Opening up GPU-Z, the ASUS TUF Gaming X3 RX 5700 reports a default GPU clock of 1275MHzwith a boost clock of 1750MHz and a memory clock of 1750MHz which is 14 GHz effective.
While monitoring with GPU Tweak II, running a Timespy benchmark with its stock settings we see the TUF Gaming X3 RX 5700 get close to its max boost clock of 1750 MHz with it maxing out at 1702 MHz.
To start our overclocking journey we selected the prebuilt OC profile in GPU Tweak which increases the boost clock from 1750 to 1770.
Running Timespy, we did see a marginal increase of 1% which equates to 73 points. Next, we will move to advanced mode and see how far we can push the ASUS TUF Gaming X3 RX 5700.
After spending about half an hour overclocking the RX 5700, we managed to max out all the sliders giving us a core clock of 1850MHZ and a memory clock of 1860MHz (14.8GHz effective).
With everything maxed out, we see an impressive 5.1% increase in performance with us scoring 8459 in Timespy, which is just above 400 points from our stock run.
Since all our sliders are maxed, we must enable “Overclocking Range Enhancement” in GPU Tweak’s settings. This will remove the lock on our sliders giving us more room to overclock.
With the sliders unlocked, we manage to get a stable core clock of an impressive 2175MHz which is +375 above the stock boost of 1750MHz. We did hit a problem with overclocking the memory in “Overclocking Range Enhancement” mode with it also resetting back to 1750MHz when a higher clock was trying to be applied. I even tried using a different piece of overclocking software and seem the same issue.
With our final overclock applied, we ran our final Timespy run on the ASUS TUF Gaming X3 RX 5700 and it garnered a mind-blowing 13.5% increase in performance. Our final run put up a score of 9135 which is over 1000 points above our stock of 8048.
To measure how much power the ASUS TUF Gaming X3 RX 5700 pulls under load, we used a combination of GPU-Z and Kill-A-Watt. GPU-Z reports the number of watts used by the RX 5700 where Kill-A-Watt reports power usage at the wall. Before running our final Timespy run system idled at 55W, during the run we saw a peak power draw a staggering 406W giving us a difference of 351W. Looking at our GPU-Z logs. we hit GPU power draw of 216.0W which seems a little odd given we are still missing about 100W. We reran the benchmark and monitored the wattage with GPU Tweak yielding a seen similar result with a max of 238W. I’m not sure if this is a bug in the drivers, but If I were to guess the wattage usage, it’s probably around 316W even taking PSU efficiency into account.
During all our benchmarks I was quite impressed with how quiet the fans were and still able to maintain a reported GPU temperature around 60C and looking at GPU-Z logs, a hotspot temperature of 70C. While overclocking the TUF Gaming X3 RX 5700, we cracked the fans to 100% which were extremely loud compared to its stock near-silent operation. In the final Timespy run, we still only observed the reported temperature reach 61C, but we did see a dramatic increase in the hotspot temperature of 92C. Now for an everyday overclock you’re probably not running your fans at 100%, so it might be a good idea to set up an aggressive fan curve when you overclock.