 GPU, resulting in an overclocked base clock of 1377MHz and more amazingly an overclocked maximum boost clock of 1515MHz. That makes this the first NVIDIA card we have tested to surpass both 1.4GHz and 1.5GHz, all in one fell swoop.
This also leaves us wondering just how much farther GM204 could overclock if we were able to truly overvolt it. At 1.25v I’m not sure too much more voltage is good for the GPU in the long term – that’s already quite a bit of voltage for a TSMC 28nm process – but I suspect there is some untapped headroom left in the GPU at higher voltages.
Memory overclocking on the other hand doesn’t end up being quite as extreme, but we’ve known from the start that at 7GHz for the stock memory clock, we were already pushing the limits for GDDR5 and NVIDIA’s memory controllers. Still, we were able to work another 800MHz (11%) out of the memory subsystem, for a final memory clock of 7.8GHz.
Before we go to our full results, in light of GTX 980’s relatively narrow memory bus and NVIDIA’s color compression improvements, we quickly broke apart our core and memory overclock testing in order to test each separately. This is to see which overclock has more effect: the core overclock or the memory overclock. One would presume that the memory overclock is the more important given the narrow memory bus, but as it turns out that is not necessarily the case.
| GeForce GTX 980 Overclocking Performance | |||||
| Core (+22%) | Memroy (+11%) | Combined | |||
| Metro: LL |
+15%
|
+4%
|
+20%
|
||
| CoH2 |
+19%
|
+5%
|
+20%
|
||
| Bioshock |
+9%
|
+4%
|
+15%
|
||
| Battlefield 4 |
+10%
|
+6%
|
+17%
|
||
| Crysis 3 |
+12%
|
+5%
|
+15%
|
||
| TW: Rome 2 |
+16%
|
+7%
|
+20%
|
||
| Thief |
+12%
|
+6%
|
+16%
|
||
While the core overclock is greater overall to begin with, what we’re also seeing is that the performance gains relative to the size of the overclock consistently favor the core overclock to the memory overclock. With a handful of exceptions our 11% memory overclock is netting us less than a 6% increase in performance. Meanwhile our 22% core overclock is netting us a 12% increase or more. This despite the fact that when it comes to core overclocking, the GTX 980 is TDP limited; in many of these games it could clock higher if the TDP budget was large enough to accommodate higher sustained clockspeeds.
Memory overclocking is still effective, and it’s clear that GTX 980 spends some of its time memory bandwidth bottlenecked (otherwise we wouldn’t be seeing even these performance gains), but it’s simply not as effective as core overclocking. And since we have more core headroom than memory headroom in the first place, it’s a double win for core overclocking.
To put it simply, the GTX 980 was already topping the charts. Now with overclocking it’s another 15-20% faster yet. With this overclock factored in the GTX 980 is routinely 2x faster than the GTX 680, if not slightly more.
But you do pay for the overclock when it comes to power consumption. NVIDIA allows you to increase the TDP by 25%, and to hit these performance numbers you are going to need every bit of that. So what was once a 165W card is now a 205W card.
Even though overclocking involves raising the temperature limit to 91C, NVIDIA's fan curve naturally tops out at 84C. So even in the case of overclocking the GTX 980 isn't going to reach temperatures higher than the mid-80s.
The noise penalty for overclocking is also pretty stiff. Since we're otherwise TDP limited, all of our workloads top out at 53.6dB, some 6.6dB higher than stock. In the big picture this means the overclocked GTX 980 is still in the middl of the pack, but it is noticably louder than before and louder than a few of NVIDIA's other cards. However interestingly enough it's no worse than the original stock GTX 680 at Crysis 3, and still better than said GTX 680 under FurMark. It's also still quieter than the stock Radeon R9 290X, not to mention the louder yet uber mode.
Facebook
Twitter
RSS
274 Comments
View All Comments
garadante - Sunday, September 21, 2014 - link
What might be interesting is doing a comparison of video cards for a specific framerate target to (ideally, perhaps it wouldn't actually work like this?) standardize the CPU usage and thus CPU power usage across greatly differing cards. And then measure the power consumed by each card. In this way, couldn't you get a better example ofgaradante - Sunday, September 21, 2014 - link
Whoops, hit tab twice and it somehow posted my comment. Continued:couldn't you get a better example of the power efficiency for a particular card and then meaningful comparisons between different cards? I see lots of people mentioning how the 980 seems to be drawing far more watts than it's rated TDP (and I'd really like someone credible to come in and state how heat dissipated and energy consumed are related. I swear they're the exact same number as any energy consumed by transistors would, after everything, be released as heat, but many people disagree here in the comments and I'd like a final say). Nvidia can slap whatever TDP they want on it and it can be justified by some marketing mumbo jumbo. Intel uses their SDPs, Nvidia using a 165 watt TDP seems highly suspect. And please, please use a nonreference 290X in your reviews, at least for a comparison standpoint. Hasn't it been proven that having cooling that isn't garbage and runs the GPU closer to high 60s/low 70s can lower power consumption (due to leakage?) something on the order of 20+ watts with the 290X? Yes there's justification in using reference products but lets face it, the only people who buy reference 290s/290Xs were either launch buyers or people who don't know better (there's the blower argument but really, better case exhaust fans and nonreference cooling destroys that argument).
