ASUS R9 290X DirectCU II - Watercooling EK FC-R9-290X vs. Aircooling
Published by Hiwa Pouri on 04.03.14Page:
Conclusion
There is acutally one crucial reason why we chose the R9 290X for this test. This chip is known to become quite hot and to then start throttling. Equipped with the reference cooler the situation is even worse than with this version of the DirectCU II cooler, which is actually a powerful aircooler for graphics cards. Nevertheless, when we started to overclock the card we reached the limit of the aircooler - apparently. At this point we're going to walk through the results gathered on previous pages and then we add a few of our thoughts.Running the card at stock clocks shows, that when we equipped it with the EK-FC R9-290X-DCII watercooler scores were on average 1.6 percent higher than with the aircooler. This is actually already a solid hint, that the card started to throttle with the powerful DirectCU II aircooler, even at stock clocks. There was actually only one game where we couldn't measure a performance difference, which is Battlefield 3. In this case fps were exactly the same. On the other hand there is Bioshock: Infinite where we see that the watercooled card is no less than 6.5 percent quicker. When it comes to stock clocks we also had a look at two theoretical benchmarks, which are 3DMark and Unigine Heaven 4.0. In 3DMark the differences between watercooled and aircooled are tiny, but in Unigine Heaven 4.0 there is a 5.4 percent gap.
Once we started overclocking the cards, we noticed that 1150 MHz on the GPU and 1500 MHz on the memory were maximum stable clocks to run the card aircooled with the DirectCU II aircooler. In order to show there is quite some potential still slumbering in the R9 290X cooler, when appropriately cooled, we overclocked the card a little bit hihger. In this case we ran itz at 1200 MHz GPU clocks and 1500 MHz memory clocks. In fact, watercooled max stable GPU clock was 1310 MHz, but we did not want to drive the chip at its limit since we believe that 1310 MHz is a rather high value and if we want to compare these results to another R9 290X watercooled that card then might not reach these clocks stable.
Let's discuss the results now. In 3DMark Fire Strike Extreme we see the scores increase by 6.4 percent in case of aircooling OC and 12.6 percent when it comes to watercoolin OC. This means there is a 6.2 percent gap inbetween aircooling and watercooling, although the difference in overclocking is only 4.34 percent. This - again - indicates that the R9 290X DirectCU II with aircooler was throttling, not much, but it was. When we run Unigine Heaven 4.0 we even notice that the result, in case of the aircooled OC tests, dropped under the non-OC value. This apparently shows there was some rather heavy throttling going on and the chip is constantly hitting the temperature target. The same card, equipped with the EK-FC R9-290X-DCII did throttle at all and the performance goes up by 11.3 percent. Overall the aircooled version is able to gain 8.7 percent performance with the overclocking and the watercooled variant got 12.9 percent faster. One value that was especially interesting at this point was Battlefield 4. In this game the watercooled and overclocked cards gains no less than 25 percent, whereas the aircooled and overclocked card only gains 10.7 percent.
On another note we had a closer look at temperatures. There is actually nothing much to say then the obvious. Even under full load and overclocked the watercooled card didn't get hotter than 53°C, whereas the aircooled version ran at 81°C.
Last but not least we want to add a few thoughts. Putting the results into persepective shows, that the R9 290X with an aircooler is running really hot. It's basically hitting the temperature limit as soon as there is heavy load and even the powerful DirectCU II cooler from ASUS is not powerful enough to eliminate throttling completely. On the other hand, if you put a watercooler on top of this card, then you can basically unleash the full potential of this card. This becomes especially clear as soon as you start overclocking the card. In our case it was possible to run the card stable even at 1310 MHz GPU clock, whereas the maximum with the aircooler was at 1150 MHz. Overall we can say if you go get yourself a high-end R9 290X and you want to get the maximum out of this card, then we'd recommend you to go for a custom watercooling loop and the EK-FC R9-290X-DCII is certainly doing a great job. Obviously if you do not own any watercooling gear at all, this is going to cost you quite some money, but at least you'd have to buy a new radiator and pump only once. If you buy an new card in a few years you just need to buy a news waterblock. The EK-FC R9-290X-DCII we've used for testing in this article is going to set you back 102.96 Euro (price 04.03.2014). To add some more numbers you can for example calculate that when you overclock the card to 1200 MHz on the GPU and 1500 MHz on the memory performance goes up by 12.9 percent. The ASUS R9 290X DirectCU II costs 532.99 Euro (price 04.03.2104 Geizhals). This means for an additional 12.9 percent performance you have to pay an additional 19 percent. If you keep in mind, that our card was capable of running even above 1300 MHz on the GPU you can imagine, that the performance gain is going to be somewhere in the ballpark of 20+ percent.
| Navigate through the articles | |
ASUS R9 280X DirectCU II - Watercooling EK FC-R9-280X vs. Aircooling
 |
|
|
|