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.

As the GM200 flagship card, GTX Titan X gets the pick of the litter as far as GM200 GPUs go. GTX Titan X needed fully-functional GM200 GPUs, and even then needed GPUs that were good enough to meet NVIDIA’s power requirements. GTX 980 Ti on the other hand, as a cut-down/salvage card, gets second pick. So we expect to see these chips be just a bit worse; to have either functional units that came out of the fab damaged, or have functional units that have been turned off due to power reasons.

GeForce GTX Titan X/980 Voltages
GTX Titan X Boost Voltage GTX 980 Ti Boost Voltage GTX 980 Boost Voltage
1.162v 1.187v 1.225v

Looking at voltages, we can see just that in our samples. GTX 980 Ti has a slightly higher boost voltage – 1.187v – than our GTX Titan X. NVIDIA sometimes bins their second-tier cards for lower voltage, but this isn’t something we’re seeing here. Nor is there necessarily a need to bin in such a manner since the 250W TDP is unchanged from GTX Titan X.

GeForce GTX 980 Ti Average Clockspeeds
Game GTX 980 Ti GTX Titan X
Max Boost Clock 1202MHz 1215MHz
Battlefield 4
1139MHz
1088MHz
Crysis 3
1177MHz
1113MHz
Mordor
1151MHz
1126MHz
Civilization: BE
1101MHz
1088MHz
Dragon Age
1189MHz
1189MHz
Talos Principle
1177MHz
1126MHz
Far Cry 4
1139MHz
1101MHz
Total War: Attila
1139MHz
1088MHz
GRID Autosport
1164MHz
1151MHz
Grand Theft Auto V
1189MHz
1189MHz

The far more interesting story here is GTX 980 Ti’s clockspeeds. As we have pointed out time and time again, GTX 980 Ti’s gaming performance trails GTX Titan X by just a few percent, this despite the fact that GTX 980 Ti is down by 2 SMMs and is clocked identically. On paper there is a 9% performance difference that in the real world we’re not seeing. So what’s going on?

The answer to that is that what GTX 980 Ti lacks in SMMs it’s making up in clockspeeds. The card’s average clockspeeds are frequently two or more bins ahead of GTX Titan X, topping out at a 64MHz advantage under Crysis 3. All of this comes despite the fact that GTX 980 Ti has a lower maximum boost clock than GTX Titan X, topping out one bin lower at 1202MHz to GTX Titan X’s 1215MHz.

Ultimately the higher clockspeeds are a result of the increased power and thermal headroom the GTX 980 Ti picks up from halving the number of VRAM chips along with disabling two SMMs. With those components no longer consuming power or generating heat, and yet the TDP staying at 250W, GTX 980 Ti can spend its power savings to boost just a bit higher. This in turn compresses the performance gap between the two cards (despite what the specs say), which coupled with the fact that performance doesn't scale lineraly with SMM count or clockspeed (you rarely lose the full theoretical performance amount when shedding frequency or functional units) leads to the GTX 980 Ti trailing the GTX Titan X by an average of just 3%.

Idle Power Consumption

Starting off with idle power consumption, there's nothing new to report here. GTX 980 Ti performs just like the GTX Titan X, which at 74W is second only to the GTX 980 by a single watt.

Load Power Consumption - Crysis 3

Load Power Consumption - FurMark

Meanwhile load power consumption is also practically identical to the GTX Titan X. With the same GPU on the same board operating at the same TDP, GTX 980 Ti ends up right where we expect it, next to GTX Titan X. GTX Titan X did very well as far as energy efficiency is concerned – setting a new bar for 250W cards – and GTX 980 Ti in turn does just as well.

Idle GPU Temperature

Load GPU Temperature - Crysis 3

Load GPU Temperature - FurMark

As was the case with power consumption, video card temperatures are similarly unchanged. NVIDIA’s metal cooler does a great job here, keeping temperatures low at idle while NVIDIA’s GPU Boost mechanism keeps temperatures from exceeding 83C under full load.

