Welcome Stranger to OCC!Login | Register

Lenovo Legion Y720 Laptop Review

   -   
» Discuss this article (0)

Battery Life:

It is not the newest of Futuremark's benchmarks, but Powermark specifically exists for testing battery life, so I pulled out the installer and the license key and put it to use. Actually I put it to much more use than I originally intended to. I mentioned before I have been running the screen at 40% brightness, which remains true for these tests. Also the Power Plan has been set to high performance, so it is possible going to another plan could increase battery life over what I report, and the NVIDIA driver settings were left at their defaults, including 'Optimal Power' for the 'Power management mode' option. Also in the NVIDIA drivers is where you can select which GPU is used.

Powermark offers three benchmarks comprised of four tests. The tests cover Web Browsing, Word Processing, Graphics, and Video Playback. The Productivity benchmark uses just the first two; the Entertainment benchmark uses the last two; and the Balanced benchmark runs through all four tests. Originally my plan was to run the benchmarks twice each, once with the computer able to decide which GPU it wanted to use (Auto), and then again with the GTX 1060 selected. The results for Auto seemed odd in comparison to the GTX 1060 numbers, so I ran the benchmark three more times, with the integrated Intel GPU selected. The results still seem kind of odd, with Auto not matching either GPU in some cases, but at least I can make some conclusions about which GPU option is best to run with.

Actually one of those conclusions I will give to you now, because it is somewhat important. Not all games like having the GPU swapped and will crash at launch as a result. Serious Sam Fusion 2017 is among these for me, so I need to have the GTX 1060 selected to even reach the game's main menu, regardless of its efficiency or power. Before getting to the other conclusions, I want to go through the data and graphs first. It is worth noting that Powermark will drain the battery down to 15% and stop there, but it can start at other charges than 100%, so the percent drained is also an important piece of data here. Also, if I only give a value in minutes, I am rounding to the nearest minute, but the data is precise to the second and I used this full precision to calculate the ratio of percent drained and length of the test or benchmark.

For the three Balanced benchmark runs I made, the Auto setting drained 80% in 156 minutes; the integrated Intel GPU drained 83% in 150 minutes; and the GTX 1060 drained 84% in 160 minutes. Working out the ratios with the unit %/second, the Auto setting was 0.0085, the GTX 1060 was 0.0088, while the integrated Intel GPU had a ratio of 0.0092. Naturally the projected battery life from these benchmark runs align with these numbers and Auto saying 3 hours flat, the GTX 1060 just less at 2 hours and 58 minutes, while the integrated Intel GPU only got an estimate of 2 hours and 47 minutes.

Instead of running through the data for the individual tests, I have a graph you can look at. (A picture is worth a thousand words they say, and I suspect most of you would prefer the pictures to the words.)

 

From this data, if you expect your workload to be comparable to Powermark's Balanced benchmark, it looks like selecting the Auto or the GTX 1060 option would be best for battery life. It is worth noting that when on the battery, the discrete GPU is almost undoubtedly being throttled, but when I look at the data from the Powermark tests, both GPUs were running at approximately the same frame rate, or 38 FPS or so.

Next up is the Entertainment benchmark, which consists of just the Graphics and Video Playback tests, repeated over and over until 15% charge is left. On the Auto setting, the benchmark drained 85% in 109 minutes with a ratio of 0.013 %/sec, while the GTX 1060 spent 82% in 105 minutes, again with a ratio of 0.013 %/sec. The integrated Intel GPU went through 82% in 111 minutes, and some quick relative math tells you this is better than the other two, if not by much, with a ratio of 0.012 %/sec, overall. The projections for Auto, GTX 1060, and integrated Intel are 2 hours and 3 minutes, 1 hour and 57 minutes, and 2 hours 4 minutes, respectively.

It looks like for these two workloads, the GPUs are more or less equivalent, which might make sense with the throttling that will be occurring when on the battery. In case you are curious, the test uses a DirectX 9 Shader Model 2.0 scene (looks like the Firefly Forest test in some older 3DMark benchmarks). It might not be the most modern or heaviest load, but it can run on a lot of hardware and does still stress the GPU.

Once more, here is a graph for the ratios for the tests, instead of more words:

 

Finally, the Productivity benchmark that only runs the Web Browsing and Word Processing tests. On the Auto setting it drained 84% over 324 minutes for a ratio of 0.0043 %/sec; the GTX 1060 expended 83% over 278 minutes for a 0.0049 %/sec; and the integrated Intel GPU spent 85% in 320 minutes for a ratio of 0.0042 %/sec. The estimated battery lives are 6 hours and 2 minutes, 5 hours and 11 minutes, and 5 hours and 58 minutes for the Auto, GTX 1060, and integrated Intel options respectively. The GTX 1060 being the least efficient here is hardly surprising and indeed the ability to switch between the integrated and discrete GPUs exists specifically because some workloads, like productivity workloads, do not need the power drain of discrete graphics.

Looking at the graph, it seems odd the GTX 1060 would be so efficient for Web Browsing and not Word Processing, but that is the result. We also saw this in the Balanced benchmark data.

 

While the individual graphs are nice, here are graphs that span all of the tests, letting you compare the results of every test from each benchmark.

 

Overall the data suggests you want to go with the Auto setting, but because of how some games and other applications can misbehave as a result of the switching, you may want to pick the GTX 1060 if you are gaming on the Y720, while running on battery. If you are not gaming, or not gaming much, sticking with Auto or the integrated GPU might be best, and switching to the GTX 1060 when necessary. It all comes down to the workloads you expect to put this through, what you will be doing when running on the battery, and how long you will go before charging. Regardless, the battery life seems appropriate or better than what I would expect from a gaming laptop.




  1. Lenovo Legoin Y720 Review - Introduction
  2. Lenovo Legoin Y720 Review - Physical Characteristics
  3. Lenovo Legoin Y720 Review - Basic-Use Experience
  4. Lenovo Legoin Y720 Review - Hardware Specs
  5. Lenovo Legoin Y720 Review - Battery Life
  6. Lenovo Legoin Y720 Review - Gaming Experience
  7. Lenovo Legoin Y720 Review - Conclusion
Related Products
Random Pic
© 2001-2018 Overclockers Club ® Privacy Policy
Elapsed: 0.1310038567   (xlweb1)