Released in March 2017, Nvidia’s GeForce GTX 1080 Ti was the top-tier card of the Pascal line-up. This is the graphics card that super-nerds and gamers drooled over. With an MSRP of $699 for the base model, board partners such as EVGA, Asus, Gigabyte, MSI, and Zotac (among others) all quickly jumped on board (pun intended) with custom designs costing well over the MSRP, as well as their own takes on the reference design.
Three years later, with the release of the RTX 2080 Ti, the 1080 Ti still holds its own, and still commands well over $400 on the used market. These are beastly cards, capable of running most games with max settings in 4K resolutions.
But, how does it fold?
Folding at home is a distributed computing project originally developed by Stanford University, where everyday users can lend their PC’s computational horsepower to help disease researchers understand and fight things like cancer, Alzheimer’s, and most recently the COVID-19 Coronavirus. User’s computers solve molecular dynamics problems in the background, which help the Folding@Home Consortium understand how proteins “misfold” to cause disease. For computer nerds, this is an awesome way to give (money–>electricity–>computer work–>fighting disease).
Folding at home (or F@H) can be run on both CPUs and GPUs. CPUs provide a good baseline of performance, and certain molecular simulations can only be done here. However, GPUs, with their massively parallel shader cores, can do certain types of single-precision math much faster than CPUs. GPUs provide the majority of the computational performance of F@H.
Geforce GTX 1080 Ti Specs
The 1080 Ti is at the top of Nvidia’s lineup of their 10-series cards.
With 3584 CUDA Cores, the 1080 Ti is an absolute beast. In benchmarks, it holds its own against the much newer RTX cards, besting even the RTX 2080 and matching the RTX 2080 Super. Only the RTX 2080 Ti is decidedly faster.
Testing is performed in my old but trusty benchmark machine, running Windows 10 Pro and using Stanford’s V7 Client. The Nvidia graphics driver version was 441.87. Power consumption measurements are taken on the system-level using a P3 Watt Meter at the wall.
- CPU: AMD FX-8320e
- Mainboard : Gigabyte GA-880GMA-USB3
- GPU: EVGA 1080 Ti (Reference Design)
- Ram: 16 GB DDR3L (low voltage)
- Power Supply: Seasonic X-650 80+ Gold
- Drives: 1x SSD, 2 x 7200 RPM HDDs, Blu-Ray Burner
- Fans: 1x CPU, 2 x 120 mm intake, 1 x 120 mm exhaust, 1 x 80 mm exhaust
- OS: Win10 64 bit
I did extensive testing of the 1080 Ti over many weeks. Folding@Home rewards donors with “Points” for their contributions, based on how much science is done and how quickly it is returned. A typical performance metric is “Points per Day” (PPD). Here, I have averaged my Points Per Day results out over many work units to provide a consistent number. Note that any given work unit can produce more or less PPD than the average, with variation of 10% being very common. For example, here are five screen shots of the client, showing five different instantaneous PPD values for the 1080 Ti.
GTX 1080 Ti Folding@Home Performance
The following plot shows just how fast the 1080 Ti is compared to other graphics cards I have tested. As you can see, with nearly 1.1 Million PPD, this card does a lot of science.
GTX 1080 Ti Power Consumption
With a board power rating of 250 Watts, this is a power hungry graphics card. Thus, it isn’t surprising to see that power consumption is at the top of the pack.
GTX 1080 Ti Efficiency
Power consumption alone isn’t the whole story. Being a blog about doing the most work possible for the least amount of power, I am all about finding Folding@Home hardware that is highly efficient. Here, efficiency is defined as Performance Out / Power In. So, for F@H, it is PPD/Watt. The best F@H hardware is gear that maximizes disease research (performance) done per watt of power consumed.
Here’s the efficiency plot.
The Geforce GTX 1080 Ti is the fastest and most efficient graphics card that I’ve tested so far for Stanford’s Folding@Home distributed computing project. With a raw performance of nearly 1.1 Million PPD in windows and an efficiency of almost 3500 PPD/Watt, this card is a good choice for doing science effectively.