Idle Noise Levels

Load Noise Levels - Crysis 3

Load Noise Levels - FurMark

Finally for noise, the situation is much the same. Unexpected but not all that surprising, the GTX 980 Ti ends up doing a hair worse than the GTX Titan X here. NVIDIA has not changed the fan curves or TDP, so this ultimately comes down to manufacturing variability in NVIDIA’s metal cooler, with our GTX 980 Ti faring ever so slightly worse than the Titan. Which is to say that it's still right at the sweet spot for noise versus power consumption, dissipating 250W at no more than 53dB, and once again proving the mettle of NVIDIA's metal cooler.

Compute Overclocking
Comments Locked

290 Comments

View All Comments

  • xenol - Monday, June 1, 2015 - link

    Transistor count means nothing. The GTX 780 Ti has 2.8 billion transistors. The GTX 980 has around 2 billion transistors, and yet the GTX 980 can dance with the GTX 780 Ti in performance.

    As the saying goes... it's not the size that matters, only how you use it.
  • Niabureth - Monday, June 1, 2015 - link

    Don't want to sound like a messer schmitt but thats 2,8K cuda cores for GK110, and 2K for the GM204. The GK110 has 7.1 billion transistors.
  • jman9295 - Tuesday, June 2, 2015 - link

    In this very article they list the transistor count of those two cards in a giant graph. The 980 has 5.2 billion transistors and the 780ti 7.1 billion. Still, your point is the same, they got more performance out of less transistors on the same manufacturing node. All 28nm means is how small the gap is between identical components, in this case the CUDA cores. Each Maxwell CUDA is clearly more efficient than each Kepler. Also helping is the double VRAM size which probably allowed them to also double the ROP count which greatly improved transistor efficiency and performance.
  • Mithan - Sunday, May 31, 2015 - link

    It matters because we are close to .16/20nm GPU's, which will destroy these.
  • dragonsqrrl - Sunday, May 31, 2015 - link

    "we are close to .16/20nm GPU's"

    People said the same thing when the 750Ti launched. I'll give give you one thing, we are closer than we were, but we are not "close".
  • Kevin G - Monday, June 1, 2015 - link

    The difference now is that there are actually 20 nm products on the market today, just none of them are GPUs. It seems that without FinFET, 20 nm looks to be optimal only for mobile.
  • felicityc - Tuesday, January 11, 2022 - link

    What if I told you we are on 8nm now?
  • LemmingOverlord - Monday, June 1, 2015 - link

    @SirMaster - The reason people care about the process node is because that right now - in mid-2015 - this is an extremely mature (ie: old but well-rehearsed) manufacturing process, which has gone through several iterations and can now yield much better results (literally) than the original 28nm process. This means that it's much cheaper to produce because there are less defective parts per wafer (ie: higher yield). Hence ComputerGuy2006 saying what he said.

    Contrary to what other people say "smaller nm" does NOT imply higher performance. Basically when a shrink comes along you can expect manufacturers to do 1 of two things:

    a) higher transistor count in a similar die size, with similar power characteristics when compared to its ancestor - and therefore higher performance
    b) same transistor count in a much smaller die size, therefore better thermals/power characteristics

    Neither of these factor in architectural enhancements (which sometimes are not that transparent, due to their immaturity).

    So ComputerGuy2006 is absolutely right. Nvidia will make a killing on a very mature process which costs them a below-average amount of money to manufacture.

    In this case Nvidia is using "defective" Titan X chips to manufacture 980 Ti. Simple as that. Their Titan X leftovers sell for $350 less and you still get almost all the performance a Titan would give you.
  • royalcrown - Wednesday, June 3, 2015 - link

    I take issue with point b) " same transistor count in a much smaller die size, therefore better thermals/power characteristics"

    I disagree because the same die shrink can also cause a rise in power density, therefore WORSE characteristics (especially thermals).
  • Gasaraki88 - Monday, June 1, 2015 - link

    Smaller nm, bigger e-peen.

Log in

Don't have an account? Sign up now