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.

The GTX Titan X represents a very interesting intersection for NVIDIA, crossing Maxwell’s unparalleled power efficiency with GTX Titan’s flagship level performance goals and similarly high power allowance. The end result is that this gives us a chance to see how well Maxwell holds up when pushed to the limit; to see how well the architecture holds up in the form of a 601mm2 GPU with a 250W TDP.

GeForce GTX Titan X Voltages
GTX Titan X Boost Voltage GTX 980 Boost Voltage GTX Titan X Idle Voltage
1.162v 1.225v 0.849v

Starting off with voltages, based on our samples we find that NVIDIA has been rather conservative in their voltage allowance, presumably to keep power consumption down. With the highest stock boost bin hitting a voltage of just 1.162v, GTX Titan X operates notably lower on the voltage curve than the GTX 980. This goes hand-in-hand with GTX Titan X’s stock clockspeeds, which are around 100MHz lower than GTX 980.

GeForce GTX Titan X Average Clockspeeds
Game GTX Titan X GTX 980
Max Boost Clock 1215MHz 1252MHz
Battlefield 4
1088MHz
1227MHz
Crysis 3
1113MHz
1177MHz
Mordor
1126MHz
1164MHz
Civilization: BE
1088MHz
1215MHz
Dragon Age
1189MHz
1215MHz
Talos Principle
1126MHz
1215MHz
Far Cry 4
1101MHz
1164MHz
Total War: Attila
1088MHz
1177MHz
GRID Autosport
1151MHz
1190MHz

Speaking of clockspeeds, taking a look at our average clockspeeds for GTX Titan X and GTX 980 showcases just why the 50% larger GM200 GPU only leads to an average performance advantage of 35% for the GTX Titan X. While the max boost bins are both over 1.2GHz, the GTX Titan has to back off far more often to stay within its power and thermal limits. The final clockspeed difference between the two cards depends on the game in question, but we’re looking at a real-world clockspeed deficit of 50-100MHz for GTX Titan X.

Idle Power Consumption

Starting off with idle power consumption, the GTX Titan X comes out strong as expected. Even at 8 billion transistors, NVIDIA is able to keep power consumption at idle very low, with all of our recent single-GPU NVIDIA cards coming in at 73-74W at the wall.

Load Power Consumption - Crysis 3

Load Power Consumption - FurMark

Meanwhile load power consumption for GTX Titan X is more or less exactly what we’d expect. With NVIDIA having nailed down their throttling mechanisms for Kepler and Maxwell, the GTX Titan X has a load power profile almost identical to the GTX 780 Ti, the closest equivalent GK110 card. Under Crysis 3 this manifests itself as a 20W increase in power consumption at the wall – generally attributable to the greater CPU load from GTX Titan X’s better GPU performance – while under FurMark the two cards are within 2W of each other.

Compared to the GTX 980 on the other hand, this is of course a sizable increase in power consumption. With a TDP difference on paper of 85W, the difference at the wall is an almost perfect match. GTX Titan X still offers Maxwell’s overall energy efficiency, delivering greatly superior performance for the power consumption, but this is a 250W card and it shows. Meanwhile the GTX Titan X’s power consumption also ends up being very close to the unrestricted R9 290X Uber, which in light of the Titan’s 44% 4K performance advantage further drives home the point about NVIDIA’s power efficiency lead at this time.

Idle GPU Temperature

With the same Titan cooler and same idle power consumption, it should come as no surprise that the GTX Titan X offers the same idle temperatures as its GK110 predecessors: a relatively cool 32C.

Load GPU Temperature - Crysis 3

Load GPU Temperature - FurMark

Moving on to load temperatures, the GTX Titan X has a stock temperature limit of 83C, just like the GTX 780 Ti. Consequently this is exactly where we see the card top out at under both FurMark and Crysis 3. 83C does lead to the card temperature throttling in most cases, though as we’ve seen in our look at average clockspeeds it’s generally not a big drop.

Idle Noise Levels

Last but not least we have our noise results. With the Titan cooler backing it, the GTX Titan X has no problem keeping quiet at idle. At 37.0db(A) it's technically the quietest card among our entire collection of high-end cards, and from a practical perspective is close to silent.

