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

  • stun - Tuesday, March 17, 2015 - link

    I hope AMD announces R9 390X fast.
    I am finally upgrading my Radeon 6870 to either GTX 980, TITAN X, or R9 390X.
  • joeh4384 - Tuesday, March 17, 2015 - link

    I do not think Nvidia will have that long with this being the only mega GPU on the market. I really wish they allowed partner models of the Titan. I think a lot of people would go nuts over a MSI Lightning Titan or something like that.
  • farealstarfareal - Tuesday, March 17, 2015 - link

    Yes, a big mistake like the last Titan to not allow custom AIB cards. Good likelihood the 390X will blow the doors off the card with many custom models like MSI Lightning, DCU2 etc.

    Also $1000 for this ??! lol is the only sensible response, none of the dual precision we saw in the original Titan to justify that price, but all of the price. Nvidia trying to cash in here, 390X will force them to do a card probably with less VRAM so people will actually buy this overpriced/overhyped card.
  • chizow - Tuesday, March 17, 2015 - link

    Titan and NVTTM are just as much about image, style and quality as much as performance. Its pretty obvious Nvidia is proud of the look and performance of this cooler, and isn't willing to strap on a hunking mass of Al/Cu to make it look like something that fell off the back of a Humvee.

    They also want to make sure it fits in the SFF and Lanboxes that have become popular. In any case I'm quite happy they dropped the DP nonsense with this card and went all gaming, no cuts, max VRAM.

    It is truly a card made for gamers, by gamers! 100% GeForce, 100% gaming, no BS compute.
  • ratzes - Tuesday, March 17, 2015 - link

    What do you think they give up when they add DP? Its the same fabrication, was for titan vs 780ti. If I'm mistaken, the only difference between cards are whether the process screwed up 1 or more of the smps, then they get sold as gaming cards at varying decreasing prices...
  • MrSpadge - Tuesday, March 17, 2015 - link

    Lot's of die space, since they used dedicated FP64 ALUs.
  • chizow - Wednesday, March 18, 2015 - link

    @ratzes, its well documented, even in the article. DP/FP64 requires extra registers for the higher precision, which means more transistors allocated to that functionality. GM200 is only 1Bn more transistors than GK210 on the same process node, yet they managed to cram in a ton more functional units. Now compare to GM204 to GK204 3.5Bn to 5.2Bn and you can see, its pretty amazing they were even able to logically increase by 1.5x over the GM204, which we know is all gaming, no DP compute also.
  • hkscfreak - Wednesday, March 18, 2015 - link

    Someone didn't read...
  • nikaldro - Tuesday, March 17, 2015 - link

    fanboysm to the Nth p0waH..
  • furthur - Wednesday, March 18, 2015 - link

    which meant fuck all when Hawaii was released

Log in

Don't have an account? Sign up now