So basically I want to see real, meaningful comparisons of efficiencies for different cards at some specific framerate target to standardize CPU usage. Perhaps even monitoring CPU usage over the course of the test and reporting average, minimum, peak usage? Even using monitoring software to measure CPU power consumption in watts (as I'm fairly sure there are reasonably accurate ways of doing this already, as I know CoreTemp reports it as its probably just voltage*amperage, but correct me if I'm wrong) and reported again average, minimum, peak usage would be handy. It would be nice to see if Maxwell is really twice as energy efficient as GCN1.1 or if it's actually much closer. If it's much closer all these naysayers prophesizing AMD's doom are in for a rude awakening. I wouldn't put it past Nvidia to use marketing language to portray artificially low TDPs.
silverblue - Sunday, September 21, 2014 - link
Apparently, compute tasks push the power usage way up; stick with gaming and it shouldn't.fm123 - Friday, September 26, 2014 - link
Don't confuse TDP with power consumption, they are not the same thing. TDP is for designing the thermal solution to maintain the chip temperature. If there is more headroom in the chip temperature, then the system can operate faster, consuming more power."Intel defines TDP as follows: The upper point of the thermal profile consists of the Thermal Design Power (TDP) and the associated Tcase value. Thermal Design Power (TDP) should be used for processor thermal solution design targets. TDP is not the maximum power that the processor can dissipate. TDP is measured at maximum TCASE"
https://www.google.com/url?sa=t&source=web&...
NeatOman - Sunday, September 21, 2014 - link
I just realized that the GTX 980 has a TDP of 165 watts, my Corsair CX430 watt PSU is almost overkill!, that's nuts. That's even enough room to give the whole system a very good stable overclock. Right now i have a pair of HD 7850's @ stock speed and a FX-8320 @ 4.5Ghz, good thing the Corsair puts out over 430 watts perfectly clean :)Nfarce - Sunday, September 21, 2014 - link
While a good power supply, you are leaving yourself little headroom with 430W. I'm surprised you are getting away with it with two 7850s and not experiencing system crashes.ET - Sunday, September 21, 2014 - link
The 980 is an impressive feat of engineering. Fewer transistors, fewer compute units, less power and better performance... NVIDIA has done a good job here. I hope that AMD has some good improvements of its own under its sleeve.garadante - Sunday, September 21, 2014 - link
One thing to remember is they probably save a -ton- of die area/transistors by giving it only what, 1/32 double precision rate? I wonder how competitive in terms of transistors/area an AMD GPU would be if they gutted double precision compute and went for a narrower, faster memory controller.Farwalker2u - Sunday, September 21, 2014 - link
I am looking forward to your review of the GTX 970 once you have a compatible sample in hand.I would like to see the results of the Folding @Home benchmarks. It seems that this site is the only one that consistently use that benchmark in its reviews.
As a "Folder" I'd like to see any indication that the GTX 970, at a cost of $330 and drawing less watts than a GTX 780; may out produce both the 780 ($420 - $470) and the 780Ti ($600). I will be studying the Folding @ Home: Explicit, Single Precision chart which contains the test results of the GTX 970.
Wolfpup - Monday, September 22, 2014 - link
Wow, this is impressive stuff. 10% more performance from 2/3 the power? That'll be great for desktops, but of course even better for notebooks. Very impressed they could pulll off that kind of leap on the same process!They've already managed to significantly bump up the top end mobile part from GTX 680 -> 880, but within a year or so I bet they can go quite a bit higher still.
Oh well, it was nice having a top of the line mobile GPU for a while LOL
If 28nm hit in 2012 though, doesn't that make 2015 its third year? At least 28nm seems to be a really good process, vs all the issues with 90/65nm, etc., since we're stuck on it so long.
Isn't this Moore's Law hitting the constraints of physical reality though? We're taking longer and longer to get to progressively smaller shrinks in die size, it seems like...
Oh well, 22nm's been great with Intel and 28's been great with everyone else!