Load Noise Levels - Crysis 3

Load Noise Levels - FurMark

Much like GTX Titan X’s power profile, GTX Titan X’s noise profile almost perfectly mirrors the GTX 780 Ti. With the card hitting 51.3dB(A) under Crysis 3 and 52.4dB(A) under FurMark, it is respectively only 0.4dB and 0.1dB off from the GTX 780 Ti. From a practical perspective what this means is that the GTX Titan X isn’t quite the hushed card that was the GTX 980 – nor with a 250W TDP would we expect it to be – but for its chart-topping gaming performance it delivers some very impressive acoustics. The Titan cooler continues to serve NVIDIA well, allowing them to dissipate 250W in a blower without making a lot of noise in the process.

Overall then, from a power/temp/noise perspective the GTX Titan X is every bit as impressive as the original GTX Titan and its GTX 780 Ti sibling. Thanks to the Maxwell architecture and Titan cooler, NVIDIA has been able to deliver a 50% increase in gaming performance over the GTX 780 Ti without an increase in power consumption or noise, leading to NVIDIA once again delivering a flagship video card that can top the performance charts without unnecessarily sacrificing power consumption or noise.

Compute Overclocking
Comments Locked

276 Comments

View All Comments

  • Kevin G - Tuesday, March 17, 2015 - link

    Last I checked, rectal limits are a bit north of 700 mm^2. However, nVidia is already in the crazy realm in terms of economics when it comes to supply/demand/yields/cost. Getting fully functional chips with die sizes over 600 mm^2 isn't easy. Then again, it isn't easy putting down $999 USD for a graphics card.

    However, harvested parts should be quiet plentiful and the retail price of such a card should be appropriately lower.
  • Michael Bay - Wednesday, March 18, 2015 - link

    >rectal limits are a bit north of 700 mm^2

    Oh wow.
  • Kevin G - Wednesday, March 18, 2015 - link

    @Michael Bay

    Intel's limit is supposed to be between 750 and 800 mm^2. They have released a 699 mm^2 product commercially (Tukwilla Itanium 2) a few years ago so it can be done.
  • Michael Bay - Wednesday, March 18, 2015 - link

    >rectal limits
  • D. Lister - Wednesday, March 18, 2015 - link

    lol
  • chizow - Tuesday, March 17, 2015 - link

    Yes its clear Nvidia had to make sacrifices somewhere to maintain advancements on 28nm and it looks like FP64/DP got the cut. I'm fine with it though, at least on GeForce products I don't want to pay a penny more for non-gaming products, if someone wants dedicated compute, go Tesla/Quadro.
  • Yojimbo - Tuesday, March 17, 2015 - link

    Kepler also has dedicated FP64 cores and from what I see in Anandtech articles, those cores are not used for FP32 calculations. How does NVIDIA save power with Maxwell by leaving FP64 cores off the die? The Maxwell GPUs seem to still be FP64 capable with respect to the number of FP64 cores placed on the die. It seems what they save by having less FP64 cores is die space and, as a result, the ability to have more FP32 cores. In other words, I haven't seen any information about Maxwell that leads me to believe they couldn't have added more FP64 cores when designing GM200 to make a GPU with superior double precision performance and inferior single precision performance compared with the configuration they actually chose for GM200. Maybe they just judged single precision performance to be more important to focus on than double precision, with a performance boost for double precision users having to wait until Pascal is released. Perhaps it was a choice between making a modest performance boost for both single and double precision calculations or making a significant performance boost for single precision calculations by forgoing double precision. Maybe they thought the efficiency gain of Maxwell could not carry sales on its own.
  • testbug00 - Tuesday, March 17, 2015 - link

    If this is a 250W card using about the same power as the 290x under gaming load, what does that make the 290x?
  • Creig - Tuesday, March 17, 2015 - link

    A very good value for the money.
  • shing3232 - Tuesday, March 17, 2015 - link

    Agree.

Log in

Don't have an account? Sign up now