Final Words

When NVIDIA launched the first Maxwell cards earlier this year, we knew that we would be in for a treat with their latest architecture. Though just a fragment of the performance of what their eventual high-end cards would be, NVIDIA’s first Maxwell cards offered an interesting look at an architecture that would be capable of doubling NVIDIA’s performance per watt on the same 28nm TSMC manufacturing process they started with over 2 years ago. To that end I don’t think there has been any doubt that NVIDIA’s eventual second generation Maxwell cards would be equally amazing when it comes to power efficiency, but I feel NVIDIA has still impressed us when it comes to performance, features, and pricing.

In many ways it feels like this latest launch has returned us to the PC video card industry of 2012. NVIDIA’s flagship consumer card is once again powered by a smaller and more potent consumer-class x04 GPU, and once again NVIDIA is swinging the one-two punch of performance and power efficiency. When GTX 680 was launched it set a new high mark for the video card industry, and now we see GTX 980 do more of the same. The GTX 980 is faster, less power hungry, and quieter than the Radeon R9 290X, so once again NVIDIA has landed the technical trifecta. Even if we’re just looking at performance and pricing the GTX 980 is the undisputed holder of the single-GPU performance crown, besting everything else AMD and NVIDIA have to offer, and offering it at a price that while no means a steal is more than reasonable given NVIDIA’s technical and performance advantage. As such GTX 980 comes very, very close to doing to Radeon R9 290X what GTX 680 did to Radeon HD 7970 over 2 years ago.

Meanwhile from a feature perspective the GTX 900 series is going to prove to be a very captivating product. Dynamic Super Resolution is a brutish-yet-clever solution of what to do about anti-aliasing on today's deferred renderer games that cannot support traditional MSAA/SSAA, and while I’m withholding my judgment on Multi-Frame sampled Anti-Aliasing until it’s made available to users in NVIDIA’s drivers, the idea at least has merit. Otherwise I am very happy to see that NVIDIA has now fully caught up to the competition in terms of baseline API features by offering everything needed to support Direct3D 11.2 and beyond.

Along those lines, NVIDIA’s focus on voxel technology for Maxwell 2 is a very interesting route to take, and I am eagerly anticipating seeing whether it gets widely adopted and what developers do with it. VXGI is a very neat concept to generate voxel based global illumination, and building in the features necessary to do significant portions of it in hardware is a wise move by NVIDIA. The catch at this point is the same catch that faces all vendor specific technologies: just because the hardware is there doesn’t mean developers will put it to good use, especially in this age of console ports. NVIDIA for their part has made the right move by making sure VXGI will run on other hardware, but I am concerned that the performance delta means that it’s only going to be viable on Maxwell 2 GPUs for now, which could discourage developers. None the less we do need better lighting in games, and I hope this encourages developers to finally adopt these kinds of high quality global illumination systems.

As for the hardware itself, is there anything left to say other than that GTX 980 is a well-built, well-engineered card? The build quality is impeccable – raising the bar over even GTX Titan – and the power efficiency gains are truly remarkable. With a TDP lower than even GTX 680, this is the lowest power consumption has been for a chart-topping card since 9800 GTX over half a decade ago. It’s really a bit of a honeymoon period since if and when NVIDIA does Big Maxwell one has to expect power consumption will go back up, but for the time being it’s very pleasing to be able to get chart-topping performance inside of 165W. And the fact that this comes from the same company responsible for GTX 480 just 2 generations ago makes this the ultimate technical turnaround.

In conclusion, the GeForce GTX 980 represents another stellar performance from NVIDIA. Their reign at the top is not going to go unchallenged – AMD can’t match NVIDIA on performance, but they can sure drive down prices – but as was the case in 2012 the crown continues to securely reside in NVIDIA’s hands, and once again they have done the technical hard work to earn it.

Finally, as a reminder we will be following up this article next week with our look at GTX 980 SLI performance and a look at the GTX 970. Of the two cards launched today the GTX 970 is without a doubt the more interesting of the two thanks to its relatively low price compared to the performance NVIDIA is offering, but due to our aforementioned board issues we will not be able to take a look at it until next week. So until then stay tuned for the rest of our GM204 coverage.

Overclocking GTX 980
Comments Locked

274 Comments

View All Comments

  • jmunjr - Friday, September 19, 2014 - link

    Wish you had done a GTX 970 review as well like many other sites since way more of us care about that card than the 980 since it is cheaper.
  • Gonemad - Friday, September 19, 2014 - link

    Apparently, if I want to run anything under the sun in 1080p cranked to full at 60fps, I will need to get me one GTX 980 and a suitable system to run with it, and forget mid-ranged priced cards.

    That should put an huge hole in my wallet.

    Oh yes, the others can run stuff at 1080p, but you have to keep tweaking drivers, turning AA on, turning AA off, what a chore. And the milennar joke, yes it RUNS Crysis, at the resolution I'd like.

    Didn't, by any chance, the card actually benefit of being fabricated at 28nm, by spreading its heat over a larger area? If the whole thing, hipothetically, just shrunk to 14nm, wouldn't all that 165W of power would be dissipated over a smaller area (1/4 area?), and this thing would hit the throttle and stay there?

    Or by being made smaller, it would actually dissipate even less heat and still get faster?
  • Yojimbo - Friday, September 19, 2014 - link

    I think that it depends on the process. If Dennard scaling were to be in effect, then it should dissipate proportionally less heat. But to my understanding, Dennard scaling has broken down somewhat in recent years, and so I think heat density could be a concern. However, I don't know if it would be accurate to say that the chip benefited from the 28nm process, since I think it was originally designed with the 20nm process in mind, and the problem with putting the chip on that process had to do with the cost and yields. So, presumably, the heat dissipation issues were already worked out for that process..?
  • AnnonymousCoward - Friday, September 26, 2014 - link

    The die size doesn't really matter for heat dissipation when the external heat sink is the same size; the thermal resistance from die to heat sink would be similar.
  • danjw - Friday, September 19, 2014 - link

    I would love to see these built on Intel's 14nm process or even the 22nm. I think both Nvidia and AMD aren't comfortable letting Intel look at their technology, despite NDAs and firewalls that would be a part of any such agreement.

    Anyway, thanks for the great review Ryan.
  • Yojimbo - Friday, September 19, 2014 - link

    Well, if one goes by Jen-Hsun Huang's (Nvidia's CEO) comments of a year or two ago, Nvidia would have liked Intel to manufacture their SOCs for them, but it seems Intel was unwilling. I don't see why they would be willing to have them manufacture SOCs and not GPUs being that at that time they must have already had the plan to put their desktop GPU technology into their SOCs, unless the one year delay between the parts makes a difference.
  • r13j13r13 - Friday, September 19, 2014 - link

    hasta que no salga la serie 300 de AMD con soporte nativo para directx 12
  • Arakageeta - Friday, September 19, 2014 - link

    No interpretation of the compute graphs whatsoever? Could you at least report the output of CUDA's deviceQuery tool?
  • texasti89 - Friday, September 19, 2014 - link

    I'm truly impressed with this new line of GPUs. To be able to acheive this leap on efficiency using the same transistor feature size is a great incremental achievement. Bravo TSMC & Nvidia. I feel comfortable to think that we will soon get this amazing 980 performance level on game laptops once we scale technology to the 10nm process. Keep up the great work.
  • stateofstatic - Friday, September 19, 2014 - link

    Spoiler alert: Intel is building a new fab in Hillsboro, OR specifically for this purpose...

Log in

Don't have an account? Sign up now