Stay tuned to see how Nvidia’s latest Turing architecture stacks up.
Pingback: How to Run Folding@Home on a Graphics Card in Windows 10 | Green Folding@Home
Would you be adding 2080 Super & 5700 XT video card to your review!?
One day I hope! I primarily test used cards because I can resell them basically for what I bought them for on eBay. There just aren’t enough used 2080s and what not out yet. However, I am in the middle of testing the 1660 (also Turing architecture), so I should have an article up at some point on that
One day I hope. I buy used cards that I can get deals on. When these become more common I hope to try them out.
I wonder how the GT 1030 would fare in performance per watt.
Don’t know how the TDP translates to real wall power drain, but the 1050 is rated at 75W while the 1030 at as low as 20W.
Since the GFLOPS dont triple, maybe the 1030 is a better low end option than the 1050.
What do you think?
I’ve wanted to test at 1030 for some time, but I haven’t been able to find a good deal on one. They’re pretty inexpensive, but I like to get them for about half price used if I can (that’s why I don’t test many new cards). I will see if I can find a good used one on eBay. I bet it does pretty well since it is a very efficient architecture
Probably pretty well…it’s a relatively modern architecture (Pascal), and will be very efficient. It should do about the same amount of work/watt as the other pascal cards. In terms of PPD/watt, it will suffer a bit from not getting as good of a quick return bonus.
This is a good article, thanks for the tests. However, im curious how the gtx 1060 is shown to get over 200,000 PPD when im running a 1660ti and i hover around 100k for my GPU with some more from CPU. How did you get these numbers and should this be expected?
It’s probably something to do with your config. I am currently testing a 1660 super and getting 600-700K PPD. First, do you have a passkey? You’ll need one to get the bonus points. See this: https://foldingathome.org/support/faq/points/passkey/
Next, if you’re doing combined CPU and GPU folding, you need to make sure you leave one CPU core free to “feed” the graphics card. If the CPU is pegged out on all cores doing CPU folding, the GPU will starve for data. See my guide here: https://greenfoldingathome.com/2020/03/15/how-to-run-foldinghome-on-a-graphics-card-in-windows-10/
Also, I thought I would mention that these numbers don’t really show how much work they do, because the faster it gets done, the more they points you get exponentially. So double the PPD is not double the amount of work done, and is probably somewhere between 1-1.5ish times more??
Yeah you’re exactly right, in terms of work done per watt (not value of work done per watt, which is the PPD/Watt metric everyone loves), you’d want to look at a metric like NS/day (nanoseconds of molecular dynamics simulation per day). The problem with this metric is that different work units have different NS/day on the same card because of solving different molecule sizes (different number of atoms). So it is very difficult to get an average value to compare to. Some people use the program FAHBENCH to benchmark a card’s performance in NS/day using the same work unit each time. I don’t like that because, like any benchmark, it doesn’t necessarily translate into real-world performance
Very interesting, great blog! I have a gaming PC I have started to use for folding and I’m thinking of building a specific rig just for folding (as I have a few parts from old PC’s lying around)
One thing I’m unsure of, I have a RTX 2070 Super and that seems to have less PPD of 923000 than the 1080 test gives a higher score. I have a passkey as well. I’ll try following your guide to leave 1 CPU core free or even turn off CPU folding as well and see if it improves.
Thanks! Let me know how that works, usually in combined CPU / GPU folding setups, a starving graphics card is the culprit of low ppd
I noticed that there was 1 thread free anyway (I have an i9-9900k) so I lowered it to 14, keeping two free, and it was jumped up with an estimated PPD of 2.2 million!
Hi, I have a closet full of 1660 super and ti cards from my mining days. Is there a “for dummies guide”? I would like to participate.
That’s great! I wrote a guide here, but it’s a bit outdated now
I think the folding forums have various guides, and some specific input for 1660s in the “hardware” section