Warning: "continue" targeting switch is equivalent to "break". Did you mean to use "continue 2"? in /homepages/9/d659054407/htdocs/app660029368/wp-content/plugins/magneto-sidebar-generator/html.php on line 83

Warning: session_start(): Cannot start session when headers already sent in /homepages/9/d659054407/htdocs/app660029368/wp-content/plugins/wpmu-dev-facebook/lib/external/facebook.php on line 60

Warning: session_start(): Cannot start session when headers already sent in /homepages/9/d659054407/htdocs/app660029368/wp-content/plugins/wpmu-dev-facebook/lib/external/facebook.php on line 60

Warning: session_start(): Cannot start session when headers already sent in /homepages/9/d659054407/htdocs/app660029368/wp-content/plugins/wpmu-dev-facebook/lib/external/facebook.php on line 60

Warning: session_start(): Cannot start session when headers already sent in /homepages/9/d659054407/htdocs/app660029368/wp-content/plugins/wpmu-dev-facebook/lib/external/facebook.php on line 60
Looking back: GeForce GTX 480 vs GTX 1070 Ti (feat. RX 560) - Page 2 of 2 - Tech Altar

Test System

To facilitate the comparison of these two GPUs, the following test setup was used:

  • Intel Core i3-8350K overclocked to 4.9 GHz
  • 16GB of DDR4 running at 3200 MHz, c16
  • AsRock Z370 KILLER SLI motherboard
  • Seasonic FOCUS Platinium 850W Power supply

 

A small note

In order to keep the games and benchmarks from hitting the relatively small VRAM cap on the GTX 480, modest settings were used. As I intended to test GPU power without running into a VRAM bottleneck. I had to balance this with settings high enough to keep the GTX 1070 Ti GPU bound.

Results

Power consumption

Power consumption is tested by running a static scene in Unigine Heaven and reading total system power consumption at the wall using a wall-outlet measurement device. Each GPU was allowed to “burn in” for 5 minutes before a measurement was taken.

Summary of results

Starting off the benchmark suite with Sleeping Dogs, we see the GTX 1070 Ti as expected, utterly demolishing the GTX 480 by producing a frame rate 385.8% higher. The Pascal-based card is nearly 5 times faster in this title. The Radeon RX 560 beats the GTX 480 by nearly 20%. Moving on to Middle-Earth: Shadow Of Mordor, we see the 1070 Ti produce 290.2% higher frame rate – nearly 4 times the performance of the once mighty GTX 480. RX 560 is 33% quicker. Unigine Superposition benchmark sees the 1070 Ti score 395.7% higher. That’s 4.9 times faster than the 1st generation Fermi flagship  and the low-power Radeon manages a 30% lead here. Batman: Arkham Knight is no different, with the GTX 1070 Ti crushing the GTX 480 with an average FPS that is 345.9% higher (4.4x). Radeon RX 560 is 62% faster in Batman, but I believe the GTX 480 was running into its VRAM limitation, likely bottle-necking it.

In Rise of the Tomb Raider we see the GTX 1070 Ti continue the trend with 364% higher frame rate (4.6x).RX 560 manages to be 46.7% faster here. In the popular 3DMark benchmark Firestrike, the GTX 1070 Ti continues the obliteration by scoring 485% higher. That’s almost 6 times faster! Even the humble Radeon RX 560 is just over twice as fast as the old flagship. Finally, we come to Unigine Heaven, a reliable DX11 benchmark. The GTX 1070 Ti completes its victory by being 354% faster. That is 4.5 times the performance of the GeForce GTX 480. However, the Tessellation heavy benchmark seems to pull the 560 down a bit, as it only manages to be a mere 12% faster here.

In addition to just blowing away the GTX 480 from a performance perspective, the GTX 1070 Ti also consumes less power. The RX 560 on the other hand in an entirely different league of power consumption. In my testing, the full system running GeForce GTX 480 in a static scene in Unigine Heaven resulted in 266.3W power consumption. The GeForce GTX 1070 Ti pulled around 241.5W. and the RX 560 needed just 128.8W. While the gap between Fermi and Pascal isn’t quite as large as I expected, it is nonetheless repeatable and impressive. I believe the considerably higher framerate on the GTX 1070 Ti resulted in higher CPU utilisation, and thus higher power consumption from the CPU. RX 560 manages to deliver higher performance across the board at literally half the power consumption.

Conclusion

NVIDIA has come a long way in seven years. An incredibly long way. In just 3 architectures (that’s 6 generations of video cards), NVIDIA has produced an average of 4.7 greater performance for less power consumption. The results would have been even more impressive if we had tested the GTX 1080 – which shares the same 180W TDP as the 1070 Ti – but alas, I do not have one on hand to test with. AMD has also made significant strides: on average the RX 560 is 43% faster than the GTX 480 for half its power consumption.

Either way, 7 years is an awful long time in the tech industry. If you’re still rocking the once-mighty GeForce GTX 480; I don’t think you’ll be disappointed with what NVIDIA (or AMD) has to offer today.