GTX 970 vs R9 390 Unigine Valley Benchmark Challenge

badrobot · 27 Sep 2015

I conducted a little benchmarking challenge with my buddies over at Google+ PC Gaming community between GTX 970 and its AMD counterpart R9 390 and I would like to share the result to those who are in the market for a mid-range GPU. If you are torn between the 970 and the 390, this may help you arrive at a decision.

GTX 970 vs R9 390 Unigine Valley Benchmark Challenge-970vs390.png

Dude · 27 Sep 2015

Thanks for sharing. Looks like team AMD won that round

KACI · 28 Sep 2015

Interesting but I'm not sure how useful your result really are. There are too many variables that I don't know if you covered, that impact on performance. Were the MBs and their chipsets the same? How about the processors, several were K processors, were any of them overclocked? Was the Ram the same type and speed? Not to mention the cards, were any of them overclocked? Were they from the same manufacturer? For the results to be significant they would all have to run on the exact same configuration.

Performance alone isn't the only deciding factor. You have to take into consideration price as well. A quick look at amazon and there was generally about a $50 difference, A person would have to decide if a little over 4FPS is worth the extra cost or not. I'm not even going to go into the great "driver" debate.

Contrary to Dude, based on what you presented, I don't think you could declare AMD a winner. You and your friends swap cards, and post the results. Still too many variables to be significant but it would be interesting to see if there's any difference.

badrobot · 28 Sep 2015

KACI said:

Interesting but I'm not sure how useful your result really are. There are too many variables that I don't know if you covered, that impact on performance. Were the MBs and their chipsets the same? How about the processors, several were K processors, were any of them overclocked? Was the Ram the same type and speed? Not to mention the cards, were any of them overclocked? Were they from the same manufacturer? For the results to be significant they would all have to run on the exact same configuration.

Performance alone isn't the only deciding factor. You have to take into consideration price as well. A quick look at amazon and there was generally about a $50 difference, A person would have to decide if a little over 4FPS is worth the extra cost or not. I'm not even going to go into the great "driver" debate.

Contrary to Dude, based on what you presented, I don't think you could declare AMD a winner. You and your friends swap cards, and post the results. Still too many variables to be significant but it would be interesting to see if there's any difference.

Swap cards? What are you talking about? This is a random user from around the world in the same community at Google+. I just called them friends because we are in the same community. We are allowed to OC all we want to get the best result using only one Unigine Valley setting. So, the "factors" you are talking about are all covered. The actual screenshots of the unigine result was also posted indicating the exact valley settings everyone used. We did not choose people with the best CPU. The objective is to make the card work at it's best no matter what the CPU is. This is actually a good basis for choosing a CPU-GPU combination for those planning to build around the 970 and 390 cards.

If you are still not convinced, Dude has a Valley Benchmark thread on this forum. On the leaderboard, so far, the two 970s are under the 390.
Here: https://www.tenforums.com/pc-custom-b...tml#post326293