Overclocking Showdown – the RTX 2060 vs. the Red Devil RX Vega 56 & vs. the GTX 1070 Ti

Intro

The RTX 2060 FE vs. the Red Devil RX Vega and vs. the GTX 1070 Ti Overclocking Showdown with 39 Games

This overclocking showdown is the follow-up to the RTX 2060 6GB Founders Edition (FE) versus the PowerColor Red Devil RX Vega 56 8GB and versus the GTX 1070 Ti 8GB review earlier this week.  Today, we have optimized and maxed-out our overclocks with all performance options set to their highest limits to get the most performance from each card.

At stock, the RTX 2060 FE won the majority of the 39 games we tested over the Red Devil RX Vega 56, and it also beat the GTX 1070 Ti.  This time, we will overclock all three cards manually each as far as they will go to see where they stand in relation to each other when fully overclocked.

The RTX 2060 OC

We ran Unigine’s Heaven Benchmark 4.0 to set our preliminary overclocks.  With everything at stock, the boost flatlined at 1860MHz after the card warmed up to around 73C.

At stock clocks but with temperature and power targets set to maximum, looping Heaven resulted in a cool-running GPU in the mid-70sC with Boost 4.0 settling in around 1875MHz and well above the 1680MHz minimum Boost.

We were able to add 65MHz to the core and we were also able to add +1000 MHz to the RTX 2060 FE’s GDDR6 memory clocks.  Performance went way up and temperatures remained in the mid- to upper-70s C with the fan speeds never resulting in any noticeable extra noise under load.

Our final RTX 2060 FE overclock is +65MHz offset to the core and +1000 MHz to the memory while Boost 4.0 settled in above or around 1965MHz to peak around 1995MHz  This appears to be a mildly disappointing core overclock as the RTX 2060 is expected to top out around 2050MHz.

The Red Devil RX Vega 56 OC

The PowerColor Red Devil RX Vega 56 uses the same two 8-pin PCIe power connectors as the reference RX Vega 56.  However, we saw the original reference RX Vega 56 get into trouble with its power delivery as being insufficient for overclocking vs. the GTX 1070 Ti. We also needed to increase the reference RX Vega’s fan speed to 100% to maintain its overclock.  In contrast, the Red Devil RX Vega 56 features a 12 phase power delivery and a huge heatsink and 3 fans with 3 BIOSes for the ultimate in stable power delivery with air cooling for enthusiasts.

First we ran Heaven 4.0 with everything at the Turbo preset.

Next we ran Heaven 4.0 with the power setting maxed out (50%) and we see that it gains in performance over the Turbo Preset (15%).

We settled on a 4.5% overclock for an average boost above 1625MHz boost, with its memory clocks overclocked +150MHz to 950MHz from 800MHz.  Overclocked and under full load, the Red Devil’s fans ramp up significantly and they can be clearly heard over the sound of our other case fans.We found that higher memory clocks gained significant performance over a slightly higher core overclock.  Adjusting the voltage to achieve the higher memory overclock required undervolting the core and overvolting the memory but we did not we achieved stability in all of our 39 tested games – some of them required lowering the core overclock to 4% with a 930MHz offset to the memory for 100% stability.

The GTX 1070 Ti OC

We devoted a separate evaluation to the GTX 1070 Ti which you can read here.  The GTX 1070 Ti delivered a sustained 1809MHz Boost as long as the GPU stayed below 84C at stock settings

In our original GTX 1070 Ti review, we decided on a preliminary core offset of +200MHz with +400MHz added to the memory clocks.  Our maximum sustained average boost was 2025MHz and our memory clocks stayed at 4404MHz but our overclock turned out to be somewhat optimistic.  Fully overclocked, the GTX 1070 Ti tends to ramp up its fan and it can be heard over the case fans although it is never intrusive.

Our final overclock to the GTX 1070 Ti turned out to be a 170MHz offset to the core and 380MHz to the memory resulting in an average core Boost of 1995MHz and 4304MHz memory overclock.  There was some slight throttling which began to occur at 84C while we were overclocking and the boost would drop to around 1975MHz.

Testing Platform

We test 39 games and 4 synthetic benchmarks at 1920×1080 and at 2560×1440.  Our testing platform is a recent install of Windows 10 64-bit Home Edition, and we are using an i7-8700K which turbos all 6 cores to 4.7GHz, an EVGA Z370 FTW motherboard, and 16GB of HyperX DDR4 3333MHz. The games, settings, and hardware are identical except for the cards being compared.

Before we run our overclocked benchmarks, let’s check out the test configuration.